libnuma-dev,
libpci-dev,
libpixman-1-dev,
+ libproxmox-backup-qemu0-dev,
librbd-dev (>= 0.48),
libsdl1.2-dev,
libseccomp-dev,
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Wed, 2 Oct 2019 10:30:03 +0200
-Subject: [PATCH] monitor/qmp: resume monitor when clearing its queue
-
-When a monitor's queue is filled up in handle_qmp_command()
-it gets suspended. It's the dispatcher bh's job currently to
-resume the monitor, which it does after processing an event
-from the queue. However, it is possible for a
-CHR_EVENT_CLOSED event to be processed before before the bh
-is scheduled, which will clear the queue without resuming
-the monitor, thereby preventing the dispatcher from reaching
-the resume() call.
-Fix this by resuming the monitor when clearing a queue which
-was filled up.
-
-Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
----
- monitor/qmp.c | 10 ++++++++++
- 1 file changed, 10 insertions(+)
-
-diff --git a/monitor/qmp.c b/monitor/qmp.c
-index e1b196217d..fb3e66c62a 100644
---- a/monitor/qmp.c
-+++ b/monitor/qmp.c
-@@ -70,9 +70,19 @@ static void qmp_request_free(QMPRequest *req)
- /* Caller must hold mon->qmp.qmp_queue_lock */
- static void monitor_qmp_cleanup_req_queue_locked(MonitorQMP *mon)
- {
-+ bool need_resume = (!qmp_oob_enabled(mon) && mon->qmp_requests->length > 0)
-+ || mon->qmp_requests->length == QMP_REQ_QUEUE_LEN_MAX;
- while (!g_queue_is_empty(mon->qmp_requests)) {
- qmp_request_free(g_queue_pop_head(mon->qmp_requests));
- }
-+ if (need_resume) {
-+ /*
-+ * Pairs with the monitor_suspend() in handle_qmp_command() in case the
-+ * queue gets cleared from a CH_EVENT_CLOSED event before the dispatch
-+ * bh got scheduled.
-+ */
-+ monitor_resume(&mon->common);
-+ }
- }
-
- static void monitor_qmp_cleanup_queues(MonitorQMP *mon)
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Sergio Lopez <slp@redhat.com>
-Date: Mon, 16 Sep 2019 13:24:12 +0200
-Subject: [PATCH] virtio-blk: schedule virtio_notify_config to run on main
- context
-
-virtio_notify_config() needs to acquire the global mutex, which isn't
-allowed from an iothread, and may lead to a deadlock like this:
-
- - main thead
- * Has acquired: qemu_global_mutex.
- * Is trying the acquire: iothread AioContext lock via
- AIO_WAIT_WHILE (after aio_poll).
-
- - iothread
- * Has acquired: AioContext lock.
- * Is trying to acquire: qemu_global_mutex (via
- virtio_notify_config->prepare_mmio_access).
-
-If virtio_blk_resize() is called from an iothread, schedule
-virtio_notify_config() to be run in the main context BH.
-
-[Removed unnecessary newline as suggested by Kevin Wolf
-<kwolf@redhat.com>.
---Stefan]
-
-Signed-off-by: Sergio Lopez <slp@redhat.com>
-Reviewed-by: Kevin Wolf <kwolf@redhat.com>
-Message-id: 20190916112411.21636-1-slp@redhat.com
-Message-Id: <20190916112411.21636-1-slp@redhat.com>
-Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
----
- hw/block/virtio-blk.c | 16 +++++++++++++++-
- 1 file changed, 15 insertions(+), 1 deletion(-)
-
-diff --git a/hw/block/virtio-blk.c b/hw/block/virtio-blk.c
-index cbb3729158..0d9adcdaff 100644
---- a/hw/block/virtio-blk.c
-+++ b/hw/block/virtio-blk.c
-@@ -16,6 +16,7 @@
- #include "qemu/iov.h"
- #include "qemu/module.h"
- #include "qemu/error-report.h"
-+#include "qemu/main-loop.h"
- #include "trace.h"
- #include "hw/block/block.h"
- #include "sysemu/blockdev.h"
-@@ -1082,11 +1083,24 @@ static int virtio_blk_load_device(VirtIODevice *vdev, QEMUFile *f,
- return 0;
- }
-
-+static void virtio_resize_cb(void *opaque)
-+{
-+ VirtIODevice *vdev = opaque;
-+
-+ assert(qemu_get_current_aio_context() == qemu_get_aio_context());
-+ virtio_notify_config(vdev);
-+}
-+
- static void virtio_blk_resize(void *opaque)
- {
- VirtIODevice *vdev = VIRTIO_DEVICE(opaque);
-
-- virtio_notify_config(vdev);
-+ /*
-+ * virtio_notify_config() needs to acquire the global mutex,
-+ * so it can't be called from an iothread. Instead, schedule
-+ * it to be run in the main context BH.
-+ */
-+ aio_bh_schedule_oneshot(qemu_get_aio_context(), virtio_resize_cb, vdev);
- }
-
- static const BlockDevOps virtio_block_ops = {
---
-2.20.1
-
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Li Qiang <liq3ea@163.com>
-Date: Sat, 31 Aug 2019 08:39:22 -0700
-Subject: [PATCH] vnc: fix memory leak when vnc disconnect
-
-Currently when qemu receives a vnc connect, it creates a 'VncState' to
-represent this connection. In 'vnc_worker_thread_loop' it creates a
-local 'VncState'. The connection 'VcnState' and local 'VncState' exchange
-data in 'vnc_async_encoding_start' and 'vnc_async_encoding_end'.
-In 'zrle_compress_data' it calls 'deflateInit2' to allocate the libz library
-opaque data. The 'VncState' used in 'zrle_compress_data' is the local
-'VncState'. In 'vnc_zrle_clear' it calls 'deflateEnd' to free the libz
-library opaque data. The 'VncState' used in 'vnc_zrle_clear' is the connection
-'VncState'. In currently implementation there will be a memory leak when the
-vnc disconnect. Following is the asan output backtrack:
-
-Direct leak of 29760 byte(s) in 5 object(s) allocated from:
- 0 0xffffa67ef3c3 in __interceptor_calloc (/lib64/libasan.so.4+0xd33c3)
- 1 0xffffa65071cb in g_malloc0 (/lib64/libglib-2.0.so.0+0x571cb)
- 2 0xffffa5e968f7 in deflateInit2_ (/lib64/libz.so.1+0x78f7)
- 3 0xaaaacec58613 in zrle_compress_data ui/vnc-enc-zrle.c:87
- 4 0xaaaacec58613 in zrle_send_framebuffer_update ui/vnc-enc-zrle.c:344
- 5 0xaaaacec34e77 in vnc_send_framebuffer_update ui/vnc.c:919
- 6 0xaaaacec5e023 in vnc_worker_thread_loop ui/vnc-jobs.c:271
- 7 0xaaaacec5e5e7 in vnc_worker_thread ui/vnc-jobs.c:340
- 8 0xaaaacee4d3c3 in qemu_thread_start util/qemu-thread-posix.c:502
- 9 0xffffa544e8bb in start_thread (/lib64/libpthread.so.0+0x78bb)
- 10 0xffffa53965cb in thread_start (/lib64/libc.so.6+0xd55cb)
-
-This is because the opaque allocated in 'deflateInit2' is not freed in
-'deflateEnd'. The reason is that the 'deflateEnd' calls 'deflateStateCheck'
-and in the latter will check whether 's->strm != strm'(libz's data structure).
-This check will be true so in 'deflateEnd' it just return 'Z_STREAM_ERROR' and
-not free the data allocated in 'deflateInit2'.
-
-The reason this happens is that the 'VncState' contains the whole 'VncZrle',
-so when calling 'deflateInit2', the 's->strm' will be the local address.
-So 's->strm != strm' will be true.
-
-To fix this issue, we need to make 'zrle' of 'VncState' to be a pointer.
-Then the connection 'VncState' and local 'VncState' exchange mechanism will
-work as expection. The 'tight' of 'VncState' has the same issue, let's also turn
-it to a pointer.
-
-Reported-by: Ying Fang <fangying1@huawei.com>
-Signed-off-by: Li Qiang <liq3ea@163.com>
-Message-id: 20190831153922.121308-1-liq3ea@163.com
-Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
-(cherry picked from commit 6bf21f3d83e95bcc4ba35a7a07cc6655e8b010b0)
-Signed-off-by: Oguz Bektas <o.bektas@proxmox.com>
----
- ui/vnc-enc-tight.c | 219 +++++++++++++++++++++---------------------
- ui/vnc-enc-zlib.c | 11 ++-
- ui/vnc-enc-zrle.c | 68 ++++++-------
- ui/vnc-enc-zrle.inc.c | 2 +-
- ui/vnc.c | 28 +++---
- ui/vnc.h | 4 +-
- 6 files changed, 170 insertions(+), 162 deletions(-)
-
-diff --git a/ui/vnc-enc-tight.c b/ui/vnc-enc-tight.c
-index 9084c2201b..1e0851826a 100644
---- a/ui/vnc-enc-tight.c
-+++ b/ui/vnc-enc-tight.c
-@@ -116,7 +116,7 @@ static int send_png_rect(VncState *vs, int x, int y, int w, int h,
-
- static bool tight_can_send_png_rect(VncState *vs, int w, int h)
- {
-- if (vs->tight.type != VNC_ENCODING_TIGHT_PNG) {
-+ if (vs->tight->type != VNC_ENCODING_TIGHT_PNG) {
- return false;
- }
-
-@@ -144,7 +144,7 @@ tight_detect_smooth_image24(VncState *vs, int w, int h)
- int pixels = 0;
- int pix, left[3];
- unsigned int errors;
-- unsigned char *buf = vs->tight.tight.buffer;
-+ unsigned char *buf = vs->tight->tight.buffer;
-
- /*
- * If client is big-endian, color samples begin from the second
-@@ -215,7 +215,7 @@ tight_detect_smooth_image24(VncState *vs, int w, int h)
- int pixels = 0; \
- int sample, sum, left[3]; \
- unsigned int errors; \
-- unsigned char *buf = vs->tight.tight.buffer; \
-+ unsigned char *buf = vs->tight->tight.buffer; \
- \
- endian = 0; /* FIXME */ \
- \
-@@ -296,8 +296,8 @@ static int
- tight_detect_smooth_image(VncState *vs, int w, int h)
- {
- unsigned int errors;
-- int compression = vs->tight.compression;
-- int quality = vs->tight.quality;
-+ int compression = vs->tight->compression;
-+ int quality = vs->tight->quality;
-
- if (!vs->vd->lossy) {
- return 0;
-@@ -309,7 +309,7 @@ tight_detect_smooth_image(VncState *vs, int w, int h)
- return 0;
- }
-
-- if (vs->tight.quality != (uint8_t)-1) {
-+ if (vs->tight->quality != (uint8_t)-1) {
- if (w * h < VNC_TIGHT_JPEG_MIN_RECT_SIZE) {
- return 0;
- }
-@@ -320,9 +320,9 @@ tight_detect_smooth_image(VncState *vs, int w, int h)
- }
-
- if (vs->client_pf.bytes_per_pixel == 4) {
-- if (vs->tight.pixel24) {
-+ if (vs->tight->pixel24) {
- errors = tight_detect_smooth_image24(vs, w, h);
-- if (vs->tight.quality != (uint8_t)-1) {
-+ if (vs->tight->quality != (uint8_t)-1) {
- return (errors < tight_conf[quality].jpeg_threshold24);
- }
- return (errors < tight_conf[compression].gradient_threshold24);
-@@ -352,7 +352,7 @@ tight_detect_smooth_image(VncState *vs, int w, int h)
- uint##bpp##_t c0, c1, ci; \
- int i, n0, n1; \
- \
-- data = (uint##bpp##_t *)vs->tight.tight.buffer; \
-+ data = (uint##bpp##_t *)vs->tight->tight.buffer; \
- \
- c0 = data[0]; \
- i = 1; \
-@@ -423,9 +423,9 @@ static int tight_fill_palette(VncState *vs, int x, int y,
- {
- int max;
-
-- max = count / tight_conf[vs->tight.compression].idx_max_colors_divisor;
-+ max = count / tight_conf[vs->tight->compression].idx_max_colors_divisor;
- if (max < 2 &&
-- count >= tight_conf[vs->tight.compression].mono_min_rect_size) {
-+ count >= tight_conf[vs->tight->compression].mono_min_rect_size) {
- max = 2;
- }
- if (max >= 256) {
-@@ -558,7 +558,7 @@ tight_filter_gradient24(VncState *vs, uint8_t *buf, int w, int h)
- int x, y, c;
-
- buf32 = (uint32_t *)buf;
-- memset(vs->tight.gradient.buffer, 0, w * 3 * sizeof(int));
-+ memset(vs->tight->gradient.buffer, 0, w * 3 * sizeof(int));
-
- if (1 /* FIXME */) {
- shift[0] = vs->client_pf.rshift;
-@@ -575,7 +575,7 @@ tight_filter_gradient24(VncState *vs, uint8_t *buf, int w, int h)
- upper[c] = 0;
- here[c] = 0;
- }
-- prev = (int *)vs->tight.gradient.buffer;
-+ prev = (int *)vs->tight->gradient.buffer;
- for (x = 0; x < w; x++) {
- pix32 = *buf32++;
- for (c = 0; c < 3; c++) {
-@@ -615,7 +615,7 @@ tight_filter_gradient24(VncState *vs, uint8_t *buf, int w, int h)
- int prediction; \
- int x, y, c; \
- \
-- memset (vs->tight.gradient.buffer, 0, w * 3 * sizeof(int)); \
-+ memset(vs->tight->gradient.buffer, 0, w * 3 * sizeof(int)); \
- \
- endian = 0; /* FIXME */ \
- \
-@@ -631,7 +631,7 @@ tight_filter_gradient24(VncState *vs, uint8_t *buf, int w, int h)
- upper[c] = 0; \
- here[c] = 0; \
- } \
-- prev = (int *)vs->tight.gradient.buffer; \
-+ prev = (int *)vs->tight->gradient.buffer; \
- for (x = 0; x < w; x++) { \
- pix = *buf; \
- if (endian) { \
-@@ -785,7 +785,7 @@ static void extend_solid_area(VncState *vs, int x, int y, int w, int h,
- static int tight_init_stream(VncState *vs, int stream_id,
- int level, int strategy)
- {
-- z_streamp zstream = &vs->tight.stream[stream_id];
-+ z_streamp zstream = &vs->tight->stream[stream_id];
-
- if (zstream->opaque == NULL) {
- int err;
-@@ -803,15 +803,15 @@ static int tight_init_stream(VncState *vs, int stream_id,
- return -1;
- }
-
-- vs->tight.levels[stream_id] = level;
-+ vs->tight->levels[stream_id] = level;
- zstream->opaque = vs;
- }
-
-- if (vs->tight.levels[stream_id] != level) {
-+ if (vs->tight->levels[stream_id] != level) {
- if (deflateParams(zstream, level, strategy) != Z_OK) {
- return -1;
- }
-- vs->tight.levels[stream_id] = level;
-+ vs->tight->levels[stream_id] = level;
- }
- return 0;
- }
-@@ -839,11 +839,11 @@ static void tight_send_compact_size(VncState *vs, size_t len)
- static int tight_compress_data(VncState *vs, int stream_id, size_t bytes,
- int level, int strategy)
- {
-- z_streamp zstream = &vs->tight.stream[stream_id];
-+ z_streamp zstream = &vs->tight->stream[stream_id];
- int previous_out;
-
- if (bytes < VNC_TIGHT_MIN_TO_COMPRESS) {
-- vnc_write(vs, vs->tight.tight.buffer, vs->tight.tight.offset);
-+ vnc_write(vs, vs->tight->tight.buffer, vs->tight->tight.offset);
- return bytes;
- }
-
-@@ -852,13 +852,13 @@ static int tight_compress_data(VncState *vs, int stream_id, size_t bytes,
- }
-
- /* reserve memory in output buffer */
-- buffer_reserve(&vs->tight.zlib, bytes + 64);
-+ buffer_reserve(&vs->tight->zlib, bytes + 64);
-
- /* set pointers */
-- zstream->next_in = vs->tight.tight.buffer;
-- zstream->avail_in = vs->tight.tight.offset;
-- zstream->next_out = vs->tight.zlib.buffer + vs->tight.zlib.offset;
-- zstream->avail_out = vs->tight.zlib.capacity - vs->tight.zlib.offset;
-+ zstream->next_in = vs->tight->tight.buffer;
-+ zstream->avail_in = vs->tight->tight.offset;
-+ zstream->next_out = vs->tight->zlib.buffer + vs->tight->zlib.offset;
-+ zstream->avail_out = vs->tight->zlib.capacity - vs->tight->zlib.offset;
- previous_out = zstream->avail_out;
- zstream->data_type = Z_BINARY;
-
-@@ -868,14 +868,14 @@ static int tight_compress_data(VncState *vs, int stream_id, size_t bytes,
- return -1;
- }
-
-- vs->tight.zlib.offset = vs->tight.zlib.capacity - zstream->avail_out;
-+ vs->tight->zlib.offset = vs->tight->zlib.capacity - zstream->avail_out;
- /* ...how much data has actually been produced by deflate() */
- bytes = previous_out - zstream->avail_out;
-
- tight_send_compact_size(vs, bytes);
-- vnc_write(vs, vs->tight.zlib.buffer, bytes);
-+ vnc_write(vs, vs->tight->zlib.buffer, bytes);
-
-- buffer_reset(&vs->tight.zlib);
-+ buffer_reset(&vs->tight->zlib);
-
- return bytes;
- }
-@@ -927,16 +927,17 @@ static int send_full_color_rect(VncState *vs, int x, int y, int w, int h)
-
- vnc_write_u8(vs, stream << 4); /* no flushing, no filter */
-
-- if (vs->tight.pixel24) {
-- tight_pack24(vs, vs->tight.tight.buffer, w * h, &vs->tight.tight.offset);
-+ if (vs->tight->pixel24) {
-+ tight_pack24(vs, vs->tight->tight.buffer, w * h,
-+ &vs->tight->tight.offset);
- bytes = 3;
- } else {
- bytes = vs->client_pf.bytes_per_pixel;
- }
-
- bytes = tight_compress_data(vs, stream, w * h * bytes,
-- tight_conf[vs->tight.compression].raw_zlib_level,
-- Z_DEFAULT_STRATEGY);
-+ tight_conf[vs->tight->compression].raw_zlib_level,
-+ Z_DEFAULT_STRATEGY);
-
- return (bytes >= 0);
- }
-@@ -947,14 +948,14 @@ static int send_solid_rect(VncState *vs)
-
- vnc_write_u8(vs, VNC_TIGHT_FILL << 4); /* no flushing, no filter */
-
-- if (vs->tight.pixel24) {
-- tight_pack24(vs, vs->tight.tight.buffer, 1, &vs->tight.tight.offset);
-+ if (vs->tight->pixel24) {
-+ tight_pack24(vs, vs->tight->tight.buffer, 1, &vs->tight->tight.offset);
- bytes = 3;
- } else {
- bytes = vs->client_pf.bytes_per_pixel;
- }
-
-- vnc_write(vs, vs->tight.tight.buffer, bytes);
-+ vnc_write(vs, vs->tight->tight.buffer, bytes);
- return 1;
- }
-
-@@ -963,7 +964,7 @@ static int send_mono_rect(VncState *vs, int x, int y,
- {
- ssize_t bytes;
- int stream = 1;
-- int level = tight_conf[vs->tight.compression].mono_zlib_level;
-+ int level = tight_conf[vs->tight->compression].mono_zlib_level;
-
- #ifdef CONFIG_VNC_PNG
- if (tight_can_send_png_rect(vs, w, h)) {
-@@ -991,26 +992,26 @@ static int send_mono_rect(VncState *vs, int x, int y,
- uint32_t buf[2] = {bg, fg};
- size_t ret = sizeof (buf);
-
-- if (vs->tight.pixel24) {
-+ if (vs->tight->pixel24) {
- tight_pack24(vs, (unsigned char*)buf, 2, &ret);
- }
- vnc_write(vs, buf, ret);
-
-- tight_encode_mono_rect32(vs->tight.tight.buffer, w, h, bg, fg);
-+ tight_encode_mono_rect32(vs->tight->tight.buffer, w, h, bg, fg);
- break;
- }
- case 2:
- vnc_write(vs, &bg, 2);
- vnc_write(vs, &fg, 2);
-- tight_encode_mono_rect16(vs->tight.tight.buffer, w, h, bg, fg);
-+ tight_encode_mono_rect16(vs->tight->tight.buffer, w, h, bg, fg);
- break;
- default:
- vnc_write_u8(vs, bg);
- vnc_write_u8(vs, fg);
-- tight_encode_mono_rect8(vs->tight.tight.buffer, w, h, bg, fg);
-+ tight_encode_mono_rect8(vs->tight->tight.buffer, w, h, bg, fg);
- break;
- }
-- vs->tight.tight.offset = bytes;
-+ vs->tight->tight.offset = bytes;
-
- bytes = tight_compress_data(vs, stream, bytes, level, Z_DEFAULT_STRATEGY);
- return (bytes >= 0);
-@@ -1040,7 +1041,7 @@ static void write_palette(int idx, uint32_t color, void *opaque)
- static bool send_gradient_rect(VncState *vs, int x, int y, int w, int h)
- {
- int stream = 3;
-- int level = tight_conf[vs->tight.compression].gradient_zlib_level;
-+ int level = tight_conf[vs->tight->compression].gradient_zlib_level;
- ssize_t bytes;
-
- if (vs->client_pf.bytes_per_pixel == 1) {
-@@ -1050,23 +1051,23 @@ static bool send_gradient_rect(VncState *vs, int x, int y, int w, int h)
- vnc_write_u8(vs, (stream | VNC_TIGHT_EXPLICIT_FILTER) << 4);
- vnc_write_u8(vs, VNC_TIGHT_FILTER_GRADIENT);
-
-- buffer_reserve(&vs->tight.gradient, w * 3 * sizeof (int));
-+ buffer_reserve(&vs->tight->gradient, w * 3 * sizeof(int));
-
-- if (vs->tight.pixel24) {
-- tight_filter_gradient24(vs, vs->tight.tight.buffer, w, h);
-+ if (vs->tight->pixel24) {
-+ tight_filter_gradient24(vs, vs->tight->tight.buffer, w, h);
- bytes = 3;
- } else if (vs->client_pf.bytes_per_pixel == 4) {
-- tight_filter_gradient32(vs, (uint32_t *)vs->tight.tight.buffer, w, h);
-+ tight_filter_gradient32(vs, (uint32_t *)vs->tight->tight.buffer, w, h);
- bytes = 4;
- } else {
-- tight_filter_gradient16(vs, (uint16_t *)vs->tight.tight.buffer, w, h);
-+ tight_filter_gradient16(vs, (uint16_t *)vs->tight->tight.buffer, w, h);
- bytes = 2;
- }
-
-- buffer_reset(&vs->tight.gradient);
-+ buffer_reset(&vs->tight->gradient);
-
- bytes = w * h * bytes;
-- vs->tight.tight.offset = bytes;
-+ vs->tight->tight.offset = bytes;
-
- bytes = tight_compress_data(vs, stream, bytes,
- level, Z_FILTERED);
-@@ -1077,7 +1078,7 @@ static int send_palette_rect(VncState *vs, int x, int y,
- int w, int h, VncPalette *palette)
- {
- int stream = 2;
-- int level = tight_conf[vs->tight.compression].idx_zlib_level;
-+ int level = tight_conf[vs->tight->compression].idx_zlib_level;
- int colors;
- ssize_t bytes;
-
-@@ -1104,12 +1105,12 @@ static int send_palette_rect(VncState *vs, int x, int y,
- palette_iter(palette, write_palette, &priv);
- vnc_write(vs, header, sizeof(header));
-
-- if (vs->tight.pixel24) {
-+ if (vs->tight->pixel24) {
- tight_pack24(vs, vs->output.buffer + old_offset, colors, &offset);
- vs->output.offset = old_offset + offset;
- }
-
-- tight_encode_indexed_rect32(vs->tight.tight.buffer, w * h, palette);
-+ tight_encode_indexed_rect32(vs->tight->tight.buffer, w * h, palette);
- break;
- }
- case 2:
-@@ -1119,7 +1120,7 @@ static int send_palette_rect(VncState *vs, int x, int y,
-
- palette_iter(palette, write_palette, &priv);
- vnc_write(vs, header, sizeof(header));
-- tight_encode_indexed_rect16(vs->tight.tight.buffer, w * h, palette);
-+ tight_encode_indexed_rect16(vs->tight->tight.buffer, w * h, palette);
- break;
- }
- default:
-@@ -1127,7 +1128,7 @@ static int send_palette_rect(VncState *vs, int x, int y,
- break;
- }
- bytes = w * h;
-- vs->tight.tight.offset = bytes;
-+ vs->tight->tight.offset = bytes;
-
- bytes = tight_compress_data(vs, stream, bytes,
- level, Z_DEFAULT_STRATEGY);
-@@ -1146,7 +1147,7 @@ static int send_palette_rect(VncState *vs, int x, int y,
- static void jpeg_init_destination(j_compress_ptr cinfo)
- {
- VncState *vs = cinfo->client_data;
-- Buffer *buffer = &vs->tight.jpeg;
-+ Buffer *buffer = &vs->tight->jpeg;
-
- cinfo->dest->next_output_byte = (JOCTET *)buffer->buffer + buffer->offset;
- cinfo->dest->free_in_buffer = (size_t)(buffer->capacity - buffer->offset);
-@@ -1156,7 +1157,7 @@ static void jpeg_init_destination(j_compress_ptr cinfo)
- static boolean jpeg_empty_output_buffer(j_compress_ptr cinfo)
- {
- VncState *vs = cinfo->client_data;
-- Buffer *buffer = &vs->tight.jpeg;
-+ Buffer *buffer = &vs->tight->jpeg;
-
- buffer->offset = buffer->capacity;
- buffer_reserve(buffer, 2048);
-@@ -1168,7 +1169,7 @@ static boolean jpeg_empty_output_buffer(j_compress_ptr cinfo)
- static void jpeg_term_destination(j_compress_ptr cinfo)
- {
- VncState *vs = cinfo->client_data;
-- Buffer *buffer = &vs->tight.jpeg;
-+ Buffer *buffer = &vs->tight->jpeg;
-
- buffer->offset = buffer->capacity - cinfo->dest->free_in_buffer;
- }
-@@ -1187,7 +1188,7 @@ static int send_jpeg_rect(VncState *vs, int x, int y, int w, int h, int quality)
- return send_full_color_rect(vs, x, y, w, h);
- }
-
-- buffer_reserve(&vs->tight.jpeg, 2048);
-+ buffer_reserve(&vs->tight->jpeg, 2048);
-
- cinfo.err = jpeg_std_error(&jerr);
- jpeg_create_compress(&cinfo);
-@@ -1222,9 +1223,9 @@ static int send_jpeg_rect(VncState *vs, int x, int y, int w, int h, int quality)
-
- vnc_write_u8(vs, VNC_TIGHT_JPEG << 4);
-
-- tight_send_compact_size(vs, vs->tight.jpeg.offset);
-- vnc_write(vs, vs->tight.jpeg.buffer, vs->tight.jpeg.offset);
-- buffer_reset(&vs->tight.jpeg);
-+ tight_send_compact_size(vs, vs->tight->jpeg.offset);
-+ vnc_write(vs, vs->tight->jpeg.buffer, vs->tight->jpeg.offset);
-+ buffer_reset(&vs->tight->jpeg);
-
- return 1;
- }
-@@ -1240,7 +1241,7 @@ static void write_png_palette(int idx, uint32_t pix, void *opaque)
- VncState *vs = priv->vs;
- png_colorp color = &priv->png_palette[idx];
-
-- if (vs->tight.pixel24)
-+ if (vs->tight->pixel24)
- {
- color->red = (pix >> vs->client_pf.rshift) & vs->client_pf.rmax;
- color->green = (pix >> vs->client_pf.gshift) & vs->client_pf.gmax;
-@@ -1267,10 +1268,10 @@ static void png_write_data(png_structp png_ptr, png_bytep data,
- {
- VncState *vs = png_get_io_ptr(png_ptr);
-
-- buffer_reserve(&vs->tight.png, vs->tight.png.offset + length);
-- memcpy(vs->tight.png.buffer + vs->tight.png.offset, data, length);
-+ buffer_reserve(&vs->tight->png, vs->tight->png.offset + length);
-+ memcpy(vs->tight->png.buffer + vs->tight->png.offset, data, length);
-
-- vs->tight.png.offset += length;
-+ vs->tight->png.offset += length;
- }
-
- static void png_flush_data(png_structp png_ptr)
-@@ -1295,8 +1296,8 @@ static int send_png_rect(VncState *vs, int x, int y, int w, int h,
- png_infop info_ptr;
- png_colorp png_palette = NULL;
- pixman_image_t *linebuf;
-- int level = tight_png_conf[vs->tight.compression].png_zlib_level;
-- int filters = tight_png_conf[vs->tight.compression].png_filters;
-+ int level = tight_png_conf[vs->tight->compression].png_zlib_level;
-+ int filters = tight_png_conf[vs->tight->compression].png_filters;
- uint8_t *buf;
- int dy;
-
-@@ -1340,21 +1341,23 @@ static int send_png_rect(VncState *vs, int x, int y, int w, int h,
- png_set_PLTE(png_ptr, info_ptr, png_palette, palette_size(palette));
-
- if (vs->client_pf.bytes_per_pixel == 4) {
-- tight_encode_indexed_rect32(vs->tight.tight.buffer, w * h, palette);
-+ tight_encode_indexed_rect32(vs->tight->tight.buffer, w * h,
-+ palette);
- } else {
-- tight_encode_indexed_rect16(vs->tight.tight.buffer, w * h, palette);
-+ tight_encode_indexed_rect16(vs->tight->tight.buffer, w * h,
-+ palette);
- }
- }
-
- png_write_info(png_ptr, info_ptr);
-
-- buffer_reserve(&vs->tight.png, 2048);
-+ buffer_reserve(&vs->tight->png, 2048);
- linebuf = qemu_pixman_linebuf_create(PIXMAN_BE_r8g8b8, w);
- buf = (uint8_t *)pixman_image_get_data(linebuf);
- for (dy = 0; dy < h; dy++)
- {
- if (color_type == PNG_COLOR_TYPE_PALETTE) {
-- memcpy(buf, vs->tight.tight.buffer + (dy * w), w);
-+ memcpy(buf, vs->tight->tight.buffer + (dy * w), w);
- } else {
- qemu_pixman_linebuf_fill(linebuf, vs->vd->server, w, x, y + dy);
- }
-@@ -1372,27 +1375,27 @@ static int send_png_rect(VncState *vs, int x, int y, int w, int h,
-
- vnc_write_u8(vs, VNC_TIGHT_PNG << 4);
-
-- tight_send_compact_size(vs, vs->tight.png.offset);
-- vnc_write(vs, vs->tight.png.buffer, vs->tight.png.offset);
-- buffer_reset(&vs->tight.png);
-+ tight_send_compact_size(vs, vs->tight->png.offset);
-+ vnc_write(vs, vs->tight->png.buffer, vs->tight->png.offset);
-+ buffer_reset(&vs->tight->png);
- return 1;
- }
- #endif /* CONFIG_VNC_PNG */
-
- static void vnc_tight_start(VncState *vs)
- {
-- buffer_reset(&vs->tight.tight);
-+ buffer_reset(&vs->tight->tight);
-
- // make the output buffer be the zlib buffer, so we can compress it later
-- vs->tight.tmp = vs->output;
-- vs->output = vs->tight.tight;
-+ vs->tight->tmp = vs->output;
-+ vs->output = vs->tight->tight;
- }
-
- static void vnc_tight_stop(VncState *vs)
- {
- // switch back to normal output/zlib buffers
-- vs->tight.tight = vs->output;
-- vs->output = vs->tight.tmp;
-+ vs->tight->tight = vs->output;
-+ vs->output = vs->tight->tmp;
- }
-
- static int send_sub_rect_nojpeg(VncState *vs, int x, int y, int w, int h,
-@@ -1426,9 +1429,9 @@ static int send_sub_rect_jpeg(VncState *vs, int x, int y, int w, int h,
- int ret;
-
- if (colors == 0) {
-- if (force || (tight_jpeg_conf[vs->tight.quality].jpeg_full &&
-+ if (force || (tight_jpeg_conf[vs->tight->quality].jpeg_full &&
- tight_detect_smooth_image(vs, w, h))) {
-- int quality = tight_conf[vs->tight.quality].jpeg_quality;
-+ int quality = tight_conf[vs->tight->quality].jpeg_quality;
-
- ret = send_jpeg_rect(vs, x, y, w, h, quality);
- } else {
-@@ -1440,9 +1443,9 @@ static int send_sub_rect_jpeg(VncState *vs, int x, int y, int w, int h,
- ret = send_mono_rect(vs, x, y, w, h, bg, fg);
- } else if (colors <= 256) {
- if (force || (colors > 96 &&
-- tight_jpeg_conf[vs->tight.quality].jpeg_idx &&
-+ tight_jpeg_conf[vs->tight->quality].jpeg_idx &&
- tight_detect_smooth_image(vs, w, h))) {
-- int quality = tight_conf[vs->tight.quality].jpeg_quality;
-+ int quality = tight_conf[vs->tight->quality].jpeg_quality;
-
- ret = send_jpeg_rect(vs, x, y, w, h, quality);
- } else {
-@@ -1480,20 +1483,20 @@ static int send_sub_rect(VncState *vs, int x, int y, int w, int h)
- qemu_thread_atexit_add(&vnc_tight_cleanup_notifier);
- }
-
-- vnc_framebuffer_update(vs, x, y, w, h, vs->tight.type);
-+ vnc_framebuffer_update(vs, x, y, w, h, vs->tight->type);
-
- vnc_tight_start(vs);
- vnc_raw_send_framebuffer_update(vs, x, y, w, h);
- vnc_tight_stop(vs);
-
- #ifdef CONFIG_VNC_JPEG
-- if (!vs->vd->non_adaptive && vs->tight.quality != (uint8_t)-1) {
-+ if (!vs->vd->non_adaptive && vs->tight->quality != (uint8_t)-1) {
- double freq = vnc_update_freq(vs, x, y, w, h);
-
-- if (freq < tight_jpeg_conf[vs->tight.quality].jpeg_freq_min) {
-+ if (freq < tight_jpeg_conf[vs->tight->quality].jpeg_freq_min) {
- allow_jpeg = false;
- }
-- if (freq >= tight_jpeg_conf[vs->tight.quality].jpeg_freq_threshold) {
-+ if (freq >= tight_jpeg_conf[vs->tight->quality].jpeg_freq_threshold) {
- force_jpeg = true;
- vnc_sent_lossy_rect(vs, x, y, w, h);
- }
-@@ -1503,7 +1506,7 @@ static int send_sub_rect(VncState *vs, int x, int y, int w, int h)
- colors = tight_fill_palette(vs, x, y, w * h, &bg, &fg, color_count_palette);
-
- #ifdef CONFIG_VNC_JPEG
-- if (allow_jpeg && vs->tight.quality != (uint8_t)-1) {
-+ if (allow_jpeg && vs->tight->quality != (uint8_t)-1) {
- ret = send_sub_rect_jpeg(vs, x, y, w, h, bg, fg, colors,
- color_count_palette, force_jpeg);
- } else {
-@@ -1520,7 +1523,7 @@ static int send_sub_rect(VncState *vs, int x, int y, int w, int h)
-
- static int send_sub_rect_solid(VncState *vs, int x, int y, int w, int h)
- {
-- vnc_framebuffer_update(vs, x, y, w, h, vs->tight.type);
-+ vnc_framebuffer_update(vs, x, y, w, h, vs->tight->type);
-
- vnc_tight_start(vs);
- vnc_raw_send_framebuffer_update(vs, x, y, w, h);
-@@ -1538,8 +1541,8 @@ static int send_rect_simple(VncState *vs, int x, int y, int w, int h,
- int rw, rh;
- int n = 0;
-
-- max_size = tight_conf[vs->tight.compression].max_rect_size;
-- max_width = tight_conf[vs->tight.compression].max_rect_width;
-+ max_size = tight_conf[vs->tight->compression].max_rect_size;
-+ max_width = tight_conf[vs->tight->compression].max_rect_width;
-
- if (split && (w > max_width || w * h > max_size)) {
- max_sub_width = (w > max_width) ? max_width : w;
-@@ -1648,16 +1651,16 @@ static int tight_send_framebuffer_update(VncState *vs, int x, int y,
-
- if (vs->client_pf.bytes_per_pixel == 4 && vs->client_pf.rmax == 0xFF &&
- vs->client_pf.bmax == 0xFF && vs->client_pf.gmax == 0xFF) {
-- vs->tight.pixel24 = true;
-+ vs->tight->pixel24 = true;
- } else {
-- vs->tight.pixel24 = false;
-+ vs->tight->pixel24 = false;
- }
-
- #ifdef CONFIG_VNC_JPEG
-- if (vs->tight.quality != (uint8_t)-1) {
-+ if (vs->tight->quality != (uint8_t)-1) {
- double freq = vnc_update_freq(vs, x, y, w, h);
-
-- if (freq > tight_jpeg_conf[vs->tight.quality].jpeg_freq_threshold) {
-+ if (freq > tight_jpeg_conf[vs->tight->quality].jpeg_freq_threshold) {
- return send_rect_simple(vs, x, y, w, h, false);
- }
- }
-@@ -1669,8 +1672,8 @@ static int tight_send_framebuffer_update(VncState *vs, int x, int y,
-
- /* Calculate maximum number of rows in one non-solid rectangle. */
-
-- max_rows = tight_conf[vs->tight.compression].max_rect_size;
-- max_rows /= MIN(tight_conf[vs->tight.compression].max_rect_width, w);
-+ max_rows = tight_conf[vs->tight->compression].max_rect_size;
-+ max_rows /= MIN(tight_conf[vs->tight->compression].max_rect_width, w);
-
- return find_large_solid_color_rect(vs, x, y, w, h, max_rows);
- }
-@@ -1678,33 +1681,33 @@ static int tight_send_framebuffer_update(VncState *vs, int x, int y,
- int vnc_tight_send_framebuffer_update(VncState *vs, int x, int y,
- int w, int h)
- {
-- vs->tight.type = VNC_ENCODING_TIGHT;
-+ vs->tight->type = VNC_ENCODING_TIGHT;
- return tight_send_framebuffer_update(vs, x, y, w, h);
- }
-
- int vnc_tight_png_send_framebuffer_update(VncState *vs, int x, int y,
- int w, int h)
- {
-- vs->tight.type = VNC_ENCODING_TIGHT_PNG;
-+ vs->tight->type = VNC_ENCODING_TIGHT_PNG;
- return tight_send_framebuffer_update(vs, x, y, w, h);
- }
-
- void vnc_tight_clear(VncState *vs)
- {
- int i;
-- for (i=0; i<ARRAY_SIZE(vs->tight.stream); i++) {
-- if (vs->tight.stream[i].opaque) {
-- deflateEnd(&vs->tight.stream[i]);
-+ for (i = 0; i < ARRAY_SIZE(vs->tight->stream); i++) {
-+ if (vs->tight->stream[i].opaque) {
-+ deflateEnd(&vs->tight->stream[i]);
- }
- }
-
-- buffer_free(&vs->tight.tight);
-- buffer_free(&vs->tight.zlib);
-- buffer_free(&vs->tight.gradient);
-+ buffer_free(&vs->tight->tight);
-+ buffer_free(&vs->tight->zlib);
-+ buffer_free(&vs->tight->gradient);
- #ifdef CONFIG_VNC_JPEG
-- buffer_free(&vs->tight.jpeg);
-+ buffer_free(&vs->tight->jpeg);
- #endif
- #ifdef CONFIG_VNC_PNG
-- buffer_free(&vs->tight.png);
-+ buffer_free(&vs->tight->png);
- #endif
- }
-diff --git a/ui/vnc-enc-zlib.c b/ui/vnc-enc-zlib.c
-index 33e9df2f6a..900ae5b30f 100644
---- a/ui/vnc-enc-zlib.c
-+++ b/ui/vnc-enc-zlib.c
-@@ -76,7 +76,8 @@ static int vnc_zlib_stop(VncState *vs)
- zstream->zalloc = vnc_zlib_zalloc;
- zstream->zfree = vnc_zlib_zfree;
-
-- err = deflateInit2(zstream, vs->tight.compression, Z_DEFLATED, MAX_WBITS,
-+ err = deflateInit2(zstream, vs->tight->compression, Z_DEFLATED,
-+ MAX_WBITS,
- MAX_MEM_LEVEL, Z_DEFAULT_STRATEGY);
-
- if (err != Z_OK) {
-@@ -84,16 +85,16 @@ static int vnc_zlib_stop(VncState *vs)
- return -1;
- }
-
-- vs->zlib.level = vs->tight.compression;
-+ vs->zlib.level = vs->tight->compression;
- zstream->opaque = vs;
- }
-
-- if (vs->tight.compression != vs->zlib.level) {
-- if (deflateParams(zstream, vs->tight.compression,
-+ if (vs->tight->compression != vs->zlib.level) {
-+ if (deflateParams(zstream, vs->tight->compression,
- Z_DEFAULT_STRATEGY) != Z_OK) {
- return -1;
- }
-- vs->zlib.level = vs->tight.compression;
-+ vs->zlib.level = vs->tight->compression;
- }
-
- // reserve memory in output buffer
-diff --git a/ui/vnc-enc-zrle.c b/ui/vnc-enc-zrle.c
-index 7493a84723..17fd28a2e2 100644
---- a/ui/vnc-enc-zrle.c
-+++ b/ui/vnc-enc-zrle.c
-@@ -37,18 +37,18 @@ static const int bits_per_packed_pixel[] = {
-
- static void vnc_zrle_start(VncState *vs)
- {
-- buffer_reset(&vs->zrle.zrle);
-+ buffer_reset(&vs->zrle->zrle);
-
- /* make the output buffer be the zlib buffer, so we can compress it later */
-- vs->zrle.tmp = vs->output;
-- vs->output = vs->zrle.zrle;
-+ vs->zrle->tmp = vs->output;
-+ vs->output = vs->zrle->zrle;
- }
-
- static void vnc_zrle_stop(VncState *vs)
- {
- /* switch back to normal output/zlib buffers */
-- vs->zrle.zrle = vs->output;
-- vs->output = vs->zrle.tmp;
-+ vs->zrle->zrle = vs->output;
-+ vs->output = vs->zrle->tmp;
- }
-
- static void *zrle_convert_fb(VncState *vs, int x, int y, int w, int h,
-@@ -56,24 +56,24 @@ static void *zrle_convert_fb(VncState *vs, int x, int y, int w, int h,
- {
- Buffer tmp;
-
-- buffer_reset(&vs->zrle.fb);
-- buffer_reserve(&vs->zrle.fb, w * h * bpp + bpp);
-+ buffer_reset(&vs->zrle->fb);
-+ buffer_reserve(&vs->zrle->fb, w * h * bpp + bpp);
-
- tmp = vs->output;
-- vs->output = vs->zrle.fb;
-+ vs->output = vs->zrle->fb;
-
- vnc_raw_send_framebuffer_update(vs, x, y, w, h);
-
-- vs->zrle.fb = vs->output;
-+ vs->zrle->fb = vs->output;
- vs->output = tmp;
-- return vs->zrle.fb.buffer;
-+ return vs->zrle->fb.buffer;
- }
-
- static int zrle_compress_data(VncState *vs, int level)
- {
-- z_streamp zstream = &vs->zrle.stream;
-+ z_streamp zstream = &vs->zrle->stream;
-
-- buffer_reset(&vs->zrle.zlib);
-+ buffer_reset(&vs->zrle->zlib);
-
- if (zstream->opaque != vs) {
- int err;
-@@ -93,13 +93,13 @@ static int zrle_compress_data(VncState *vs, int level)
- }
-
- /* reserve memory in output buffer */
-- buffer_reserve(&vs->zrle.zlib, vs->zrle.zrle.offset + 64);
-+ buffer_reserve(&vs->zrle->zlib, vs->zrle->zrle.offset + 64);
-
- /* set pointers */
-- zstream->next_in = vs->zrle.zrle.buffer;
-- zstream->avail_in = vs->zrle.zrle.offset;
-- zstream->next_out = vs->zrle.zlib.buffer + vs->zrle.zlib.offset;
-- zstream->avail_out = vs->zrle.zlib.capacity - vs->zrle.zlib.offset;
-+ zstream->next_in = vs->zrle->zrle.buffer;
-+ zstream->avail_in = vs->zrle->zrle.offset;
-+ zstream->next_out = vs->zrle->zlib.buffer + vs->zrle->zlib.offset;
-+ zstream->avail_out = vs->zrle->zlib.capacity - vs->zrle->zlib.offset;
- zstream->data_type = Z_BINARY;
-
- /* start encoding */
-@@ -108,8 +108,8 @@ static int zrle_compress_data(VncState *vs, int level)
- return -1;
- }
-
-- vs->zrle.zlib.offset = vs->zrle.zlib.capacity - zstream->avail_out;
-- return vs->zrle.zlib.offset;
-+ vs->zrle->zlib.offset = vs->zrle->zlib.capacity - zstream->avail_out;
-+ return vs->zrle->zlib.offset;
- }
-
- /* Try to work out whether to use RLE and/or a palette. We do this by
-@@ -259,14 +259,14 @@ static int zrle_send_framebuffer_update(VncState *vs, int x, int y,
- size_t bytes;
- int zywrle_level;
-
-- if (vs->zrle.type == VNC_ENCODING_ZYWRLE) {
-- if (!vs->vd->lossy || vs->tight.quality == (uint8_t)-1
-- || vs->tight.quality == 9) {
-+ if (vs->zrle->type == VNC_ENCODING_ZYWRLE) {
-+ if (!vs->vd->lossy || vs->tight->quality == (uint8_t)-1
-+ || vs->tight->quality == 9) {
- zywrle_level = 0;
-- vs->zrle.type = VNC_ENCODING_ZRLE;
-- } else if (vs->tight.quality < 3) {
-+ vs->zrle->type = VNC_ENCODING_ZRLE;
-+ } else if (vs->tight->quality < 3) {
- zywrle_level = 3;
-- } else if (vs->tight.quality < 6) {
-+ } else if (vs->tight->quality < 6) {
- zywrle_level = 2;
- } else {
- zywrle_level = 1;
-@@ -337,30 +337,30 @@ static int zrle_send_framebuffer_update(VncState *vs, int x, int y,
-
- vnc_zrle_stop(vs);
- bytes = zrle_compress_data(vs, Z_DEFAULT_COMPRESSION);
-- vnc_framebuffer_update(vs, x, y, w, h, vs->zrle.type);
-+ vnc_framebuffer_update(vs, x, y, w, h, vs->zrle->type);
- vnc_write_u32(vs, bytes);
-- vnc_write(vs, vs->zrle.zlib.buffer, vs->zrle.zlib.offset);
-+ vnc_write(vs, vs->zrle->zlib.buffer, vs->zrle->zlib.offset);
- return 1;
- }
-
- int vnc_zrle_send_framebuffer_update(VncState *vs, int x, int y, int w, int h)
- {
-- vs->zrle.type = VNC_ENCODING_ZRLE;
-+ vs->zrle->type = VNC_ENCODING_ZRLE;
- return zrle_send_framebuffer_update(vs, x, y, w, h);
- }
-
- int vnc_zywrle_send_framebuffer_update(VncState *vs, int x, int y, int w, int h)
- {
-- vs->zrle.type = VNC_ENCODING_ZYWRLE;
-+ vs->zrle->type = VNC_ENCODING_ZYWRLE;
- return zrle_send_framebuffer_update(vs, x, y, w, h);
- }
-
- void vnc_zrle_clear(VncState *vs)
- {
-- if (vs->zrle.stream.opaque) {
-- deflateEnd(&vs->zrle.stream);
-+ if (vs->zrle->stream.opaque) {
-+ deflateEnd(&vs->zrle->stream);
- }
-- buffer_free(&vs->zrle.zrle);
-- buffer_free(&vs->zrle.fb);
-- buffer_free(&vs->zrle.zlib);
-+ buffer_free(&vs->zrle->zrle);
-+ buffer_free(&vs->zrle->fb);
-+ buffer_free(&vs->zrle->zlib);
- }
-diff --git a/ui/vnc-enc-zrle.inc.c b/ui/vnc-enc-zrle.inc.c
-index abf6b86e4e..c107d8affc 100644
---- a/ui/vnc-enc-zrle.inc.c
-+++ b/ui/vnc-enc-zrle.inc.c
-@@ -96,7 +96,7 @@ static void ZRLE_ENCODE(VncState *vs, int x, int y, int w, int h,
- static void ZRLE_ENCODE_TILE(VncState *vs, ZRLE_PIXEL *data, int w, int h,
- int zywrle_level)
- {
-- VncPalette *palette = &vs->zrle.palette;
-+ VncPalette *palette = &vs->zrle->palette;
-
- int runs = 0;
- int single_pixels = 0;
-diff --git a/ui/vnc.c b/ui/vnc.c
-index 38f92bfca3..e0756ca53a 100644
---- a/ui/vnc.c
-+++ b/ui/vnc.c
-@@ -1304,6 +1304,8 @@ void vnc_disconnect_finish(VncState *vs)
- object_unref(OBJECT(vs->sioc));
- vs->sioc = NULL;
- vs->magic = 0;
-+ g_free(vs->zrle);
-+ g_free(vs->tight);
- g_free(vs);
- }
-
-@@ -2055,8 +2057,8 @@ static void set_encodings(VncState *vs, int32_t *encodings, size_t n_encodings)
-
- vs->features = 0;
- vs->vnc_encoding = 0;
-- vs->tight.compression = 9;
-- vs->tight.quality = -1; /* Lossless by default */
-+ vs->tight->compression = 9;
-+ vs->tight->quality = -1; /* Lossless by default */
- vs->absolute = -1;
-
- /*
-@@ -2124,11 +2126,11 @@ static void set_encodings(VncState *vs, int32_t *encodings, size_t n_encodings)
- vs->features |= VNC_FEATURE_LED_STATE_MASK;
- break;
- case VNC_ENCODING_COMPRESSLEVEL0 ... VNC_ENCODING_COMPRESSLEVEL0 + 9:
-- vs->tight.compression = (enc & 0x0F);
-+ vs->tight->compression = (enc & 0x0F);
- break;
- case VNC_ENCODING_QUALITYLEVEL0 ... VNC_ENCODING_QUALITYLEVEL0 + 9:
- if (vs->vd->lossy) {
-- vs->tight.quality = (enc & 0x0F);
-+ vs->tight->quality = (enc & 0x0F);
- }
- break;
- default:
-@@ -3031,6 +3033,8 @@ static void vnc_connect(VncDisplay *vd, QIOChannelSocket *sioc,
- int i;
-
- trace_vnc_client_connect(vs, sioc);
-+ vs->zrle = g_new0(VncZrle, 1);
-+ vs->tight = g_new0(VncTight, 1);
- vs->magic = VNC_MAGIC;
- vs->sioc = sioc;
- object_ref(OBJECT(vs->sioc));
-@@ -3042,19 +3046,19 @@ static void vnc_connect(VncDisplay *vd, QIOChannelSocket *sioc,
- buffer_init(&vs->output, "vnc-output/%p", sioc);
- buffer_init(&vs->jobs_buffer, "vnc-jobs_buffer/%p", sioc);
-
-- buffer_init(&vs->tight.tight, "vnc-tight/%p", sioc);
-- buffer_init(&vs->tight.zlib, "vnc-tight-zlib/%p", sioc);
-- buffer_init(&vs->tight.gradient, "vnc-tight-gradient/%p", sioc);
-+ buffer_init(&vs->tight->tight, "vnc-tight/%p", sioc);
-+ buffer_init(&vs->tight->zlib, "vnc-tight-zlib/%p", sioc);
-+ buffer_init(&vs->tight->gradient, "vnc-tight-gradient/%p", sioc);
- #ifdef CONFIG_VNC_JPEG
-- buffer_init(&vs->tight.jpeg, "vnc-tight-jpeg/%p", sioc);
-+ buffer_init(&vs->tight->jpeg, "vnc-tight-jpeg/%p", sioc);
- #endif
- #ifdef CONFIG_VNC_PNG
-- buffer_init(&vs->tight.png, "vnc-tight-png/%p", sioc);
-+ buffer_init(&vs->tight->png, "vnc-tight-png/%p", sioc);
- #endif
- buffer_init(&vs->zlib.zlib, "vnc-zlib/%p", sioc);
-- buffer_init(&vs->zrle.zrle, "vnc-zrle/%p", sioc);
-- buffer_init(&vs->zrle.fb, "vnc-zrle-fb/%p", sioc);
-- buffer_init(&vs->zrle.zlib, "vnc-zrle-zlib/%p", sioc);
-+ buffer_init(&vs->zrle->zrle, "vnc-zrle/%p", sioc);
-+ buffer_init(&vs->zrle->fb, "vnc-zrle-fb/%p", sioc);
-+ buffer_init(&vs->zrle->zlib, "vnc-zrle-zlib/%p", sioc);
-
- if (skipauth) {
- vs->auth = VNC_AUTH_NONE;
-diff --git a/ui/vnc.h b/ui/vnc.h
-index 2f84db3142..75888e1558 100644
---- a/ui/vnc.h
-+++ b/ui/vnc.h
-@@ -337,10 +337,10 @@ struct VncState
- /* Encoding specific, if you add something here, don't forget to
- * update vnc_async_encoding_start()
- */
-- VncTight tight;
-+ VncTight *tight;
- VncZlib zlib;
- VncHextile hextile;
-- VncZrle zrle;
-+ VncZrle *zrle;
- VncZywrle zywrle;
-
- Notifier mouse_mode_notifier;
---
-2.20.1
-
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: =?UTF-8?q?Marc-Andr=C3=A9=20Lureau?= <marcandre.lureau@redhat.com>
-Date: Mon, 27 Jan 2020 10:24:09 +0100
-Subject: [PATCH 1/2] util: add slirp_fmt() helpers
-MIME-Version: 1.0
-Content-Type: text/plain; charset=UTF-8
-Content-Transfer-Encoding: 8bit
-
-Various calls to snprintf() in libslirp assume that snprintf() returns
-"only" the number of bytes written (excluding terminating NUL).
-
-https://pubs.opengroup.org/onlinepubs/9699919799/functions/snprintf.html#tag_16_159_04
-
-"Upon successful completion, the snprintf() function shall return the
-number of bytes that would be written to s had n been sufficiently
-large excluding the terminating null byte."
-
-Introduce slirp_fmt() that handles several pathological cases the
-way libslirp usually expect:
-
-- treat error as fatal (instead of silently returning -1)
-
-- fmt0() will always \0 end
-
-- return the number of bytes actually written (instead of what would
-have been written, which would usually result in OOB later), including
-the ending \0 for fmt0()
-
-- warn if truncation happened (instead of ignoring)
-
-Other less common cases can still be handled with strcpy/snprintf() etc.
-
-Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
-Reviewed-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
-Message-Id: <20200127092414.169796-2-marcandre.lureau@redhat.com>
-Signed-off-by: Oguz Bektas <o.bektas@proxmox.com>
----
- slirp/src/util.c | 62 ++++++++++++++++++++++++++++++++++++++++++++++++++++++
- slirp/src/util.h | 3 +++
- 2 files changed, 65 insertions(+)
-
-diff --git a/slirp/src/util.c b/slirp/src/util.c
-index e596087..e3b6257 100644
---- a/slirp/src/util.c
-+++ b/slirp/src/util.c
-@@ -364,3 +364,65 @@ void slirp_pstrcpy(char *buf, int buf_size, const char *str)
- }
- *q = '\0';
- }
-+
-+static int slirp_vsnprintf(char *str, size_t size,
-+ const char *format, va_list args)
-+{
-+ int rv = vsnprintf(str, size, format, args);
-+
-+ if (rv < 0) {
-+ g_error("vsnprintf() failed: %s", g_strerror(errno));
-+ }
-+
-+ return rv;
-+}
-+
-+/*
-+ * A snprintf()-like function that:
-+ * - returns the number of bytes written (excluding optional \0-ending)
-+ * - dies on error
-+ * - warn on truncation
-+ */
-+int slirp_fmt(char *str, size_t size, const char *format, ...)
-+{
-+ va_list args;
-+ int rv;
-+
-+ va_start(args, format);
-+ rv = slirp_vsnprintf(str, size, format, args);
-+ va_end(args);
-+
-+ if (rv > size) {
-+ g_critical("vsnprintf() truncation");
-+ }
-+
-+ return MIN(rv, size);
-+}
-+
-+/*
-+ * A snprintf()-like function that:
-+ * - always \0-end (unless size == 0)
-+ * - returns the number of bytes actually written, including \0 ending
-+ * - dies on error
-+ * - warn on truncation
-+ */
-+int slirp_fmt0(char *str, size_t size, const char *format, ...)
-+{
-+ va_list args;
-+ int rv;
-+
-+ va_start(args, format);
-+ rv = slirp_vsnprintf(str, size, format, args);
-+ va_end(args);
-+
-+ if (rv >= size) {
-+ g_critical("vsnprintf() truncation");
-+ if (size > 0)
-+ str[size - 1] = '\0';
-+ rv = size;
-+ } else {
-+ rv += 1; /* include \0 */
-+ }
-+
-+ return rv;
-+}
-diff --git a/slirp/src/util.h b/slirp/src/util.h
-index 3c6223c..0558dfc 100644
---- a/slirp/src/util.h
-+++ b/slirp/src/util.h
-@@ -177,4 +177,7 @@ static inline int slirp_socket_set_fast_reuse(int fd)
-
- void slirp_pstrcpy(char *buf, int buf_size, const char *str);
-
-+int slirp_fmt(char *str, size_t size, const char *format, ...);
-+int slirp_fmt0(char *str, size_t size, const char *format, ...);
-+
- #endif
---
-2.20.1
-
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: =?UTF-8?q?Marc-Andr=C3=A9=20Lureau?= <marcandre.lureau@redhat.com>
-Date: Mon, 27 Jan 2020 10:24:14 +0100
-Subject: [PATCH 2/2] tcp_emu: fix unsafe snprintf() usages
-MIME-Version: 1.0
-Content-Type: text/plain; charset=UTF-8
-Content-Transfer-Encoding: 8bit
-
-Various calls to snprintf() assume that snprintf() returns "only" the
-number of bytes written (excluding terminating NUL).
-
-https://pubs.opengroup.org/onlinepubs/9699919799/functions/snprintf.html#tag_16_159_04
-
-"Upon successful completion, the snprintf() function shall return the
-number of bytes that would be written to s had n been sufficiently
-large excluding the terminating null byte."
-
-Before patch ce131029, if there isn't enough room in "m_data" for the
-"DCC ..." message, we overflow "m_data".
-
-After the patch, if there isn't enough room for the same, we don't
-overflow "m_data", but we set "m_len" out-of-bounds. The next time an
-access is bounded by "m_len", we'll have a buffer overflow then.
-
-Use slirp_fmt*() to fix potential OOB memory access.
-
-Reported-by: Laszlo Ersek <lersek@redhat.com>
-Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
-Reviewed-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
-Message-Id: <20200127092414.169796-7-marcandre.lureau@redhat.com>
-Signed-off-by: Oguz Bektas <o.bektas@proxmox.com>
----
- slirp/src/tcp_subr.c | 44 +++++++++++++++++++++-----------------------
- 1 file changed, 21 insertions(+), 23 deletions(-)
-
-diff --git a/slirp/src/tcp_subr.c b/slirp/src/tcp_subr.c
-index d6dd133..93b3488 100644
---- a/slirp/src/tcp_subr.c
-+++ b/slirp/src/tcp_subr.c
-@@ -655,8 +655,7 @@ int tcp_emu(struct socket *so, struct mbuf *m)
- NTOHS(n1);
- NTOHS(n2);
- m_inc(m, snprintf(NULL, 0, "%d,%d\r\n", n1, n2) + 1);
-- m->m_len = snprintf(m->m_data, M_ROOM(m), "%d,%d\r\n", n1, n2);
-- assert(m->m_len < M_ROOM(m));
-+ m->m_len = slirp_fmt(m->m_data, M_ROOM(m), "%d,%d\r\n", n1, n2);
- } else {
- *eol = '\r';
- }
-@@ -696,9 +695,9 @@ int tcp_emu(struct socket *so, struct mbuf *m)
- n4 = (laddr & 0xff);
-
- m->m_len = bptr - m->m_data; /* Adjust length */
-- m->m_len += snprintf(bptr, m->m_size - m->m_len,
-- "ORT %d,%d,%d,%d,%d,%d\r\n%s", n1, n2, n3, n4,
-- n5, n6, x == 7 ? buff : "");
-+ m->m_len += slirp_fmt(bptr, M_FREEROOM(m),
-+ "ORT %d,%d,%d,%d,%d,%d\r\n%s",
-+ n1, n2, n3, n4, n5, n6, x == 7 ? buff : "");
- return 1;
- } else if ((bptr = (char *)strstr(m->m_data, "27 Entering")) != NULL) {
- /*
-@@ -731,11 +730,9 @@ int tcp_emu(struct socket *so, struct mbuf *m)
- n4 = (laddr & 0xff);
-
- m->m_len = bptr - m->m_data; /* Adjust length */
-- m->m_len +=
-- snprintf(bptr, m->m_size - m->m_len,
-- "27 Entering Passive Mode (%d,%d,%d,%d,%d,%d)\r\n%s",
-- n1, n2, n3, n4, n5, n6, x == 7 ? buff : "");
--
-+ m->m_len += slirp_fmt(bptr, M_FREEROOM(m),
-+ "27 Entering Passive Mode (%d,%d,%d,%d,%d,%d)\r\n%s",
-+ n1, n2, n3, n4, n5, n6, x == 7 ? buff : "");
- return 1;
- }
-
-@@ -758,8 +755,8 @@ int tcp_emu(struct socket *so, struct mbuf *m)
- if (m->m_data[m->m_len - 1] == '\0' && lport != 0 &&
- (so = tcp_listen(slirp, INADDR_ANY, 0, so->so_laddr.s_addr,
- htons(lport), SS_FACCEPTONCE)) != NULL)
-- m->m_len =
-- snprintf(m->m_data, m->m_size, "%d", ntohs(so->so_fport)) + 1;
-+ m->m_len = slirp_fmt0(m->m_data, M_ROOM(m),
-+ "%d", ntohs(so->so_fport));
- return 1;
-
- case EMU_IRC:
-@@ -778,9 +775,10 @@ int tcp_emu(struct socket *so, struct mbuf *m)
- return 1;
- }
- m->m_len = bptr - m->m_data; /* Adjust length */
-- m->m_len += snprintf(bptr, m->m_size, "DCC CHAT chat %lu %u%c\n",
-- (unsigned long)ntohl(so->so_faddr.s_addr),
-- ntohs(so->so_fport), 1);
-+ m->m_len += slirp_fmt(bptr, M_FREEROOM(m),
-+ "DCC CHAT chat %lu %u%c\n",
-+ (unsigned long)ntohl(so->so_faddr.s_addr),
-+ ntohs(so->so_fport), 1);
- } else if (sscanf(bptr, "DCC SEND %256s %u %u %u", buff, &laddr, &lport,
- &n1) == 4) {
- if ((so = tcp_listen(slirp, INADDR_ANY, 0, htonl(laddr),
-@@ -788,10 +786,10 @@ int tcp_emu(struct socket *so, struct mbuf *m)
- return 1;
- }
- m->m_len = bptr - m->m_data; /* Adjust length */
-- m->m_len +=
-- snprintf(bptr, m->m_size, "DCC SEND %s %lu %u %u%c\n", buff,
-- (unsigned long)ntohl(so->so_faddr.s_addr),
-- ntohs(so->so_fport), n1, 1);
-+ m->m_len += slirp_fmt(bptr, M_FREEROOM(m),
-+ "DCC SEND %s %lu %u %u%c\n", buff,
-+ (unsigned long)ntohl(so->so_faddr.s_addr),
-+ ntohs(so->so_fport), n1, 1);
- } else if (sscanf(bptr, "DCC MOVE %256s %u %u %u", buff, &laddr, &lport,
- &n1) == 4) {
- if ((so = tcp_listen(slirp, INADDR_ANY, 0, htonl(laddr),
-@@ -799,10 +797,10 @@ int tcp_emu(struct socket *so, struct mbuf *m)
- return 1;
- }
- m->m_len = bptr - m->m_data; /* Adjust length */
-- m->m_len +=
-- snprintf(bptr, m->m_size, "DCC MOVE %s %lu %u %u%c\n", buff,
-- (unsigned long)ntohl(so->so_faddr.s_addr),
-- ntohs(so->so_fport), n1, 1);
-+ m->m_len += slirp_fmt(bptr, M_FREEROOM(m),
-+ "DCC MOVE %s %lu %u %u%c\n", buff,
-+ (unsigned long)ntohl(so->so_faddr.s_addr),
-+ ntohs(so->so_fport), n1, 1);
- }
- return 1;
-
---
-2.20.1
-
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Wed, 29 Nov 2017 11:11:46 +0100
-Subject: [PATCH] PVE: [Config] block/file: change locking default to off
+Date: Tue, 10 Mar 2020 12:54:58 +0100
+Subject: [PATCH 01/32] PVE: [Config] block/file: change locking default to off
'auto' only checks whether the system generally supports OFD
locks but not whether the storage the file resides on
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/block/file-posix.c b/block/file-posix.c
-index 2184aa980c..71aa45ce5d 100644
+index 1b805bd938..44b49265ae 100644
--- a/block/file-posix.c
+++ b/block/file-posix.c
-@@ -444,7 +444,7 @@ static QemuOptsList raw_runtime_opts = {
+@@ -449,7 +449,7 @@ static QemuOptsList raw_runtime_opts = {
{
.name = "locking",
.type = QEMU_OPT_STRING,
},
{
.name = "pr-manager",
-@@ -533,7 +533,7 @@ static int raw_open_common(BlockDriverState *bs, QDict *options,
+@@ -538,7 +538,7 @@ static int raw_open_common(BlockDriverState *bs, QDict *options,
s->use_lock = false;
break;
case ON_OFF_AUTO_AUTO:
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Wed, 9 Dec 2015 14:16:49 +0100
-Subject: [PATCH] PVE: [Config] Adjust network script path to /etc/kvm/
+Date: Tue, 10 Mar 2020 12:54:59 +0100
+Subject: [PATCH 02/32] PVE: [Config] Adjust network script path to /etc/kvm/
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
---
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/include/net/net.h b/include/net/net.h
-index acf0451fc4..4a64633577 100644
+index e175ba9677..5b9f099d21 100644
--- a/include/net/net.h
+++ b/include/net/net.h
-@@ -209,8 +209,9 @@ void qmp_netdev_add(QDict *qdict, QObject **ret, Error **errp);
+@@ -208,8 +208,9 @@ void qmp_netdev_add(QDict *qdict, QObject **ret, Error **errp);
int net_hub_id_for_client(NetClientState *nc, int *id);
NetClientState *net_hub_port_find(int hub_id);
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Wed, 9 Dec 2015 14:30:21 +0100
-Subject: [PATCH] PVE: [Config] set the CPU model to kvm64/32 instead of
+Date: Tue, 10 Mar 2020 12:55:00 +0100
+Subject: [PATCH 03/32] PVE: [Config] set the CPU model to kvm64/32 instead of
qemu64/32
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/target/i386/cpu.h b/target/i386/cpu.h
-index 8b3dc5533e..1fea162e02 100644
+index cde2a16b94..3e73104bf9 100644
--- a/target/i386/cpu.h
+++ b/target/i386/cpu.h
-@@ -1725,9 +1725,9 @@ uint64_t cpu_get_tsc(CPUX86State *env);
+@@ -1940,9 +1940,9 @@ uint64_t cpu_get_tsc(CPUX86State *env);
#define CPU_RESOLVING_TYPE TYPE_X86_CPU
#ifdef TARGET_X86_64
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Wed, 9 Dec 2015 14:33:34 +0100
-Subject: [PATCH] PVE: [Config] ui/spice: default to pve certificates
+Date: Tue, 10 Mar 2020 12:55:01 +0100
+Subject: [PATCH 04/32] PVE: [Config] ui/spice: default to pve certificates
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
---
1 file changed, 9 insertions(+), 6 deletions(-)
diff --git a/ui/spice-core.c b/ui/spice-core.c
-index 2ffc3335f0..c95bbd6c77 100644
+index ecc2ec2c55..ca04965ead 100644
--- a/ui/spice-core.c
+++ b/ui/spice-core.c
@@ -668,32 +668,35 @@ void qemu_spice_init(void)
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Alexandre Derumier <aderumier@odiso.com>
-Date: Tue, 29 Sep 2015 15:37:44 +0200
-Subject: [PATCH] PVE: [Config] smm_available = false
+Date: Tue, 10 Mar 2020 12:55:02 +0100
+Subject: [PATCH 05/32] PVE: [Config] smm_available = false
Signed-off-by: Alexandre Derumier <aderumier@odiso.com>
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/hw/i386/pc.c b/hw/i386/pc.c
-index d011733ff7..e20e189a5f 100644
+index ac08e63604..4bd9ab52a0 100644
--- a/hw/i386/pc.c
+++ b/hw/i386/pc.c
-@@ -2723,7 +2723,7 @@ bool pc_machine_is_smm_enabled(PCMachineState *pcms)
+@@ -2040,7 +2040,7 @@ bool pc_machine_is_smm_enabled(PCMachineState *pcms)
if (tcg_enabled() || qtest_enabled()) {
smm_available = true;
} else if (kvm_enabled()) {
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Mon, 24 Oct 2016 09:32:36 +0200
-Subject: [PATCH] PVE: [Config] glusterfs: no default logfile if daemonized
+Date: Tue, 10 Mar 2020 12:55:03 +0100
+Subject: [PATCH 06/32] PVE: [Config] glusterfs: no default logfile if
+ daemonized
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
---
1 file changed, 11 insertions(+), 4 deletions(-)
diff --git a/block/gluster.c b/block/gluster.c
-index f64dc5b01e..061cab48c0 100644
+index 4fa4a77a47..bfb57ba098 100644
--- a/block/gluster.c
+++ b/block/gluster.c
@@ -42,7 +42,7 @@
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Fri, 31 Aug 2018 09:32:55 +0200
-Subject: [PATCH] PVE: [Config] rbd: block: rbd: disable
+Date: Tue, 10 Mar 2020 12:55:04 +0100
+Subject: [PATCH 07/32] PVE: [Config] rbd: block: rbd: disable
rbd_cache_writethrough_until_flush
Either the cache mode asks for a cache or not. There's no
1 file changed, 2 insertions(+)
diff --git a/block/rbd.c b/block/rbd.c
-index 59757b3120..d00c9d2d12 100644
+index 027cbcc695..3ac7ff7bd5 100644
--- a/block/rbd.c
+++ b/block/rbd.c
-@@ -636,6 +636,8 @@ static int qemu_rbd_connect(rados_t *cluster, rados_ioctx_t *io_ctx,
+@@ -637,6 +637,8 @@ static int qemu_rbd_connect(rados_t *cluster, rados_ioctx_t *io_ctx,
rados_conf_set(*cluster, "rbd_cache", "false");
}
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Wed, 9 Dec 2015 16:34:41 +0100
-Subject: [PATCH] PVE: [Up] qmp: add get_link_status
+Date: Tue, 10 Mar 2020 12:55:05 +0100
+Subject: [PATCH 08/32] PVE: [Up] qmp: add get_link_status
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
---
3 files changed, 43 insertions(+)
diff --git a/net/net.c b/net/net.c
-index 7d4098254f..c6d5e66bfc 100644
+index 84aa6d8d00..f548202ec6 100644
--- a/net/net.c
+++ b/net/net.c
-@@ -1347,6 +1347,33 @@ void hmp_info_network(Monitor *mon, const QDict *qdict)
+@@ -1349,6 +1349,33 @@ void hmp_info_network(Monitor *mon, const QDict *qdict)
}
}
{
NetClientState *nc;
diff --git a/qapi/net.json b/qapi/net.json
-index 728990f4fb..d53c66320b 100644
+index 335295be50..7f3ea194c8 100644
--- a/qapi/net.json
+++ b/qapi/net.json
@@ -34,6 +34,21 @@
# @netdev_add:
#
diff --git a/qapi/qapi-schema.json b/qapi/qapi-schema.json
-index 38af54d6b3..d6a4177935 100644
+index 9751b11f8f..a449f158e1 100644
--- a/qapi/qapi-schema.json
+++ b/qapi/qapi-schema.json
@@ -61,6 +61,7 @@
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Wed, 30 Nov 2016 10:27:47 +0100
-Subject: [PATCH] PVE: [Up] glusterfs: allow partial reads
+Date: Tue, 10 Mar 2020 12:55:06 +0100
+Subject: [PATCH 09/32] PVE: [Up] glusterfs: allow partial reads
This should deal with qemu bug #1644754 until upstream
decides which way to go. The general direction seems to be
1 file changed, 9 insertions(+), 1 deletion(-)
diff --git a/block/gluster.c b/block/gluster.c
-index 061cab48c0..0ed1ec5856 100644
+index bfb57ba098..81fff09c6c 100644
--- a/block/gluster.c
+++ b/block/gluster.c
@@ -57,6 +57,7 @@ typedef struct GlusterAIOCB {
ret = glfs_preadv_async(s->fd, qiov->iov, qiov->niov, offset, 0,
gluster_finish_aiocb, &acb);
}
-@@ -1279,6 +1285,7 @@ static coroutine_fn int qemu_gluster_co_flush_to_disk(BlockDriverState *bs)
+@@ -1280,6 +1286,7 @@ static coroutine_fn int qemu_gluster_co_flush_to_disk(BlockDriverState *bs)
acb.ret = 0;
acb.coroutine = qemu_coroutine_self();
acb.aio_context = bdrv_get_aio_context(bs);
ret = glfs_fsync_async(s->fd, gluster_finish_aiocb, &acb);
if (ret < 0) {
-@@ -1325,6 +1332,7 @@ static coroutine_fn int qemu_gluster_co_pdiscard(BlockDriverState *bs,
+@@ -1326,6 +1333,7 @@ static coroutine_fn int qemu_gluster_co_pdiscard(BlockDriverState *bs,
acb.ret = 0;
acb.coroutine = qemu_coroutine_self();
acb.aio_context = bdrv_get_aio_context(bs);
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Wed, 9 Dec 2015 14:18:46 +0100
-Subject: [PATCH] PVE: [Up] qemu-img: return success on info without snapshots
+Date: Tue, 10 Mar 2020 12:55:07 +0100
+Subject: [PATCH 10/32] PVE: [Up] qemu-img: return success on info without
+ snapshots
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
---
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/qemu-img.c b/qemu-img.c
-index 79983772de..c64f260876 100644
+index 95a24b9762..12211bed76 100644
--- a/qemu-img.c
+++ b/qemu-img.c
-@@ -2773,7 +2773,8 @@ static int img_info(int argc, char **argv)
+@@ -2791,7 +2791,8 @@ static int img_info(int argc, char **argv)
list = collect_image_info_list(image_opts, filename, fmt, chain,
force_share);
if (!list) {
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Fri, 23 Jun 2017 12:01:43 +0200
-Subject: [PATCH] PVE: [Up] qemu-img dd: add osize and read from/to
+Date: Tue, 10 Mar 2020 12:55:08 +0100
+Subject: [PATCH 11/32] PVE: [Up] qemu-img dd: add osize and read from/to
stdin/stdout
Neither convert nor dd were previously able to write to or
DEF("info", img_info,
diff --git a/qemu-img.c b/qemu-img.c
-index c64f260876..8129677d7a 100644
+index 12211bed76..d2516968c6 100644
--- a/qemu-img.c
+++ b/qemu-img.c
-@@ -4413,10 +4413,12 @@ out:
+@@ -4405,10 +4405,12 @@ out:
#define C_IF 04
#define C_OF 010
#define C_SKIP 020
};
struct DdIo {
-@@ -4495,6 +4497,20 @@ static int img_dd_skip(const char *arg,
+@@ -4487,6 +4489,20 @@ static int img_dd_skip(const char *arg,
return 0;
}
static int img_dd(int argc, char **argv)
{
int ret = 0;
-@@ -4535,6 +4551,7 @@ static int img_dd(int argc, char **argv)
+@@ -4527,6 +4543,7 @@ static int img_dd(int argc, char **argv)
{ "if", img_dd_if, C_IF },
{ "of", img_dd_of, C_OF },
{ "skip", img_dd_skip, C_SKIP },
{ NULL, NULL, 0 }
};
const struct option long_options[] = {
-@@ -4613,8 +4630,13 @@ static int img_dd(int argc, char **argv)
+@@ -4605,8 +4622,13 @@ static int img_dd(int argc, char **argv)
arg = NULL;
}
ret = -1;
goto out;
}
-@@ -4626,85 +4648,101 @@ static int img_dd(int argc, char **argv)
+@@ -4618,85 +4640,101 @@ static int img_dd(int argc, char **argv)
goto out;
}
}
if (dd.flags & C_SKIP && (in.offset > INT64_MAX / in.bsz ||
-@@ -4722,11 +4760,17 @@ static int img_dd(int argc, char **argv)
+@@ -4714,11 +4752,17 @@ static int img_dd(int argc, char **argv)
for (out_pos = 0; in_pos < size; block_count++) {
int in_ret, out_ret;
}
if (in_ret < 0) {
error_report("error while reading from input image file: %s",
-@@ -4736,9 +4780,13 @@ static int img_dd(int argc, char **argv)
+@@ -4728,9 +4772,13 @@ static int img_dd(int argc, char **argv)
}
in_pos += in_ret;
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Fri, 23 Feb 2018 08:43:18 +0100
-Subject: [PATCH] PVE: [Up] qemu-img dd: add isize parameter
+Date: Tue, 10 Mar 2020 12:55:09 +0100
+Subject: [PATCH 12/32] PVE: [Up] qemu-img dd: add isize parameter
for writing small images from stdin to bigger ones
1 file changed, 26 insertions(+), 3 deletions(-)
diff --git a/qemu-img.c b/qemu-img.c
-index 8129677d7a..1b7f211368 100644
+index d2516968c6..8da1ea3951 100644
--- a/qemu-img.c
+++ b/qemu-img.c
-@@ -4414,11 +4414,13 @@ out:
+@@ -4406,11 +4406,13 @@ out:
#define C_OF 010
#define C_SKIP 020
#define C_OSIZE 040
};
struct DdIo {
-@@ -4511,6 +4513,20 @@ static int img_dd_osize(const char *arg,
+@@ -4503,6 +4505,20 @@ static int img_dd_osize(const char *arg,
return 0;
}
static int img_dd(int argc, char **argv)
{
int ret = 0;
-@@ -4525,12 +4541,14 @@ static int img_dd(int argc, char **argv)
+@@ -4517,12 +4533,14 @@ static int img_dd(int argc, char **argv)
int c, i;
const char *out_fmt = "raw";
const char *fmt = NULL;
};
struct DdIo in = {
.bsz = 512, /* Block size is by default 512 bytes */
-@@ -4552,6 +4570,7 @@ static int img_dd(int argc, char **argv)
+@@ -4544,6 +4562,7 @@ static int img_dd(int argc, char **argv)
{ "of", img_dd_of, C_OF },
{ "skip", img_dd_skip, C_SKIP },
{ "osize", img_dd_osize, C_OSIZE },
{ NULL, NULL, 0 }
};
const struct option long_options[] = {
-@@ -4758,14 +4777,18 @@ static int img_dd(int argc, char **argv)
+@@ -4750,14 +4769,18 @@ static int img_dd(int argc, char **argv)
in.buf = g_new(uint8_t, in.bsz);
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Alexandre Derumier <aderumier@odiso.com>
-Date: Wed, 21 Mar 2018 08:51:34 +0100
-Subject: [PATCH] PVE: [Up] qemu-img dd : add -n skip_create
+Date: Tue, 10 Mar 2020 12:55:10 +0100
+Subject: [PATCH 13/32] PVE: [Up] qemu-img dd : add -n skip_create
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
---
1 file changed, 14 insertions(+), 9 deletions(-)
diff --git a/qemu-img.c b/qemu-img.c
-index 1b7f211368..e14d2370c4 100644
+index 8da1ea3951..ea3edb4f04 100644
--- a/qemu-img.c
+++ b/qemu-img.c
-@@ -4543,7 +4543,7 @@ static int img_dd(int argc, char **argv)
+@@ -4535,7 +4535,7 @@ static int img_dd(int argc, char **argv)
const char *fmt = NULL;
int64_t size = 0, readsize = 0;
int64_t block_count = 0, out_pos, in_pos;
struct DdInfo dd = {
.flags = 0,
.count = 0,
-@@ -4581,7 +4581,7 @@ static int img_dd(int argc, char **argv)
+@@ -4573,7 +4573,7 @@ static int img_dd(int argc, char **argv)
{ 0, 0, 0, 0 }
};
if (c == EOF) {
break;
}
-@@ -4601,6 +4601,9 @@ static int img_dd(int argc, char **argv)
+@@ -4593,6 +4593,9 @@ static int img_dd(int argc, char **argv)
case 'h':
help();
break;
case 'U':
force_share = true;
break;
-@@ -4741,13 +4744,15 @@ static int img_dd(int argc, char **argv)
+@@ -4733,13 +4736,15 @@ static int img_dd(int argc, char **argv)
size - in.bsz * in.offset, &error_abort);
}
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Wed, 9 Dec 2015 14:27:49 +0100
-Subject: [PATCH] PVE: virtio-balloon: improve query-balloon
+Date: Tue, 10 Mar 2020 12:55:11 +0100
+Subject: [PATCH 14/32] PVE: virtio-balloon: improve query-balloon
Actually provide memory information via the query-balloon
command.
3 files changed, 81 insertions(+), 4 deletions(-)
diff --git a/hw/virtio/virtio-balloon.c b/hw/virtio/virtio-balloon.c
-index 25de154307..7c09716035 100644
+index 40b04f5180..76e907e628 100644
--- a/hw/virtio/virtio-balloon.c
+++ b/hw/virtio/virtio-balloon.c
-@@ -712,8 +712,37 @@ static uint64_t virtio_balloon_get_features(VirtIODevice *vdev, uint64_t f,
+@@ -713,8 +713,37 @@ static uint64_t virtio_balloon_get_features(VirtIODevice *vdev, uint64_t f,
static void virtio_balloon_stat(void *opaque, BalloonInfo *info)
{
VirtIOBalloon *dev = opaque;
static void virtio_balloon_to_target(void *opaque, ram_addr_t target)
diff --git a/monitor/hmp-cmds.c b/monitor/hmp-cmds.c
-index 5ca3ebe942..1b32c59329 100644
+index b2551c16d1..2e725ed818 100644
--- a/monitor/hmp-cmds.c
+++ b/monitor/hmp-cmds.c
-@@ -870,7 +870,35 @@ void hmp_info_balloon(Monitor *mon, const QDict *qdict)
+@@ -854,7 +854,35 @@ void hmp_info_balloon(Monitor *mon, const QDict *qdict)
return;
}
qapi_free_BalloonInfo(info);
}
diff --git a/qapi/misc.json b/qapi/misc.json
-index a7fba7230c..2445c950cc 100644
+index 33b94e3589..ed65ed27e3 100644
--- a/qapi/misc.json
+++ b/qapi/misc.json
@@ -408,10 +408,30 @@
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Wed, 9 Dec 2015 14:31:18 +0100
-Subject: [PATCH] PVE: qapi: modify query machines
+From: Dietmar Maurer <dietmar@proxmox.com>
+Date: Tue, 10 Mar 2020 12:55:12 +0100
+Subject: [PATCH 15/32] PVE: qapi: modify query machines
provide '*is-current' in MachineInfo struct
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+Signed-off-by: Dietmar Maurer <dietmar@proxmox.com>
---
- hw/core/machine-qmp-cmds.c | 5 +++++
+ hw/core/machine-qmp-cmds.c | 6 ++++++
qapi/machine.json | 4 +++-
- 2 files changed, 8 insertions(+), 1 deletion(-)
+ 2 files changed, 9 insertions(+), 1 deletion(-)
diff --git a/hw/core/machine-qmp-cmds.c b/hw/core/machine-qmp-cmds.c
-index 5bd95b8ab0..fd68f9baf8 100644
+index eed5aeb2f7..1953633e82 100644
--- a/hw/core/machine-qmp-cmds.c
+++ b/hw/core/machine-qmp-cmds.c
-@@ -229,6 +229,11 @@ MachineInfoList *qmp_query_machines(Error **errp)
+@@ -230,6 +230,12 @@ MachineInfoList *qmp_query_machines(Error **errp)
+ info->hotpluggable_cpus = mc->has_hotpluggable_cpus;
info->numa_mem_supported = mc->numa_mem_supported;
info->deprecated = !!mc->deprecation_reason;
-
++
+ if (strcmp(mc->name, MACHINE_GET_CLASS(current_machine)->name) == 0) {
+ info->has_is_current = true;
+ info->is_current = true;
+ }
+
- entry = g_malloc0(sizeof(*entry));
- entry->value = info;
- entry->next = mach_list;
+ if (mc->default_cpu_type) {
+ info->default_cpu_type = g_strdup(mc->default_cpu_type);
+ info->has_default_cpu_type = true;
diff --git a/qapi/machine.json b/qapi/machine.json
-index 6db8a7e2ec..7b82c5f7f5 100644
+index ca26779f1a..cbdb6f6d66 100644
--- a/qapi/machine.json
+++ b/qapi/machine.json
-@@ -313,6 +313,8 @@
+@@ -336,6 +336,8 @@
#
# @is-default: whether the machine is default
#
# @cpu-max: maximum number of CPUs supported by the machine type
# (since 1.5.0)
#
-@@ -329,7 +331,7 @@
+@@ -355,7 +357,7 @@
##
{ 'struct': 'MachineInfo',
'data': { 'name': 'str', '*alias': 'str',
- '*is-default': 'bool', 'cpu-max': 'int',
+ '*is-default': 'bool', '*is-current': 'bool', 'cpu-max': 'int',
'hotpluggable-cpus': 'bool', 'numa-mem-supported': 'bool',
- 'deprecated': 'bool' } }
+ 'deprecated': 'bool', '*default-cpu-type': 'str' } }
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Wed, 9 Dec 2015 14:32:11 +0100
-Subject: [PATCH] PVE: qapi: modify spice query
+Date: Tue, 10 Mar 2020 12:55:13 +0100
+Subject: [PATCH 16/32] PVE: qapi: modify spice query
Provide the last ticket in the SpiceInfo struct optionally.
2 files changed, 8 insertions(+)
diff --git a/qapi/ui.json b/qapi/ui.json
-index 59e412139a..bcd781a1b9 100644
+index e04525d8b4..6127990e23 100644
--- a/qapi/ui.json
+++ b/qapi/ui.json
@@ -211,11 +211,14 @@
'if': 'defined(CONFIG_SPICE)' }
diff --git a/ui/spice-core.c b/ui/spice-core.c
-index c95bbd6c77..ccba92a6ed 100644
+index ca04965ead..243466c13d 100644
--- a/ui/spice-core.c
+++ b/ui/spice-core.c
@@ -539,6 +539,11 @@ SpiceInfo *qmp_query_spice(Error **errp)
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Wed, 9 Dec 2015 16:04:32 +0100
-Subject: [PATCH] PVE: internal snapshot async
+From: Dietmar Maurer <dietmar@proxmox.com>
+Date: Tue, 10 Mar 2020 13:49:27 +0100
+Subject: [PATCH 17/32] PVE: internal snapshot async
+
+Truncate at 1024 boundary (Fabian Ebner will send a patch for stable)
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+Signed-off-by: Dietmar Maurer <dietmar@proxmox.com>
---
Makefile.objs | 1 +
hmp-commands-info.hx | 13 +
qapi/migration.json | 34 +++
qapi/misc.json | 32 +++
qemu-options.hx | 13 +
- savevm-async.c | 460 +++++++++++++++++++++++++++++++++++
+ savevm-async.c | 464 +++++++++++++++++++++++++++++++++++
vl.c | 10 +
- 11 files changed, 658 insertions(+)
+ 11 files changed, 662 insertions(+)
create mode 100644 savevm-async.c
diff --git a/Makefile.objs b/Makefile.objs
-index 6a143dcd57..21dd93b58c 100644
+index 11ba1a36bd..f97b40f232 100644
--- a/Makefile.objs
+++ b/Makefile.objs
@@ -48,6 +48,7 @@ common-obj-y += bootdevice.o iothread.o
common-obj-y += qdev-monitor.o device-hotplug.o
common-obj-$(CONFIG_WIN32) += os-win32.o
diff --git a/hmp-commands-info.hx b/hmp-commands-info.hx
-index c59444c461..444bd8e43d 100644
+index 257ee7d7a3..139e673bea 100644
--- a/hmp-commands-info.hx
+++ b/hmp-commands-info.hx
@@ -608,6 +608,19 @@ STEXI
{
diff --git a/hmp-commands.hx b/hmp-commands.hx
-index bfa5681dd2..e075d413c0 100644
+index cfcc044ce4..104288322d 100644
--- a/hmp-commands.hx
+++ b/hmp-commands.hx
-@@ -1944,3 +1944,35 @@ ETEXI
+@@ -1945,3 +1945,35 @@ ETEXI
STEXI
@end table
ETEXI
void hmp_screendump(Monitor *mon, const QDict *qdict);
void hmp_nbd_server_start(Monitor *mon, const QDict *qdict);
diff --git a/monitor/hmp-cmds.c b/monitor/hmp-cmds.c
-index 1b32c59329..39a8020367 100644
+index 2e725ed818..90aa34be25 100644
--- a/monitor/hmp-cmds.c
+++ b/monitor/hmp-cmds.c
-@@ -2640,6 +2640,63 @@ void hmp_info_memory_devices(Monitor *mon, const QDict *qdict)
+@@ -2607,6 +2607,63 @@ void hmp_info_memory_devices(Monitor *mon, const QDict *qdict)
hmp_handle_error(mon, &err);
}
{
IOThreadInfoList *info_list = qmp_query_iothreads(NULL);
diff --git a/qapi/migration.json b/qapi/migration.json
-index 9cfbaf8c6c..e206355d56 100644
+index b7348d0c8b..2792409977 100644
--- a/qapi/migration.json
+++ b/qapi/migration.json
-@@ -219,6 +219,40 @@
+@@ -222,6 +222,40 @@
'*compression': 'CompressionStats',
'*socket-address': ['SocketAddress'] } }
# @query-migrate:
#
diff --git a/qapi/misc.json b/qapi/misc.json
-index 2445c950cc..31029e3132 100644
+index ed65ed27e3..4c4618a574 100644
--- a/qapi/misc.json
+++ b/qapi/misc.json
-@@ -1384,6 +1384,38 @@
+@@ -1368,6 +1368,38 @@
##
- { 'command': 'query-target', 'returns': 'TargetInfo' }
+ { 'command': 'query-fdsets', 'returns': ['FdsetInfo'] }
+##
+# @savevm-start:
# @AcpiTableOptions:
#
diff --git a/qemu-options.hx b/qemu-options.hx
-index 9621e934c0..34994daafd 100644
+index 65c9473b73..4cb2681bfc 100644
--- a/qemu-options.hx
+++ b/qemu-options.hx
-@@ -3731,6 +3731,19 @@ STEXI
+@@ -3818,6 +3818,19 @@ STEXI
Start right away with a saved state (@code{loadvm} in monitor)
ETEXI
"-daemonize daemonize QEMU after initializing\n", QEMU_ARCH_ALL)
diff --git a/savevm-async.c b/savevm-async.c
new file mode 100644
-index 0000000000..2149010bb8
+index 0000000000..54ceeae26c
--- /dev/null
+++ b/savevm-async.c
-@@ -0,0 +1,460 @@
+@@ -0,0 +1,464 @@
+#include "qemu/osdep.h"
+#include "migration/migration.h"
+#include "migration/savevm.h"
+#include "migration/ram.h"
+#include "migration/qemu-file.h"
+#include "sysemu/sysemu.h"
++#include "sysemu/runstate.h"
+#include "block/block.h"
+#include "sysemu/block-backend.h"
+#include "qapi/error.h"
+#include "qapi/qapi-commands-misc.h"
+#include "qapi/qapi-commands-block.h"
+#include "qemu/cutils.h"
++#include "qemu/main-loop.h"
++#include "qemu/rcu.h"
+
+/* #define DEBUG_SAVEVM_STATE */
+
+
+ if (snap_state.target) {
+ /* try to truncate, but ignore errors (will fail on block devices).
-+ * note: bdrv_read() need whole blocks, so we round up
++ * note1: bdrv_read() need whole blocks, so we need to round up
++ * note2: PVE requires 1024 (BDRV_SECTOR_SIZE*2) alignment
+ */
-+ size_t size = (snap_state.bs_pos + BDRV_SECTOR_SIZE) & BDRV_SECTOR_MASK;
-+ blk_truncate(snap_state.target, size, PREALLOC_MODE_OFF, NULL);
++ size_t size = QEMU_ALIGN_UP(snap_state.bs_pos, BDRV_SECTOR_SIZE*2);
++ blk_truncate(snap_state.target, size, false, PREALLOC_MODE_OFF, NULL);
+ blk_op_unblock_all(snap_state.target, snap_state.blocker);
+ error_free(snap_state.blocker);
+ snap_state.blocker = NULL;
+ snap_state.state = SAVE_STATE_ERROR;
+}
+
-+static int block_state_close(void *opaque)
++static int block_state_close(void *opaque, Error **errp)
+{
+ snap_state.file = NULL;
+ return blk_flush(snap_state.target);
+}
+
+static ssize_t block_state_writev_buffer(void *opaque, struct iovec *iov,
-+ int iovcnt, int64_t pos)
++ int iovcnt, int64_t pos, Error **errp)
+{
+ QEMUIOVector qiov;
+ BlkRwCo rwco;
+}
+
+static ssize_t loadstate_get_buffer(void *opaque, uint8_t *buf, int64_t pos,
-+ size_t size)
++ size_t size, Error **errp)
+{
+ BlockBackend *be = opaque;
+ int64_t maxlen = blk_getlength(be);
+ return ret;
+}
diff --git a/vl.c b/vl.c
-index b426b32134..1c5536e5bb 100644
+index 6a65a64bfd..1616f55a38 100644
--- a/vl.c
+++ b/vl.c
-@@ -2869,6 +2869,7 @@ int main(int argc, char **argv, char **envp)
+@@ -2840,6 +2840,7 @@ int main(int argc, char **argv, char **envp)
int optind;
const char *optarg;
const char *loadvm = NULL;
MachineClass *machine_class;
const char *cpu_option;
const char *vga_model = NULL;
-@@ -3445,6 +3446,9 @@ int main(int argc, char **argv, char **envp)
+@@ -3430,6 +3431,9 @@ int main(int argc, char **argv, char **envp)
case QEMU_OPTION_loadvm:
loadvm = optarg;
break;
case QEMU_OPTION_full_screen:
dpy.has_full_screen = true;
dpy.full_screen = true;
-@@ -4444,6 +4448,12 @@ int main(int argc, char **argv, char **envp)
+@@ -4442,6 +4446,12 @@ int main(int argc, char **argv, char **envp)
autostart = 0;
exit(1);
}
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Thu, 17 Mar 2016 11:33:37 +0100
-Subject: [PATCH] PVE: block: add the zeroinit block driver filter
+Date: Tue, 10 Mar 2020 12:55:15 +0100
+Subject: [PATCH 18/32] PVE: block: add the zeroinit block driver filter
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
---
create mode 100644 block/zeroinit.c
diff --git a/block/Makefile.objs b/block/Makefile.objs
-index 35f3bca4d9..6022242c3f 100644
+index e394fe0b6c..a10ceabf5b 100644
--- a/block/Makefile.objs
+++ b/block/Makefile.objs
@@ -11,6 +11,7 @@ block-obj-$(CONFIG_QED) += qed.o qed-l2-cache.o qed-table.o qed-cluster.o
block-obj-y += blklogwrites.o
diff --git a/block/zeroinit.c b/block/zeroinit.c
new file mode 100644
-index 0000000000..e78511d36c
+index 0000000000..b74a78ece6
--- /dev/null
+++ b/block/zeroinit.c
@@ -0,0 +1,204 @@
+}
+
+static int zeroinit_co_truncate(BlockDriverState *bs, int64_t offset,
-+ PreallocMode prealloc, Error **errp)
++ _Bool exact, PreallocMode prealloc, Error **errp)
+{
-+ return bdrv_co_truncate(bs->file, offset, prealloc, errp);
++ return bdrv_co_truncate(bs->file, offset, exact, prealloc, errp);
+}
+
+static int zeroinit_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Wed, 9 Dec 2015 15:04:57 +0100
-Subject: [PATCH] PVE: backup: modify job api
+Date: Tue, 10 Mar 2020 12:55:16 +0100
+Subject: [PATCH 19/32] PVE: backup: modify job api
Introduce a pause_count parameter to start a backup in
paused mode. This way backups of multiple drives can be
Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
---
- block/backup.c | 2 ++
+ block/backup.c | 3 +++
block/replication.c | 2 +-
- blockdev.c | 4 ++--
+ blockdev.c | 3 ++-
include/block/block_int.h | 1 +
job.c | 2 +-
- 5 files changed, 7 insertions(+), 4 deletions(-)
+ 5 files changed, 8 insertions(+), 3 deletions(-)
diff --git a/block/backup.c b/block/backup.c
-index 8761f1f9a7..30008fcc34 100644
+index cf62b1a38c..c155081de2 100644
--- a/block/backup.c
+++ b/block/backup.c
-@@ -550,6 +550,7 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
+@@ -347,6 +347,7 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
BlockdevOnError on_target_error,
int creation_flags,
BlockCompletionFunc *cb, void *opaque,
JobTxn *txn, Error **errp)
{
int64_t len;
-@@ -675,6 +676,7 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
+@@ -468,6 +469,8 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
block_job_add_bdrv(&job->common, "target", target, 0, BLK_PERM_ALL,
&error_abort);
- job->len = len;
-+ job->common.job.pause_count += pause_count;
++ job->common.job.pause_count += pause_count;
++
return &job->common;
+ error:
diff --git a/block/replication.c b/block/replication.c
-index 23b2993d74..e70a6cf2bd 100644
+index 99532ce521..ec8de7b427 100644
--- a/block/replication.c
+++ b/block/replication.c
@@ -546,7 +546,7 @@ static void replication_start(ReplicationState *rs, ReplicationMode mode,
- 0, MIRROR_SYNC_MODE_NONE, NULL, false,
+ 0, MIRROR_SYNC_MODE_NONE, NULL, 0, false, NULL,
BLOCKDEV_ON_ERROR_REPORT,
BLOCKDEV_ON_ERROR_REPORT, JOB_INTERNAL,
- backup_job_completed, bs, NULL, &local_err);
error_propagate(errp, local_err);
backup_job_cleanup(bs);
diff --git a/blockdev.c b/blockdev.c
-index 4d141e9a1f..a7c97b1585 100644
+index 8e029e9c01..c7fa663ebf 100644
--- a/blockdev.c
+++ b/blockdev.c
-@@ -3574,7 +3574,7 @@ static BlockJob *do_drive_backup(DriveBackup *backup, JobTxn *txn,
- job = backup_job_create(backup->job_id, bs, target_bs, backup->speed,
- backup->sync, bmap, backup->compress,
- backup->on_source_error, backup->on_target_error,
-- job_flags, NULL, NULL, txn, &local_err);
-+ job_flags, NULL, NULL, 0, txn, &local_err);
- if (local_err != NULL) {
- error_propagate(errp, local_err);
- goto unref;
-@@ -3679,7 +3679,7 @@ BlockJob *do_blockdev_backup(BlockdevBackup *backup, JobTxn *txn,
- job = backup_job_create(backup->job_id, bs, target_bs, backup->speed,
- backup->sync, bmap, backup->compress,
- backup->on_source_error, backup->on_target_error,
-- job_flags, NULL, NULL, txn, &local_err);
-+ job_flags, NULL, NULL, 0, txn, &local_err);
- if (local_err != NULL) {
- error_propagate(errp, local_err);
- }
+@@ -3583,7 +3583,8 @@ static BlockJob *do_backup_common(BackupCommon *backup,
+ backup->filter_node_name,
+ backup->on_source_error,
+ backup->on_target_error,
+- job_flags, NULL, NULL, txn, errp);
++ job_flags, NULL, NULL, 0, txn, errp);
++
+ return job;
+ }
+
diff --git a/include/block/block_int.h b/include/block/block_int.h
-index 05ee6b4866..bb2dddca83 100644
+index dd033d0b37..b0d5eb9485 100644
--- a/include/block/block_int.h
+++ b/include/block/block_int.h
-@@ -1173,6 +1173,7 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
+@@ -1215,6 +1215,7 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
BlockdevOnError on_target_error,
int creation_flags,
BlockCompletionFunc *cb, void *opaque,
void hmp_drive_add_node(Monitor *mon, const char *optstr);
diff --git a/job.c b/job.c
-index 28dd48f8a5..7a21e83780 100644
+index 04409b40aa..7554f735e3 100644
--- a/job.c
+++ b/job.c
-@@ -898,7 +898,7 @@ void job_start(Job *job)
+@@ -888,7 +888,7 @@ void job_start(Job *job)
job->co = qemu_coroutine_create(job_co_entry, job);
job->pause_count--;
job->busy = true;
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Wolfgang Bumiller <w.bumiller@proxmox.com>
+Date: Tue, 10 Mar 2020 12:55:17 +0100
+Subject: [PATCH 20/32] PVE: Add dummy -id command line parameter
+
+This used to be part of the qemu-side PVE authentication for
+VNC. Now this does nothing.
+
+Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ qemu-options.hx | 3 +++
+ vl.c | 8 ++++++++
+ 2 files changed, 11 insertions(+)
+
+diff --git a/qemu-options.hx b/qemu-options.hx
+index 4cb2681bfc..b84e260fa5 100644
+--- a/qemu-options.hx
++++ b/qemu-options.hx
+@@ -826,6 +826,9 @@ STEXI
+ @table @option
+ ETEXI
+
++DEF("id", HAS_ARG, QEMU_OPTION_id,
++ "-id n set the VMID", QEMU_ARCH_ALL)
++
+ DEF("fda", HAS_ARG, QEMU_OPTION_fda,
+ "-fda/-fdb file use 'file' as floppy disk 0/1 image\n", QEMU_ARCH_ALL)
+ DEF("fdb", HAS_ARG, QEMU_OPTION_fdb, "", QEMU_ARCH_ALL)
+diff --git a/vl.c b/vl.c
+index 1616f55a38..4df15640c5 100644
+--- a/vl.c
++++ b/vl.c
+@@ -2828,6 +2828,7 @@ static void user_register_global_props(void)
+ int main(int argc, char **argv, char **envp)
+ {
+ int i;
++ long vm_id;
+ int snapshot, linux_boot;
+ const char *initrd_filename;
+ const char *kernel_filename, *kernel_cmdline;
+@@ -3560,6 +3561,13 @@ int main(int argc, char **argv, char **envp)
+ exit(1);
+ }
+ break;
++ case QEMU_OPTION_id:
++ vm_id = strtol(optarg, (char **)&optarg, 10);
++ if (*optarg != 0 || vm_id < 100 || vm_id > INT_MAX) {
++ error_report("invalid -id argument %s", optarg);
++ exit(1);
++ }
++ break;
+ case QEMU_OPTION_vnc:
+ vnc_parse(optarg, &error_fatal);
+ break;
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Wed, 2 Aug 2017 13:51:02 +0200
-Subject: [PATCH] PVE: backup: introduce vma archive format
-
-TODO: Move to a libvma block backend.
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- MAINTAINERS | 6 +
- block/Makefile.objs | 3 +
- block/vma.c | 503 +++++++++++++++++++++++++++++++++++++++
- blockdev.c | 536 ++++++++++++++++++++++++++++++++++++++++++
- configure | 30 +++
- hmp-commands-info.hx | 13 +
- hmp-commands.hx | 31 +++
- include/monitor/hmp.h | 3 +
- monitor/hmp-cmds.c | 63 +++++
- qapi/block-core.json | 109 ++++++++-
- qapi/common.json | 13 +
- qapi/misc.json | 13 -
- 12 files changed, 1309 insertions(+), 14 deletions(-)
- create mode 100644 block/vma.c
-
-diff --git a/MAINTAINERS b/MAINTAINERS
-index d6de200453..de09d099f6 100644
---- a/MAINTAINERS
-+++ b/MAINTAINERS
-@@ -2563,6 +2563,12 @@ L: qemu-block@nongnu.org
- S: Supported
- F: block/vvfat.c
-
-+VMA
-+M: Wolfgang Bumiller <w.bumiller@proxmox.com>.
-+L: pve-devel@proxmox.com
-+S: Supported
-+F: block/vma.c
-+
- Image format fuzzer
- M: Stefan Hajnoczi <stefanha@redhat.com>
- L: qemu-block@nongnu.org
-diff --git a/block/Makefile.objs b/block/Makefile.objs
-index 6022242c3f..86b10d8ea7 100644
---- a/block/Makefile.objs
-+++ b/block/Makefile.objs
-@@ -33,6 +33,7 @@ block-obj-$(CONFIG_RBD) += rbd.o
- block-obj-$(CONFIG_GLUSTERFS) += gluster.o
- block-obj-$(CONFIG_VXHS) += vxhs.o
- block-obj-$(CONFIG_LIBSSH) += ssh.o
-+block-obj-$(CONFIG_VMA) += vma.o
- block-obj-y += accounting.o dirty-bitmap.o
- block-obj-y += write-threshold.o
- block-obj-y += backup.o
-@@ -64,3 +65,5 @@ qcow.o-libs := -lz
- linux-aio.o-libs := -laio
- parallels.o-cflags := $(LIBXML2_CFLAGS)
- parallels.o-libs := $(LIBXML2_LIBS)
-+vma.o-cflags := $(VMA_CFLAGS)
-+vma.o-libs := $(VMA_LIBS)
-diff --git a/block/vma.c b/block/vma.c
-new file mode 100644
-index 0000000000..b911b198dc
---- /dev/null
-+++ b/block/vma.c
-@@ -0,0 +1,503 @@
-+/*
-+ * VMA archive backend for QEMU, container object
-+ *
-+ * Copyright (C) 2017 Proxmox Server Solutions GmbH
-+ *
-+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
-+ * See the COPYING file in the top-level directory.
-+ *
-+ */
-+#include <vma/vma.h>
-+
-+#include "qemu/osdep.h"
-+#include "qemu/uuid.h"
-+#include "qemu/option.h"
-+#include "qemu-common.h"
-+#include "qapi/error.h"
-+#include "qapi/qmp/qerror.h"
-+#include "qapi/qmp/qstring.h"
-+#include "qapi/qmp/qdict.h"
-+#include "qom/object.h"
-+#include "qom/object_interfaces.h"
-+#include "block/block_int.h"
-+
-+/* exported interface */
-+void vma_object_add_config_file(Object *obj, const char *name,
-+ const char *contents, size_t len,
-+ Error **errp);
-+
-+#define TYPE_VMA_OBJECT "vma"
-+#define VMA_OBJECT(obj) \
-+ OBJECT_CHECK(VMAObjectState, (obj), TYPE_VMA_OBJECT)
-+#define VMA_OBJECT_GET_CLASS(obj) \
-+ OBJECT_GET_CLASS(VMAObjectClass, (obj), TYPE_VMA_OBJECT)
-+
-+typedef struct VMAObjectClass {
-+ ObjectClass parent_class;
-+} VMAObjectClass;
-+
-+typedef struct VMAObjectState {
-+ Object parent;
-+
-+ char *filename;
-+
-+ QemuUUID uuid;
-+ bool blocked;
-+ VMAWriter *vma;
-+ QemuMutex mutex;
-+} VMAObjectState;
-+
-+static VMAObjectState *vma_by_id(const char *name)
-+{
-+ Object *container;
-+ Object *obj;
-+
-+ container = object_get_objects_root();
-+ obj = object_resolve_path_component(container, name);
-+
-+ return VMA_OBJECT(obj);
-+}
-+
-+static void vma_object_class_complete(UserCreatable *uc, Error **errp)
-+{
-+ int rc;
-+ VMAObjectState *vo = VMA_OBJECT(uc);
-+ VMAObjectClass *voc = VMA_OBJECT_GET_CLASS(uc);
-+ (void)!vo;
-+ (void)!voc;
-+
-+ if (!vo->filename) {
-+ error_setg(errp, "Parameter 'filename' is required");
-+ return;
-+ }
-+
-+ vo->vma = VMAWriter_fopen(vo->filename);
-+ if (!vo->vma) {
-+ error_setg_errno(errp, errno, "failed to create VMA archive");
-+ return;
-+ }
-+
-+ rc = VMAWriter_set_uuid(vo->vma, vo->uuid.data, sizeof(vo->uuid.data));
-+ if (rc < 0) {
-+ error_setg_errno(errp, -rc, "failed to set UUID of VMA archive");
-+ return;
-+ }
-+
-+ qemu_mutex_init(&vo->mutex);
-+}
-+
-+static bool vma_object_can_be_deleted(UserCreatable *uc)
-+{
-+ //VMAObjectState *vo = VMA_OBJECT(uc);
-+ //if (!vo->vma) {
-+ // return true;
-+ //}
-+ //return false;
-+ return true;
-+}
-+
-+static void vma_object_class_init(ObjectClass *oc, void *data)
-+{
-+ UserCreatableClass *ucc = USER_CREATABLE_CLASS(oc);
-+
-+ ucc->can_be_deleted = vma_object_can_be_deleted;
-+ ucc->complete = vma_object_class_complete;
-+}
-+
-+static char *vma_object_get_filename(Object *obj, Error **errp)
-+{
-+ VMAObjectState *vo = VMA_OBJECT(obj);
-+
-+ return g_strdup(vo->filename);
-+}
-+
-+static void vma_object_set_filename(Object *obj, const char *str, Error **errp)
-+{
-+ VMAObjectState *vo = VMA_OBJECT(obj);
-+
-+ if (vo->vma) {
-+ error_setg(errp, "filename cannot be changed after creation");
-+ return;
-+ }
-+
-+ g_free(vo->filename);
-+ vo->filename = g_strdup(str);
-+}
-+
-+static char *vma_object_get_uuid(Object *obj, Error **errp)
-+{
-+ VMAObjectState *vo = VMA_OBJECT(obj);
-+
-+ return qemu_uuid_unparse_strdup(&vo->uuid);
-+}
-+
-+static void vma_object_set_uuid(Object *obj, const char *str, Error **errp)
-+{
-+ VMAObjectState *vo = VMA_OBJECT(obj);
-+
-+ if (vo->vma) {
-+ error_setg(errp, "uuid cannot be changed after creation");
-+ return;
-+ }
-+
-+ qemu_uuid_parse(str, &vo->uuid);
-+}
-+
-+static bool vma_object_get_blocked(Object *obj, Error **errp)
-+{
-+ VMAObjectState *vo = VMA_OBJECT(obj);
-+
-+ return vo->blocked;
-+}
-+
-+static void vma_object_set_blocked(Object *obj, bool blocked, Error **errp)
-+{
-+ VMAObjectState *vo = VMA_OBJECT(obj);
-+
-+ (void)errp;
-+
-+ vo->blocked = blocked;
-+}
-+
-+void vma_object_add_config_file(Object *obj, const char *name,
-+ const char *contents, size_t len,
-+ Error **errp)
-+{
-+ int rc;
-+ VMAObjectState *vo = VMA_OBJECT(obj);
-+
-+ if (!vo || !vo->vma) {
-+ error_setg(errp, "not a valid vma object to add config files to");
-+ return;
-+ }
-+
-+ rc = VMAWriter_addConfigFile(vo->vma, name, contents, len);
-+ if (rc < 0) {
-+ error_setg_errno(errp, -rc, "failed to add config file to VMA");
-+ return;
-+ }
-+}
-+
-+static void vma_object_init(Object *obj)
-+{
-+ VMAObjectState *vo = VMA_OBJECT(obj);
-+ (void)!vo;
-+
-+ object_property_add_str(obj, "filename",
-+ vma_object_get_filename, vma_object_set_filename,
-+ NULL);
-+ object_property_add_str(obj, "uuid",
-+ vma_object_get_uuid, vma_object_set_uuid,
-+ NULL);
-+ object_property_add_bool(obj, "blocked",
-+ vma_object_get_blocked, vma_object_set_blocked,
-+ NULL);
-+}
-+
-+static void vma_object_finalize(Object *obj)
-+{
-+ VMAObjectState *vo = VMA_OBJECT(obj);
-+ VMAObjectClass *voc = VMA_OBJECT_GET_CLASS(obj);
-+ (void)!voc;
-+
-+ qemu_mutex_destroy(&vo->mutex);
-+
-+ VMAWriter_destroy(vo->vma, true);
-+ g_free(vo->filename);
-+}
-+
-+static const TypeInfo vma_object_info = {
-+ .name = TYPE_VMA_OBJECT,
-+ .parent = TYPE_OBJECT,
-+ .class_size = sizeof(VMAObjectClass),
-+ .class_init = vma_object_class_init,
-+ .instance_size = sizeof(VMAObjectState),
-+ .instance_init = vma_object_init,
-+ .instance_finalize = vma_object_finalize,
-+ .interfaces = (InterfaceInfo[]) {
-+ { TYPE_USER_CREATABLE },
-+ { }
-+ }
-+};
-+
-+static void register_types(void)
-+{
-+ type_register_static(&vma_object_info);
-+}
-+
-+type_init(register_types);
-+
-+typedef struct {
-+ VMAObjectState *vma_obj;
-+ char *name;
-+ size_t device_id;
-+ uint64_t byte_size;
-+} BDRVVMAState;
-+
-+static void qemu_vma_parse_filename(const char *filename, QDict *options,
-+ Error **errp)
-+{
-+ char *sep;
-+
-+ if (strncmp(filename, "vma:", sizeof("vma:")-1) == 0) {
-+ filename += sizeof("vma:")-1;
-+ }
-+
-+ sep = strchr(filename, '/');
-+ if (!sep || sep == filename) {
-+ error_setg(errp, "VMA file should be <vma-obj>/<name>/<size>");
-+ return;
-+ }
-+
-+ qdict_put(options, "vma", qstring_from_substr(filename, 0, sep-filename));
-+
-+ while (*sep && *sep == '/')
-+ ++sep;
-+ if (!*sep) {
-+ error_setg(errp, "missing device name\n");
-+ return;
-+ }
-+
-+ filename = sep;
-+ sep = strchr(filename, '/');
-+ if (!sep || sep == filename) {
-+ error_setg(errp, "VMA file should be <vma-obj>/<name>/<size>");
-+ return;
-+ }
-+
-+ qdict_put(options, "name", qstring_from_substr(filename, 0, sep-filename));
-+
-+ while (*sep && *sep == '/')
-+ ++sep;
-+ if (!*sep) {
-+ error_setg(errp, "missing device size\n");
-+ return;
-+ }
-+
-+ filename = sep;
-+ qdict_put_str(options, "size", filename);
-+}
-+
-+static QemuOptsList runtime_opts = {
-+ .name = "vma-drive",
-+ .head = QTAILQ_HEAD_INITIALIZER(runtime_opts.head),
-+ .desc = {
-+ {
-+ .name = "vma",
-+ .type = QEMU_OPT_STRING,
-+ .help = "VMA Object name",
-+ },
-+ {
-+ .name = "name",
-+ .type = QEMU_OPT_STRING,
-+ .help = "VMA device name",
-+ },
-+ {
-+ .name = BLOCK_OPT_SIZE,
-+ .type = QEMU_OPT_SIZE,
-+ .help = "Virtual disk size"
-+ },
-+ { /* end of list */ }
-+ },
-+};
-+static int qemu_vma_open(BlockDriverState *bs, QDict *options, int flags,
-+ Error **errp)
-+{
-+ Error *local_err = NULL;
-+ BDRVVMAState *s = bs->opaque;
-+ QemuOpts *opts;
-+ const char *vma_id, *device_name;
-+ ssize_t dev_id;
-+ int64_t bytes = 0;
-+ int ret;
-+
-+ opts = qemu_opts_create(&runtime_opts, NULL, 0, &error_abort);
-+ qemu_opts_absorb_qdict(opts, options, &local_err);
-+ if (local_err) {
-+ error_propagate(errp, local_err);
-+ ret = -EINVAL;
-+ goto failed_opts;
-+ }
-+
-+ bytes = ROUND_UP(qemu_opt_get_size_del(opts, BLOCK_OPT_SIZE, 0),
-+ BDRV_SECTOR_SIZE);
-+
-+ vma_id = qemu_opt_get(opts, "vma");
-+ if (!vma_id) {
-+ ret = -EINVAL;
-+ error_setg(errp, "missing 'vma' property");
-+ goto failed_opts;
-+ }
-+
-+ device_name = qemu_opt_get(opts, "name");
-+ if (!device_name) {
-+ ret = -EINVAL;
-+ error_setg(errp, "missing 'name' property");
-+ goto failed_opts;
-+ }
-+
-+ VMAObjectState *vma = vma_by_id(vma_id);
-+ if (!vma) {
-+ ret = -EINVAL;
-+ error_setg(errp, "no such VMA object: %s", vma_id);
-+ goto failed_opts;
-+ }
-+
-+ dev_id = VMAWriter_findDevice(vma->vma, device_name);
-+ if (dev_id >= 0) {
-+ error_setg(errp, "drive already exists in VMA object");
-+ ret = -EIO;
-+ goto failed_opts;
-+ }
-+
-+ dev_id = VMAWriter_addDevice(vma->vma, device_name, (uint64_t)bytes);
-+ if (dev_id < 0) {
-+ error_setg_errno(errp, -dev_id, "failed to add VMA device");
-+ ret = -EIO;
-+ goto failed_opts;
-+ }
-+
-+ object_ref(OBJECT(vma));
-+ s->vma_obj = vma;
-+ s->name = g_strdup(device_name);
-+ s->device_id = (size_t)dev_id;
-+ s->byte_size = bytes;
-+
-+ ret = 0;
-+
-+failed_opts:
-+ qemu_opts_del(opts);
-+ return ret;
-+}
-+
-+static void qemu_vma_close(BlockDriverState *bs)
-+{
-+ BDRVVMAState *s = bs->opaque;
-+
-+ (void)VMAWriter_finishDevice(s->vma_obj->vma, s->device_id);
-+ object_unref(OBJECT(s->vma_obj));
-+
-+ g_free(s->name);
-+}
-+
-+static int64_t qemu_vma_getlength(BlockDriverState *bs)
-+{
-+ BDRVVMAState *s = bs->opaque;
-+
-+ return s->byte_size;
-+}
-+
-+static coroutine_fn int qemu_vma_co_writev(BlockDriverState *bs,
-+ int64_t sector_num,
-+ int nb_sectors,
-+ QEMUIOVector *qiov,
-+ int flags)
-+{
-+ size_t i;
-+ ssize_t rc;
-+ BDRVVMAState *s = bs->opaque;
-+ VMAObjectState *vo = s->vma_obj;
-+ off_t offset = sector_num * BDRV_SECTOR_SIZE;
-+ /* flags can be only values we set in supported_write_flags */
-+ assert(flags == 0);
-+
-+ qemu_mutex_lock(&vo->mutex);
-+ if (vo->blocked) {
-+ return -EPERM;
-+ }
-+ for (i = 0; i != qiov->niov; ++i) {
-+ const struct iovec *v = &qiov->iov[i];
-+ size_t blocks = v->iov_len / VMA_BLOCK_SIZE;
-+ if (blocks * VMA_BLOCK_SIZE != v->iov_len) {
-+ return -EIO;
-+ }
-+ rc = VMAWriter_writeBlocks(vo->vma, s->device_id,
-+ v->iov_base, blocks, offset);
-+ if (errno) {
-+ return -errno;
-+ }
-+ if (rc != blocks) {
-+ return -EIO;
-+ }
-+ offset += v->iov_len;
-+ }
-+ qemu_mutex_unlock(&vo->mutex);
-+ return 0;
-+}
-+
-+static int qemu_vma_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
-+{
-+ bdi->cluster_size = VMA_CLUSTER_SIZE;
-+ bdi->unallocated_blocks_are_zero = true;
-+ return 0;
-+}
-+
-+static int qemu_vma_check_perm(BlockDriverState *bs,
-+ uint64_t perm,
-+ uint64_t shared,
-+ Error **errp)
-+{
-+ /* Nothing to do. */
-+ return 0;
-+}
-+
-+static void qemu_vma_set_perm(BlockDriverState *bs,
-+ uint64_t perm,
-+ uint64_t shared)
-+{
-+ /* Nothing to do. */
-+}
-+
-+static void qemu_vma_abort_perm_update(BlockDriverState *bs)
-+{
-+ /* Nothing to do. */
-+}
-+
-+static void qemu_vma_refresh_limits(BlockDriverState *bs, Error **errp)
-+{
-+ bs->bl.request_alignment = BDRV_SECTOR_SIZE; /* No sub-sector I/O */
-+}
-+static void qemu_vma_child_perm(BlockDriverState *bs, BdrvChild *c,
-+ const BdrvChildRole *role,
-+ BlockReopenQueue *reopen_queue,
-+ uint64_t perm, uint64_t shared,
-+ uint64_t *nperm, uint64_t *nshared)
-+{
-+ *nperm = BLK_PERM_ALL;
-+ *nshared = BLK_PERM_ALL;
-+}
-+
-+static BlockDriver bdrv_vma_drive = {
-+ .format_name = "vma-drive",
-+ .protocol_name = "vma",
-+ .instance_size = sizeof(BDRVVMAState),
-+
-+#if 0
-+ .bdrv_create = qemu_vma_create,
-+ .create_opts = &qemu_vma_create_opts,
-+#endif
-+
-+ .bdrv_parse_filename = qemu_vma_parse_filename,
-+ .bdrv_file_open = qemu_vma_open,
-+
-+ .bdrv_close = qemu_vma_close,
-+ .bdrv_has_zero_init = bdrv_has_zero_init_1,
-+ .bdrv_getlength = qemu_vma_getlength,
-+ .bdrv_get_info = qemu_vma_get_info,
-+
-+ //.bdrv_co_preadv = qemu_vma_co_preadv,
-+ .bdrv_co_writev = qemu_vma_co_writev,
-+
-+ .bdrv_refresh_limits = qemu_vma_refresh_limits,
-+ .bdrv_check_perm = qemu_vma_check_perm,
-+ .bdrv_set_perm = qemu_vma_set_perm,
-+ .bdrv_abort_perm_update = qemu_vma_abort_perm_update,
-+ .bdrv_child_perm = qemu_vma_child_perm,
-+};
-+
-+static void bdrv_vma_init(void)
-+{
-+ bdrv_register(&bdrv_vma_drive);
-+}
-+
-+block_init(bdrv_vma_init);
-diff --git a/blockdev.c b/blockdev.c
-index a7c97b1585..7047475a3c 100644
---- a/blockdev.c
-+++ b/blockdev.c
-@@ -31,11 +31,13 @@
- */
-
- #include "qemu/osdep.h"
-+#include "qemu/uuid.h"
- #include "sysemu/block-backend.h"
- #include "sysemu/blockdev.h"
- #include "hw/block/block.h"
- #include "block/blockjob.h"
- #include "block/qdict.h"
-+#include "block/blockjob_int.h"
- #include "block/throttle-groups.h"
- #include "monitor/monitor.h"
- #include "qemu/error-report.h"
-@@ -45,6 +47,7 @@
- #include "qapi/qapi-commands-block.h"
- #include "qapi/qapi-commands-transaction.h"
- #include "qapi/qapi-visit-block-core.h"
-+#include "qapi/qapi-types-misc.h"
- #include "qapi/qmp/qdict.h"
- #include "qapi/qmp/qnum.h"
- #include "qapi/qmp/qstring.h"
-@@ -3168,6 +3171,539 @@ out:
- aio_context_release(aio_context);
- }
-
-+/* PVE backup related function */
-+
-+static struct PVEBackupState {
-+ Error *error;
-+ bool cancel;
-+ QemuUUID uuid;
-+ char uuid_str[37];
-+ int64_t speed;
-+ time_t start_time;
-+ time_t end_time;
-+ char *backup_file;
-+ Object *vmaobj;
-+ GList *di_list;
-+ size_t next_job;
-+ size_t total;
-+ size_t transferred;
-+ size_t zero_bytes;
-+ QemuMutex backup_mutex;
-+ bool backup_mutex_initialized;
-+} backup_state;
-+
-+typedef struct PVEBackupDevInfo {
-+ BlockDriverState *bs;
-+ size_t size;
-+ uint8_t dev_id;
-+ bool completed;
-+ char targetfile[PATH_MAX];
-+ BlockDriverState *target;
-+} PVEBackupDevInfo;
-+
-+static void pvebackup_run_next_job(void);
-+
-+static void pvebackup_cleanup(void)
-+{
-+ qemu_mutex_lock(&backup_state.backup_mutex);
-+ // Avoid race between block jobs and backup-cancel command:
-+ if (!backup_state.vmaw) {
-+ qemu_mutex_unlock(&backup_state.backup_mutex);
-+ return;
-+ }
-+
-+ backup_state.end_time = time(NULL);
-+
-+ if (backup_state.vmaobj) {
-+ object_unparent(backup_state.vmaobj);
-+ backup_state.vmaobj = NULL;
-+ }
-+
-+ g_list_free(backup_state.di_list);
-+ backup_state.di_list = NULL;
-+ qemu_mutex_unlock(&backup_state.backup_mutex);
-+}
-+
-+static void pvebackup_complete_cb(void *opaque, int ret)
-+{
-+ // This always runs in the main loop
-+
-+ PVEBackupDevInfo *di = opaque;
-+
-+ di->completed = true;
-+
-+ if (ret < 0 && !backup_state.error) {
-+ error_setg(&backup_state.error, "job failed with err %d - %s",
-+ ret, strerror(-ret));
-+ }
-+
-+ di->bs = NULL;
-+ di->target = NULL;
-+
-+ if (backup_state.vmaobj) {
-+ object_unparent(backup_state.vmaobj);
-+ backup_state.vmaobj = NULL;
-+ }
-+
-+ // remove self from job queue
-+ qemu_mutex_lock(&backup_state.backup_mutex);
-+ backup_state.di_list = g_list_remove(backup_state.di_list, di);
-+ g_free(di);
-+ qemu_mutex_unlock(&backup_state.backup_mutex);
-+
-+ if (!backup_state.cancel) {
-+ pvebackup_run_next_job();
-+ }
-+}
-+
-+static void pvebackup_cancel(void *opaque)
-+{
-+ backup_state.cancel = true;
-+ qemu_mutex_lock(&backup_state.backup_mutex);
-+ // Avoid race between block jobs and backup-cancel command:
-+ if (!backup_state.vmaw) {
-+ qemu_mutex_unlock(&backup_state.backup_mutex);
-+ return;
-+ }
-+
-+ if (!backup_state.error) {
-+ error_setg(&backup_state.error, "backup cancelled");
-+ }
-+
-+ if (backup_state.vmaobj) {
-+ Error *err;
-+ /* make sure vma writer does not block anymore */
-+ if (!object_set_props(backup_state.vmaobj, &err, "blocked", "yes", NULL)) {
-+ if (err) {
-+ error_report_err(err);
-+ }
-+ }
-+ }
-+
-+ GList *l = backup_state.di_list;
-+ while (l) {
-+ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
-+ l = g_list_next(l);
-+ if (!di->completed && di->bs) {
-+ BlockJob *job = di->bs->job;
-+ if (job) {
-+ AioContext *aio_context = blk_get_aio_context(job->blk);
-+ aio_context_acquire(aio_context);
-+ if (!di->completed) {
-+ job_cancel(&job->job, false);
-+ }
-+ aio_context_release(aio_context);
-+ }
-+ }
-+ }
-+
-+ qemu_mutex_unlock(&backup_state.backup_mutex);
-+ pvebackup_cleanup();
-+}
-+
-+void qmp_backup_cancel(Error **errp)
-+{
-+ if (!backup_state.backup_mutex_initialized)
-+ return;
-+ Coroutine *co = qemu_coroutine_create(pvebackup_cancel, NULL);
-+ qemu_coroutine_enter(co);
-+
-+ while (backup_state.vmaobj) {
-+ /* FIXME: Find something better for this */
-+ aio_poll(qemu_get_aio_context(), true);
-+ }
-+}
-+
-+void vma_object_add_config_file(Object *obj, const char *name,
-+ const char *contents, size_t len,
-+ Error **errp);
-+static int config_to_vma(const char *file, BackupFormat format,
-+ Object *vmaobj,
-+ const char *backup_dir,
-+ Error **errp)
-+{
-+ char *cdata = NULL;
-+ gsize clen = 0;
-+ GError *err = NULL;
-+ if (!g_file_get_contents(file, &cdata, &clen, &err)) {
-+ error_setg(errp, "unable to read file '%s'", file);
-+ return 1;
-+ }
-+
-+ char *basename = g_path_get_basename(file);
-+
-+ if (format == BACKUP_FORMAT_VMA) {
-+ vma_object_add_config_file(vmaobj, basename, cdata, clen, errp);
-+ } else if (format == BACKUP_FORMAT_DIR) {
-+ char config_path[PATH_MAX];
-+ snprintf(config_path, PATH_MAX, "%s/%s", backup_dir, basename);
-+ if (!g_file_set_contents(config_path, cdata, clen, &err)) {
-+ error_setg(errp, "unable to write config file '%s'", config_path);
-+ g_free(cdata);
-+ g_free(basename);
-+ return 1;
-+ }
-+ }
-+
-+ g_free(basename);
-+ g_free(cdata);
-+ return 0;
-+}
-+
-+static void pvebackup_run_next_job(void)
-+{
-+ qemu_mutex_lock(&backup_state.backup_mutex);
-+
-+ GList *next = g_list_nth(backup_state.di_list, backup_state.next_job);
-+ while (next) {
-+ PVEBackupDevInfo *di = (PVEBackupDevInfo *)next->data;
-+ backup_state.next_job++;
-+ if (!di->completed && di->bs && di->bs->job) {
-+ BlockJob *job = di->bs->job;
-+ AioContext *aio_context = blk_get_aio_context(job->blk);
-+ aio_context_acquire(aio_context);
-+ qemu_mutex_unlock(&backup_state.backup_mutex);
-+ if (backup_state.error || backup_state.cancel) {
-+ job_cancel_sync(job);
-+ } else {
-+ job_resume(job);
-+ }
-+ aio_context_release(aio_context);
-+ return;
-+ }
-+ next = g_list_next(next);
-+ }
-+ qemu_mutex_unlock(&backup_state.backup_mutex);
-+
-+ // no more jobs, run the cleanup
-+ pvebackup_cleanup();
-+}
-+
-+UuidInfo *qmp_backup(const char *backup_file, bool has_format,
-+ BackupFormat format,
-+ bool has_config_file, const char *config_file,
-+ bool has_firewall_file, const char *firewall_file,
-+ bool has_devlist, const char *devlist,
-+ bool has_speed, int64_t speed, Error **errp)
-+{
-+ BlockBackend *blk;
-+ BlockDriverState *bs = NULL;
-+ const char *backup_dir = NULL;
-+ Error *local_err = NULL;
-+ QemuUUID uuid;
-+ gchar **devs = NULL;
-+ GList *di_list = NULL;
-+ GList *l;
-+ UuidInfo *uuid_info;
-+ BlockJob *job;
-+
-+ if (!backup_state.backup_mutex_initialized) {
-+ qemu_mutex_init(&backup_state.backup_mutex);
-+ backup_state.backup_mutex_initialized = true;
-+ }
-+
-+ if (backup_state.di_list || backup_state.vmaobj) {
-+ error_set(errp, ERROR_CLASS_GENERIC_ERROR,
-+ "previous backup not finished");
-+ return NULL;
-+ }
-+
-+ /* Todo: try to auto-detect format based on file name */
-+ format = has_format ? format : BACKUP_FORMAT_VMA;
-+
-+ if (has_devlist) {
-+ devs = g_strsplit_set(devlist, ",;:", -1);
-+
-+ gchar **d = devs;
-+ while (d && *d) {
-+ blk = blk_by_name(*d);
-+ if (blk) {
-+ bs = blk_bs(blk);
-+ if (bdrv_is_read_only(bs)) {
-+ error_setg(errp, "Node '%s' is read only", *d);
-+ goto err;
-+ }
-+ if (!bdrv_is_inserted(bs)) {
-+ error_setg(errp, QERR_DEVICE_HAS_NO_MEDIUM, *d);
-+ goto err;
-+ }
-+ PVEBackupDevInfo *di = g_new0(PVEBackupDevInfo, 1);
-+ di->bs = bs;
-+ di_list = g_list_append(di_list, di);
-+ } else {
-+ error_set(errp, ERROR_CLASS_DEVICE_NOT_FOUND,
-+ "Device '%s' not found", *d);
-+ goto err;
-+ }
-+ d++;
-+ }
-+
-+ } else {
-+ BdrvNextIterator it;
-+
-+ bs = NULL;
-+ for (bs = bdrv_first(&it); bs; bs = bdrv_next(&it)) {
-+ if (!bdrv_is_inserted(bs) || bdrv_is_read_only(bs)) {
-+ continue;
-+ }
-+
-+ PVEBackupDevInfo *di = g_new0(PVEBackupDevInfo, 1);
-+ di->bs = bs;
-+ di_list = g_list_append(di_list, di);
-+ }
-+ }
-+
-+ if (!di_list) {
-+ error_set(errp, ERROR_CLASS_GENERIC_ERROR, "empty device list");
-+ goto err;
-+ }
-+
-+ size_t total = 0;
-+
-+ l = di_list;
-+ while (l) {
-+ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
-+ l = g_list_next(l);
-+ if (bdrv_op_is_blocked(di->bs, BLOCK_OP_TYPE_BACKUP_SOURCE, errp)) {
-+ goto err;
-+ }
-+
-+ ssize_t size = bdrv_getlength(di->bs);
-+ if (size < 0) {
-+ error_setg_errno(errp, -di->size, "bdrv_getlength failed");
-+ goto err;
-+ }
-+ di->size = size;
-+ total += size;
-+ }
-+
-+ qemu_uuid_generate(&uuid);
-+
-+ if (format == BACKUP_FORMAT_VMA) {
-+ char uuidstr[UUID_FMT_LEN+1];
-+ qemu_uuid_unparse(&uuid, uuidstr);
-+ uuidstr[UUID_FMT_LEN] = 0;
-+ backup_state.vmaobj =
-+ object_new_with_props("vma", object_get_objects_root(),
-+ "vma-backup-obj", &local_err,
-+ "filename", backup_file,
-+ "uuid", uuidstr,
-+ NULL);
-+ if (!backup_state.vmaobj) {
-+ if (local_err) {
-+ error_propagate(errp, local_err);
-+ }
-+ goto err;
-+ }
-+
-+ l = di_list;
-+ while (l) {
-+ QDict *options = qdict_new();
-+
-+ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
-+ l = g_list_next(l);
-+
-+ const char *devname = bdrv_get_device_name(di->bs);
-+ snprintf(di->targetfile, PATH_MAX, "vma-backup-obj/%s.raw", devname);
-+
-+ qdict_put(options, "driver", qstring_from_str("vma-drive"));
-+ qdict_put(options, "size", qint_from_int(di->size));
-+ di->target = bdrv_open(di->targetfile, NULL, options, BDRV_O_RDWR, &local_err);
-+ if (!di->target) {
-+ error_propagate(errp, local_err);
-+ goto err;
-+ }
-+ }
-+ } else if (format == BACKUP_FORMAT_DIR) {
-+ if (mkdir(backup_file, 0640) != 0) {
-+ error_setg_errno(errp, errno, "can't create directory '%s'\n",
-+ backup_file);
-+ goto err;
-+ }
-+ backup_dir = backup_file;
-+
-+ l = di_list;
-+ while (l) {
-+ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
-+ l = g_list_next(l);
-+
-+ const char *devname = bdrv_get_device_name(di->bs);
-+ snprintf(di->targetfile, PATH_MAX, "%s/%s.raw", backup_dir, devname);
-+
-+ int flags = BDRV_O_RDWR;
-+ bdrv_img_create(di->targetfile, "raw", NULL, NULL, NULL,
-+ di->size, flags, false, &local_err);
-+ if (local_err) {
-+ error_propagate(errp, local_err);
-+ goto err;
-+ }
-+
-+ di->target = bdrv_open(di->targetfile, NULL, NULL, flags, &local_err);
-+ if (!di->target) {
-+ error_propagate(errp, local_err);
-+ goto err;
-+ }
-+ }
-+ } else {
-+ error_set(errp, ERROR_CLASS_GENERIC_ERROR, "unknown backup format");
-+ goto err;
-+ }
-+
-+ /* add configuration file to archive */
-+ if (has_config_file) {
-+ if(config_to_vma(config_file, format, backup_state.vmaobj, backup_dir, errp) != 0) {
-+ goto err;
-+ }
-+ }
-+
-+ /* add firewall file to archive */
-+ if (has_firewall_file) {
-+ if(config_to_vma(firewall_file, format, backup_state.vmaobj, backup_dir, errp) != 0) {
-+ goto err;
-+ }
-+ }
-+ /* initialize global backup_state now */
-+
-+ backup_state.cancel = false;
-+
-+ if (backup_state.error) {
-+ error_free(backup_state.error);
-+ backup_state.error = NULL;
-+ }
-+
-+ backup_state.speed = (has_speed && speed > 0) ? speed : 0;
-+
-+ backup_state.start_time = time(NULL);
-+ backup_state.end_time = 0;
-+
-+ if (backup_state.backup_file) {
-+ g_free(backup_state.backup_file);
-+ }
-+ backup_state.backup_file = g_strdup(backup_file);
-+
-+ memcpy(&backup_state.uuid, &uuid, sizeof(uuid));
-+ qemu_uuid_unparse(&uuid, backup_state.uuid_str);
-+
-+ qemu_mutex_lock(&backup_state.backup_mutex);
-+ backup_state.di_list = di_list;
-+ backup_state.next_job = 0;
-+
-+ backup_state.total = total;
-+ backup_state.transferred = 0;
-+ backup_state.zero_bytes = 0;
-+
-+ /* start all jobs (paused state) */
-+ l = di_list;
-+ while (l) {
-+ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
-+ l = g_list_next(l);
-+
-+ job = backup_job_create(NULL, di->bs, di->target, speed, MIRROR_SYNC_MODE_FULL, NULL,
-+ false, BLOCKDEV_ON_ERROR_REPORT, BLOCKDEV_ON_ERROR_REPORT,
-+ JOB_DEFAULT,
-+ pvebackup_complete_cb, di, 2, NULL, &local_err);
-+ if (di->target) {
-+ bdrv_unref(di->target);
-+ di->target = NULL;
-+ }
-+ if (!job || local_err != NULL) {
-+ error_setg(&backup_state.error, "backup_job_create failed");
-+ pvebackup_cancel(NULL);
-+ } else {
-+ job_start(&job->job);
-+ }
-+ }
-+
-+ qemu_mutex_unlock(&backup_state.backup_mutex);
-+
-+ if (!backup_state.error) {
-+ pvebackup_run_next_job(); // run one job
-+ }
-+
-+ uuid_info = g_malloc0(sizeof(*uuid_info));
-+ uuid_info->UUID = g_strdup(backup_state.uuid_str);
-+
-+ return uuid_info;
-+
-+err:
-+
-+ l = di_list;
-+ while (l) {
-+ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
-+ l = g_list_next(l);
-+
-+ if (di->target) {
-+ bdrv_unref(di->target);
-+ }
-+
-+ if (di->targetfile[0]) {
-+ unlink(di->targetfile);
-+ }
-+ g_free(di);
-+ }
-+ g_list_free(di_list);
-+
-+ if (devs) {
-+ g_strfreev(devs);
-+ }
-+
-+ if (backup_state.vmaobj) {
-+ object_unparent(backup_state.vmaobj);
-+ backup_state.vmaobj = NULL;
-+ }
-+
-+ if (backup_dir) {
-+ rmdir(backup_dir);
-+ }
-+
-+ return NULL;
-+}
-+
-+BackupStatus *qmp_query_backup(Error **errp)
-+{
-+ BackupStatus *info = g_malloc0(sizeof(*info));
-+
-+ if (!backup_state.start_time) {
-+ /* not started, return {} */
-+ return info;
-+ }
-+
-+ info->has_status = true;
-+ info->has_start_time = true;
-+ info->start_time = backup_state.start_time;
-+
-+ if (backup_state.backup_file) {
-+ info->has_backup_file = true;
-+ info->backup_file = g_strdup(backup_state.backup_file);
-+ }
-+
-+ info->has_uuid = true;
-+ info->uuid = g_strdup(backup_state.uuid_str);
-+
-+ if (backup_state.end_time) {
-+ if (backup_state.error) {
-+ info->status = g_strdup("error");
-+ info->has_errmsg = true;
-+ info->errmsg = g_strdup(error_get_pretty(backup_state.error));
-+ } else {
-+ info->status = g_strdup("done");
-+ }
-+ info->has_end_time = true;
-+ info->end_time = backup_state.end_time;
-+ } else {
-+ info->status = g_strdup("active");
-+ }
-+
-+ info->has_total = true;
-+ info->total = backup_state.total;
-+ info->has_zero_bytes = true;
-+ info->zero_bytes = backup_state.zero_bytes;
-+ info->has_transferred = true;
-+ info->transferred = backup_state.transferred;
-+
-+ return info;
-+}
-+
- void qmp_block_stream(bool has_job_id, const char *job_id, const char *device,
- bool has_base, const char *base,
- bool has_base_node, const char *base_node,
-diff --git a/configure b/configure
-index 714e7fb6a1..5e16d8e562 100755
---- a/configure
-+++ b/configure
-@@ -499,6 +499,8 @@ docker="no"
- debug_mutex="no"
- libpmem=""
- default_devices="yes"
-+libudev="no"
-+vma=""
-
- # cross compilers defaults, can be overridden with --cross-cc-ARCH
- cross_cc_aarch64="aarch64-linux-gnu-gcc"
-@@ -1543,6 +1545,10 @@ for opt do
- ;;
- --disable-libpmem) libpmem=no
- ;;
-+ --enable-vma) vma=yes
-+ ;;
-+ --disable-vma) vma=no
-+ ;;
- *)
- echo "ERROR: unknown option $opt"
- echo "Try '$0 --help' for more information"
-@@ -1842,6 +1848,7 @@ disabled with --disable-FEATURE, default is enabled if available:
- capstone capstone disassembler support
- debug-mutex mutex debugging support
- libpmem libpmem support
-+ vma VMA archive backend
-
- NOTE: The object files are built at the place where configure is launched
- EOF
-@@ -4402,6 +4409,22 @@ EOF
- fi
- fi
-
-+##########################################
-+# vma probe
-+if test "$vma" != "no" ; then
-+ if $pkg_config --exact-version=0.1.0 vma; then
-+ vma="yes"
-+ vma_cflags=$($pkg_config --cflags vma)
-+ vma_libs=$($pkg_config --libs vma)
-+ else
-+ if test "$vma" = "yes" ; then
-+ feature_not_found "VMA Archive backend support" \
-+ "Install libvma devel"
-+ fi
-+ vma="no"
-+ fi
-+fi
-+
- ##########################################
- # signalfd probe
- signalfd="no"
-@@ -6481,6 +6504,7 @@ echo "docker $docker"
- echo "libpmem support $libpmem"
- echo "libudev $libudev"
- echo "default devices $default_devices"
-+echo "VMA support $vma"
-
- if test "$supported_cpu" = "no"; then
- echo
-@@ -6983,6 +7007,12 @@ if test "$usb_redir" = "yes" ; then
- echo "USB_REDIR_LIBS=$usb_redir_libs" >> $config_host_mak
- fi
-
-+if test "$vma" = "yes" ; then
-+ echo "CONFIG_VMA=y" >> $config_host_mak
-+ echo "VMA_CFLAGS=$vma_cflags" >> $config_host_mak
-+ echo "VMA_LIBS=$vma_libs" >> $config_host_mak
-+fi
-+
- if test "$opengl" = "yes" ; then
- echo "CONFIG_OPENGL=y" >> $config_host_mak
- echo "OPENGL_LIBS=$opengl_libs" >> $config_host_mak
-diff --git a/hmp-commands-info.hx b/hmp-commands-info.hx
-index 444bd8e43d..21106bcbe6 100644
---- a/hmp-commands-info.hx
-+++ b/hmp-commands-info.hx
-@@ -536,6 +536,19 @@ STEXI
- @item info cpustats
- @findex info cpustats
- Show CPU statistics.
-+ETEXI
-+
-+ {
-+ .name = "backup",
-+ .args_type = "",
-+ .params = "",
-+ .help = "show backup status",
-+ .cmd = hmp_info_backup,
-+ },
-+
-+STEXI
-+@item info backup
-+show backup status
- ETEXI
-
- #if defined(CONFIG_SLIRP)
-diff --git a/hmp-commands.hx b/hmp-commands.hx
-index e075d413c0..5bdfdeaf57 100644
---- a/hmp-commands.hx
-+++ b/hmp-commands.hx
-@@ -105,6 +105,37 @@ STEXI
- @item block_stream
- @findex block_stream
- Copy data from a backing file into a block device.
-+ETEXI
-+
-+ {
-+ .name = "backup",
-+ .args_type = "directory:-d,backupfile:s,speed:o?,devlist:s?",
-+ .params = "[-d] backupfile [speed [devlist]]",
-+ .help = "create a VM Backup."
-+ "\n\t\t\t Use -d to dump data into a directory instead"
-+ "\n\t\t\t of using VMA format.",
-+ .cmd = hmp_backup,
-+ },
-+
-+STEXI
-+@item backup
-+@findex backup
-+Create a VM backup.
-+ETEXI
-+
-+ {
-+ .name = "backup_cancel",
-+ .args_type = "",
-+ .params = "",
-+ .help = "cancel the current VM backup",
-+ .cmd = hmp_backup_cancel,
-+ },
-+
-+STEXI
-+@item backup_cancel
-+@findex backup_cancel
-+Cancel the current VM backup.
-+
- ETEXI
-
- {
-diff --git a/include/monitor/hmp.h b/include/monitor/hmp.h
-index c6ee8295f0..0f2f96c4af 100644
---- a/include/monitor/hmp.h
-+++ b/include/monitor/hmp.h
-@@ -30,6 +30,7 @@ void hmp_info_migrate(Monitor *mon, const QDict *qdict);
- void hmp_info_migrate_capabilities(Monitor *mon, const QDict *qdict);
- void hmp_info_migrate_parameters(Monitor *mon, const QDict *qdict);
- void hmp_info_migrate_cache_size(Monitor *mon, const QDict *qdict);
-+void hmp_info_backup(Monitor *mon, const QDict *qdict);
- void hmp_info_cpus(Monitor *mon, const QDict *qdict);
- void hmp_info_block(Monitor *mon, const QDict *qdict);
- void hmp_info_blockstats(Monitor *mon, const QDict *qdict);
-@@ -90,6 +91,8 @@ void hmp_eject(Monitor *mon, const QDict *qdict);
- void hmp_change(Monitor *mon, const QDict *qdict);
- void hmp_block_set_io_throttle(Monitor *mon, const QDict *qdict);
- void hmp_block_stream(Monitor *mon, const QDict *qdict);
-+void hmp_backup(Monitor *mon, const QDict *qdict);
-+void hmp_backup_cancel(Monitor *mon, const QDict *qdict);
- void hmp_block_job_set_speed(Monitor *mon, const QDict *qdict);
- void hmp_block_job_cancel(Monitor *mon, const QDict *qdict);
- void hmp_block_job_pause(Monitor *mon, const QDict *qdict);
-diff --git a/monitor/hmp-cmds.c b/monitor/hmp-cmds.c
-index 39a8020367..bc9ca346f7 100644
---- a/monitor/hmp-cmds.c
-+++ b/monitor/hmp-cmds.c
-@@ -196,6 +196,44 @@ void hmp_info_mice(Monitor *mon, const QDict *qdict)
- qapi_free_MouseInfoList(mice_list);
- }
-
-+void hmp_info_backup(Monitor *mon, const QDict *qdict)
-+{
-+ BackupStatus *info;
-+
-+ info = qmp_query_backup(NULL);
-+ if (info->has_status) {
-+ if (info->has_errmsg) {
-+ monitor_printf(mon, "Backup status: %s - %s\n",
-+ info->status, info->errmsg);
-+ } else {
-+ monitor_printf(mon, "Backup status: %s\n", info->status);
-+ }
-+ }
-+
-+ if (info->has_backup_file) {
-+ monitor_printf(mon, "Start time: %s", ctime(&info->start_time));
-+ if (info->end_time) {
-+ monitor_printf(mon, "End time: %s", ctime(&info->end_time));
-+ }
-+
-+ int per = (info->has_total && info->total &&
-+ info->has_transferred && info->transferred) ?
-+ (info->transferred * 100)/info->total : 0;
-+ int zero_per = (info->has_total && info->total &&
-+ info->has_zero_bytes && info->zero_bytes) ?
-+ (info->zero_bytes * 100)/info->total : 0;
-+ monitor_printf(mon, "Backup file: %s\n", info->backup_file);
-+ monitor_printf(mon, "Backup uuid: %s\n", info->uuid);
-+ monitor_printf(mon, "Total size: %zd\n", info->total);
-+ monitor_printf(mon, "Transferred bytes: %zd (%d%%)\n",
-+ info->transferred, per);
-+ monitor_printf(mon, "Zero bytes: %zd (%d%%)\n",
-+ info->zero_bytes, zero_per);
-+ }
-+
-+ qapi_free_BackupStatus(info);
-+}
-+
- static char *SocketAddress_to_str(SocketAddress *addr)
- {
- switch (addr->type) {
-@@ -2078,6 +2116,31 @@ void hmp_block_stream(Monitor *mon, const QDict *qdict)
- hmp_handle_error(mon, &error);
- }
-
-+void hmp_backup_cancel(Monitor *mon, const QDict *qdict)
-+{
-+ Error *error = NULL;
-+
-+ qmp_backup_cancel(&error);
-+
-+ hmp_handle_error(mon, &error);
-+}
-+
-+void hmp_backup(Monitor *mon, const QDict *qdict)
-+{
-+ Error *error = NULL;
-+
-+ int dir = qdict_get_try_bool(qdict, "directory", 0);
-+ const char *backup_file = qdict_get_str(qdict, "backupfile");
-+ const char *devlist = qdict_get_try_str(qdict, "devlist");
-+ int64_t speed = qdict_get_try_int(qdict, "speed", 0);
-+
-+ qmp_backup(backup_file, true, dir ? BACKUP_FORMAT_DIR : BACKUP_FORMAT_VMA,
-+ false, NULL, false, NULL, !!devlist,
-+ devlist, qdict_haskey(qdict, "speed"), speed, &error);
-+
-+ hmp_handle_error(mon, &error);
-+}
-+
- void hmp_block_job_set_speed(Monitor *mon, const QDict *qdict)
- {
- Error *error = NULL;
-diff --git a/qapi/block-core.json b/qapi/block-core.json
-index 0d43d4f37c..97cb7ec41c 100644
---- a/qapi/block-core.json
-+++ b/qapi/block-core.json
-@@ -796,6 +796,97 @@
- { 'command': 'query-block', 'returns': ['BlockInfo'] }
-
-
-+##
-+# @BackupStatus:
-+#
-+# Detailed backup status.
-+#
-+# @status: string describing the current backup status.
-+# This can be 'active', 'done', 'error'. If this field is not
-+# returned, no backup process has been initiated
-+#
-+# @errmsg: error message (only returned if status is 'error')
-+#
-+# @total: total amount of bytes involved in the backup process
-+#
-+# @transferred: amount of bytes already backed up.
-+#
-+# @zero-bytes: amount of 'zero' bytes detected.
-+#
-+# @start-time: time (epoch) when backup job started.
-+#
-+# @end-time: time (epoch) when backup job finished.
-+#
-+# @backup-file: backup file name
-+#
-+# @uuid: uuid for this backup job
-+#
-+##
-+{ 'struct': 'BackupStatus',
-+ 'data': {'*status': 'str', '*errmsg': 'str', '*total': 'int',
-+ '*transferred': 'int', '*zero-bytes': 'int',
-+ '*start-time': 'int', '*end-time': 'int',
-+ '*backup-file': 'str', '*uuid': 'str' } }
-+
-+##
-+# @BackupFormat:
-+#
-+# An enumeration of supported backup formats.
-+#
-+# @vma: Proxmox vma backup format
-+##
-+{ 'enum': 'BackupFormat',
-+ 'data': [ 'vma', 'dir' ] }
-+
-+##
-+# @backup:
-+#
-+# Starts a VM backup.
-+#
-+# @backup-file: the backup file name
-+#
-+# @format: format of the backup file
-+#
-+# @config-file: a configuration file to include into
-+# the backup archive.
-+#
-+# @speed: the maximum speed, in bytes per second
-+#
-+# @devlist: list of block device names (separated by ',', ';'
-+# or ':'). By default the backup includes all writable block devices.
-+#
-+# Returns: the uuid of the backup job
-+#
-+##
-+{ 'command': 'backup', 'data': { 'backup-file': 'str',
-+ '*format': 'BackupFormat',
-+ '*config-file': 'str',
-+ '*firewall-file': 'str',
-+ '*devlist': 'str', '*speed': 'int' },
-+ 'returns': 'UuidInfo' }
-+
-+##
-+# @query-backup:
-+#
-+# Returns information about current/last backup task.
-+#
-+# Returns: @BackupStatus
-+#
-+##
-+{ 'command': 'query-backup', 'returns': 'BackupStatus' }
-+
-+##
-+# @backup-cancel:
-+#
-+# Cancel the current executing backup process.
-+#
-+# Returns: nothing on success
-+#
-+# Notes: This command succeeds even if there is no backup process running.
-+#
-+##
-+{ 'command': 'backup-cancel' }
-+
- ##
- # @BlockDeviceTimedStats:
- #
-@@ -2835,7 +2926,7 @@
- 'qcow2', 'qed', 'quorum', 'raw', 'rbd',
- { 'name': 'replication', 'if': 'defined(CONFIG_REPLICATION)' },
- 'sheepdog',
-- 'ssh', 'throttle', 'vdi', 'vhdx', 'vmdk', 'vpc', 'vvfat', 'vxhs' ] }
-+ 'ssh', 'throttle', 'vdi', 'vhdx', 'vma-drive', 'vmdk', 'vpc', 'vvfat', 'vxhs' ] }
-
- ##
- # @BlockdevOptionsFile:
-@@ -3898,6 +3989,21 @@
- 'server': 'InetSocketAddressBase',
- '*tls-creds': 'str' } }
-
-+##
-+# @BlockdevOptionsVMADrive:
-+#
-+# Driver specific block device options for VMA Drives
-+#
-+# @filename: vma-drive path
-+#
-+# @size: drive size in bytes
-+#
-+# Since: 2.9
-+##
-+{ 'struct': 'BlockdevOptionsVMADrive',
-+ 'data': { 'filename': 'str',
-+ 'size': 'int' } }
-+
- ##
- # @BlockdevOptionsThrottle:
- #
-@@ -3993,6 +4099,7 @@
- 'throttle': 'BlockdevOptionsThrottle',
- 'vdi': 'BlockdevOptionsGenericFormat',
- 'vhdx': 'BlockdevOptionsGenericFormat',
-+ 'vma-drive': 'BlockdevOptionsVMADrive',
- 'vmdk': 'BlockdevOptionsGenericCOWFormat',
- 'vpc': 'BlockdevOptionsGenericFormat',
- 'vvfat': 'BlockdevOptionsVVFAT',
-diff --git a/qapi/common.json b/qapi/common.json
-index 99d313ef3b..bae0650c51 100644
---- a/qapi/common.json
-+++ b/qapi/common.json
-@@ -193,3 +193,16 @@
- 'ppc64', 'riscv32', 'riscv64', 's390x', 'sh4',
- 'sh4eb', 'sparc', 'sparc64', 'tricore', 'unicore32',
- 'x86_64', 'xtensa', 'xtensaeb' ] }
-+
-+##
-+# @UuidInfo:
-+#
-+# Guest UUID information (Universally Unique Identifier).
-+#
-+# @UUID: the UUID of the guest
-+#
-+# Since: 0.14.0
-+#
-+# Notes: If no UUID was specified for the guest, a null UUID is returned.
-+##
-+{ 'struct': 'UuidInfo', 'data': {'UUID': 'str'} }
-diff --git a/qapi/misc.json b/qapi/misc.json
-index 31029e3132..335035c04d 100644
---- a/qapi/misc.json
-+++ b/qapi/misc.json
-@@ -270,19 +270,6 @@
- ##
- { 'command': 'query-kvm', 'returns': 'KvmInfo' }
-
--##
--# @UuidInfo:
--#
--# Guest UUID information (Universally Unique Identifier).
--#
--# @UUID: the UUID of the guest
--#
--# Since: 0.14.0
--#
--# Notes: If no UUID was specified for the guest, a null UUID is returned.
--##
--{ 'struct': 'UuidInfo', 'data': {'UUID': 'str'} }
--
- ##
- # @query-uuid:
- #
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Wolfgang Bumiller <w.bumiller@proxmox.com>
+Date: Tue, 10 Mar 2020 12:55:18 +0100
+Subject: [PATCH 21/32] PVE: [Config] Revert "target-i386: disable LINT0 after
+ reset"
+
+This reverts commit b8eb5512fd8a115f164edbbe897cdf8884920ccb.
+
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ hw/intc/apic_common.c | 9 +++++++++
+ 1 file changed, 9 insertions(+)
+
+diff --git a/hw/intc/apic_common.c b/hw/intc/apic_common.c
+index 375cb6abe9..e7d479c7e9 100644
+--- a/hw/intc/apic_common.c
++++ b/hw/intc/apic_common.c
+@@ -259,6 +259,15 @@ static void apic_reset_common(DeviceState *dev)
+ info->vapic_base_update(s);
+
+ apic_init_reset(dev);
++
++ if (bsp) {
++ /*
++ * LINT0 delivery mode on CPU #0 is set to ExtInt at initialization
++ * time typically by BIOS, so PIC interrupt can be delivered to the
++ * processor when local APIC is enabled.
++ */
++ s->lvt[APIC_LVT_LINT0] = 0x700;
++ }
+ }
+
+ static const VMStateDescription vmstate_apic_common;
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Mon, 7 Aug 2017 08:51:16 +0200
-Subject: [PATCH] PVE: [Deprecated] adding old vma files
-
-TODO: Move to using a libvma block backend
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- Makefile | 3 +-
- Makefile.objs | 1 +
- block/backup.c | 88 ++--
- block/replication.c | 1 +
- blockdev.c | 208 +++++----
- include/block/block_int.h | 4 +
- job.c | 3 +-
- vma-reader.c | 857 ++++++++++++++++++++++++++++++++++++++
- vma-writer.c | 771 ++++++++++++++++++++++++++++++++++
- vma.c | 756 +++++++++++++++++++++++++++++++++
- vma.h | 150 +++++++
- 11 files changed, 2737 insertions(+), 105 deletions(-)
- create mode 100644 vma-reader.c
- create mode 100644 vma-writer.c
- create mode 100644 vma.c
- create mode 100644 vma.h
-
-diff --git a/Makefile b/Makefile
-index 85862fb81a..421e7b486c 100644
---- a/Makefile
-+++ b/Makefile
-@@ -436,7 +436,7 @@ dummy := $(call unnest-vars,, \
-
- include $(SRC_PATH)/tests/Makefile.include
-
--all: $(DOCS) $(if $(BUILD_DOCS),sphinxdocs) $(TOOLS) $(HELPERS-y) recurse-all modules $(vhost-user-json-y)
-+all: $(DOCS) $(if $(BUILD_DOCS),sphinxdocs) $(TOOLS) vma$(EXESUF) $(HELPERS-y) recurse-all modules $(vhost-user-json-y)
-
- qemu-version.h: FORCE
- $(call quiet-command, \
-@@ -544,6 +544,7 @@ qemu-img.o: qemu-img-cmds.h
- qemu-img$(EXESUF): qemu-img.o $(authz-obj-y) $(block-obj-y) $(crypto-obj-y) $(io-obj-y) $(qom-obj-y) $(COMMON_LDADDS)
- qemu-nbd$(EXESUF): qemu-nbd.o $(authz-obj-y) $(block-obj-y) $(crypto-obj-y) $(io-obj-y) $(qom-obj-y) $(COMMON_LDADDS)
- qemu-io$(EXESUF): qemu-io.o $(authz-obj-y) $(block-obj-y) $(crypto-obj-y) $(io-obj-y) $(qom-obj-y) $(COMMON_LDADDS)
-+vma$(EXESUF): vma.o vma-reader.o $(authz-obj-y) $(block-obj-y) $(crypto-obj-y) $(io-obj-y) $(qom-obj-y) $(COMMON_LDADDS)
-
- qemu-bridge-helper$(EXESUF): qemu-bridge-helper.o $(COMMON_LDADDS)
-
-diff --git a/Makefile.objs b/Makefile.objs
-index 21dd93b58c..87c6033bc1 100644
---- a/Makefile.objs
-+++ b/Makefile.objs
-@@ -18,6 +18,7 @@ block-obj-y += block.o blockjob.o job.o
- block-obj-y += block/ scsi/
- block-obj-y += qemu-io-cmds.o
- block-obj-$(CONFIG_REPLICATION) += replication.o
-+block-obj-y += vma-writer.o
-
- block-obj-m = block/
-
-diff --git a/block/backup.c b/block/backup.c
-index 30008fcc34..fdcfb0529c 100644
---- a/block/backup.c
-+++ b/block/backup.c
-@@ -41,6 +41,7 @@ typedef struct BackupBlockJob {
- /* bitmap for sync=incremental */
- BdrvDirtyBitmap *sync_bitmap;
- MirrorSyncMode sync_mode;
-+ BackupDumpFunc *dump_cb;
- BlockdevOnError on_source_error;
- BlockdevOnError on_target_error;
- CoRwlock flush_rwlock;
-@@ -129,12 +130,20 @@ static int coroutine_fn backup_cow_with_bounce_buffer(BackupBlockJob *job,
- }
-
- if (buffer_is_zero(*bounce_buffer, nbytes)) {
-- ret = blk_co_pwrite_zeroes(job->target, start,
-- nbytes, write_flags | BDRV_REQ_MAY_UNMAP);
-+ if (job->dump_cb) {
-+ ret = job->dump_cb(job->common.job.opaque, job->target, start, nbytes, NULL);
-+ } else {
-+ ret = blk_co_pwrite_zeroes(job->target, start,
-+ nbytes, write_flags | BDRV_REQ_MAY_UNMAP);
-+ }
- } else {
-- ret = blk_co_pwrite(job->target, start,
-- nbytes, *bounce_buffer, write_flags |
-- (job->compress ? BDRV_REQ_WRITE_COMPRESSED : 0));
-+ if (job->dump_cb) {
-+ ret = job->dump_cb(job->common.job.opaque, job->target, start, nbytes, *bounce_buffer);
-+ } else {
-+ ret = blk_co_pwrite(job->target, start,
-+ nbytes, *bounce_buffer, write_flags |
-+ (job->compress ? BDRV_REQ_WRITE_COMPRESSED : 0));
-+ }
- }
- if (ret < 0) {
- trace_backup_do_cow_write_fail(job, start, ret);
-@@ -218,8 +227,11 @@ static int coroutine_fn backup_do_cow(BackupBlockJob *job,
- trace_backup_do_cow_process(job, start);
-
- if (job->use_copy_range) {
-- ret = backup_cow_with_offload(job, start, dirty_end,
-- is_write_notifier);
-+ if (job->dump_cb) {
-+ ret = - 1;
-+ } else {
-+ ret = backup_cow_with_offload(job, start, dirty_end, is_write_notifier);
-+ }
- if (ret < 0) {
- job->use_copy_range = false;
- }
-@@ -304,7 +316,9 @@ static void backup_abort(Job *job)
- static void backup_clean(Job *job)
- {
- BackupBlockJob *s = container_of(job, BackupBlockJob, common.job);
-- assert(s->target);
-+ if (!s->target) {
-+ return;
-+ }
- blk_unref(s->target);
- s->target = NULL;
-
-@@ -350,9 +364,11 @@ static BlockErrorAction backup_error_action(BackupBlockJob *job,
- if (read) {
- return block_job_error_action(&job->common, job->on_source_error,
- true, error);
-- } else {
-+ } else if (job->target) {
- return block_job_error_action(&job->common, job->on_target_error,
- false, error);
-+ } else {
-+ return BLOCK_ERROR_ACTION_REPORT;
- }
- }
-
-@@ -549,6 +565,7 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
- BlockdevOnError on_source_error,
- BlockdevOnError on_target_error,
- int creation_flags,
-+ BackupDumpFunc *dump_cb,
- BlockCompletionFunc *cb, void *opaque,
- int pause_count,
- JobTxn *txn, Error **errp)
-@@ -560,7 +577,7 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
- HBitmap *copy_bitmap = NULL;
-
- assert(bs);
-- assert(target);
-+ assert(target || dump_cb);
-
- if (bs == target) {
- error_setg(errp, "Source and target cannot be the same");
-@@ -573,13 +590,13 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
- return NULL;
- }
-
-- if (!bdrv_is_inserted(target)) {
-+ if (target && !bdrv_is_inserted(target)) {
- error_setg(errp, "Device is not inserted: %s",
- bdrv_get_device_name(target));
- return NULL;
- }
-
-- if (compress && target->drv->bdrv_co_pwritev_compressed == NULL) {
-+ if (target && compress && target->drv->bdrv_co_pwritev_compressed == NULL) {
- error_setg(errp, "Compression is not supported for this drive %s",
- bdrv_get_device_name(target));
- return NULL;
-@@ -589,7 +606,7 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
- return NULL;
- }
-
-- if (bdrv_op_is_blocked(target, BLOCK_OP_TYPE_BACKUP_TARGET, errp)) {
-+ if (target && bdrv_op_is_blocked(target, BLOCK_OP_TYPE_BACKUP_TARGET, errp)) {
- return NULL;
- }
-
-@@ -619,7 +636,7 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
- goto error;
- }
-
-- cluster_size = backup_calculate_cluster_size(target, errp);
-+ cluster_size = backup_calculate_cluster_size(target ? target : bs, errp);
- if (cluster_size < 0) {
- goto error;
- }
-@@ -636,16 +653,19 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
- goto error;
- }
-
-- /* The target must match the source in size, so no resize here either */
-- job->target = blk_new(job->common.job.aio_context,
-- BLK_PERM_WRITE,
-- BLK_PERM_CONSISTENT_READ | BLK_PERM_WRITE |
-- BLK_PERM_WRITE_UNCHANGED | BLK_PERM_GRAPH_MOD);
-- ret = blk_insert_bs(job->target, target, errp);
-- if (ret < 0) {
-- goto error;
-+ if (target) {
-+ /* The target must match the source in size, so no resize here either */
-+ job->target = blk_new(job->common.job.aio_context,
-+ BLK_PERM_WRITE,
-+ BLK_PERM_CONSISTENT_READ | BLK_PERM_WRITE |
-+ BLK_PERM_WRITE_UNCHANGED | BLK_PERM_GRAPH_MOD);
-+ ret = blk_insert_bs(job->target, target, errp);
-+ if (ret < 0) {
-+ goto error;
-+ }
- }
-
-+ job->dump_cb = dump_cb;
- job->on_source_error = on_source_error;
- job->on_target_error = on_target_error;
- job->sync_mode = sync_mode;
-@@ -658,10 +678,14 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
- job->cluster_size = cluster_size;
- job->copy_bitmap = copy_bitmap;
- copy_bitmap = NULL;
-- job->copy_range_size = MIN_NON_ZERO(blk_get_max_transfer(job->common.blk),
-- blk_get_max_transfer(job->target));
-- job->copy_range_size = QEMU_ALIGN_DOWN(job->copy_range_size,
-- job->cluster_size);
-+
-+ if (target) {
-+ job->copy_range_size = MIN_NON_ZERO(blk_get_max_transfer(job->common.blk),
-+ blk_get_max_transfer(job->target));
-+ job->copy_range_size = QEMU_ALIGN_DOWN(job->copy_range_size,
-+ job->cluster_size);
-+ }
-+
- /*
- * Set use_copy_range, consider the following:
- * 1. Compression is not supported for copy_range.
-@@ -669,12 +693,16 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
- * that in here. If max_transfer is smaller than the job->cluster_size,
- * we do not use copy_range (in that case it's zero after aligning down
- * above).
-+ * 3. If !target, we're using PVE dump_cb callback
- */
-- job->use_copy_range = !compress && job->copy_range_size > 0;
-+ job->use_copy_range = target && !compress && job->copy_range_size > 0;
-+
-+ if (target) {
-+ /* Required permissions are already taken with target's blk_new() */
-+ block_job_add_bdrv(&job->common, "target", target, 0, BLK_PERM_ALL,
-+ &error_abort);
-+ }
-
-- /* Required permissions are already taken with target's blk_new() */
-- block_job_add_bdrv(&job->common, "target", target, 0, BLK_PERM_ALL,
-- &error_abort);
- job->len = len;
- job->common.job.pause_count += pause_count;
-
-diff --git a/block/replication.c b/block/replication.c
-index e70a6cf2bd..f060755713 100644
---- a/block/replication.c
-+++ b/block/replication.c
-@@ -546,6 +546,7 @@ static void replication_start(ReplicationState *rs, ReplicationMode mode,
- 0, MIRROR_SYNC_MODE_NONE, NULL, false,
- BLOCKDEV_ON_ERROR_REPORT,
- BLOCKDEV_ON_ERROR_REPORT, JOB_INTERNAL,
-+ NULL,
- backup_job_completed, bs, 0, NULL, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
-diff --git a/blockdev.c b/blockdev.c
-index 7047475a3c..cee7952bbb 100644
---- a/blockdev.c
-+++ b/blockdev.c
-@@ -31,7 +31,6 @@
- */
-
- #include "qemu/osdep.h"
--#include "qemu/uuid.h"
- #include "sysemu/block-backend.h"
- #include "sysemu/blockdev.h"
- #include "hw/block/block.h"
-@@ -64,6 +63,7 @@
- #include "qemu/cutils.h"
- #include "qemu/help_option.h"
- #include "qemu/throttle-options.h"
-+#include "vma.h"
-
- static QTAILQ_HEAD(, BlockDriverState) monitor_bdrv_states =
- QTAILQ_HEAD_INITIALIZER(monitor_bdrv_states);
-@@ -3176,15 +3176,14 @@ out:
- static struct PVEBackupState {
- Error *error;
- bool cancel;
-- QemuUUID uuid;
-+ uuid_t uuid;
- char uuid_str[37];
- int64_t speed;
- time_t start_time;
- time_t end_time;
- char *backup_file;
-- Object *vmaobj;
-+ VmaWriter *vmaw;
- GList *di_list;
-- size_t next_job;
- size_t total;
- size_t transferred;
- size_t zero_bytes;
-@@ -3203,6 +3202,71 @@ typedef struct PVEBackupDevInfo {
-
- static void pvebackup_run_next_job(void);
-
-+static int pvebackup_dump_cb(void *opaque, BlockBackend *target,
-+ uint64_t start, uint64_t bytes,
-+ const void *pbuf)
-+{
-+ const uint64_t size = bytes;
-+ const unsigned char *buf = pbuf;
-+ PVEBackupDevInfo *di = opaque;
-+
-+ if (backup_state.cancel) {
-+ return size; // return success
-+ }
-+
-+ uint64_t cluster_num = start / VMA_CLUSTER_SIZE;
-+ if ((cluster_num * VMA_CLUSTER_SIZE) != start) {
-+ if (!backup_state.error) {
-+ error_setg(&backup_state.error,
-+ "got unaligned write inside backup dump "
-+ "callback (sector %ld)", start);
-+ }
-+ return -1; // not aligned to cluster size
-+ }
-+
-+ int ret = -1;
-+
-+ if (backup_state.vmaw) {
-+ size_t zero_bytes = 0;
-+ uint64_t remaining = size;
-+ while (remaining > 0) {
-+ ret = vma_writer_write(backup_state.vmaw, di->dev_id, cluster_num,
-+ buf, &zero_bytes);
-+ ++cluster_num;
-+ if (buf) {
-+ buf += VMA_CLUSTER_SIZE;
-+ }
-+ if (ret < 0) {
-+ if (!backup_state.error) {
-+ vma_writer_error_propagate(backup_state.vmaw, &backup_state.error);
-+ }
-+ if (di->bs && di->bs->job) {
-+ job_cancel(&di->bs->job->job, true);
-+ }
-+ break;
-+ } else {
-+ backup_state.zero_bytes += zero_bytes;
-+ if (remaining >= VMA_CLUSTER_SIZE) {
-+ backup_state.transferred += VMA_CLUSTER_SIZE;
-+ remaining -= VMA_CLUSTER_SIZE;
-+ } else {
-+ backup_state.transferred += remaining;
-+ remaining = 0;
-+ }
-+ }
-+ }
-+ } else {
-+ if (!buf) {
-+ backup_state.zero_bytes += size;
-+ }
-+ backup_state.transferred += size;
-+ }
-+
-+ // Note: always return success, because we want that writes succeed anyways.
-+
-+ return size;
-+}
-+
- static void pvebackup_cleanup(void)
- {
- qemu_mutex_lock(&backup_state.backup_mutex);
-@@ -3214,9 +3278,11 @@ static void pvebackup_cleanup(void)
-
- backup_state.end_time = time(NULL);
-
-- if (backup_state.vmaobj) {
-- object_unparent(backup_state.vmaobj);
-- backup_state.vmaobj = NULL;
-+ if (backup_state.vmaw) {
-+ Error *local_err = NULL;
-+ vma_writer_close(backup_state.vmaw, &local_err);
-+ error_propagate(&backup_state.error, local_err);
-+ backup_state.vmaw = NULL;
- }
-
- g_list_free(backup_state.di_list);
-@@ -3224,6 +3290,13 @@ static void pvebackup_cleanup(void)
- qemu_mutex_unlock(&backup_state.backup_mutex);
- }
-
-+static void coroutine_fn backup_close_vma_stream(void *opaque)
-+{
-+ PVEBackupDevInfo *di = opaque;
-+
-+ vma_writer_close_stream(backup_state.vmaw, di->dev_id);
-+}
-+
- static void pvebackup_complete_cb(void *opaque, int ret)
- {
- // This always runs in the main loop
-@@ -3240,9 +3313,9 @@ static void pvebackup_complete_cb(void *opaque, int ret)
- di->bs = NULL;
- di->target = NULL;
-
-- if (backup_state.vmaobj) {
-- object_unparent(backup_state.vmaobj);
-- backup_state.vmaobj = NULL;
-+ if (backup_state.vmaw) {
-+ Coroutine *co = qemu_coroutine_create(backup_close_vma_stream, di);
-+ qemu_coroutine_enter(co);
- }
-
- // remove self from job queue
-@@ -3270,14 +3343,9 @@ static void pvebackup_cancel(void *opaque)
- error_setg(&backup_state.error, "backup cancelled");
- }
-
-- if (backup_state.vmaobj) {
-- Error *err;
-+ if (backup_state.vmaw) {
- /* make sure vma writer does not block anymore */
-- if (!object_set_props(backup_state.vmaobj, &err, "blocked", "yes", NULL)) {
-- if (err) {
-- error_report_err(err);
-- }
-- }
-+ vma_writer_set_error(backup_state.vmaw, "backup cancelled");
- }
-
- GList *l = backup_state.di_list;
-@@ -3308,18 +3376,14 @@ void qmp_backup_cancel(Error **errp)
- Coroutine *co = qemu_coroutine_create(pvebackup_cancel, NULL);
- qemu_coroutine_enter(co);
-
-- while (backup_state.vmaobj) {
-- /* FIXME: Find something better for this */
-+ while (backup_state.vmaw) {
-+ /* vma writer use main aio context */
- aio_poll(qemu_get_aio_context(), true);
- }
- }
-
--void vma_object_add_config_file(Object *obj, const char *name,
-- const char *contents, size_t len,
-- Error **errp);
- static int config_to_vma(const char *file, BackupFormat format,
-- Object *vmaobj,
-- const char *backup_dir,
-+ const char *backup_dir, VmaWriter *vmaw,
- Error **errp)
- {
- char *cdata = NULL;
-@@ -3333,7 +3397,12 @@ static int config_to_vma(const char *file, BackupFormat format,
- char *basename = g_path_get_basename(file);
-
- if (format == BACKUP_FORMAT_VMA) {
-- vma_object_add_config_file(vmaobj, basename, cdata, clen, errp);
-+ if (vma_writer_add_config(vmaw, basename, cdata, clen) != 0) {
-+ error_setg(errp, "unable to add %s config data to vma archive", file);
-+ g_free(cdata);
-+ g_free(basename);
-+ return 1;
-+ }
- } else if (format == BACKUP_FORMAT_DIR) {
- char config_path[PATH_MAX];
- snprintf(config_path, PATH_MAX, "%s/%s", backup_dir, basename);
-@@ -3350,28 +3419,30 @@ static int config_to_vma(const char *file, BackupFormat format,
- return 0;
- }
-
-+bool job_should_pause(Job *job);
- static void pvebackup_run_next_job(void)
- {
- qemu_mutex_lock(&backup_state.backup_mutex);
-
-- GList *next = g_list_nth(backup_state.di_list, backup_state.next_job);
-- while (next) {
-- PVEBackupDevInfo *di = (PVEBackupDevInfo *)next->data;
-- backup_state.next_job++;
-+ GList *l = backup_state.di_list;
-+ while (l) {
-+ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
-+ l = g_list_next(l);
- if (!di->completed && di->bs && di->bs->job) {
- BlockJob *job = di->bs->job;
- AioContext *aio_context = blk_get_aio_context(job->blk);
- aio_context_acquire(aio_context);
- qemu_mutex_unlock(&backup_state.backup_mutex);
-- if (backup_state.error || backup_state.cancel) {
-- job_cancel_sync(job);
-- } else {
-- job_resume(job);
-+ if (job_should_pause(&job->job)) {
-+ if (backup_state.error || backup_state.cancel) {
-+ job_cancel_sync(&job->job);
-+ } else {
-+ job_resume(&job->job);
-+ }
- }
- aio_context_release(aio_context);
- return;
- }
-- next = g_list_next(next);
- }
- qemu_mutex_unlock(&backup_state.backup_mutex);
-
-@@ -3382,7 +3453,7 @@ static void pvebackup_run_next_job(void)
- UuidInfo *qmp_backup(const char *backup_file, bool has_format,
- BackupFormat format,
- bool has_config_file, const char *config_file,
-- bool has_firewall_file, const char *firewall_file,
-+ bool has_firewall_file, const char *firewall_file,
- bool has_devlist, const char *devlist,
- bool has_speed, int64_t speed, Error **errp)
- {
-@@ -3390,7 +3461,8 @@ UuidInfo *qmp_backup(const char *backup_file, bool has_format,
- BlockDriverState *bs = NULL;
- const char *backup_dir = NULL;
- Error *local_err = NULL;
-- QemuUUID uuid;
-+ uuid_t uuid;
-+ VmaWriter *vmaw = NULL;
- gchar **devs = NULL;
- GList *di_list = NULL;
- GList *l;
-@@ -3402,7 +3474,7 @@ UuidInfo *qmp_backup(const char *backup_file, bool has_format,
- backup_state.backup_mutex_initialized = true;
- }
-
-- if (backup_state.di_list || backup_state.vmaobj) {
-+ if (backup_state.di_list) {
- error_set(errp, ERROR_CLASS_GENERIC_ERROR,
- "previous backup not finished");
- return NULL;
-@@ -3477,40 +3549,28 @@ UuidInfo *qmp_backup(const char *backup_file, bool has_format,
- total += size;
- }
-
-- qemu_uuid_generate(&uuid);
-+ uuid_generate(uuid);
-
- if (format == BACKUP_FORMAT_VMA) {
-- char uuidstr[UUID_FMT_LEN+1];
-- qemu_uuid_unparse(&uuid, uuidstr);
-- uuidstr[UUID_FMT_LEN] = 0;
-- backup_state.vmaobj =
-- object_new_with_props("vma", object_get_objects_root(),
-- "vma-backup-obj", &local_err,
-- "filename", backup_file,
-- "uuid", uuidstr,
-- NULL);
-- if (!backup_state.vmaobj) {
-+ vmaw = vma_writer_create(backup_file, uuid, &local_err);
-+ if (!vmaw) {
- if (local_err) {
- error_propagate(errp, local_err);
- }
- goto err;
- }
-
-+ /* register all devices for vma writer */
- l = di_list;
- while (l) {
-- QDict *options = qdict_new();
--
- PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
- l = g_list_next(l);
-
- const char *devname = bdrv_get_device_name(di->bs);
-- snprintf(di->targetfile, PATH_MAX, "vma-backup-obj/%s.raw", devname);
--
-- qdict_put(options, "driver", qstring_from_str("vma-drive"));
-- qdict_put(options, "size", qint_from_int(di->size));
-- di->target = bdrv_open(di->targetfile, NULL, options, BDRV_O_RDWR, &local_err);
-- if (!di->target) {
-- error_propagate(errp, local_err);
-+ di->dev_id = vma_writer_register_stream(vmaw, devname, di->size);
-+ if (di->dev_id <= 0) {
-+ error_set(errp, ERROR_CLASS_GENERIC_ERROR,
-+ "register_stream failed");
- goto err;
- }
- }
-@@ -3551,14 +3611,14 @@ UuidInfo *qmp_backup(const char *backup_file, bool has_format,
-
- /* add configuration file to archive */
- if (has_config_file) {
-- if(config_to_vma(config_file, format, backup_state.vmaobj, backup_dir, errp) != 0) {
-+ if (config_to_vma(config_file, format, backup_dir, vmaw, errp) != 0) {
- goto err;
- }
- }
-
- /* add firewall file to archive */
- if (has_firewall_file) {
-- if(config_to_vma(firewall_file, format, backup_state.vmaobj, backup_dir, errp) != 0) {
-+ if (config_to_vma(firewall_file, format, backup_dir, vmaw, errp) != 0) {
- goto err;
- }
- }
-@@ -3581,12 +3641,13 @@ UuidInfo *qmp_backup(const char *backup_file, bool has_format,
- }
- backup_state.backup_file = g_strdup(backup_file);
-
-- memcpy(&backup_state.uuid, &uuid, sizeof(uuid));
-- qemu_uuid_unparse(&uuid, backup_state.uuid_str);
-+ backup_state.vmaw = vmaw;
-+
-+ uuid_copy(backup_state.uuid, uuid);
-+ uuid_unparse_lower(uuid, backup_state.uuid_str);
-
- qemu_mutex_lock(&backup_state.backup_mutex);
- backup_state.di_list = di_list;
-- backup_state.next_job = 0;
-
- backup_state.total = total;
- backup_state.transferred = 0;
-@@ -3597,21 +3658,21 @@ UuidInfo *qmp_backup(const char *backup_file, bool has_format,
- while (l) {
- PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
- l = g_list_next(l);
--
- job = backup_job_create(NULL, di->bs, di->target, speed, MIRROR_SYNC_MODE_FULL, NULL,
- false, BLOCKDEV_ON_ERROR_REPORT, BLOCKDEV_ON_ERROR_REPORT,
- JOB_DEFAULT,
-- pvebackup_complete_cb, di, 2, NULL, &local_err);
-- if (di->target) {
-- bdrv_unref(di->target);
-- di->target = NULL;
-- }
-+ pvebackup_dump_cb, pvebackup_complete_cb, di,
-+ 1, NULL, &local_err);
- if (!job || local_err != NULL) {
- error_setg(&backup_state.error, "backup_job_create failed");
- pvebackup_cancel(NULL);
- } else {
- job_start(&job->job);
- }
-+ if (di->target) {
-+ bdrv_unref(di->target);
-+ di->target = NULL;
-+ }
- }
-
- qemu_mutex_unlock(&backup_state.backup_mutex);
-@@ -3647,9 +3708,10 @@ err:
- g_strfreev(devs);
- }
-
-- if (backup_state.vmaobj) {
-- object_unparent(backup_state.vmaobj);
-- backup_state.vmaobj = NULL;
-+ if (vmaw) {
-+ Error *err = NULL;
-+ vma_writer_close(vmaw, &err);
-+ unlink(backup_file);
- }
-
- if (backup_dir) {
-@@ -4110,7 +4172,7 @@ static BlockJob *do_drive_backup(DriveBackup *backup, JobTxn *txn,
- job = backup_job_create(backup->job_id, bs, target_bs, backup->speed,
- backup->sync, bmap, backup->compress,
- backup->on_source_error, backup->on_target_error,
-- job_flags, NULL, NULL, 0, txn, &local_err);
-+ job_flags, NULL, NULL, NULL, 0, txn, &local_err);
- if (local_err != NULL) {
- error_propagate(errp, local_err);
- goto unref;
-@@ -4215,7 +4277,7 @@ BlockJob *do_blockdev_backup(BlockdevBackup *backup, JobTxn *txn,
- job = backup_job_create(backup->job_id, bs, target_bs, backup->speed,
- backup->sync, bmap, backup->compress,
- backup->on_source_error, backup->on_target_error,
-- job_flags, NULL, NULL, 0, txn, &local_err);
-+ job_flags, NULL, NULL, NULL, 0, txn, &local_err);
- if (local_err != NULL) {
- error_propagate(errp, local_err);
- }
-diff --git a/include/block/block_int.h b/include/block/block_int.h
-index bb2dddca83..5a8b2e06c1 100644
---- a/include/block/block_int.h
-+++ b/include/block/block_int.h
-@@ -61,6 +61,9 @@
-
- #define BLOCK_PROBE_BUF_SIZE 512
-
-+typedef int BackupDumpFunc(void *opaque, BlockBackend *be,
-+ uint64_t offset, uint64_t bytes, const void *buf);
-+
- enum BdrvTrackedRequestType {
- BDRV_TRACKED_READ,
- BDRV_TRACKED_WRITE,
-@@ -1172,6 +1175,7 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
- BlockdevOnError on_source_error,
- BlockdevOnError on_target_error,
- int creation_flags,
-+ BackupDumpFunc *dump_cb,
- BlockCompletionFunc *cb, void *opaque,
- int pause_count,
- JobTxn *txn, Error **errp);
-diff --git a/job.c b/job.c
-index 7a21e83780..9d27999678 100644
---- a/job.c
-+++ b/job.c
-@@ -248,7 +248,8 @@ static bool job_started(Job *job)
- return job->co;
- }
-
--static bool job_should_pause(Job *job)
-+bool job_should_pause(Job *job);
-+bool job_should_pause(Job *job)
- {
- return job->pause_count > 0;
- }
-diff --git a/vma-reader.c b/vma-reader.c
-new file mode 100644
-index 0000000000..2b1d1cdab3
---- /dev/null
-+++ b/vma-reader.c
-@@ -0,0 +1,857 @@
-+/*
-+ * VMA: Virtual Machine Archive
-+ *
-+ * Copyright (C) 2012 Proxmox Server Solutions
-+ *
-+ * Authors:
-+ * Dietmar Maurer (dietmar@proxmox.com)
-+ *
-+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
-+ * See the COPYING file in the top-level directory.
-+ *
-+ */
-+
-+#include "qemu/osdep.h"
-+#include <glib.h>
-+#include <uuid/uuid.h>
-+
-+#include "qemu-common.h"
-+#include "qemu/timer.h"
-+#include "qemu/ratelimit.h"
-+#include "vma.h"
-+#include "block/block.h"
-+#include "sysemu/block-backend.h"
-+
-+static unsigned char zero_vma_block[VMA_BLOCK_SIZE];
-+
-+typedef struct VmaRestoreState {
-+ BlockBackend *target;
-+ bool write_zeroes;
-+ unsigned long *bitmap;
-+ int bitmap_size;
-+} VmaRestoreState;
-+
-+struct VmaReader {
-+ int fd;
-+ GChecksum *md5csum;
-+ GHashTable *blob_hash;
-+ unsigned char *head_data;
-+ VmaDeviceInfo devinfo[256];
-+ VmaRestoreState rstate[256];
-+ GList *cdata_list;
-+ guint8 vmstate_stream;
-+ uint32_t vmstate_clusters;
-+ /* to show restore percentage if run with -v */
-+ time_t start_time;
-+ int64_t cluster_count;
-+ int64_t clusters_read;
-+ int64_t zero_cluster_data;
-+ int64_t partial_zero_cluster_data;
-+ int clusters_read_per;
-+};
-+
-+static guint
-+g_int32_hash(gconstpointer v)
-+{
-+ return *(const uint32_t *)v;
-+}
-+
-+static gboolean
-+g_int32_equal(gconstpointer v1, gconstpointer v2)
-+{
-+ return *((const uint32_t *)v1) == *((const uint32_t *)v2);
-+}
-+
-+static int vma_reader_get_bitmap(VmaRestoreState *rstate, int64_t cluster_num)
-+{
-+ assert(rstate);
-+ assert(rstate->bitmap);
-+
-+ unsigned long val, idx, bit;
-+
-+ idx = cluster_num / BITS_PER_LONG;
-+
-+ assert(rstate->bitmap_size > idx);
-+
-+ bit = cluster_num % BITS_PER_LONG;
-+ val = rstate->bitmap[idx];
-+
-+ return !!(val & (1UL << bit));
-+}
-+
-+static void vma_reader_set_bitmap(VmaRestoreState *rstate, int64_t cluster_num,
-+ int dirty)
-+{
-+ assert(rstate);
-+ assert(rstate->bitmap);
-+
-+ unsigned long val, idx, bit;
-+
-+ idx = cluster_num / BITS_PER_LONG;
-+
-+ assert(rstate->bitmap_size > idx);
-+
-+ bit = cluster_num % BITS_PER_LONG;
-+ val = rstate->bitmap[idx];
-+ if (dirty) {
-+ if (!(val & (1UL << bit))) {
-+ val |= 1UL << bit;
-+ }
-+ } else {
-+ if (val & (1UL << bit)) {
-+ val &= ~(1UL << bit);
-+ }
-+ }
-+ rstate->bitmap[idx] = val;
-+}
-+
-+typedef struct VmaBlob {
-+ uint32_t start;
-+ uint32_t len;
-+ void *data;
-+} VmaBlob;
-+
-+static const VmaBlob *get_header_blob(VmaReader *vmar, uint32_t pos)
-+{
-+ assert(vmar);
-+ assert(vmar->blob_hash);
-+
-+ return g_hash_table_lookup(vmar->blob_hash, &pos);
-+}
-+
-+static const char *get_header_str(VmaReader *vmar, uint32_t pos)
-+{
-+ const VmaBlob *blob = get_header_blob(vmar, pos);
-+ if (!blob) {
-+ return NULL;
-+ }
-+ const char *res = (char *)blob->data;
-+ if (res[blob->len-1] != '\0') {
-+ return NULL;
-+ }
-+ return res;
-+}
-+
-+static ssize_t
-+safe_read(int fd, unsigned char *buf, size_t count)
-+{
-+ ssize_t n;
-+
-+ do {
-+ n = read(fd, buf, count);
-+ } while (n < 0 && errno == EINTR);
-+
-+ return n;
-+}
-+
-+static ssize_t
-+full_read(int fd, unsigned char *buf, size_t len)
-+{
-+ ssize_t n;
-+ size_t total;
-+
-+ total = 0;
-+
-+ while (len > 0) {
-+ n = safe_read(fd, buf, len);
-+
-+ if (n == 0) {
-+ return total;
-+ }
-+
-+ if (n <= 0) {
-+ break;
-+ }
-+
-+ buf += n;
-+ total += n;
-+ len -= n;
-+ }
-+
-+ if (len) {
-+ return -1;
-+ }
-+
-+ return total;
-+}
-+
-+void vma_reader_destroy(VmaReader *vmar)
-+{
-+ assert(vmar);
-+
-+ if (vmar->fd >= 0) {
-+ close(vmar->fd);
-+ }
-+
-+ if (vmar->cdata_list) {
-+ g_list_free(vmar->cdata_list);
-+ }
-+
-+ int i;
-+ for (i = 1; i < 256; i++) {
-+ if (vmar->rstate[i].bitmap) {
-+ g_free(vmar->rstate[i].bitmap);
-+ }
-+ }
-+
-+ if (vmar->md5csum) {
-+ g_checksum_free(vmar->md5csum);
-+ }
-+
-+ if (vmar->blob_hash) {
-+ g_hash_table_destroy(vmar->blob_hash);
-+ }
-+
-+ if (vmar->head_data) {
-+ g_free(vmar->head_data);
-+ }
-+
-+ g_free(vmar);
-+
-+};
-+
-+static int vma_reader_read_head(VmaReader *vmar, Error **errp)
-+{
-+ assert(vmar);
-+ assert(errp);
-+ assert(*errp == NULL);
-+
-+ unsigned char md5sum[16];
-+ int i;
-+ int ret = 0;
-+
-+ vmar->head_data = g_malloc(sizeof(VmaHeader));
-+
-+ if (full_read(vmar->fd, vmar->head_data, sizeof(VmaHeader)) !=
-+ sizeof(VmaHeader)) {
-+ error_setg(errp, "can't read vma header - %s",
-+ errno ? g_strerror(errno) : "got EOF");
-+ return -1;
-+ }
-+
-+ VmaHeader *h = (VmaHeader *)vmar->head_data;
-+
-+ if (h->magic != VMA_MAGIC) {
-+ error_setg(errp, "not a vma file - wrong magic number");
-+ return -1;
-+ }
-+
-+ uint32_t header_size = GUINT32_FROM_BE(h->header_size);
-+ int need = header_size - sizeof(VmaHeader);
-+ if (need <= 0) {
-+ error_setg(errp, "wrong vma header size %d", header_size);
-+ return -1;
-+ }
-+
-+ vmar->head_data = g_realloc(vmar->head_data, header_size);
-+ h = (VmaHeader *)vmar->head_data;
-+
-+ if (full_read(vmar->fd, vmar->head_data + sizeof(VmaHeader), need) !=
-+ need) {
-+ error_setg(errp, "can't read vma header data - %s",
-+ errno ? g_strerror(errno) : "got EOF");
-+ return -1;
-+ }
-+
-+ memcpy(md5sum, h->md5sum, 16);
-+ memset(h->md5sum, 0, 16);
-+
-+ g_checksum_reset(vmar->md5csum);
-+ g_checksum_update(vmar->md5csum, vmar->head_data, header_size);
-+ gsize csize = 16;
-+ g_checksum_get_digest(vmar->md5csum, (guint8 *)(h->md5sum), &csize);
-+
-+ if (memcmp(md5sum, h->md5sum, 16) != 0) {
-+ error_setg(errp, "wrong vma header chechsum");
-+ return -1;
-+ }
-+
-+ /* we can modify header data after checksum verify */
-+ h->header_size = header_size;
-+
-+ h->version = GUINT32_FROM_BE(h->version);
-+ if (h->version != 1) {
-+ error_setg(errp, "wrong vma version %d", h->version);
-+ return -1;
-+ }
-+
-+ h->ctime = GUINT64_FROM_BE(h->ctime);
-+ h->blob_buffer_offset = GUINT32_FROM_BE(h->blob_buffer_offset);
-+ h->blob_buffer_size = GUINT32_FROM_BE(h->blob_buffer_size);
-+
-+ uint32_t bstart = h->blob_buffer_offset + 1;
-+ uint32_t bend = h->blob_buffer_offset + h->blob_buffer_size;
-+
-+ if (bstart <= sizeof(VmaHeader)) {
-+ error_setg(errp, "wrong vma blob buffer offset %d",
-+ h->blob_buffer_offset);
-+ return -1;
-+ }
-+
-+ if (bend > header_size) {
-+ error_setg(errp, "wrong vma blob buffer size %d/%d",
-+ h->blob_buffer_offset, h->blob_buffer_size);
-+ return -1;
-+ }
-+
-+ while ((bstart + 2) <= bend) {
-+ uint32_t size = vmar->head_data[bstart] +
-+ (vmar->head_data[bstart+1] << 8);
-+ if ((bstart + size + 2) <= bend) {
-+ VmaBlob *blob = g_new0(VmaBlob, 1);
-+ blob->start = bstart - h->blob_buffer_offset;
-+ blob->len = size;
-+ blob->data = vmar->head_data + bstart + 2;
-+ g_hash_table_insert(vmar->blob_hash, &blob->start, blob);
-+ }
-+ bstart += size + 2;
-+ }
-+
-+
-+ int count = 0;
-+ for (i = 1; i < 256; i++) {
-+ VmaDeviceInfoHeader *dih = &h->dev_info[i];
-+ uint32_t devname_ptr = GUINT32_FROM_BE(dih->devname_ptr);
-+ uint64_t size = GUINT64_FROM_BE(dih->size);
-+ const char *devname = get_header_str(vmar, devname_ptr);
-+
-+ if (size && devname) {
-+ count++;
-+ vmar->devinfo[i].size = size;
-+ vmar->devinfo[i].devname = devname;
-+
-+ if (strcmp(devname, "vmstate") == 0) {
-+ vmar->vmstate_stream = i;
-+ }
-+ }
-+ }
-+
-+ for (i = 0; i < VMA_MAX_CONFIGS; i++) {
-+ uint32_t name_ptr = GUINT32_FROM_BE(h->config_names[i]);
-+ uint32_t data_ptr = GUINT32_FROM_BE(h->config_data[i]);
-+
-+ if (!(name_ptr && data_ptr)) {
-+ continue;
-+ }
-+ const char *name = get_header_str(vmar, name_ptr);
-+ const VmaBlob *blob = get_header_blob(vmar, data_ptr);
-+
-+ if (!(name && blob)) {
-+ error_setg(errp, "vma contains invalid data pointers");
-+ return -1;
-+ }
-+
-+ VmaConfigData *cdata = g_new0(VmaConfigData, 1);
-+ cdata->name = name;
-+ cdata->data = blob->data;
-+ cdata->len = blob->len;
-+
-+ vmar->cdata_list = g_list_append(vmar->cdata_list, cdata);
-+ }
-+
-+ return ret;
-+};
-+
-+VmaReader *vma_reader_create(const char *filename, Error **errp)
-+{
-+ assert(filename);
-+ assert(errp);
-+
-+ VmaReader *vmar = g_new0(VmaReader, 1);
-+
-+ if (strcmp(filename, "-") == 0) {
-+ vmar->fd = dup(0);
-+ } else {
-+ vmar->fd = open(filename, O_RDONLY);
-+ }
-+
-+ if (vmar->fd < 0) {
-+ error_setg(errp, "can't open file %s - %s\n", filename,
-+ g_strerror(errno));
-+ goto err;
-+ }
-+
-+ vmar->md5csum = g_checksum_new(G_CHECKSUM_MD5);
-+ if (!vmar->md5csum) {
-+ error_setg(errp, "can't allocate cmsum\n");
-+ goto err;
-+ }
-+
-+ vmar->blob_hash = g_hash_table_new_full(g_int32_hash, g_int32_equal,
-+ NULL, g_free);
-+
-+ if (vma_reader_read_head(vmar, errp) < 0) {
-+ goto err;
-+ }
-+
-+ return vmar;
-+
-+err:
-+ if (vmar) {
-+ vma_reader_destroy(vmar);
-+ }
-+
-+ return NULL;
-+}
-+
-+VmaHeader *vma_reader_get_header(VmaReader *vmar)
-+{
-+ assert(vmar);
-+ assert(vmar->head_data);
-+
-+ return (VmaHeader *)(vmar->head_data);
-+}
-+
-+GList *vma_reader_get_config_data(VmaReader *vmar)
-+{
-+ assert(vmar);
-+ assert(vmar->head_data);
-+
-+ return vmar->cdata_list;
-+}
-+
-+VmaDeviceInfo *vma_reader_get_device_info(VmaReader *vmar, guint8 dev_id)
-+{
-+ assert(vmar);
-+ assert(dev_id);
-+
-+ if (vmar->devinfo[dev_id].size && vmar->devinfo[dev_id].devname) {
-+ return &vmar->devinfo[dev_id];
-+ }
-+
-+ return NULL;
-+}
-+
-+static void allocate_rstate(VmaReader *vmar, guint8 dev_id,
-+ BlockBackend *target, bool write_zeroes)
-+{
-+ assert(vmar);
-+ assert(dev_id);
-+
-+ vmar->rstate[dev_id].target = target;
-+ vmar->rstate[dev_id].write_zeroes = write_zeroes;
-+
-+ int64_t size = vmar->devinfo[dev_id].size;
-+
-+ int64_t bitmap_size = (size/BDRV_SECTOR_SIZE) +
-+ (VMA_CLUSTER_SIZE/BDRV_SECTOR_SIZE) * BITS_PER_LONG - 1;
-+ bitmap_size /= (VMA_CLUSTER_SIZE/BDRV_SECTOR_SIZE) * BITS_PER_LONG;
-+
-+ vmar->rstate[dev_id].bitmap_size = bitmap_size;
-+ vmar->rstate[dev_id].bitmap = g_new0(unsigned long, bitmap_size);
-+
-+ vmar->cluster_count += size/VMA_CLUSTER_SIZE;
-+}
-+
-+int vma_reader_register_bs(VmaReader *vmar, guint8 dev_id, BlockBackend *target,
-+ bool write_zeroes, Error **errp)
-+{
-+ assert(vmar);
-+ assert(target != NULL);
-+ assert(dev_id);
-+ assert(vmar->rstate[dev_id].target == NULL);
-+
-+ int64_t size = blk_getlength(target);
-+ int64_t size_diff = size - vmar->devinfo[dev_id].size;
-+
-+ /* storage types can have different size restrictions, so it
-+ * is not always possible to create an image with exact size.
-+ * So we tolerate a size difference up to 4MB.
-+ */
-+ if ((size_diff < 0) || (size_diff > 4*1024*1024)) {
-+ error_setg(errp, "vma_reader_register_bs for stream %s failed - "
-+ "unexpected size %zd != %zd", vmar->devinfo[dev_id].devname,
-+ size, vmar->devinfo[dev_id].size);
-+ return -1;
-+ }
-+
-+ allocate_rstate(vmar, dev_id, target, write_zeroes);
-+
-+ return 0;
-+}
-+
-+static ssize_t safe_write(int fd, void *buf, size_t count)
-+{
-+ ssize_t n;
-+
-+ do {
-+ n = write(fd, buf, count);
-+ } while (n < 0 && errno == EINTR);
-+
-+ return n;
-+}
-+
-+static size_t full_write(int fd, void *buf, size_t len)
-+{
-+ ssize_t n;
-+ size_t total;
-+
-+ total = 0;
-+
-+ while (len > 0) {
-+ n = safe_write(fd, buf, len);
-+ if (n < 0) {
-+ return n;
-+ }
-+ buf += n;
-+ total += n;
-+ len -= n;
-+ }
-+
-+ if (len) {
-+ /* incomplete write ? */
-+ return -1;
-+ }
-+
-+ return total;
-+}
-+
-+static int restore_write_data(VmaReader *vmar, guint8 dev_id,
-+ BlockBackend *target, int vmstate_fd,
-+ unsigned char *buf, int64_t sector_num,
-+ int nb_sectors, Error **errp)
-+{
-+ assert(vmar);
-+
-+ if (dev_id == vmar->vmstate_stream) {
-+ if (vmstate_fd >= 0) {
-+ int len = nb_sectors * BDRV_SECTOR_SIZE;
-+ int res = full_write(vmstate_fd, buf, len);
-+ if (res < 0) {
-+ error_setg(errp, "write vmstate failed %d", res);
-+ return -1;
-+ }
-+ }
-+ } else {
-+ int res = blk_pwrite(target, sector_num * BDRV_SECTOR_SIZE, buf, nb_sectors * BDRV_SECTOR_SIZE, 0);
-+ if (res < 0) {
-+ error_setg(errp, "blk_pwrite to %s failed (%d)",
-+ bdrv_get_device_name(blk_bs(target)), res);
-+ return -1;
-+ }
-+ }
-+ return 0;
-+}
-+
-+static int restore_extent(VmaReader *vmar, unsigned char *buf,
-+ int extent_size, int vmstate_fd,
-+ bool verbose, bool verify, Error **errp)
-+{
-+ assert(vmar);
-+ assert(buf);
-+
-+ VmaExtentHeader *ehead = (VmaExtentHeader *)buf;
-+ int start = VMA_EXTENT_HEADER_SIZE;
-+ int i;
-+
-+ for (i = 0; i < VMA_BLOCKS_PER_EXTENT; i++) {
-+ uint64_t block_info = GUINT64_FROM_BE(ehead->blockinfo[i]);
-+ uint64_t cluster_num = block_info & 0xffffffff;
-+ uint8_t dev_id = (block_info >> 32) & 0xff;
-+ uint16_t mask = block_info >> (32+16);
-+ int64_t max_sector;
-+
-+ if (!dev_id) {
-+ continue;
-+ }
-+
-+ VmaRestoreState *rstate = &vmar->rstate[dev_id];
-+ BlockBackend *target = NULL;
-+
-+ if (dev_id != vmar->vmstate_stream) {
-+ target = rstate->target;
-+ if (!verify && !target) {
-+ error_setg(errp, "got wrong dev id %d", dev_id);
-+ return -1;
-+ }
-+
-+ if (vma_reader_get_bitmap(rstate, cluster_num)) {
-+ error_setg(errp, "found duplicated cluster %zd for stream %s",
-+ cluster_num, vmar->devinfo[dev_id].devname);
-+ return -1;
-+ }
-+ vma_reader_set_bitmap(rstate, cluster_num, 1);
-+
-+ max_sector = vmar->devinfo[dev_id].size/BDRV_SECTOR_SIZE;
-+ } else {
-+ max_sector = G_MAXINT64;
-+ if (cluster_num != vmar->vmstate_clusters) {
-+ error_setg(errp, "found out of order vmstate data");
-+ return -1;
-+ }
-+ vmar->vmstate_clusters++;
-+ }
-+
-+ vmar->clusters_read++;
-+
-+ if (verbose) {
-+ time_t duration = time(NULL) - vmar->start_time;
-+ int percent = (vmar->clusters_read*100)/vmar->cluster_count;
-+ if (percent != vmar->clusters_read_per) {
-+ printf("progress %d%% (read %zd bytes, duration %zd sec)\n",
-+ percent, vmar->clusters_read*VMA_CLUSTER_SIZE,
-+ duration);
-+ fflush(stdout);
-+ vmar->clusters_read_per = percent;
-+ }
-+ }
-+
-+ /* try to write whole clusters to speedup restore */
-+ if (mask == 0xffff) {
-+ if ((start + VMA_CLUSTER_SIZE) > extent_size) {
-+ error_setg(errp, "short vma extent - too many blocks");
-+ return -1;
-+ }
-+ int64_t sector_num = (cluster_num * VMA_CLUSTER_SIZE) /
-+ BDRV_SECTOR_SIZE;
-+ int64_t end_sector = sector_num +
-+ VMA_CLUSTER_SIZE/BDRV_SECTOR_SIZE;
-+
-+ if (end_sector > max_sector) {
-+ end_sector = max_sector;
-+ }
-+
-+ if (end_sector <= sector_num) {
-+ error_setg(errp, "got wrong block address - write beyond end");
-+ return -1;
-+ }
-+
-+ if (!verify) {
-+ int nb_sectors = end_sector - sector_num;
-+ if (restore_write_data(vmar, dev_id, target, vmstate_fd,
-+ buf + start, sector_num, nb_sectors,
-+ errp) < 0) {
-+ return -1;
-+ }
-+ }
-+
-+ start += VMA_CLUSTER_SIZE;
-+ } else {
-+ int j;
-+ int bit = 1;
-+
-+ for (j = 0; j < 16; j++) {
-+ int64_t sector_num = (cluster_num*VMA_CLUSTER_SIZE +
-+ j*VMA_BLOCK_SIZE)/BDRV_SECTOR_SIZE;
-+
-+ int64_t end_sector = sector_num +
-+ VMA_BLOCK_SIZE/BDRV_SECTOR_SIZE;
-+ if (end_sector > max_sector) {
-+ end_sector = max_sector;
-+ }
-+
-+ if (mask & bit) {
-+ if ((start + VMA_BLOCK_SIZE) > extent_size) {
-+ error_setg(errp, "short vma extent - too many blocks");
-+ return -1;
-+ }
-+
-+ if (end_sector <= sector_num) {
-+ error_setg(errp, "got wrong block address - "
-+ "write beyond end");
-+ return -1;
-+ }
-+
-+ if (!verify) {
-+ int nb_sectors = end_sector - sector_num;
-+ if (restore_write_data(vmar, dev_id, target, vmstate_fd,
-+ buf + start, sector_num,
-+ nb_sectors, errp) < 0) {
-+ return -1;
-+ }
-+ }
-+
-+ start += VMA_BLOCK_SIZE;
-+
-+ } else {
-+
-+
-+ if (end_sector > sector_num) {
-+ /* Todo: use bdrv_co_write_zeroes (but that need to
-+ * be run inside coroutine?)
-+ */
-+ int nb_sectors = end_sector - sector_num;
-+ int zero_size = BDRV_SECTOR_SIZE*nb_sectors;
-+ vmar->zero_cluster_data += zero_size;
-+ if (mask != 0) {
-+ vmar->partial_zero_cluster_data += zero_size;
-+ }
-+
-+ if (rstate->write_zeroes && !verify) {
-+ if (restore_write_data(vmar, dev_id, target, vmstate_fd,
-+ zero_vma_block, sector_num,
-+ nb_sectors, errp) < 0) {
-+ return -1;
-+ }
-+ }
-+ }
-+ }
-+
-+ bit = bit << 1;
-+ }
-+ }
-+ }
-+
-+ if (start != extent_size) {
-+ error_setg(errp, "vma extent error - missing blocks");
-+ return -1;
-+ }
-+
-+ return 0;
-+}
-+
-+static int vma_reader_restore_full(VmaReader *vmar, int vmstate_fd,
-+ bool verbose, bool verify,
-+ Error **errp)
-+{
-+ assert(vmar);
-+ assert(vmar->head_data);
-+
-+ int ret = 0;
-+ unsigned char buf[VMA_MAX_EXTENT_SIZE];
-+ int buf_pos = 0;
-+ unsigned char md5sum[16];
-+ VmaHeader *h = (VmaHeader *)vmar->head_data;
-+
-+ vmar->start_time = time(NULL);
-+
-+ while (1) {
-+ int bytes = full_read(vmar->fd, buf + buf_pos, sizeof(buf) - buf_pos);
-+ if (bytes < 0) {
-+ error_setg(errp, "read failed - %s", g_strerror(errno));
-+ return -1;
-+ }
-+
-+ buf_pos += bytes;
-+
-+ if (!buf_pos) {
-+ break; /* EOF */
-+ }
-+
-+ if (buf_pos < VMA_EXTENT_HEADER_SIZE) {
-+ error_setg(errp, "read short extent (%d bytes)", buf_pos);
-+ return -1;
-+ }
-+
-+ VmaExtentHeader *ehead = (VmaExtentHeader *)buf;
-+
-+ /* extract md5sum */
-+ memcpy(md5sum, ehead->md5sum, sizeof(ehead->md5sum));
-+ memset(ehead->md5sum, 0, sizeof(ehead->md5sum));
-+
-+ g_checksum_reset(vmar->md5csum);
-+ g_checksum_update(vmar->md5csum, buf, VMA_EXTENT_HEADER_SIZE);
-+ gsize csize = 16;
-+ g_checksum_get_digest(vmar->md5csum, ehead->md5sum, &csize);
-+
-+ if (memcmp(md5sum, ehead->md5sum, 16) != 0) {
-+ error_setg(errp, "wrong vma extent header chechsum");
-+ return -1;
-+ }
-+
-+ if (memcmp(h->uuid, ehead->uuid, sizeof(ehead->uuid)) != 0) {
-+ error_setg(errp, "wrong vma extent uuid");
-+ return -1;
-+ }
-+
-+ if (ehead->magic != VMA_EXTENT_MAGIC || ehead->reserved1 != 0) {
-+ error_setg(errp, "wrong vma extent header magic");
-+ return -1;
-+ }
-+
-+ int block_count = GUINT16_FROM_BE(ehead->block_count);
-+ int extent_size = VMA_EXTENT_HEADER_SIZE + block_count*VMA_BLOCK_SIZE;
-+
-+ if (buf_pos < extent_size) {
-+ error_setg(errp, "short vma extent (%d < %d)", buf_pos,
-+ extent_size);
-+ return -1;
-+ }
-+
-+ if (restore_extent(vmar, buf, extent_size, vmstate_fd, verbose,
-+ verify, errp) < 0) {
-+ return -1;
-+ }
-+
-+ if (buf_pos > extent_size) {
-+ memmove(buf, buf + extent_size, buf_pos - extent_size);
-+ buf_pos = buf_pos - extent_size;
-+ } else {
-+ buf_pos = 0;
-+ }
-+ }
-+
-+ bdrv_drain_all();
-+
-+ int i;
-+ for (i = 1; i < 256; i++) {
-+ VmaRestoreState *rstate = &vmar->rstate[i];
-+ if (!rstate->target) {
-+ continue;
-+ }
-+
-+ if (blk_flush(rstate->target) < 0) {
-+ error_setg(errp, "vma blk_flush %s failed",
-+ vmar->devinfo[i].devname);
-+ return -1;
-+ }
-+
-+ if (vmar->devinfo[i].size &&
-+ (strcmp(vmar->devinfo[i].devname, "vmstate") != 0)) {
-+ assert(rstate->bitmap);
-+
-+ int64_t cluster_num, end;
-+
-+ end = (vmar->devinfo[i].size + VMA_CLUSTER_SIZE - 1) /
-+ VMA_CLUSTER_SIZE;
-+
-+ for (cluster_num = 0; cluster_num < end; cluster_num++) {
-+ if (!vma_reader_get_bitmap(rstate, cluster_num)) {
-+ error_setg(errp, "detected missing cluster %zd "
-+ "for stream %s", cluster_num,
-+ vmar->devinfo[i].devname);
-+ return -1;
-+ }
-+ }
-+ }
-+ }
-+
-+ if (verbose) {
-+ if (vmar->clusters_read) {
-+ printf("total bytes read %zd, sparse bytes %zd (%.3g%%)\n",
-+ vmar->clusters_read*VMA_CLUSTER_SIZE,
-+ vmar->zero_cluster_data,
-+ (double)(100.0*vmar->zero_cluster_data)/
-+ (vmar->clusters_read*VMA_CLUSTER_SIZE));
-+
-+ int64_t datasize = vmar->clusters_read*VMA_CLUSTER_SIZE-vmar->zero_cluster_data;
-+ if (datasize) { // this does not make sense for empty files
-+ printf("space reduction due to 4K zero blocks %.3g%%\n",
-+ (double)(100.0*vmar->partial_zero_cluster_data) / datasize);
-+ }
-+ } else {
-+ printf("vma archive contains no image data\n");
-+ }
-+ }
-+ return ret;
-+}
-+
-+int vma_reader_restore(VmaReader *vmar, int vmstate_fd, bool verbose,
-+ Error **errp)
-+{
-+ return vma_reader_restore_full(vmar, vmstate_fd, verbose, false, errp);
-+}
-+
-+int vma_reader_verify(VmaReader *vmar, bool verbose, Error **errp)
-+{
-+ guint8 dev_id;
-+
-+ for (dev_id = 1; dev_id < 255; dev_id++) {
-+ if (vma_reader_get_device_info(vmar, dev_id)) {
-+ allocate_rstate(vmar, dev_id, NULL, false);
-+ }
-+ }
-+
-+ return vma_reader_restore_full(vmar, -1, verbose, true, errp);
-+}
-+
-diff --git a/vma-writer.c b/vma-writer.c
-new file mode 100644
-index 0000000000..fd9567634d
---- /dev/null
-+++ b/vma-writer.c
-@@ -0,0 +1,771 @@
-+/*
-+ * VMA: Virtual Machine Archive
-+ *
-+ * Copyright (C) 2012 Proxmox Server Solutions
-+ *
-+ * Authors:
-+ * Dietmar Maurer (dietmar@proxmox.com)
-+ *
-+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
-+ * See the COPYING file in the top-level directory.
-+ *
-+ */
-+
-+#include "qemu/osdep.h"
-+#include <glib.h>
-+#include <uuid/uuid.h>
-+
-+#include "vma.h"
-+#include "block/block.h"
-+#include "monitor/monitor.h"
-+#include "qemu/main-loop.h"
-+#include "qemu/coroutine.h"
-+#include "qemu/cutils.h"
-+
-+#define DEBUG_VMA 0
-+
-+#define DPRINTF(fmt, ...)\
-+ do { if (DEBUG_VMA) { printf("vma: " fmt, ## __VA_ARGS__); } } while (0)
-+
-+#define WRITE_BUFFERS 5
-+#define HEADER_CLUSTERS 8
-+#define HEADERBUF_SIZE (VMA_CLUSTER_SIZE*HEADER_CLUSTERS)
-+
-+struct VmaWriter {
-+ int fd;
-+ FILE *cmd;
-+ int status;
-+ char errmsg[8192];
-+ uuid_t uuid;
-+ bool header_written;
-+ bool closed;
-+
-+ /* we always write extents */
-+ unsigned char *outbuf;
-+ int outbuf_pos; /* in bytes */
-+ int outbuf_count; /* in VMA_BLOCKS */
-+ uint64_t outbuf_block_info[VMA_BLOCKS_PER_EXTENT];
-+
-+ unsigned char *headerbuf;
-+
-+ GChecksum *md5csum;
-+ CoMutex flush_lock;
-+ Coroutine *co_writer;
-+
-+ /* drive informations */
-+ VmaStreamInfo stream_info[256];
-+ guint stream_count;
-+
-+ guint8 vmstate_stream;
-+ uint32_t vmstate_clusters;
-+
-+ /* header blob table */
-+ char *header_blob_table;
-+ uint32_t header_blob_table_size;
-+ uint32_t header_blob_table_pos;
-+
-+ /* store for config blobs */
-+ uint32_t config_names[VMA_MAX_CONFIGS]; /* offset into blob_buffer table */
-+ uint32_t config_data[VMA_MAX_CONFIGS]; /* offset into blob_buffer table */
-+ uint32_t config_count;
-+};
-+
-+void vma_writer_set_error(VmaWriter *vmaw, const char *fmt, ...)
-+{
-+ va_list ap;
-+
-+ if (vmaw->status < 0) {
-+ return;
-+ }
-+
-+ vmaw->status = -1;
-+
-+ va_start(ap, fmt);
-+ g_vsnprintf(vmaw->errmsg, sizeof(vmaw->errmsg), fmt, ap);
-+ va_end(ap);
-+
-+ DPRINTF("vma_writer_set_error: %s\n", vmaw->errmsg);
-+}
-+
-+static uint32_t allocate_header_blob(VmaWriter *vmaw, const char *data,
-+ size_t len)
-+{
-+ if (len > 65535) {
-+ return 0;
-+ }
-+
-+ if (!vmaw->header_blob_table ||
-+ (vmaw->header_blob_table_size <
-+ (vmaw->header_blob_table_pos + len + 2))) {
-+ int newsize = vmaw->header_blob_table_size + ((len + 2 + 511)/512)*512;
-+
-+ vmaw->header_blob_table = g_realloc(vmaw->header_blob_table, newsize);
-+ memset(vmaw->header_blob_table + vmaw->header_blob_table_size,
-+ 0, newsize - vmaw->header_blob_table_size);
-+ vmaw->header_blob_table_size = newsize;
-+ }
-+
-+ uint32_t cpos = vmaw->header_blob_table_pos;
-+ vmaw->header_blob_table[cpos] = len & 255;
-+ vmaw->header_blob_table[cpos+1] = (len >> 8) & 255;
-+ memcpy(vmaw->header_blob_table + cpos + 2, data, len);
-+ vmaw->header_blob_table_pos += len + 2;
-+ return cpos;
-+}
-+
-+static uint32_t allocate_header_string(VmaWriter *vmaw, const char *str)
-+{
-+ assert(vmaw);
-+
-+ size_t len = strlen(str) + 1;
-+
-+ return allocate_header_blob(vmaw, str, len);
-+}
-+
-+int vma_writer_add_config(VmaWriter *vmaw, const char *name, gpointer data,
-+ gsize len)
-+{
-+ assert(vmaw);
-+ assert(!vmaw->header_written);
-+ assert(vmaw->config_count < VMA_MAX_CONFIGS);
-+ assert(name);
-+ assert(data);
-+
-+ gchar *basename = g_path_get_basename(name);
-+ uint32_t name_ptr = allocate_header_string(vmaw, basename);
-+ g_free(basename);
-+
-+ if (!name_ptr) {
-+ return -1;
-+ }
-+
-+ uint32_t data_ptr = allocate_header_blob(vmaw, data, len);
-+ if (!data_ptr) {
-+ return -1;
-+ }
-+
-+ vmaw->config_names[vmaw->config_count] = name_ptr;
-+ vmaw->config_data[vmaw->config_count] = data_ptr;
-+
-+ vmaw->config_count++;
-+
-+ return 0;
-+}
-+
-+int vma_writer_register_stream(VmaWriter *vmaw, const char *devname,
-+ size_t size)
-+{
-+ assert(vmaw);
-+ assert(devname);
-+ assert(!vmaw->status);
-+
-+ if (vmaw->header_written) {
-+ vma_writer_set_error(vmaw, "vma_writer_register_stream: header "
-+ "already written");
-+ return -1;
-+ }
-+
-+ guint n = vmaw->stream_count + 1;
-+
-+ /* we can have dev_ids form 1 to 255 (0 reserved)
-+ * 255(-1) reseverd for safety
-+ */
-+ if (n > 254) {
-+ vma_writer_set_error(vmaw, "vma_writer_register_stream: "
-+ "too many drives");
-+ return -1;
-+ }
-+
-+ if (size <= 0) {
-+ vma_writer_set_error(vmaw, "vma_writer_register_stream: "
-+ "got strange size %zd", size);
-+ return -1;
-+ }
-+
-+ DPRINTF("vma_writer_register_stream %s %zu %d\n", devname, size, n);
-+
-+ vmaw->stream_info[n].devname = g_strdup(devname);
-+ vmaw->stream_info[n].size = size;
-+
-+ vmaw->stream_info[n].cluster_count = (size + VMA_CLUSTER_SIZE - 1) /
-+ VMA_CLUSTER_SIZE;
-+
-+ vmaw->stream_count = n;
-+
-+ if (strcmp(devname, "vmstate") == 0) {
-+ vmaw->vmstate_stream = n;
-+ }
-+
-+ return n;
-+}
-+
-+static void vma_co_continue_write(void *opaque)
-+{
-+ VmaWriter *vmaw = opaque;
-+
-+ DPRINTF("vma_co_continue_write\n");
-+ qemu_coroutine_enter(vmaw->co_writer);
-+}
-+
-+static ssize_t coroutine_fn
-+vma_queue_write(VmaWriter *vmaw, const void *buf, size_t bytes)
-+{
-+ DPRINTF("vma_queue_write enter %zd\n", bytes);
-+
-+ assert(vmaw);
-+ assert(buf);
-+ assert(bytes <= VMA_MAX_EXTENT_SIZE);
-+
-+ size_t done = 0;
-+ ssize_t ret;
-+
-+ assert(vmaw->co_writer == NULL);
-+
-+ vmaw->co_writer = qemu_coroutine_self();
-+
-+ while (done < bytes) {
-+ aio_set_fd_handler(qemu_get_aio_context(), vmaw->fd, false, NULL, vma_co_continue_write, NULL, vmaw);
-+ qemu_coroutine_yield();
-+ aio_set_fd_handler(qemu_get_aio_context(), vmaw->fd, false, NULL, NULL, NULL, NULL);
-+ if (vmaw->status < 0) {
-+ DPRINTF("vma_queue_write detected canceled backup\n");
-+ done = -1;
-+ break;
-+ }
-+ ret = write(vmaw->fd, buf + done, bytes - done);
-+ if (ret > 0) {
-+ done += ret;
-+ DPRINTF("vma_queue_write written %zd %zd\n", done, ret);
-+ } else if (ret < 0) {
-+ if (errno == EAGAIN || errno == EWOULDBLOCK) {
-+ /* try again */
-+ } else {
-+ vma_writer_set_error(vmaw, "vma_queue_write: write error - %s",
-+ g_strerror(errno));
-+ done = -1; /* always return failure for partial writes */
-+ break;
-+ }
-+ } else if (ret == 0) {
-+ /* should not happen - simply try again */
-+ }
-+ }
-+
-+ vmaw->co_writer = NULL;
-+
-+ return (done == bytes) ? bytes : -1;
-+}
-+
-+VmaWriter *vma_writer_create(const char *filename, uuid_t uuid, Error **errp)
-+{
-+ const char *p;
-+
-+ assert(sizeof(VmaHeader) == (4096 + 8192));
-+ assert(G_STRUCT_OFFSET(VmaHeader, config_names) == 2044);
-+ assert(G_STRUCT_OFFSET(VmaHeader, config_data) == 3068);
-+ assert(G_STRUCT_OFFSET(VmaHeader, dev_info) == 4096);
-+ assert(sizeof(VmaExtentHeader) == 512);
-+
-+ VmaWriter *vmaw = g_new0(VmaWriter, 1);
-+ vmaw->fd = -1;
-+
-+ vmaw->md5csum = g_checksum_new(G_CHECKSUM_MD5);
-+ if (!vmaw->md5csum) {
-+ error_setg(errp, "can't allocate cmsum\n");
-+ goto err;
-+ }
-+
-+ if (strstart(filename, "exec:", &p)) {
-+ vmaw->cmd = popen(p, "w");
-+ if (vmaw->cmd == NULL) {
-+ error_setg(errp, "can't popen command '%s' - %s\n", p,
-+ g_strerror(errno));
-+ goto err;
-+ }
-+ vmaw->fd = fileno(vmaw->cmd);
-+
-+ /* try to use O_NONBLOCK */
-+ fcntl(vmaw->fd, F_SETFL, fcntl(vmaw->fd, F_GETFL)|O_NONBLOCK);
-+
-+ } else {
-+ struct stat st;
-+ int oflags;
-+ const char *tmp_id_str;
-+
-+ if ((stat(filename, &st) == 0) && S_ISFIFO(st.st_mode)) {
-+ oflags = O_NONBLOCK|O_WRONLY;
-+ vmaw->fd = qemu_open(filename, oflags, 0644);
-+ } else if (strstart(filename, "/dev/fdset/", &tmp_id_str)) {
-+ oflags = O_NONBLOCK|O_WRONLY;
-+ vmaw->fd = qemu_open(filename, oflags, 0644);
-+ } else if (strstart(filename, "/dev/fdname/", &tmp_id_str)) {
-+ vmaw->fd = monitor_get_fd(cur_mon, tmp_id_str, errp);
-+ if (vmaw->fd < 0) {
-+ goto err;
-+ }
-+ /* try to use O_NONBLOCK */
-+ fcntl(vmaw->fd, F_SETFL, fcntl(vmaw->fd, F_GETFL)|O_NONBLOCK);
-+ } else {
-+ oflags = O_NONBLOCK|O_DIRECT|O_WRONLY|O_CREAT|O_EXCL;
-+ vmaw->fd = qemu_open(filename, oflags, 0644);
-+ }
-+
-+ if (vmaw->fd < 0) {
-+ error_setg(errp, "can't open file %s - %s\n", filename,
-+ g_strerror(errno));
-+ goto err;
-+ }
-+ }
-+
-+ /* we use O_DIRECT, so we need to align IO buffers */
-+
-+ vmaw->outbuf = qemu_memalign(512, VMA_MAX_EXTENT_SIZE);
-+ vmaw->headerbuf = qemu_memalign(512, HEADERBUF_SIZE);
-+
-+ vmaw->outbuf_count = 0;
-+ vmaw->outbuf_pos = VMA_EXTENT_HEADER_SIZE;
-+
-+ vmaw->header_blob_table_pos = 1; /* start at pos 1 */
-+
-+ qemu_co_mutex_init(&vmaw->flush_lock);
-+
-+ uuid_copy(vmaw->uuid, uuid);
-+
-+ return vmaw;
-+
-+err:
-+ if (vmaw) {
-+ if (vmaw->cmd) {
-+ pclose(vmaw->cmd);
-+ } else if (vmaw->fd >= 0) {
-+ close(vmaw->fd);
-+ }
-+
-+ if (vmaw->md5csum) {
-+ g_checksum_free(vmaw->md5csum);
-+ }
-+
-+ g_free(vmaw);
-+ }
-+
-+ return NULL;
-+}
-+
-+static int coroutine_fn vma_write_header(VmaWriter *vmaw)
-+{
-+ assert(vmaw);
-+ unsigned char *buf = vmaw->headerbuf;
-+ VmaHeader *head = (VmaHeader *)buf;
-+
-+ int i;
-+
-+ DPRINTF("VMA WRITE HEADER\n");
-+
-+ if (vmaw->status < 0) {
-+ return vmaw->status;
-+ }
-+
-+ memset(buf, 0, HEADERBUF_SIZE);
-+
-+ head->magic = VMA_MAGIC;
-+ head->version = GUINT32_TO_BE(1); /* v1 */
-+ memcpy(head->uuid, vmaw->uuid, 16);
-+
-+ time_t ctime = time(NULL);
-+ head->ctime = GUINT64_TO_BE(ctime);
-+
-+ for (i = 0; i < VMA_MAX_CONFIGS; i++) {
-+ head->config_names[i] = GUINT32_TO_BE(vmaw->config_names[i]);
-+ head->config_data[i] = GUINT32_TO_BE(vmaw->config_data[i]);
-+ }
-+
-+ /* 32 bytes per device (12 used currently) = 8192 bytes max */
-+ for (i = 1; i <= 254; i++) {
-+ VmaStreamInfo *si = &vmaw->stream_info[i];
-+ if (si->size) {
-+ assert(si->devname);
-+ uint32_t devname_ptr = allocate_header_string(vmaw, si->devname);
-+ if (!devname_ptr) {
-+ return -1;
-+ }
-+ head->dev_info[i].devname_ptr = GUINT32_TO_BE(devname_ptr);
-+ head->dev_info[i].size = GUINT64_TO_BE(si->size);
-+ }
-+ }
-+
-+ uint32_t header_size = sizeof(VmaHeader) + vmaw->header_blob_table_size;
-+ head->header_size = GUINT32_TO_BE(header_size);
-+
-+ if (header_size > HEADERBUF_SIZE) {
-+ return -1; /* just to be sure */
-+ }
-+
-+ uint32_t blob_buffer_offset = sizeof(VmaHeader);
-+ memcpy(buf + blob_buffer_offset, vmaw->header_blob_table,
-+ vmaw->header_blob_table_size);
-+ head->blob_buffer_offset = GUINT32_TO_BE(blob_buffer_offset);
-+ head->blob_buffer_size = GUINT32_TO_BE(vmaw->header_blob_table_pos);
-+
-+ g_checksum_reset(vmaw->md5csum);
-+ g_checksum_update(vmaw->md5csum, (const guchar *)buf, header_size);
-+ gsize csize = 16;
-+ g_checksum_get_digest(vmaw->md5csum, (guint8 *)(head->md5sum), &csize);
-+
-+ return vma_queue_write(vmaw, buf, header_size);
-+}
-+
-+static int coroutine_fn vma_writer_flush(VmaWriter *vmaw)
-+{
-+ assert(vmaw);
-+
-+ int ret;
-+ int i;
-+
-+ if (vmaw->status < 0) {
-+ return vmaw->status;
-+ }
-+
-+ if (!vmaw->header_written) {
-+ vmaw->header_written = true;
-+ ret = vma_write_header(vmaw);
-+ if (ret < 0) {
-+ vma_writer_set_error(vmaw, "vma_writer_flush: write header failed");
-+ return ret;
-+ }
-+ }
-+
-+ DPRINTF("VMA WRITE FLUSH %d %d\n", vmaw->outbuf_count, vmaw->outbuf_pos);
-+
-+
-+ VmaExtentHeader *ehead = (VmaExtentHeader *)vmaw->outbuf;
-+
-+ ehead->magic = VMA_EXTENT_MAGIC;
-+ ehead->reserved1 = 0;
-+
-+ for (i = 0; i < VMA_BLOCKS_PER_EXTENT; i++) {
-+ ehead->blockinfo[i] = GUINT64_TO_BE(vmaw->outbuf_block_info[i]);
-+ }
-+
-+ guint16 block_count = (vmaw->outbuf_pos - VMA_EXTENT_HEADER_SIZE) /
-+ VMA_BLOCK_SIZE;
-+
-+ ehead->block_count = GUINT16_TO_BE(block_count);
-+
-+ memcpy(ehead->uuid, vmaw->uuid, sizeof(ehead->uuid));
-+ memset(ehead->md5sum, 0, sizeof(ehead->md5sum));
-+
-+ g_checksum_reset(vmaw->md5csum);
-+ g_checksum_update(vmaw->md5csum, vmaw->outbuf, VMA_EXTENT_HEADER_SIZE);
-+ gsize csize = 16;
-+ g_checksum_get_digest(vmaw->md5csum, ehead->md5sum, &csize);
-+
-+ int bytes = vmaw->outbuf_pos;
-+ ret = vma_queue_write(vmaw, vmaw->outbuf, bytes);
-+ if (ret != bytes) {
-+ vma_writer_set_error(vmaw, "vma_writer_flush: failed write");
-+ }
-+
-+ vmaw->outbuf_count = 0;
-+ vmaw->outbuf_pos = VMA_EXTENT_HEADER_SIZE;
-+
-+ for (i = 0; i < VMA_BLOCKS_PER_EXTENT; i++) {
-+ vmaw->outbuf_block_info[i] = 0;
-+ }
-+
-+ return vmaw->status;
-+}
-+
-+static int vma_count_open_streams(VmaWriter *vmaw)
-+{
-+ g_assert(vmaw != NULL);
-+
-+ int i;
-+ int open_drives = 0;
-+ for (i = 0; i <= 255; i++) {
-+ if (vmaw->stream_info[i].size && !vmaw->stream_info[i].finished) {
-+ open_drives++;
-+ }
-+ }
-+
-+ return open_drives;
-+}
-+
-+
-+/**
-+ * You need to call this if the vma archive does not contain
-+ * any data stream.
-+ */
-+int coroutine_fn
-+vma_writer_flush_output(VmaWriter *vmaw)
-+{
-+ qemu_co_mutex_lock(&vmaw->flush_lock);
-+ int ret = vma_writer_flush(vmaw);
-+ qemu_co_mutex_unlock(&vmaw->flush_lock);
-+ if (ret < 0) {
-+ vma_writer_set_error(vmaw, "vma_writer_flush_header failed");
-+ }
-+ return ret;
-+}
-+
-+/**
-+ * all jobs should call this when there is no more data
-+ * Returns: number of remaining stream (0 ==> finished)
-+ */
-+int coroutine_fn
-+vma_writer_close_stream(VmaWriter *vmaw, uint8_t dev_id)
-+{
-+ g_assert(vmaw != NULL);
-+
-+ DPRINTF("vma_writer_set_status %d\n", dev_id);
-+ if (!vmaw->stream_info[dev_id].size) {
-+ vma_writer_set_error(vmaw, "vma_writer_close_stream: "
-+ "no such stream %d", dev_id);
-+ return -1;
-+ }
-+ if (vmaw->stream_info[dev_id].finished) {
-+ vma_writer_set_error(vmaw, "vma_writer_close_stream: "
-+ "stream already closed %d", dev_id);
-+ return -1;
-+ }
-+
-+ vmaw->stream_info[dev_id].finished = true;
-+
-+ int open_drives = vma_count_open_streams(vmaw);
-+
-+ if (open_drives <= 0) {
-+ DPRINTF("vma_writer_set_status all drives completed\n");
-+ vma_writer_flush_output(vmaw);
-+ }
-+
-+ return open_drives;
-+}
-+
-+int vma_writer_get_status(VmaWriter *vmaw, VmaStatus *status)
-+{
-+ int i;
-+
-+ g_assert(vmaw != NULL);
-+
-+ if (status) {
-+ status->status = vmaw->status;
-+ g_strlcpy(status->errmsg, vmaw->errmsg, sizeof(status->errmsg));
-+ for (i = 0; i <= 255; i++) {
-+ status->stream_info[i] = vmaw->stream_info[i];
-+ }
-+
-+ uuid_unparse_lower(vmaw->uuid, status->uuid_str);
-+ }
-+
-+ status->closed = vmaw->closed;
-+
-+ return vmaw->status;
-+}
-+
-+static int vma_writer_get_buffer(VmaWriter *vmaw)
-+{
-+ int ret = 0;
-+
-+ qemu_co_mutex_lock(&vmaw->flush_lock);
-+
-+ /* wait until buffer is available */
-+ while (vmaw->outbuf_count >= (VMA_BLOCKS_PER_EXTENT - 1)) {
-+ ret = vma_writer_flush(vmaw);
-+ if (ret < 0) {
-+ vma_writer_set_error(vmaw, "vma_writer_get_buffer: flush failed");
-+ break;
-+ }
-+ }
-+
-+ qemu_co_mutex_unlock(&vmaw->flush_lock);
-+
-+ return ret;
-+}
-+
-+
-+int64_t coroutine_fn
-+vma_writer_write(VmaWriter *vmaw, uint8_t dev_id, int64_t cluster_num,
-+ const unsigned char *buf, size_t *zero_bytes)
-+{
-+ g_assert(vmaw != NULL);
-+ g_assert(zero_bytes != NULL);
-+
-+ *zero_bytes = 0;
-+
-+ if (vmaw->status < 0) {
-+ return vmaw->status;
-+ }
-+
-+ if (!dev_id || !vmaw->stream_info[dev_id].size) {
-+ vma_writer_set_error(vmaw, "vma_writer_write: "
-+ "no such stream %d", dev_id);
-+ return -1;
-+ }
-+
-+ if (vmaw->stream_info[dev_id].finished) {
-+ vma_writer_set_error(vmaw, "vma_writer_write: "
-+ "stream already closed %d", dev_id);
-+ return -1;
-+ }
-+
-+
-+ if (cluster_num >= (((uint64_t)1)<<32)) {
-+ vma_writer_set_error(vmaw, "vma_writer_write: "
-+ "cluster number out of range");
-+ return -1;
-+ }
-+
-+ if (dev_id == vmaw->vmstate_stream) {
-+ if (cluster_num != vmaw->vmstate_clusters) {
-+ vma_writer_set_error(vmaw, "vma_writer_write: "
-+ "non sequential vmstate write");
-+ }
-+ vmaw->vmstate_clusters++;
-+ } else if (cluster_num >= vmaw->stream_info[dev_id].cluster_count) {
-+ vma_writer_set_error(vmaw, "vma_writer_write: cluster number too big");
-+ return -1;
-+ }
-+
-+ /* wait until buffer is available */
-+ if (vma_writer_get_buffer(vmaw) < 0) {
-+ vma_writer_set_error(vmaw, "vma_writer_write: "
-+ "vma_writer_get_buffer failed");
-+ return -1;
-+ }
-+
-+ DPRINTF("VMA WRITE %d %zd\n", dev_id, cluster_num);
-+
-+ uint16_t mask = 0;
-+
-+ if (buf) {
-+ int i;
-+ int bit = 1;
-+ for (i = 0; i < 16; i++) {
-+ const unsigned char *vmablock = buf + (i*VMA_BLOCK_SIZE);
-+ if (!buffer_is_zero(vmablock, VMA_BLOCK_SIZE)) {
-+ mask |= bit;
-+ memcpy(vmaw->outbuf + vmaw->outbuf_pos, vmablock,
-+ VMA_BLOCK_SIZE);
-+ vmaw->outbuf_pos += VMA_BLOCK_SIZE;
-+ } else {
-+ DPRINTF("VMA WRITE %zd ZERO BLOCK %d\n", cluster_num, i);
-+ vmaw->stream_info[dev_id].zero_bytes += VMA_BLOCK_SIZE;
-+ *zero_bytes += VMA_BLOCK_SIZE;
-+ }
-+
-+ bit = bit << 1;
-+ }
-+ } else {
-+ DPRINTF("VMA WRITE %zd ZERO CLUSTER\n", cluster_num);
-+ vmaw->stream_info[dev_id].zero_bytes += VMA_CLUSTER_SIZE;
-+ *zero_bytes += VMA_CLUSTER_SIZE;
-+ }
-+
-+ uint64_t block_info = ((uint64_t)mask) << (32+16);
-+ block_info |= ((uint64_t)dev_id) << 32;
-+ block_info |= (cluster_num & 0xffffffff);
-+ vmaw->outbuf_block_info[vmaw->outbuf_count] = block_info;
-+
-+ DPRINTF("VMA WRITE MASK %zd %zx\n", cluster_num, block_info);
-+
-+ vmaw->outbuf_count++;
-+
-+ /** NOTE: We allways write whole clusters, but we correctly set
-+ * transferred bytes. So transferred == size when when everything
-+ * went OK.
-+ */
-+ size_t transferred = VMA_CLUSTER_SIZE;
-+
-+ if (dev_id != vmaw->vmstate_stream) {
-+ uint64_t last = (cluster_num + 1) * VMA_CLUSTER_SIZE;
-+ if (last > vmaw->stream_info[dev_id].size) {
-+ uint64_t diff = last - vmaw->stream_info[dev_id].size;
-+ if (diff >= VMA_CLUSTER_SIZE) {
-+ vma_writer_set_error(vmaw, "vma_writer_write: "
-+ "read after last cluster");
-+ return -1;
-+ }
-+ transferred -= diff;
-+ }
-+ }
-+
-+ vmaw->stream_info[dev_id].transferred += transferred;
-+
-+ return transferred;
-+}
-+
-+void vma_writer_error_propagate(VmaWriter *vmaw, Error **errp)
-+{
-+ if (vmaw->status < 0 && *errp == NULL) {
-+ error_setg(errp, "%s", vmaw->errmsg);
-+ }
-+}
-+
-+int vma_writer_close(VmaWriter *vmaw, Error **errp)
-+{
-+ g_assert(vmaw != NULL);
-+
-+ int i;
-+
-+ while (vmaw->co_writer) {
-+ aio_poll(qemu_get_aio_context(), true);
-+ }
-+
-+ assert(vmaw->co_writer == NULL);
-+
-+ if (vmaw->cmd) {
-+ if (pclose(vmaw->cmd) < 0) {
-+ vma_writer_set_error(vmaw, "vma_writer_close: "
-+ "pclose failed - %s", g_strerror(errno));
-+ }
-+ } else {
-+ if (close(vmaw->fd) < 0) {
-+ vma_writer_set_error(vmaw, "vma_writer_close: "
-+ "close failed - %s", g_strerror(errno));
-+ }
-+ }
-+
-+ for (i = 0; i <= 255; i++) {
-+ VmaStreamInfo *si = &vmaw->stream_info[i];
-+ if (si->size) {
-+ if (!si->finished) {
-+ vma_writer_set_error(vmaw, "vma_writer_close: "
-+ "detected open stream '%s'", si->devname);
-+ } else if ((si->transferred != si->size) &&
-+ (i != vmaw->vmstate_stream)) {
-+ vma_writer_set_error(vmaw, "vma_writer_close: "
-+ "incomplete stream '%s' (%zd != %zd)",
-+ si->devname, si->transferred, si->size);
-+ }
-+ }
-+ }
-+
-+ for (i = 0; i <= 255; i++) {
-+ vmaw->stream_info[i].finished = 1; /* mark as closed */
-+ }
-+
-+ vmaw->closed = 1;
-+
-+ if (vmaw->status < 0 && *errp == NULL) {
-+ error_setg(errp, "%s", vmaw->errmsg);
-+ }
-+
-+ return vmaw->status;
-+}
-+
-+void vma_writer_destroy(VmaWriter *vmaw)
-+{
-+ assert(vmaw);
-+
-+ int i;
-+
-+ for (i = 0; i <= 255; i++) {
-+ if (vmaw->stream_info[i].devname) {
-+ g_free(vmaw->stream_info[i].devname);
-+ }
-+ }
-+
-+ if (vmaw->md5csum) {
-+ g_checksum_free(vmaw->md5csum);
-+ }
-+
-+ g_free(vmaw);
-+}
-diff --git a/vma.c b/vma.c
-new file mode 100644
-index 0000000000..1b59fd1555
---- /dev/null
-+++ b/vma.c
-@@ -0,0 +1,756 @@
-+/*
-+ * VMA: Virtual Machine Archive
-+ *
-+ * Copyright (C) 2012-2013 Proxmox Server Solutions
-+ *
-+ * Authors:
-+ * Dietmar Maurer (dietmar@proxmox.com)
-+ *
-+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
-+ * See the COPYING file in the top-level directory.
-+ *
-+ */
-+
-+#include "qemu/osdep.h"
-+#include <glib.h>
-+
-+#include "vma.h"
-+#include "qemu-common.h"
-+#include "qemu/error-report.h"
-+#include "qemu/main-loop.h"
-+#include "qapi/qmp/qstring.h"
-+#include "sysemu/block-backend.h"
-+
-+static void help(void)
-+{
-+ const char *help_msg =
-+ "usage: vma command [command options]\n"
-+ "\n"
-+ "vma list <filename>\n"
-+ "vma config <filename> [-c config]\n"
-+ "vma create <filename> [-c config] pathname ...\n"
-+ "vma extract <filename> [-r <fifo>] <targetdir>\n"
-+ "vma verify <filename> [-v]\n"
-+ ;
-+
-+ printf("%s", help_msg);
-+ exit(1);
-+}
-+
-+static const char *extract_devname(const char *path, char **devname, int index)
-+{
-+ assert(path);
-+
-+ const char *sep = strchr(path, '=');
-+
-+ if (sep) {
-+ *devname = g_strndup(path, sep - path);
-+ path = sep + 1;
-+ } else {
-+ if (index >= 0) {
-+ *devname = g_strdup_printf("disk%d", index);
-+ } else {
-+ *devname = NULL;
-+ }
-+ }
-+
-+ return path;
-+}
-+
-+static void print_content(VmaReader *vmar)
-+{
-+ assert(vmar);
-+
-+ VmaHeader *head = vma_reader_get_header(vmar);
-+
-+ GList *l = vma_reader_get_config_data(vmar);
-+ while (l && l->data) {
-+ VmaConfigData *cdata = (VmaConfigData *)l->data;
-+ l = g_list_next(l);
-+ printf("CFG: size: %d name: %s\n", cdata->len, cdata->name);
-+ }
-+
-+ int i;
-+ VmaDeviceInfo *di;
-+ for (i = 1; i < 255; i++) {
-+ di = vma_reader_get_device_info(vmar, i);
-+ if (di) {
-+ if (strcmp(di->devname, "vmstate") == 0) {
-+ printf("VMSTATE: dev_id=%d memory: %zd\n", i, di->size);
-+ } else {
-+ printf("DEV: dev_id=%d size: %zd devname: %s\n",
-+ i, di->size, di->devname);
-+ }
-+ }
-+ }
-+ /* ctime is the last entry we print */
-+ printf("CTIME: %s", ctime(&head->ctime));
-+ fflush(stdout);
-+}
-+
-+static int list_content(int argc, char **argv)
-+{
-+ int c, ret = 0;
-+ const char *filename;
-+
-+ for (;;) {
-+ c = getopt(argc, argv, "h");
-+ if (c == -1) {
-+ break;
-+ }
-+ switch (c) {
-+ case '?':
-+ case 'h':
-+ help();
-+ break;
-+ default:
-+ g_assert_not_reached();
-+ }
-+ }
-+
-+ /* Get the filename */
-+ if ((optind + 1) != argc) {
-+ help();
-+ }
-+ filename = argv[optind++];
-+
-+ Error *errp = NULL;
-+ VmaReader *vmar = vma_reader_create(filename, &errp);
-+
-+ if (!vmar) {
-+ g_error("%s", error_get_pretty(errp));
-+ }
-+
-+ print_content(vmar);
-+
-+ vma_reader_destroy(vmar);
-+
-+ return ret;
-+}
-+
-+typedef struct RestoreMap {
-+ char *devname;
-+ char *path;
-+ char *format;
-+ bool write_zero;
-+} RestoreMap;
-+
-+static int extract_content(int argc, char **argv)
-+{
-+ int c, ret = 0;
-+ int verbose = 0;
-+ const char *filename;
-+ const char *dirname;
-+ const char *readmap = NULL;
-+
-+ for (;;) {
-+ c = getopt(argc, argv, "hvr:");
-+ if (c == -1) {
-+ break;
-+ }
-+ switch (c) {
-+ case '?':
-+ case 'h':
-+ help();
-+ break;
-+ case 'r':
-+ readmap = optarg;
-+ break;
-+ case 'v':
-+ verbose = 1;
-+ break;
-+ default:
-+ help();
-+ }
-+ }
-+
-+ /* Get the filename */
-+ if ((optind + 2) != argc) {
-+ help();
-+ }
-+ filename = argv[optind++];
-+ dirname = argv[optind++];
-+
-+ Error *errp = NULL;
-+ VmaReader *vmar = vma_reader_create(filename, &errp);
-+
-+ if (!vmar) {
-+ g_error("%s", error_get_pretty(errp));
-+ }
-+
-+ if (mkdir(dirname, 0777) < 0) {
-+ g_error("unable to create target directory %s - %s",
-+ dirname, g_strerror(errno));
-+ }
-+
-+ GList *l = vma_reader_get_config_data(vmar);
-+ while (l && l->data) {
-+ VmaConfigData *cdata = (VmaConfigData *)l->data;
-+ l = g_list_next(l);
-+ char *cfgfn = g_strdup_printf("%s/%s", dirname, cdata->name);
-+ GError *err = NULL;
-+ if (!g_file_set_contents(cfgfn, (gchar *)cdata->data, cdata->len,
-+ &err)) {
-+ g_error("unable to write file: %s", err->message);
-+ }
-+ }
-+
-+ GHashTable *devmap = g_hash_table_new(g_str_hash, g_str_equal);
-+
-+ if (readmap) {
-+ print_content(vmar);
-+
-+ FILE *map = fopen(readmap, "r");
-+ if (!map) {
-+ g_error("unable to open fifo %s - %s", readmap, g_strerror(errno));
-+ }
-+
-+ while (1) {
-+ char inbuf[8192];
-+ char *line = fgets(inbuf, sizeof(inbuf), map);
-+ if (!line || line[0] == '\0' || !strcmp(line, "done\n")) {
-+ break;
-+ }
-+ int len = strlen(line);
-+ if (line[len - 1] == '\n') {
-+ line[len - 1] = '\0';
-+ if (len == 1) {
-+ break;
-+ }
-+ }
-+
-+ char *format = NULL;
-+ if (strncmp(line, "format=", sizeof("format=")-1) == 0) {
-+ format = line + sizeof("format=")-1;
-+ char *colon = strchr(format, ':');
-+ if (!colon) {
-+ g_error("read map failed - found only a format ('%s')", inbuf);
-+ }
-+ format = g_strndup(format, colon - format);
-+ line = colon+1;
-+ }
-+
-+ const char *path;
-+ bool write_zero;
-+ if (line[0] == '0' && line[1] == ':') {
-+ path = line + 2;
-+ write_zero = false;
-+ } else if (line[0] == '1' && line[1] == ':') {
-+ path = line + 2;
-+ write_zero = true;
-+ } else {
-+ g_error("read map failed - parse error ('%s')", inbuf);
-+ }
-+
-+ char *devname = NULL;
-+ path = extract_devname(path, &devname, -1);
-+ if (!devname) {
-+ g_error("read map failed - no dev name specified ('%s')",
-+ inbuf);
-+ }
-+
-+ RestoreMap *map = g_new0(RestoreMap, 1);
-+ map->devname = g_strdup(devname);
-+ map->path = g_strdup(path);
-+ map->format = format;
-+ map->write_zero = write_zero;
-+
-+ g_hash_table_insert(devmap, map->devname, map);
-+
-+ };
-+ }
-+
-+ int i;
-+ int vmstate_fd = -1;
-+ guint8 vmstate_stream = 0;
-+
-+ BlockBackend *blk = NULL;
-+
-+ for (i = 1; i < 255; i++) {
-+ VmaDeviceInfo *di = vma_reader_get_device_info(vmar, i);
-+ if (di && (strcmp(di->devname, "vmstate") == 0)) {
-+ vmstate_stream = i;
-+ char *statefn = g_strdup_printf("%s/vmstate.bin", dirname);
-+ vmstate_fd = open(statefn, O_WRONLY|O_CREAT|O_EXCL, 0644);
-+ if (vmstate_fd < 0) {
-+ g_error("create vmstate file '%s' failed - %s", statefn,
-+ g_strerror(errno));
-+ }
-+ g_free(statefn);
-+ } else if (di) {
-+ char *devfn = NULL;
-+ const char *format = NULL;
-+ int flags = BDRV_O_RDWR | BDRV_O_NO_FLUSH;
-+ bool write_zero = true;
-+
-+ if (readmap) {
-+ RestoreMap *map;
-+ map = (RestoreMap *)g_hash_table_lookup(devmap, di->devname);
-+ if (map == NULL) {
-+ g_error("no device name mapping for %s", di->devname);
-+ }
-+ devfn = map->path;
-+ format = map->format;
-+ write_zero = map->write_zero;
-+ } else {
-+ devfn = g_strdup_printf("%s/tmp-disk-%s.raw",
-+ dirname, di->devname);
-+ printf("DEVINFO %s %zd\n", devfn, di->size);
-+
-+ bdrv_img_create(devfn, "raw", NULL, NULL, NULL, di->size,
-+ flags, true, &errp);
-+ if (errp) {
-+ g_error("can't create file %s: %s", devfn,
-+ error_get_pretty(errp));
-+ }
-+
-+ /* Note: we created an empty file above, so there is no
-+ * need to write zeroes (so we generate a sparse file)
-+ */
-+ write_zero = false;
-+ }
-+
-+ size_t devlen = strlen(devfn);
-+ QDict *options = NULL;
-+ if (format) {
-+ /* explicit format from commandline */
-+ options = qdict_new();
-+ qdict_put(options, "driver", qstring_from_str(format));
-+ } else if ((devlen > 4 && strcmp(devfn+devlen-4, ".raw") == 0) ||
-+ strncmp(devfn, "/dev/", 5) == 0)
-+ {
-+ /* This part is now deprecated for PVE as well (just as qemu
-+ * deprecated not specifying an explicit raw format, too.
-+ */
-+ /* explicit raw format */
-+ options = qdict_new();
-+ qdict_put(options, "driver", qstring_from_str("raw"));
-+ }
-+
-+
-+ if (errp || !(blk = blk_new_open(devfn, NULL, options, flags, &errp))) {
-+ g_error("can't open file %s - %s", devfn,
-+ error_get_pretty(errp));
-+ }
-+
-+ if (vma_reader_register_bs(vmar, i, blk, write_zero, &errp) < 0) {
-+ g_error("%s", error_get_pretty(errp));
-+ }
-+
-+ if (!readmap) {
-+ g_free(devfn);
-+ }
-+ }
-+ }
-+
-+ if (vma_reader_restore(vmar, vmstate_fd, verbose, &errp) < 0) {
-+ g_error("restore failed - %s", error_get_pretty(errp));
-+ }
-+
-+ if (!readmap) {
-+ for (i = 1; i < 255; i++) {
-+ VmaDeviceInfo *di = vma_reader_get_device_info(vmar, i);
-+ if (di && (i != vmstate_stream)) {
-+ char *tmpfn = g_strdup_printf("%s/tmp-disk-%s.raw",
-+ dirname, di->devname);
-+ char *fn = g_strdup_printf("%s/disk-%s.raw",
-+ dirname, di->devname);
-+ if (rename(tmpfn, fn) != 0) {
-+ g_error("rename %s to %s failed - %s",
-+ tmpfn, fn, g_strerror(errno));
-+ }
-+ }
-+ }
-+ }
-+
-+ vma_reader_destroy(vmar);
-+
-+ blk_unref(blk);
-+
-+ bdrv_close_all();
-+
-+ return ret;
-+}
-+
-+static int verify_content(int argc, char **argv)
-+{
-+ int c, ret = 0;
-+ int verbose = 0;
-+ const char *filename;
-+
-+ for (;;) {
-+ c = getopt(argc, argv, "hv");
-+ if (c == -1) {
-+ break;
-+ }
-+ switch (c) {
-+ case '?':
-+ case 'h':
-+ help();
-+ break;
-+ case 'v':
-+ verbose = 1;
-+ break;
-+ default:
-+ help();
-+ }
-+ }
-+
-+ /* Get the filename */
-+ if ((optind + 1) != argc) {
-+ help();
-+ }
-+ filename = argv[optind++];
-+
-+ Error *errp = NULL;
-+ VmaReader *vmar = vma_reader_create(filename, &errp);
-+
-+ if (!vmar) {
-+ g_error("%s", error_get_pretty(errp));
-+ }
-+
-+ if (verbose) {
-+ print_content(vmar);
-+ }
-+
-+ if (vma_reader_verify(vmar, verbose, &errp) < 0) {
-+ g_error("verify failed - %s", error_get_pretty(errp));
-+ }
-+
-+ vma_reader_destroy(vmar);
-+
-+ bdrv_close_all();
-+
-+ return ret;
-+}
-+
-+typedef struct BackupJob {
-+ BlockBackend *target;
-+ int64_t len;
-+ VmaWriter *vmaw;
-+ uint8_t dev_id;
-+} BackupJob;
-+
-+#define BACKUP_SECTORS_PER_CLUSTER (VMA_CLUSTER_SIZE / BDRV_SECTOR_SIZE)
-+
-+static void coroutine_fn backup_run_empty(void *opaque)
-+{
-+ VmaWriter *vmaw = (VmaWriter *)opaque;
-+
-+ vma_writer_flush_output(vmaw);
-+
-+ Error *err = NULL;
-+ if (vma_writer_close(vmaw, &err) != 0) {
-+ g_warning("vma_writer_close failed %s", error_get_pretty(err));
-+ }
-+}
-+
-+static void coroutine_fn backup_run(void *opaque)
-+{
-+ BackupJob *job = (BackupJob *)opaque;
-+ struct iovec iov;
-+ QEMUIOVector qiov;
-+
-+ int64_t start, end;
-+ int ret = 0;
-+
-+ unsigned char *buf = blk_blockalign(job->target, VMA_CLUSTER_SIZE);
-+
-+ start = 0;
-+ end = DIV_ROUND_UP(job->len / BDRV_SECTOR_SIZE,
-+ BACKUP_SECTORS_PER_CLUSTER);
-+
-+ for (; start < end; start++) {
-+ iov.iov_base = buf;
-+ iov.iov_len = VMA_CLUSTER_SIZE;
-+ qemu_iovec_init_external(&qiov, &iov, 1);
-+
-+ ret = blk_co_preadv(job->target, start * VMA_CLUSTER_SIZE,
-+ VMA_CLUSTER_SIZE, &qiov, 0);
-+ if (ret < 0) {
-+ vma_writer_set_error(job->vmaw, "read error", -1);
-+ goto out;
-+ }
-+
-+ size_t zb = 0;
-+ if (vma_writer_write(job->vmaw, job->dev_id, start, buf, &zb) < 0) {
-+ vma_writer_set_error(job->vmaw, "backup_dump_cb vma_writer_write failed", -1);
-+ goto out;
-+ }
-+ }
-+
-+
-+out:
-+ if (vma_writer_close_stream(job->vmaw, job->dev_id) <= 0) {
-+ Error *err = NULL;
-+ if (vma_writer_close(job->vmaw, &err) != 0) {
-+ g_warning("vma_writer_close failed %s", error_get_pretty(err));
-+ }
-+ }
-+}
-+
-+static int create_archive(int argc, char **argv)
-+{
-+ int i, c;
-+ int verbose = 0;
-+ const char *archivename;
-+ GList *config_files = NULL;
-+
-+ for (;;) {
-+ c = getopt(argc, argv, "hvc:");
-+ if (c == -1) {
-+ break;
-+ }
-+ switch (c) {
-+ case '?':
-+ case 'h':
-+ help();
-+ break;
-+ case 'c':
-+ config_files = g_list_append(config_files, optarg);
-+ break;
-+ case 'v':
-+ verbose = 1;
-+ break;
-+ default:
-+ g_assert_not_reached();
-+ }
-+ }
-+
-+
-+ /* make sure we an archive name */
-+ if ((optind + 1) > argc) {
-+ help();
-+ }
-+
-+ archivename = argv[optind++];
-+
-+ uuid_t uuid;
-+ uuid_generate(uuid);
-+
-+ Error *local_err = NULL;
-+ VmaWriter *vmaw = vma_writer_create(archivename, uuid, &local_err);
-+
-+ if (vmaw == NULL) {
-+ g_error("%s", error_get_pretty(local_err));
-+ }
-+
-+ GList *l = config_files;
-+ while (l && l->data) {
-+ char *name = l->data;
-+ char *cdata = NULL;
-+ gsize clen = 0;
-+ GError *err = NULL;
-+ if (!g_file_get_contents(name, &cdata, &clen, &err)) {
-+ unlink(archivename);
-+ g_error("Unable to read file: %s", err->message);
-+ }
-+
-+ if (vma_writer_add_config(vmaw, name, cdata, clen) != 0) {
-+ unlink(archivename);
-+ g_error("Unable to append config data %s (len = %zd)",
-+ name, clen);
-+ }
-+ l = g_list_next(l);
-+ }
-+
-+ int devcount = 0;
-+ while (optind < argc) {
-+ const char *path = argv[optind++];
-+ char *devname = NULL;
-+ path = extract_devname(path, &devname, devcount++);
-+
-+ Error *errp = NULL;
-+ BlockBackend *target;
-+
-+ target = blk_new_open(path, NULL, NULL, 0, &errp);
-+ if (!target) {
-+ unlink(archivename);
-+ g_error("bdrv_open '%s' failed - %s", path, error_get_pretty(errp));
-+ }
-+ int64_t size = blk_getlength(target);
-+ int dev_id = vma_writer_register_stream(vmaw, devname, size);
-+ if (dev_id <= 0) {
-+ unlink(archivename);
-+ g_error("vma_writer_register_stream '%s' failed", devname);
-+ }
-+
-+ BackupJob *job = g_new0(BackupJob, 1);
-+ job->len = size;
-+ job->target = target;
-+ job->vmaw = vmaw;
-+ job->dev_id = dev_id;
-+
-+ Coroutine *co = qemu_coroutine_create(backup_run, job);
-+ qemu_coroutine_enter(co);
-+ }
-+
-+ VmaStatus vmastat;
-+ int percent = 0;
-+ int last_percent = -1;
-+
-+ if (devcount) {
-+ while (1) {
-+ main_loop_wait(false);
-+ vma_writer_get_status(vmaw, &vmastat);
-+
-+ if (verbose) {
-+
-+ uint64_t total = 0;
-+ uint64_t transferred = 0;
-+ uint64_t zero_bytes = 0;
-+
-+ int i;
-+ for (i = 0; i < 256; i++) {
-+ if (vmastat.stream_info[i].size) {
-+ total += vmastat.stream_info[i].size;
-+ transferred += vmastat.stream_info[i].transferred;
-+ zero_bytes += vmastat.stream_info[i].zero_bytes;
-+ }
-+ }
-+ percent = (transferred*100)/total;
-+ if (percent != last_percent) {
-+ fprintf(stderr, "progress %d%% %zd/%zd %zd\n", percent,
-+ transferred, total, zero_bytes);
-+ fflush(stderr);
-+
-+ last_percent = percent;
-+ }
-+ }
-+
-+ if (vmastat.closed) {
-+ break;
-+ }
-+ }
-+ } else {
-+ Coroutine *co = qemu_coroutine_create(backup_run_empty, vmaw);
-+ qemu_coroutine_enter(co);
-+ while (1) {
-+ main_loop_wait(false);
-+ vma_writer_get_status(vmaw, &vmastat);
-+ if (vmastat.closed) {
-+ break;
-+ }
-+ }
-+ }
-+
-+ bdrv_drain_all();
-+
-+ vma_writer_get_status(vmaw, &vmastat);
-+
-+ if (verbose) {
-+ for (i = 0; i < 256; i++) {
-+ VmaStreamInfo *si = &vmastat.stream_info[i];
-+ if (si->size) {
-+ fprintf(stderr, "image %s: size=%zd zeros=%zd saved=%zd\n",
-+ si->devname, si->size, si->zero_bytes,
-+ si->size - si->zero_bytes);
-+ }
-+ }
-+ }
-+
-+ if (vmastat.status < 0) {
-+ unlink(archivename);
-+ g_error("creating vma archive failed");
-+ }
-+
-+ return 0;
-+}
-+
-+static int dump_config(int argc, char **argv)
-+{
-+ int c, ret = 0;
-+ const char *filename;
-+ const char *config_name = "qemu-server.conf";
-+
-+ for (;;) {
-+ c = getopt(argc, argv, "hc:");
-+ if (c == -1) {
-+ break;
-+ }
-+ switch (c) {
-+ case '?':
-+ case 'h':
-+ help();
-+ break;
-+ case 'c':
-+ config_name = optarg;
-+ break;
-+ default:
-+ help();
-+ }
-+ }
-+
-+ /* Get the filename */
-+ if ((optind + 1) != argc) {
-+ help();
-+ }
-+ filename = argv[optind++];
-+
-+ Error *errp = NULL;
-+ VmaReader *vmar = vma_reader_create(filename, &errp);
-+
-+ if (!vmar) {
-+ g_error("%s", error_get_pretty(errp));
-+ }
-+
-+ int found = 0;
-+ GList *l = vma_reader_get_config_data(vmar);
-+ while (l && l->data) {
-+ VmaConfigData *cdata = (VmaConfigData *)l->data;
-+ l = g_list_next(l);
-+ if (strcmp(cdata->name, config_name) == 0) {
-+ found = 1;
-+ fwrite(cdata->data, cdata->len, 1, stdout);
-+ break;
-+ }
-+ }
-+
-+ vma_reader_destroy(vmar);
-+
-+ bdrv_close_all();
-+
-+ if (!found) {
-+ fprintf(stderr, "unable to find configuration data '%s'\n", config_name);
-+ return -1;
-+ }
-+
-+ return ret;
-+}
-+
-+int main(int argc, char **argv)
-+{
-+ const char *cmdname;
-+ Error *main_loop_err = NULL;
-+
-+ error_set_progname(argv[0]);
-+
-+ if (qemu_init_main_loop(&main_loop_err)) {
-+ g_error("%s", error_get_pretty(main_loop_err));
-+ }
-+
-+ bdrv_init();
-+
-+ if (argc < 2) {
-+ help();
-+ }
-+
-+ cmdname = argv[1];
-+ argc--; argv++;
-+
-+
-+ if (!strcmp(cmdname, "list")) {
-+ return list_content(argc, argv);
-+ } else if (!strcmp(cmdname, "create")) {
-+ return create_archive(argc, argv);
-+ } else if (!strcmp(cmdname, "extract")) {
-+ return extract_content(argc, argv);
-+ } else if (!strcmp(cmdname, "verify")) {
-+ return verify_content(argc, argv);
-+ } else if (!strcmp(cmdname, "config")) {
-+ return dump_config(argc, argv);
-+ }
-+
-+ help();
-+ return 0;
-+}
-diff --git a/vma.h b/vma.h
-new file mode 100644
-index 0000000000..c895c97f6d
---- /dev/null
-+++ b/vma.h
-@@ -0,0 +1,150 @@
-+/*
-+ * VMA: Virtual Machine Archive
-+ *
-+ * Copyright (C) Proxmox Server Solutions
-+ *
-+ * Authors:
-+ * Dietmar Maurer (dietmar@proxmox.com)
-+ *
-+ * This work is licensed under the terms of the GNU GPL, version 2 or later.
-+ * See the COPYING file in the top-level directory.
-+ *
-+ */
-+
-+#ifndef BACKUP_VMA_H
-+#define BACKUP_VMA_H
-+
-+#include <uuid/uuid.h>
-+#include "qapi/error.h"
-+#include "block/block.h"
-+
-+#define VMA_BLOCK_BITS 12
-+#define VMA_BLOCK_SIZE (1<<VMA_BLOCK_BITS)
-+#define VMA_CLUSTER_BITS (VMA_BLOCK_BITS+4)
-+#define VMA_CLUSTER_SIZE (1<<VMA_CLUSTER_BITS)
-+
-+#if VMA_CLUSTER_SIZE != 65536
-+#error unexpected cluster size
-+#endif
-+
-+#define VMA_EXTENT_HEADER_SIZE 512
-+#define VMA_BLOCKS_PER_EXTENT 59
-+#define VMA_MAX_CONFIGS 256
-+
-+#define VMA_MAX_EXTENT_SIZE \
-+ (VMA_EXTENT_HEADER_SIZE+VMA_CLUSTER_SIZE*VMA_BLOCKS_PER_EXTENT)
-+#if VMA_MAX_EXTENT_SIZE != 3867136
-+#error unexpected VMA_EXTENT_SIZE
-+#endif
-+
-+/* File Format Definitions */
-+
-+#define VMA_MAGIC (GUINT32_TO_BE(('V'<<24)|('M'<<16)|('A'<<8)|0x00))
-+#define VMA_EXTENT_MAGIC (GUINT32_TO_BE(('V'<<24)|('M'<<16)|('A'<<8)|'E'))
-+
-+typedef struct VmaDeviceInfoHeader {
-+ uint32_t devname_ptr; /* offset into blob_buffer table */
-+ uint32_t reserved0;
-+ uint64_t size; /* device size in bytes */
-+ uint64_t reserved1;
-+ uint64_t reserved2;
-+} VmaDeviceInfoHeader;
-+
-+typedef struct VmaHeader {
-+ uint32_t magic;
-+ uint32_t version;
-+ unsigned char uuid[16];
-+ int64_t ctime;
-+ unsigned char md5sum[16];
-+
-+ uint32_t blob_buffer_offset;
-+ uint32_t blob_buffer_size;
-+ uint32_t header_size;
-+
-+ unsigned char reserved[1984];
-+
-+ uint32_t config_names[VMA_MAX_CONFIGS]; /* offset into blob_buffer table */
-+ uint32_t config_data[VMA_MAX_CONFIGS]; /* offset into blob_buffer table */
-+
-+ uint32_t reserved1;
-+
-+ VmaDeviceInfoHeader dev_info[256];
-+} VmaHeader;
-+
-+typedef struct VmaExtentHeader {
-+ uint32_t magic;
-+ uint16_t reserved1;
-+ uint16_t block_count;
-+ unsigned char uuid[16];
-+ unsigned char md5sum[16];
-+ uint64_t blockinfo[VMA_BLOCKS_PER_EXTENT];
-+} VmaExtentHeader;
-+
-+/* functions/definitions to read/write vma files */
-+
-+typedef struct VmaReader VmaReader;
-+
-+typedef struct VmaWriter VmaWriter;
-+
-+typedef struct VmaConfigData {
-+ const char *name;
-+ const void *data;
-+ uint32_t len;
-+} VmaConfigData;
-+
-+typedef struct VmaStreamInfo {
-+ uint64_t size;
-+ uint64_t cluster_count;
-+ uint64_t transferred;
-+ uint64_t zero_bytes;
-+ int finished;
-+ char *devname;
-+} VmaStreamInfo;
-+
-+typedef struct VmaStatus {
-+ int status;
-+ bool closed;
-+ char errmsg[8192];
-+ char uuid_str[37];
-+ VmaStreamInfo stream_info[256];
-+} VmaStatus;
-+
-+typedef struct VmaDeviceInfo {
-+ uint64_t size; /* device size in bytes */
-+ const char *devname;
-+} VmaDeviceInfo;
-+
-+VmaWriter *vma_writer_create(const char *filename, uuid_t uuid, Error **errp);
-+int vma_writer_close(VmaWriter *vmaw, Error **errp);
-+void vma_writer_error_propagate(VmaWriter *vmaw, Error **errp);
-+void vma_writer_destroy(VmaWriter *vmaw);
-+int vma_writer_add_config(VmaWriter *vmaw, const char *name, gpointer data,
-+ size_t len);
-+int vma_writer_register_stream(VmaWriter *vmaw, const char *devname,
-+ size_t size);
-+
-+int64_t coroutine_fn vma_writer_write(VmaWriter *vmaw, uint8_t dev_id,
-+ int64_t cluster_num,
-+ const unsigned char *buf,
-+ size_t *zero_bytes);
-+
-+int coroutine_fn vma_writer_close_stream(VmaWriter *vmaw, uint8_t dev_id);
-+int coroutine_fn vma_writer_flush_output(VmaWriter *vmaw);
-+
-+int vma_writer_get_status(VmaWriter *vmaw, VmaStatus *status);
-+void vma_writer_set_error(VmaWriter *vmaw, const char *fmt, ...);
-+
-+
-+VmaReader *vma_reader_create(const char *filename, Error **errp);
-+void vma_reader_destroy(VmaReader *vmar);
-+VmaHeader *vma_reader_get_header(VmaReader *vmar);
-+GList *vma_reader_get_config_data(VmaReader *vmar);
-+VmaDeviceInfo *vma_reader_get_device_info(VmaReader *vmar, guint8 dev_id);
-+int vma_reader_register_bs(VmaReader *vmar, guint8 dev_id,
-+ BlockBackend *target, bool write_zeroes,
-+ Error **errp);
-+int vma_reader_restore(VmaReader *vmar, int vmstate_fd, bool verbose,
-+ Error **errp);
-+int vma_reader_verify(VmaReader *vmar, bool verbose, Error **errp);
-+
-+#endif /* BACKUP_VMA_H */
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Wolfgang Bumiller <w.bumiller@proxmox.com>
+Date: Tue, 10 Mar 2020 12:55:19 +0100
+Subject: [PATCH 22/32] PVE: [Up+Config] file-posix: make locking optiono on
+ create
+
+Otherwise creating images on nfs/cifs can be problematic.
+
+Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ block/file-posix.c | 61 +++++++++++++++++++++++++++++---------------
+ qapi/block-core.json | 3 ++-
+ 2 files changed, 43 insertions(+), 21 deletions(-)
+
+diff --git a/block/file-posix.c b/block/file-posix.c
+index 44b49265ae..0722b0f529 100644
+--- a/block/file-posix.c
++++ b/block/file-posix.c
+@@ -2250,6 +2250,7 @@ raw_co_create(BlockdevCreateOptions *options, Error **errp)
+ int fd;
+ uint64_t perm, shared;
+ int result = 0;
++ bool locked = false;
+
+ /* Validate options and set default values */
+ assert(options->driver == BLOCKDEV_DRIVER_FILE);
+@@ -2283,19 +2284,22 @@ raw_co_create(BlockdevCreateOptions *options, Error **errp)
+ perm = BLK_PERM_WRITE | BLK_PERM_RESIZE;
+ shared = BLK_PERM_ALL & ~BLK_PERM_RESIZE;
+
+- /* Step one: Take locks */
+- result = raw_apply_lock_bytes(NULL, fd, perm, ~shared, false, errp);
+- if (result < 0) {
+- goto out_close;
+- }
++ if (file_opts->locking != ON_OFF_AUTO_OFF) {
++ /* Step one: Take locks */
++ result = raw_apply_lock_bytes(NULL, fd, perm, ~shared, false, errp);
++ if (result < 0) {
++ goto out_close;
++ }
++ locked = true;
+
+- /* Step two: Check that nobody else has taken conflicting locks */
+- result = raw_check_lock_bytes(fd, perm, shared, errp);
+- if (result < 0) {
+- error_append_hint(errp,
+- "Is another process using the image [%s]?\n",
+- file_opts->filename);
+- goto out_unlock;
++ /* Step two: Check that nobody else has taken conflicting locks */
++ result = raw_check_lock_bytes(fd, perm, shared, errp);
++ if (result < 0) {
++ error_append_hint(errp,
++ "Is another process using the image [%s]?\n",
++ file_opts->filename);
++ goto out_unlock;
++ }
+ }
+
+ /* Clear the file by truncating it to 0 */
+@@ -2328,13 +2332,15 @@ raw_co_create(BlockdevCreateOptions *options, Error **errp)
+ }
+
+ out_unlock:
+- raw_apply_lock_bytes(NULL, fd, 0, 0, true, &local_err);
+- if (local_err) {
+- /* The above call should not fail, and if it does, that does
+- * not mean the whole creation operation has failed. So
+- * report it the user for their convenience, but do not report
+- * it to the caller. */
+- warn_report_err(local_err);
++ if (locked) {
++ raw_apply_lock_bytes(NULL, fd, 0, 0, true, &local_err);
++ if (local_err) {
++ /* The above call should not fail, and if it does, that does
++ * not mean the whole creation operation has failed. So
++ * report it the user for their convenience, but do not report
++ * it to the caller. */
++ warn_report_err(local_err);
++ }
+ }
+
+ out_close:
+@@ -2355,6 +2361,7 @@ static int coroutine_fn raw_co_create_opts(const char *filename, QemuOpts *opts,
+ PreallocMode prealloc;
+ char *buf = NULL;
+ Error *local_err = NULL;
++ OnOffAuto locking;
+
+ /* Skip file: protocol prefix */
+ strstart(filename, "file:", &filename);
+@@ -2372,6 +2379,18 @@ static int coroutine_fn raw_co_create_opts(const char *filename, QemuOpts *opts,
+ return -EINVAL;
+ }
+
++ locking = qapi_enum_parse(&OnOffAuto_lookup,
++ qemu_opt_get(opts, "locking"),
++ ON_OFF_AUTO_AUTO, &local_err);
++ if (local_err) {
++ error_propagate(errp, local_err);
++ return -EINVAL;
++ }
++
++ if (locking == ON_OFF_AUTO_AUTO) {
++ locking = ON_OFF_AUTO_OFF;
++ }
++
+ options = (BlockdevCreateOptions) {
+ .driver = BLOCKDEV_DRIVER_FILE,
+ .u.file = {
+@@ -2381,6 +2400,8 @@ static int coroutine_fn raw_co_create_opts(const char *filename, QemuOpts *opts,
+ .preallocation = prealloc,
+ .has_nocow = true,
+ .nocow = nocow,
++ .has_locking = true,
++ .locking = locking,
+ },
+ };
+ return raw_co_create(&options, errp);
+@@ -2901,7 +2922,7 @@ static int raw_check_perm(BlockDriverState *bs, uint64_t perm, uint64_t shared,
+ }
+
+ /* Copy locks to the new fd */
+- if (s->perm_change_fd) {
++ if (s->use_lock && s->perm_change_fd) {
+ ret = raw_apply_lock_bytes(NULL, s->perm_change_fd, perm, ~shared,
+ false, errp);
+ if (ret < 0) {
+diff --git a/qapi/block-core.json b/qapi/block-core.json
+index 0cf68fea14..783a868eb2 100644
+--- a/qapi/block-core.json
++++ b/qapi/block-core.json
+@@ -4259,7 +4259,8 @@
+ 'data': { 'filename': 'str',
+ 'size': 'size',
+ '*preallocation': 'PreallocMode',
+- '*nocow': 'bool' } }
++ '*nocow': 'bool',
++ '*locking': 'OnOffAuto' } }
+
+ ##
+ # @BlockdevCreateOptionsGluster:
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Thu, 15 Feb 2018 11:07:56 +0100
-Subject: [PATCH] PVE: vma: add throttling options to drive mapping fifo
- protocol
-
-We now need to call initialize the qom module as well.
-
-Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- vma.c | 88 +++++++++++++++++++++++++++++++++++++++++++++++++++--------
- 1 file changed, 76 insertions(+), 12 deletions(-)
-
-diff --git a/vma.c b/vma.c
-index 1b59fd1555..f9f5c308fe 100644
---- a/vma.c
-+++ b/vma.c
-@@ -18,7 +18,8 @@
- #include "qemu-common.h"
- #include "qemu/error-report.h"
- #include "qemu/main-loop.h"
--#include "qapi/qmp/qstring.h"
-+#include "qemu/cutils.h"
-+#include "qapi/qmp/qdict.h"
- #include "sysemu/block-backend.h"
-
- static void help(void)
-@@ -132,9 +133,39 @@ typedef struct RestoreMap {
- char *devname;
- char *path;
- char *format;
-+ uint64_t throttling_bps;
-+ char *throttling_group;
- bool write_zero;
- } RestoreMap;
-
-+static bool try_parse_option(char **line, const char *optname, char **out, const char *inbuf) {
-+ size_t optlen = strlen(optname);
-+ if (strncmp(*line, optname, optlen) != 0 || (*line)[optlen] != '=') {
-+ return false;
-+ }
-+ if (*out) {
-+ g_error("read map failed - duplicate value for option '%s'", optname);
-+ }
-+ char *value = (*line) + optlen + 1; /* including a '=' */
-+ char *colon = strchr(value, ':');
-+ if (!colon) {
-+ g_error("read map failed - option '%s' not terminated ('%s')",
-+ optname, inbuf);
-+ }
-+ *line = colon+1;
-+ *out = g_strndup(value, colon - value);
-+ return true;
-+}
-+
-+static uint64_t verify_u64(const char *text) {
-+ uint64_t value;
-+ const char *endptr = NULL;
-+ if (qemu_strtou64(text, &endptr, 0, &value) != 0 || !endptr || *endptr) {
-+ g_error("read map failed - not a number: %s", text);
-+ }
-+ return value;
-+}
-+
- static int extract_content(int argc, char **argv)
- {
- int c, ret = 0;
-@@ -208,6 +239,9 @@ static int extract_content(int argc, char **argv)
- while (1) {
- char inbuf[8192];
- char *line = fgets(inbuf, sizeof(inbuf), map);
-+ char *format = NULL;
-+ char *bps = NULL;
-+ char *group = NULL;
- if (!line || line[0] == '\0' || !strcmp(line, "done\n")) {
- break;
- }
-@@ -219,15 +253,19 @@ static int extract_content(int argc, char **argv)
- }
- }
-
-- char *format = NULL;
-- if (strncmp(line, "format=", sizeof("format=")-1) == 0) {
-- format = line + sizeof("format=")-1;
-- char *colon = strchr(format, ':');
-- if (!colon) {
-- g_error("read map failed - found only a format ('%s')", inbuf);
-+ while (1) {
-+ if (!try_parse_option(&line, "format", &format, inbuf) &&
-+ !try_parse_option(&line, "throttling.bps", &bps, inbuf) &&
-+ !try_parse_option(&line, "throttling.group", &group, inbuf))
-+ {
-+ break;
- }
-- format = g_strndup(format, colon - format);
-- line = colon+1;
-+ }
-+
-+ uint64_t bps_value = 0;
-+ if (bps) {
-+ bps_value = verify_u64(bps);
-+ g_free(bps);
- }
-
- const char *path;
-@@ -253,6 +291,8 @@ static int extract_content(int argc, char **argv)
- map->devname = g_strdup(devname);
- map->path = g_strdup(path);
- map->format = format;
-+ map->throttling_bps = bps_value;
-+ map->throttling_group = group;
- map->write_zero = write_zero;
-
- g_hash_table_insert(devmap, map->devname, map);
-@@ -280,6 +320,8 @@ static int extract_content(int argc, char **argv)
- } else if (di) {
- char *devfn = NULL;
- const char *format = NULL;
-+ uint64_t throttling_bps = 0;
-+ const char *throttling_group = NULL;
- int flags = BDRV_O_RDWR | BDRV_O_NO_FLUSH;
- bool write_zero = true;
-
-@@ -291,6 +333,8 @@ static int extract_content(int argc, char **argv)
- }
- devfn = map->path;
- format = map->format;
-+ throttling_bps = map->throttling_bps;
-+ throttling_group = map->throttling_group;
- write_zero = map->write_zero;
- } else {
- devfn = g_strdup_printf("%s/tmp-disk-%s.raw",
-@@ -315,7 +359,7 @@ static int extract_content(int argc, char **argv)
- if (format) {
- /* explicit format from commandline */
- options = qdict_new();
-- qdict_put(options, "driver", qstring_from_str(format));
-+ qdict_put_str(options, "driver", format);
- } else if ((devlen > 4 && strcmp(devfn+devlen-4, ".raw") == 0) ||
- strncmp(devfn, "/dev/", 5) == 0)
- {
-@@ -324,15 +368,34 @@ static int extract_content(int argc, char **argv)
- */
- /* explicit raw format */
- options = qdict_new();
-- qdict_put(options, "driver", qstring_from_str("raw"));
-+ qdict_put_str(options, "driver", "raw");
- }
-
--
- if (errp || !(blk = blk_new_open(devfn, NULL, options, flags, &errp))) {
- g_error("can't open file %s - %s", devfn,
- error_get_pretty(errp));
- }
-
-+ if (throttling_group) {
-+ blk_io_limits_enable(blk, throttling_group);
-+ }
-+
-+ if (throttling_bps) {
-+ if (!throttling_group) {
-+ blk_io_limits_enable(blk, devfn);
-+ }
-+
-+ ThrottleConfig cfg;
-+ throttle_config_init(&cfg);
-+ cfg.buckets[THROTTLE_BPS_WRITE].avg = throttling_bps;
-+ Error *err = NULL;
-+ if (!throttle_is_valid(&cfg, &err)) {
-+ error_report_err(err);
-+ g_error("failed to apply throttling");
-+ }
-+ blk_set_io_limits(blk, &cfg);
-+ }
-+
- if (vma_reader_register_bs(vmar, i, blk, write_zero, &errp) < 0) {
- g_error("%s", error_get_pretty(errp));
- }
-@@ -730,6 +793,7 @@ int main(int argc, char **argv)
- }
-
- bdrv_init();
-+ module_call_init(MODULE_INIT_QOM);
-
- if (argc < 2) {
- help();
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Thomas Lamprecht <t.lamprecht@proxmox.com>
+Date: Tue, 10 Mar 2020 12:55:20 +0100
+Subject: [PATCH 23/32] PVE: savevm-async: kick AIO wait on block state write
+
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ savevm-async.c | 1 +
+ 1 file changed, 1 insertion(+)
+
+diff --git a/savevm-async.c b/savevm-async.c
+index 54ceeae26c..393d55af2a 100644
+--- a/savevm-async.c
++++ b/savevm-async.c
+@@ -158,6 +158,7 @@ static void coroutine_fn block_state_write_entry(void *opaque) {
+ BlkRwCo *rwco = opaque;
+ rwco->ret = blk_co_pwritev(snap_state.target, rwco->offset, rwco->qiov->size,
+ rwco->qiov, 0);
++ aio_wait_kick();
+ }
+
+ static ssize_t block_state_writev_buffer(void *opaque, struct iovec *iov,
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Thu, 22 Mar 2018 15:32:04 +0100
-Subject: [PATCH] PVE: vma: add cache option to device map
-
-Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- vma.c | 16 +++++++++++++++-
- 1 file changed, 15 insertions(+), 1 deletion(-)
-
-diff --git a/vma.c b/vma.c
-index f9f5c308fe..476b7bee00 100644
---- a/vma.c
-+++ b/vma.c
-@@ -135,6 +135,7 @@ typedef struct RestoreMap {
- char *format;
- uint64_t throttling_bps;
- char *throttling_group;
-+ char *cache;
- bool write_zero;
- } RestoreMap;
-
-@@ -242,6 +243,7 @@ static int extract_content(int argc, char **argv)
- char *format = NULL;
- char *bps = NULL;
- char *group = NULL;
-+ char *cache = NULL;
- if (!line || line[0] == '\0' || !strcmp(line, "done\n")) {
- break;
- }
-@@ -256,7 +258,8 @@ static int extract_content(int argc, char **argv)
- while (1) {
- if (!try_parse_option(&line, "format", &format, inbuf) &&
- !try_parse_option(&line, "throttling.bps", &bps, inbuf) &&
-- !try_parse_option(&line, "throttling.group", &group, inbuf))
-+ !try_parse_option(&line, "throttling.group", &group, inbuf) &&
-+ !try_parse_option(&line, "cache", &cache, inbuf))
- {
- break;
- }
-@@ -293,6 +296,7 @@ static int extract_content(int argc, char **argv)
- map->format = format;
- map->throttling_bps = bps_value;
- map->throttling_group = group;
-+ map->cache = cache;
- map->write_zero = write_zero;
-
- g_hash_table_insert(devmap, map->devname, map);
-@@ -322,6 +326,7 @@ static int extract_content(int argc, char **argv)
- const char *format = NULL;
- uint64_t throttling_bps = 0;
- const char *throttling_group = NULL;
-+ const char *cache = NULL;
- int flags = BDRV_O_RDWR | BDRV_O_NO_FLUSH;
- bool write_zero = true;
-
-@@ -335,6 +340,7 @@ static int extract_content(int argc, char **argv)
- format = map->format;
- throttling_bps = map->throttling_bps;
- throttling_group = map->throttling_group;
-+ cache = map->cache;
- write_zero = map->write_zero;
- } else {
- devfn = g_strdup_printf("%s/tmp-disk-%s.raw",
-@@ -356,6 +362,7 @@ static int extract_content(int argc, char **argv)
-
- size_t devlen = strlen(devfn);
- QDict *options = NULL;
-+ bool writethrough;
- if (format) {
- /* explicit format from commandline */
- options = qdict_new();
-@@ -370,12 +377,19 @@ static int extract_content(int argc, char **argv)
- options = qdict_new();
- qdict_put_str(options, "driver", "raw");
- }
-+ if (cache && bdrv_parse_cache_mode(cache, &flags, &writethrough)) {
-+ g_error("invalid cache option: %s\n", cache);
-+ }
-
- if (errp || !(blk = blk_new_open(devfn, NULL, options, flags, &errp))) {
- g_error("can't open file %s - %s", devfn,
- error_get_pretty(errp));
- }
-
-+ if (cache) {
-+ blk_set_enable_write_cache(blk, !writethrough);
-+ }
-+
- if (throttling_group) {
- blk_io_limits_enable(blk, throttling_group);
- }
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Wolfgang Bumiller <w.bumiller@proxmox.com>
+Date: Tue, 10 Mar 2020 12:55:21 +0100
+Subject: [PATCH 24/32] PVE: move snapshot cleanup into bottom half
+
+as per:
+ (0ceccd858a8d) migration: qemu_savevm_state_cleanup() in cleanup
+
+may affect held locks and therefore change assumptions made
+by that function!
+
+Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ savevm-async.c | 3 ++-
+ 1 file changed, 2 insertions(+), 1 deletion(-)
+
+diff --git a/savevm-async.c b/savevm-async.c
+index 393d55af2a..790e27ae37 100644
+--- a/savevm-async.c
++++ b/savevm-async.c
+@@ -201,6 +201,8 @@ static void process_savevm_cleanup(void *opaque)
+ int ret;
+ qemu_bh_delete(snap_state.cleanup_bh);
+ snap_state.cleanup_bh = NULL;
++ qemu_savevm_state_cleanup();
++
+ qemu_mutex_unlock_iothread();
+ qemu_thread_join(&snap_state.thread);
+ qemu_mutex_lock_iothread();
+@@ -277,7 +279,6 @@ static void *process_savevm_thread(void *opaque)
+ save_snapshot_error("qemu_savevm_state_iterate error %d", ret);
+ break;
+ }
+- qemu_savevm_state_cleanup();
+ DPRINTF("save complete\n");
+ break;
+ }
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Tue, 27 Mar 2018 10:49:03 +0200
-Subject: [PATCH] PVE: vma: remove forced NO_FLUSH option
-
-This one's rbd specific and in no way a sane choice for all
-types storages. Instead, we want to honor the cache option
-passed along.
-
-Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- vma.c | 2 +-
- 1 file changed, 1 insertion(+), 1 deletion(-)
-
-diff --git a/vma.c b/vma.c
-index 476b7bee00..3289fd722f 100644
---- a/vma.c
-+++ b/vma.c
-@@ -327,7 +327,7 @@ static int extract_content(int argc, char **argv)
- uint64_t throttling_bps = 0;
- const char *throttling_group = NULL;
- const char *cache = NULL;
-- int flags = BDRV_O_RDWR | BDRV_O_NO_FLUSH;
-+ int flags = BDRV_O_RDWR;
- bool write_zero = true;
-
- if (readmap) {
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Thu, 30 Aug 2018 14:52:56 +0200
-Subject: [PATCH] PVE: Add dummy -id command line parameter
-
-This used to be part of the qemu-side PVE authentication for
-VNC. Now this does nothing.
-
-Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- qemu-options.hx | 3 +++
- vl.c | 8 ++++++++
- 2 files changed, 11 insertions(+)
-
-diff --git a/qemu-options.hx b/qemu-options.hx
-index 34994daafd..6db02fe903 100644
---- a/qemu-options.hx
-+++ b/qemu-options.hx
-@@ -802,6 +802,9 @@ STEXI
- @table @option
- ETEXI
-
-+DEF("id", HAS_ARG, QEMU_OPTION_id,
-+ "-id n set the VMID", QEMU_ARCH_ALL)
-+
- DEF("fda", HAS_ARG, QEMU_OPTION_fda,
- "-fda/-fdb file use 'file' as floppy disk 0/1 image\n", QEMU_ARCH_ALL)
- DEF("fdb", HAS_ARG, QEMU_OPTION_fdb, "", QEMU_ARCH_ALL)
-diff --git a/vl.c b/vl.c
-index 1c5536e5bb..7ffcd271f1 100644
---- a/vl.c
-+++ b/vl.c
-@@ -2857,6 +2857,7 @@ static void user_register_global_props(void)
- int main(int argc, char **argv, char **envp)
- {
- int i;
-+ long vm_id;
- int snapshot, linux_boot;
- const char *initrd_filename;
- const char *kernel_filename, *kernel_cmdline;
-@@ -3570,6 +3571,13 @@ int main(int argc, char **argv, char **envp)
- exit(1);
- }
- break;
-+ case QEMU_OPTION_id:
-+ vm_id = strtol(optarg, (char **)&optarg, 10);
-+ if (*optarg != 0 || vm_id < 100 || vm_id > INT_MAX) {
-+ error_report("invalid -id argument %s", optarg);
-+ exit(1);
-+ }
-+ break;
- case QEMU_OPTION_vnc:
- vnc_parse(optarg, &error_fatal);
- break;
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Wolfgang Bumiller <w.bumiller@proxmox.com>
+Date: Tue, 10 Mar 2020 12:55:22 +0100
+Subject: [PATCH 25/32] PVE: monitor: disable oob capability
+
+A bisect revealed that commit 8258292e18c3
+("monitor: Remove "x-oob", offer capability "oob" unconditionally")
+causes unexpected hangs when restoring live snapshots from some
+types of block devices (particularly RBD).
+We need to figure out what's happnening there. For now, since we
+had this disabled before and probably don't need it now either,
+disable oob, so we can get a functioning qemu out...
+
+Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ monitor/qmp.c | 3 +--
+ 1 file changed, 1 insertion(+), 2 deletions(-)
+
+diff --git a/monitor/qmp.c b/monitor/qmp.c
+index b67a8e7d1f..8f44fed944 100644
+--- a/monitor/qmp.c
++++ b/monitor/qmp.c
+@@ -395,8 +395,7 @@ void monitor_init_qmp(Chardev *chr, bool pretty)
+ MonitorQMP *mon = g_new0(MonitorQMP, 1);
+
+ /* Note: we run QMP monitor in I/O thread when @chr supports that */
+- monitor_data_init(&mon->common, true, false,
+- qemu_chr_has_feature(chr, QEMU_CHAR_FEATURE_GCONTEXT));
++ monitor_data_init(&mon->common, true, false, false);
+
+ mon->pretty = pretty;
+
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Stefan Reiter <s.reiter@proxmox.com>
+Date: Tue, 10 Mar 2020 12:55:23 +0100
+Subject: [PATCH 26/32] PVE: Acquire aio_context before calling
+ block_job_add_bdrv
+
+Otherwise backups immediately fail with 'permission denied' since
+_add_bdrv tries to release a lock we don't own.
+
+Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
+---
+ blockjob.c | 10 ++++++++++
+ 1 file changed, 10 insertions(+)
+
+diff --git a/blockjob.c b/blockjob.c
+index c6e20e2fcd..4e6074f18c 100644
+--- a/blockjob.c
++++ b/blockjob.c
+@@ -436,10 +436,20 @@ void *block_job_create(const char *job_id, const BlockJobDriver *driver,
+ notifier_list_add(&job->job.on_ready, &job->ready_notifier);
+ notifier_list_add(&job->job.on_idle, &job->idle_notifier);
+
++ /* block_job_add_bdrv expects us to hold the aio context lock, so acquire it
++ * before calling if we're not in the main context anyway. */
++ if (job->job.aio_context != qemu_get_aio_context()) {
++ aio_context_acquire(job->job.aio_context);
++ }
++
+ error_setg(&job->blocker, "block device is in use by block job: %s",
+ job_type_str(&job->job));
+ block_job_add_bdrv(job, "main node", bs, 0, BLK_PERM_ALL, &error_abort);
+
++ if (job->job.aio_context != qemu_get_aio_context()) {
++ aio_context_release(job->job.aio_context);
++ }
++
+ bdrv_op_unblock(bs, BLOCK_OP_TYPE_DATAPLANE, job->blocker);
+
+ /* Disable request queuing in the BlockBackend to avoid deadlocks on drain:
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Mon, 4 Jul 2016 15:02:26 +0200
-Subject: [PATCH] PVE: [Config] Revert "target-i386: disable LINT0 after reset"
-
-This reverts commit b8eb5512fd8a115f164edbbe897cdf8884920ccb.
-
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- hw/intc/apic_common.c | 9 +++++++++
- 1 file changed, 9 insertions(+)
-
-diff --git a/hw/intc/apic_common.c b/hw/intc/apic_common.c
-index e764a2bb03..f66ea72d78 100644
---- a/hw/intc/apic_common.c
-+++ b/hw/intc/apic_common.c
-@@ -258,6 +258,15 @@ static void apic_reset_common(DeviceState *dev)
- info->vapic_base_update(s);
-
- apic_init_reset(dev);
-+
-+ if (bsp) {
-+ /*
-+ * LINT0 delivery mode on CPU #0 is set to ExtInt at initialization
-+ * time typically by BIOS, so PIC interrupt can be delivered to the
-+ * processor when local APIC is enabled.
-+ */
-+ s->lvt[APIC_LVT_LINT0] = 0x700;
-+ }
- }
-
- /* This function is only used for old state version 1 and 2 */
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Thomas Lamprecht <t.lamprecht@proxmox.com>
+Date: Tue, 10 Mar 2020 12:55:24 +0100
+Subject: [PATCH 27/32] PVE: [Compat]: 4.0 used balloon qemu-4-0-config-size
+ false here
+
+The underlying issue why this change from upstream to us arised in
+the first place is that QEMU 4.0 was already released at the point we
+run into this migration issue, so we did the then obvious fallback to
+false for virtio-balloon-device qemu-4-0-config-size.
+
+QEMU made that switch back in 4.1, where it now uses a backward
+compatible mechanism to detect if the bigger CFG sizes should be
+used, i.e., checking the VIRTIO_BALLOON_F_PAGE_POISON or
+VIRTIO_BALLOON_F_FREE_PAGE_HINT balloon feature flags.
+As for them, upstream released version 4.0 had this to true they keep
+it to true in their compatibility record for the 4.0 machine, to
+allow live migrations from 4.0 to 4.1.
+As for us, downstream released version 4.0 (first public release of
+this QEMU) had this to false, we change it back to false again, for
+the same reason.
+
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+---
+ hw/core/machine.c | 3 ++-
+ 1 file changed, 2 insertions(+), 1 deletion(-)
+
+diff --git a/hw/core/machine.c b/hw/core/machine.c
+index 1689ad3bf8..bdcb351ede 100644
+--- a/hw/core/machine.c
++++ b/hw/core/machine.c
+@@ -39,7 +39,8 @@ GlobalProperty hw_compat_4_0[] = {
+ { "virtio-vga", "edid", "false" },
+ { "virtio-gpu", "edid", "false" },
+ { "virtio-device", "use-started", "false" },
+- { "virtio-balloon-device", "qemu-4-0-config-size", "true" },
++ // PVE differed from upstream for 4.0 balloon cfg size
++ { "virtio-balloon-device", "qemu-4-0-config-size", "false" },
+ { "pl031", "migrate-tick-offset", "false" },
+ };
+ const size_t hw_compat_4_0_len = G_N_ELEMENTS(hw_compat_4_0);
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Fri, 26 Apr 2019 09:15:28 +0200
-Subject: [PATCH] PVE: [Up+Config] file-posix: make locking optiono on create
-
-Otherwise creating images on nfs/cifs can be problematic.
-
-Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- block/file-posix.c | 61 +++++++++++++++++++++++++++++---------------
- qapi/block-core.json | 3 ++-
- 2 files changed, 43 insertions(+), 21 deletions(-)
-
-diff --git a/block/file-posix.c b/block/file-posix.c
-index 71aa45ce5d..ae1cd3980e 100644
---- a/block/file-posix.c
-+++ b/block/file-posix.c
-@@ -2230,6 +2230,7 @@ raw_co_create(BlockdevCreateOptions *options, Error **errp)
- int fd;
- uint64_t perm, shared;
- int result = 0;
-+ bool locked = false;
-
- /* Validate options and set default values */
- assert(options->driver == BLOCKDEV_DRIVER_FILE);
-@@ -2263,19 +2264,22 @@ raw_co_create(BlockdevCreateOptions *options, Error **errp)
- perm = BLK_PERM_WRITE | BLK_PERM_RESIZE;
- shared = BLK_PERM_ALL & ~BLK_PERM_RESIZE;
-
-- /* Step one: Take locks */
-- result = raw_apply_lock_bytes(NULL, fd, perm, ~shared, false, errp);
-- if (result < 0) {
-- goto out_close;
-- }
-+ if (file_opts->locking != ON_OFF_AUTO_OFF) {
-+ /* Step one: Take locks */
-+ result = raw_apply_lock_bytes(NULL, fd, perm, ~shared, false, errp);
-+ if (result < 0) {
-+ goto out_close;
-+ }
-+ locked = true;
-
-- /* Step two: Check that nobody else has taken conflicting locks */
-- result = raw_check_lock_bytes(fd, perm, shared, errp);
-- if (result < 0) {
-- error_append_hint(errp,
-- "Is another process using the image [%s]?\n",
-- file_opts->filename);
-- goto out_unlock;
-+ /* Step two: Check that nobody else has taken conflicting locks */
-+ result = raw_check_lock_bytes(fd, perm, shared, errp);
-+ if (result < 0) {
-+ error_append_hint(errp,
-+ "Is another process using the image [%s]?\n",
-+ file_opts->filename);
-+ goto out_unlock;
-+ }
- }
-
- /* Clear the file by truncating it to 0 */
-@@ -2308,13 +2312,15 @@ raw_co_create(BlockdevCreateOptions *options, Error **errp)
- }
-
- out_unlock:
-- raw_apply_lock_bytes(NULL, fd, 0, 0, true, &local_err);
-- if (local_err) {
-- /* The above call should not fail, and if it does, that does
-- * not mean the whole creation operation has failed. So
-- * report it the user for their convenience, but do not report
-- * it to the caller. */
-- warn_report_err(local_err);
-+ if (locked) {
-+ raw_apply_lock_bytes(NULL, fd, 0, 0, true, &local_err);
-+ if (local_err) {
-+ /* The above call should not fail, and if it does, that does
-+ * not mean the whole creation operation has failed. So
-+ * report it the user for their convenience, but do not report
-+ * it to the caller. */
-+ warn_report_err(local_err);
-+ }
- }
-
- out_close:
-@@ -2335,6 +2341,7 @@ static int coroutine_fn raw_co_create_opts(const char *filename, QemuOpts *opts,
- PreallocMode prealloc;
- char *buf = NULL;
- Error *local_err = NULL;
-+ OnOffAuto locking;
-
- /* Skip file: protocol prefix */
- strstart(filename, "file:", &filename);
-@@ -2352,6 +2359,18 @@ static int coroutine_fn raw_co_create_opts(const char *filename, QemuOpts *opts,
- return -EINVAL;
- }
-
-+ locking = qapi_enum_parse(&OnOffAuto_lookup,
-+ qemu_opt_get(opts, "locking"),
-+ ON_OFF_AUTO_AUTO, &local_err);
-+ if (local_err) {
-+ error_propagate(errp, local_err);
-+ return -EINVAL;
-+ }
-+
-+ if (locking == ON_OFF_AUTO_AUTO) {
-+ locking = ON_OFF_AUTO_OFF;
-+ }
-+
- options = (BlockdevCreateOptions) {
- .driver = BLOCKDEV_DRIVER_FILE,
- .u.file = {
-@@ -2361,6 +2380,8 @@ static int coroutine_fn raw_co_create_opts(const char *filename, QemuOpts *opts,
- .preallocation = prealloc,
- .has_nocow = true,
- .nocow = nocow,
-+ .has_locking = true,
-+ .locking = locking,
- },
- };
- return raw_co_create(&options, errp);
-@@ -2838,7 +2859,7 @@ static int raw_check_perm(BlockDriverState *bs, uint64_t perm, uint64_t shared,
- }
-
- /* Copy locks to the new fd */
-- if (s->perm_change_fd) {
-+ if (s->use_lock && s->perm_change_fd) {
- ret = raw_apply_lock_bytes(NULL, s->perm_change_fd, perm, ~shared,
- false, errp);
- if (ret < 0) {
-diff --git a/qapi/block-core.json b/qapi/block-core.json
-index 97cb7ec41c..f432d4f3ec 100644
---- a/qapi/block-core.json
-+++ b/qapi/block-core.json
-@@ -4284,7 +4284,8 @@
- 'data': { 'filename': 'str',
- 'size': 'size',
- '*preallocation': 'PreallocMode',
-- '*nocow': 'bool' } }
-+ '*nocow': 'bool',
-+ '*locking': 'OnOffAuto' } }
-
- ##
- # @BlockdevCreateOptionsGluster:
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Dietmar Maurer <dietmar@proxmox.com>
+Date: Tue, 10 Mar 2020 12:55:25 +0100
+Subject: [PATCH 28/32] PVE: Allow version code in machine type
+
+E.g. pc-i440fx-4.0+pve3 would print 'pve3' as version code while
+selecting pc-i440fx-4.0 as machine type.
+
+Version is made available as 'pve-version' in query-machines (same as,
+and only if 'is-current').
+
+Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
+---
+ hw/core/machine-qmp-cmds.c | 6 ++++++
+ include/hw/boards.h | 2 ++
+ qapi/machine.json | 3 ++-
+ vl.c | 15 ++++++++++++++-
+ 4 files changed, 24 insertions(+), 2 deletions(-)
+
+diff --git a/hw/core/machine-qmp-cmds.c b/hw/core/machine-qmp-cmds.c
+index 1953633e82..ca8c0dc53d 100644
+--- a/hw/core/machine-qmp-cmds.c
++++ b/hw/core/machine-qmp-cmds.c
+@@ -234,6 +234,12 @@ MachineInfoList *qmp_query_machines(Error **errp)
+ if (strcmp(mc->name, MACHINE_GET_CLASS(current_machine)->name) == 0) {
+ info->has_is_current = true;
+ info->is_current = true;
++
++ // PVE version string only exists for current machine
++ if (mc->pve_version) {
++ info->has_pve_version = true;
++ info->pve_version = g_strdup(mc->pve_version);
++ }
+ }
+
+ if (mc->default_cpu_type) {
+diff --git a/include/hw/boards.h b/include/hw/boards.h
+index de45087f34..e24d2134c0 100644
+--- a/include/hw/boards.h
++++ b/include/hw/boards.h
+@@ -185,6 +185,8 @@ struct MachineClass {
+ const char *desc;
+ const char *deprecation_reason;
+
++ const char *pve_version;
++
+ void (*init)(MachineState *state);
+ void (*reset)(MachineState *state);
+ void (*wakeup)(MachineState *state);
+diff --git a/qapi/machine.json b/qapi/machine.json
+index cbdb6f6d66..a2bd4dd304 100644
+--- a/qapi/machine.json
++++ b/qapi/machine.json
+@@ -359,7 +359,8 @@
+ 'data': { 'name': 'str', '*alias': 'str',
+ '*is-default': 'bool', '*is-current': 'bool', 'cpu-max': 'int',
+ 'hotpluggable-cpus': 'bool', 'numa-mem-supported': 'bool',
+- 'deprecated': 'bool', '*default-cpu-type': 'str' } }
++ 'deprecated': 'bool', '*default-cpu-type': 'str',
++ '*pve-version': 'str' } }
+
+ ##
+ # @query-machines:
+diff --git a/vl.c b/vl.c
+index 4df15640c5..e7f3ce7607 100644
+--- a/vl.c
++++ b/vl.c
+@@ -2475,6 +2475,8 @@ static MachineClass *machine_parse(const char *name, GSList *machines)
+ {
+ MachineClass *mc;
+ GSList *el;
++ size_t pvever_index = 0;
++ gchar *name_clean;
+
+ if (is_help_option(name)) {
+ printf("Supported machines are:\n");
+@@ -2491,12 +2493,23 @@ static MachineClass *machine_parse(const char *name, GSList *machines)
+ exit(0);
+ }
+
+- mc = find_machine(name, machines);
++ // PVE version is specified with '+' as seperator, e.g. pc-i440fx+pvever
++ pvever_index = strcspn(name, "+");
++
++ name_clean = g_strndup(name, pvever_index);
++ mc = find_machine(name_clean, machines);
++ g_free(name_clean);
++
+ if (!mc) {
+ error_report("unsupported machine type");
+ error_printf("Use -machine help to list supported machines\n");
+ exit(1);
+ }
++
++ if (pvever_index < strlen(name)) {
++ mc->pve_version = &name[pvever_index+1];
++ }
++
+ return mc;
+ }
+
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Thomas Lamprecht <t.lamprecht@proxmox.com>
-Date: Fri, 21 Jun 2019 10:38:46 +0200
-Subject: [PATCH] PVE: savevm-async: kick AIO wait on block state write
-
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- savevm-async.c | 1 +
- 1 file changed, 1 insertion(+)
-
-diff --git a/savevm-async.c b/savevm-async.c
-index 2149010bb8..0bbbbf51ba 100644
---- a/savevm-async.c
-+++ b/savevm-async.c
-@@ -154,6 +154,7 @@ static void coroutine_fn block_state_write_entry(void *opaque) {
- BlkRwCo *rwco = opaque;
- rwco->ret = blk_co_pwritev(snap_state.target, rwco->offset, rwco->qiov->size,
- rwco->qiov, 0);
-+ aio_wait_kick();
- }
-
- static ssize_t block_state_writev_buffer(void *opaque, struct iovec *iov,
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Dietmar Maurer <dietmar@proxmox.com>
+Date: Tue, 10 Mar 2020 12:55:26 +0100
+Subject: [PATCH 29/32] PVE-Backup: add vma backup format code
+
+---
+ Makefile | 3 +-
+ Makefile.objs | 1 +
+ vma-reader.c | 857 ++++++++++++++++++++++++++++++++++++++++++++++++++
+ vma-writer.c | 771 +++++++++++++++++++++++++++++++++++++++++++++
+ vma.c | 837 ++++++++++++++++++++++++++++++++++++++++++++++++
+ vma.h | 150 +++++++++
+ 6 files changed, 2618 insertions(+), 1 deletion(-)
+ create mode 100644 vma-reader.c
+ create mode 100644 vma-writer.c
+ create mode 100644 vma.c
+ create mode 100644 vma.h
+
+diff --git a/Makefile b/Makefile
+index b437a346d7..18d2dba2e4 100644
+--- a/Makefile
++++ b/Makefile
+@@ -453,7 +453,7 @@ dummy := $(call unnest-vars,, \
+
+ include $(SRC_PATH)/tests/Makefile.include
+
+-all: $(DOCS) $(if $(BUILD_DOCS),sphinxdocs) $(TOOLS) $(HELPERS-y) recurse-all modules $(vhost-user-json-y)
++all: $(DOCS) $(if $(BUILD_DOCS),sphinxdocs) $(TOOLS) vma$(EXESUF) $(HELPERS-y) recurse-all modules $(vhost-user-json-y)
+
+ qemu-version.h: FORCE
+ $(call quiet-command, \
+@@ -567,6 +567,7 @@ qemu-img.o: qemu-img-cmds.h
+ qemu-img$(EXESUF): qemu-img.o $(authz-obj-y) $(block-obj-y) $(crypto-obj-y) $(io-obj-y) $(qom-obj-y) $(COMMON_LDADDS)
+ qemu-nbd$(EXESUF): qemu-nbd.o $(authz-obj-y) $(block-obj-y) $(crypto-obj-y) $(io-obj-y) $(qom-obj-y) $(COMMON_LDADDS)
+ qemu-io$(EXESUF): qemu-io.o $(authz-obj-y) $(block-obj-y) $(crypto-obj-y) $(io-obj-y) $(qom-obj-y) $(COMMON_LDADDS)
++vma$(EXESUF): vma.o vma-reader.o $(authz-obj-y) $(block-obj-y) $(crypto-obj-y) $(io-obj-y) $(qom-obj-y) $(COMMON_LDADDS)
+
+ qemu-bridge-helper$(EXESUF): qemu-bridge-helper.o $(COMMON_LDADDS)
+
+diff --git a/Makefile.objs b/Makefile.objs
+index f97b40f232..db7fbbe73b 100644
+--- a/Makefile.objs
++++ b/Makefile.objs
+@@ -18,6 +18,7 @@ block-obj-y += block.o blockjob.o job.o
+ block-obj-y += block/ scsi/
+ block-obj-y += qemu-io-cmds.o
+ block-obj-$(CONFIG_REPLICATION) += replication.o
++block-obj-y += vma-writer.o
+
+ block-obj-m = block/
+
+diff --git a/vma-reader.c b/vma-reader.c
+new file mode 100644
+index 0000000000..2b1d1cdab3
+--- /dev/null
++++ b/vma-reader.c
+@@ -0,0 +1,857 @@
++/*
++ * VMA: Virtual Machine Archive
++ *
++ * Copyright (C) 2012 Proxmox Server Solutions
++ *
++ * Authors:
++ * Dietmar Maurer (dietmar@proxmox.com)
++ *
++ * This work is licensed under the terms of the GNU GPL, version 2 or later.
++ * See the COPYING file in the top-level directory.
++ *
++ */
++
++#include "qemu/osdep.h"
++#include <glib.h>
++#include <uuid/uuid.h>
++
++#include "qemu-common.h"
++#include "qemu/timer.h"
++#include "qemu/ratelimit.h"
++#include "vma.h"
++#include "block/block.h"
++#include "sysemu/block-backend.h"
++
++static unsigned char zero_vma_block[VMA_BLOCK_SIZE];
++
++typedef struct VmaRestoreState {
++ BlockBackend *target;
++ bool write_zeroes;
++ unsigned long *bitmap;
++ int bitmap_size;
++} VmaRestoreState;
++
++struct VmaReader {
++ int fd;
++ GChecksum *md5csum;
++ GHashTable *blob_hash;
++ unsigned char *head_data;
++ VmaDeviceInfo devinfo[256];
++ VmaRestoreState rstate[256];
++ GList *cdata_list;
++ guint8 vmstate_stream;
++ uint32_t vmstate_clusters;
++ /* to show restore percentage if run with -v */
++ time_t start_time;
++ int64_t cluster_count;
++ int64_t clusters_read;
++ int64_t zero_cluster_data;
++ int64_t partial_zero_cluster_data;
++ int clusters_read_per;
++};
++
++static guint
++g_int32_hash(gconstpointer v)
++{
++ return *(const uint32_t *)v;
++}
++
++static gboolean
++g_int32_equal(gconstpointer v1, gconstpointer v2)
++{
++ return *((const uint32_t *)v1) == *((const uint32_t *)v2);
++}
++
++static int vma_reader_get_bitmap(VmaRestoreState *rstate, int64_t cluster_num)
++{
++ assert(rstate);
++ assert(rstate->bitmap);
++
++ unsigned long val, idx, bit;
++
++ idx = cluster_num / BITS_PER_LONG;
++
++ assert(rstate->bitmap_size > idx);
++
++ bit = cluster_num % BITS_PER_LONG;
++ val = rstate->bitmap[idx];
++
++ return !!(val & (1UL << bit));
++}
++
++static void vma_reader_set_bitmap(VmaRestoreState *rstate, int64_t cluster_num,
++ int dirty)
++{
++ assert(rstate);
++ assert(rstate->bitmap);
++
++ unsigned long val, idx, bit;
++
++ idx = cluster_num / BITS_PER_LONG;
++
++ assert(rstate->bitmap_size > idx);
++
++ bit = cluster_num % BITS_PER_LONG;
++ val = rstate->bitmap[idx];
++ if (dirty) {
++ if (!(val & (1UL << bit))) {
++ val |= 1UL << bit;
++ }
++ } else {
++ if (val & (1UL << bit)) {
++ val &= ~(1UL << bit);
++ }
++ }
++ rstate->bitmap[idx] = val;
++}
++
++typedef struct VmaBlob {
++ uint32_t start;
++ uint32_t len;
++ void *data;
++} VmaBlob;
++
++static const VmaBlob *get_header_blob(VmaReader *vmar, uint32_t pos)
++{
++ assert(vmar);
++ assert(vmar->blob_hash);
++
++ return g_hash_table_lookup(vmar->blob_hash, &pos);
++}
++
++static const char *get_header_str(VmaReader *vmar, uint32_t pos)
++{
++ const VmaBlob *blob = get_header_blob(vmar, pos);
++ if (!blob) {
++ return NULL;
++ }
++ const char *res = (char *)blob->data;
++ if (res[blob->len-1] != '\0') {
++ return NULL;
++ }
++ return res;
++}
++
++static ssize_t
++safe_read(int fd, unsigned char *buf, size_t count)
++{
++ ssize_t n;
++
++ do {
++ n = read(fd, buf, count);
++ } while (n < 0 && errno == EINTR);
++
++ return n;
++}
++
++static ssize_t
++full_read(int fd, unsigned char *buf, size_t len)
++{
++ ssize_t n;
++ size_t total;
++
++ total = 0;
++
++ while (len > 0) {
++ n = safe_read(fd, buf, len);
++
++ if (n == 0) {
++ return total;
++ }
++
++ if (n <= 0) {
++ break;
++ }
++
++ buf += n;
++ total += n;
++ len -= n;
++ }
++
++ if (len) {
++ return -1;
++ }
++
++ return total;
++}
++
++void vma_reader_destroy(VmaReader *vmar)
++{
++ assert(vmar);
++
++ if (vmar->fd >= 0) {
++ close(vmar->fd);
++ }
++
++ if (vmar->cdata_list) {
++ g_list_free(vmar->cdata_list);
++ }
++
++ int i;
++ for (i = 1; i < 256; i++) {
++ if (vmar->rstate[i].bitmap) {
++ g_free(vmar->rstate[i].bitmap);
++ }
++ }
++
++ if (vmar->md5csum) {
++ g_checksum_free(vmar->md5csum);
++ }
++
++ if (vmar->blob_hash) {
++ g_hash_table_destroy(vmar->blob_hash);
++ }
++
++ if (vmar->head_data) {
++ g_free(vmar->head_data);
++ }
++
++ g_free(vmar);
++
++};
++
++static int vma_reader_read_head(VmaReader *vmar, Error **errp)
++{
++ assert(vmar);
++ assert(errp);
++ assert(*errp == NULL);
++
++ unsigned char md5sum[16];
++ int i;
++ int ret = 0;
++
++ vmar->head_data = g_malloc(sizeof(VmaHeader));
++
++ if (full_read(vmar->fd, vmar->head_data, sizeof(VmaHeader)) !=
++ sizeof(VmaHeader)) {
++ error_setg(errp, "can't read vma header - %s",
++ errno ? g_strerror(errno) : "got EOF");
++ return -1;
++ }
++
++ VmaHeader *h = (VmaHeader *)vmar->head_data;
++
++ if (h->magic != VMA_MAGIC) {
++ error_setg(errp, "not a vma file - wrong magic number");
++ return -1;
++ }
++
++ uint32_t header_size = GUINT32_FROM_BE(h->header_size);
++ int need = header_size - sizeof(VmaHeader);
++ if (need <= 0) {
++ error_setg(errp, "wrong vma header size %d", header_size);
++ return -1;
++ }
++
++ vmar->head_data = g_realloc(vmar->head_data, header_size);
++ h = (VmaHeader *)vmar->head_data;
++
++ if (full_read(vmar->fd, vmar->head_data + sizeof(VmaHeader), need) !=
++ need) {
++ error_setg(errp, "can't read vma header data - %s",
++ errno ? g_strerror(errno) : "got EOF");
++ return -1;
++ }
++
++ memcpy(md5sum, h->md5sum, 16);
++ memset(h->md5sum, 0, 16);
++
++ g_checksum_reset(vmar->md5csum);
++ g_checksum_update(vmar->md5csum, vmar->head_data, header_size);
++ gsize csize = 16;
++ g_checksum_get_digest(vmar->md5csum, (guint8 *)(h->md5sum), &csize);
++
++ if (memcmp(md5sum, h->md5sum, 16) != 0) {
++ error_setg(errp, "wrong vma header chechsum");
++ return -1;
++ }
++
++ /* we can modify header data after checksum verify */
++ h->header_size = header_size;
++
++ h->version = GUINT32_FROM_BE(h->version);
++ if (h->version != 1) {
++ error_setg(errp, "wrong vma version %d", h->version);
++ return -1;
++ }
++
++ h->ctime = GUINT64_FROM_BE(h->ctime);
++ h->blob_buffer_offset = GUINT32_FROM_BE(h->blob_buffer_offset);
++ h->blob_buffer_size = GUINT32_FROM_BE(h->blob_buffer_size);
++
++ uint32_t bstart = h->blob_buffer_offset + 1;
++ uint32_t bend = h->blob_buffer_offset + h->blob_buffer_size;
++
++ if (bstart <= sizeof(VmaHeader)) {
++ error_setg(errp, "wrong vma blob buffer offset %d",
++ h->blob_buffer_offset);
++ return -1;
++ }
++
++ if (bend > header_size) {
++ error_setg(errp, "wrong vma blob buffer size %d/%d",
++ h->blob_buffer_offset, h->blob_buffer_size);
++ return -1;
++ }
++
++ while ((bstart + 2) <= bend) {
++ uint32_t size = vmar->head_data[bstart] +
++ (vmar->head_data[bstart+1] << 8);
++ if ((bstart + size + 2) <= bend) {
++ VmaBlob *blob = g_new0(VmaBlob, 1);
++ blob->start = bstart - h->blob_buffer_offset;
++ blob->len = size;
++ blob->data = vmar->head_data + bstart + 2;
++ g_hash_table_insert(vmar->blob_hash, &blob->start, blob);
++ }
++ bstart += size + 2;
++ }
++
++
++ int count = 0;
++ for (i = 1; i < 256; i++) {
++ VmaDeviceInfoHeader *dih = &h->dev_info[i];
++ uint32_t devname_ptr = GUINT32_FROM_BE(dih->devname_ptr);
++ uint64_t size = GUINT64_FROM_BE(dih->size);
++ const char *devname = get_header_str(vmar, devname_ptr);
++
++ if (size && devname) {
++ count++;
++ vmar->devinfo[i].size = size;
++ vmar->devinfo[i].devname = devname;
++
++ if (strcmp(devname, "vmstate") == 0) {
++ vmar->vmstate_stream = i;
++ }
++ }
++ }
++
++ for (i = 0; i < VMA_MAX_CONFIGS; i++) {
++ uint32_t name_ptr = GUINT32_FROM_BE(h->config_names[i]);
++ uint32_t data_ptr = GUINT32_FROM_BE(h->config_data[i]);
++
++ if (!(name_ptr && data_ptr)) {
++ continue;
++ }
++ const char *name = get_header_str(vmar, name_ptr);
++ const VmaBlob *blob = get_header_blob(vmar, data_ptr);
++
++ if (!(name && blob)) {
++ error_setg(errp, "vma contains invalid data pointers");
++ return -1;
++ }
++
++ VmaConfigData *cdata = g_new0(VmaConfigData, 1);
++ cdata->name = name;
++ cdata->data = blob->data;
++ cdata->len = blob->len;
++
++ vmar->cdata_list = g_list_append(vmar->cdata_list, cdata);
++ }
++
++ return ret;
++};
++
++VmaReader *vma_reader_create(const char *filename, Error **errp)
++{
++ assert(filename);
++ assert(errp);
++
++ VmaReader *vmar = g_new0(VmaReader, 1);
++
++ if (strcmp(filename, "-") == 0) {
++ vmar->fd = dup(0);
++ } else {
++ vmar->fd = open(filename, O_RDONLY);
++ }
++
++ if (vmar->fd < 0) {
++ error_setg(errp, "can't open file %s - %s\n", filename,
++ g_strerror(errno));
++ goto err;
++ }
++
++ vmar->md5csum = g_checksum_new(G_CHECKSUM_MD5);
++ if (!vmar->md5csum) {
++ error_setg(errp, "can't allocate cmsum\n");
++ goto err;
++ }
++
++ vmar->blob_hash = g_hash_table_new_full(g_int32_hash, g_int32_equal,
++ NULL, g_free);
++
++ if (vma_reader_read_head(vmar, errp) < 0) {
++ goto err;
++ }
++
++ return vmar;
++
++err:
++ if (vmar) {
++ vma_reader_destroy(vmar);
++ }
++
++ return NULL;
++}
++
++VmaHeader *vma_reader_get_header(VmaReader *vmar)
++{
++ assert(vmar);
++ assert(vmar->head_data);
++
++ return (VmaHeader *)(vmar->head_data);
++}
++
++GList *vma_reader_get_config_data(VmaReader *vmar)
++{
++ assert(vmar);
++ assert(vmar->head_data);
++
++ return vmar->cdata_list;
++}
++
++VmaDeviceInfo *vma_reader_get_device_info(VmaReader *vmar, guint8 dev_id)
++{
++ assert(vmar);
++ assert(dev_id);
++
++ if (vmar->devinfo[dev_id].size && vmar->devinfo[dev_id].devname) {
++ return &vmar->devinfo[dev_id];
++ }
++
++ return NULL;
++}
++
++static void allocate_rstate(VmaReader *vmar, guint8 dev_id,
++ BlockBackend *target, bool write_zeroes)
++{
++ assert(vmar);
++ assert(dev_id);
++
++ vmar->rstate[dev_id].target = target;
++ vmar->rstate[dev_id].write_zeroes = write_zeroes;
++
++ int64_t size = vmar->devinfo[dev_id].size;
++
++ int64_t bitmap_size = (size/BDRV_SECTOR_SIZE) +
++ (VMA_CLUSTER_SIZE/BDRV_SECTOR_SIZE) * BITS_PER_LONG - 1;
++ bitmap_size /= (VMA_CLUSTER_SIZE/BDRV_SECTOR_SIZE) * BITS_PER_LONG;
++
++ vmar->rstate[dev_id].bitmap_size = bitmap_size;
++ vmar->rstate[dev_id].bitmap = g_new0(unsigned long, bitmap_size);
++
++ vmar->cluster_count += size/VMA_CLUSTER_SIZE;
++}
++
++int vma_reader_register_bs(VmaReader *vmar, guint8 dev_id, BlockBackend *target,
++ bool write_zeroes, Error **errp)
++{
++ assert(vmar);
++ assert(target != NULL);
++ assert(dev_id);
++ assert(vmar->rstate[dev_id].target == NULL);
++
++ int64_t size = blk_getlength(target);
++ int64_t size_diff = size - vmar->devinfo[dev_id].size;
++
++ /* storage types can have different size restrictions, so it
++ * is not always possible to create an image with exact size.
++ * So we tolerate a size difference up to 4MB.
++ */
++ if ((size_diff < 0) || (size_diff > 4*1024*1024)) {
++ error_setg(errp, "vma_reader_register_bs for stream %s failed - "
++ "unexpected size %zd != %zd", vmar->devinfo[dev_id].devname,
++ size, vmar->devinfo[dev_id].size);
++ return -1;
++ }
++
++ allocate_rstate(vmar, dev_id, target, write_zeroes);
++
++ return 0;
++}
++
++static ssize_t safe_write(int fd, void *buf, size_t count)
++{
++ ssize_t n;
++
++ do {
++ n = write(fd, buf, count);
++ } while (n < 0 && errno == EINTR);
++
++ return n;
++}
++
++static size_t full_write(int fd, void *buf, size_t len)
++{
++ ssize_t n;
++ size_t total;
++
++ total = 0;
++
++ while (len > 0) {
++ n = safe_write(fd, buf, len);
++ if (n < 0) {
++ return n;
++ }
++ buf += n;
++ total += n;
++ len -= n;
++ }
++
++ if (len) {
++ /* incomplete write ? */
++ return -1;
++ }
++
++ return total;
++}
++
++static int restore_write_data(VmaReader *vmar, guint8 dev_id,
++ BlockBackend *target, int vmstate_fd,
++ unsigned char *buf, int64_t sector_num,
++ int nb_sectors, Error **errp)
++{
++ assert(vmar);
++
++ if (dev_id == vmar->vmstate_stream) {
++ if (vmstate_fd >= 0) {
++ int len = nb_sectors * BDRV_SECTOR_SIZE;
++ int res = full_write(vmstate_fd, buf, len);
++ if (res < 0) {
++ error_setg(errp, "write vmstate failed %d", res);
++ return -1;
++ }
++ }
++ } else {
++ int res = blk_pwrite(target, sector_num * BDRV_SECTOR_SIZE, buf, nb_sectors * BDRV_SECTOR_SIZE, 0);
++ if (res < 0) {
++ error_setg(errp, "blk_pwrite to %s failed (%d)",
++ bdrv_get_device_name(blk_bs(target)), res);
++ return -1;
++ }
++ }
++ return 0;
++}
++
++static int restore_extent(VmaReader *vmar, unsigned char *buf,
++ int extent_size, int vmstate_fd,
++ bool verbose, bool verify, Error **errp)
++{
++ assert(vmar);
++ assert(buf);
++
++ VmaExtentHeader *ehead = (VmaExtentHeader *)buf;
++ int start = VMA_EXTENT_HEADER_SIZE;
++ int i;
++
++ for (i = 0; i < VMA_BLOCKS_PER_EXTENT; i++) {
++ uint64_t block_info = GUINT64_FROM_BE(ehead->blockinfo[i]);
++ uint64_t cluster_num = block_info & 0xffffffff;
++ uint8_t dev_id = (block_info >> 32) & 0xff;
++ uint16_t mask = block_info >> (32+16);
++ int64_t max_sector;
++
++ if (!dev_id) {
++ continue;
++ }
++
++ VmaRestoreState *rstate = &vmar->rstate[dev_id];
++ BlockBackend *target = NULL;
++
++ if (dev_id != vmar->vmstate_stream) {
++ target = rstate->target;
++ if (!verify && !target) {
++ error_setg(errp, "got wrong dev id %d", dev_id);
++ return -1;
++ }
++
++ if (vma_reader_get_bitmap(rstate, cluster_num)) {
++ error_setg(errp, "found duplicated cluster %zd for stream %s",
++ cluster_num, vmar->devinfo[dev_id].devname);
++ return -1;
++ }
++ vma_reader_set_bitmap(rstate, cluster_num, 1);
++
++ max_sector = vmar->devinfo[dev_id].size/BDRV_SECTOR_SIZE;
++ } else {
++ max_sector = G_MAXINT64;
++ if (cluster_num != vmar->vmstate_clusters) {
++ error_setg(errp, "found out of order vmstate data");
++ return -1;
++ }
++ vmar->vmstate_clusters++;
++ }
++
++ vmar->clusters_read++;
++
++ if (verbose) {
++ time_t duration = time(NULL) - vmar->start_time;
++ int percent = (vmar->clusters_read*100)/vmar->cluster_count;
++ if (percent != vmar->clusters_read_per) {
++ printf("progress %d%% (read %zd bytes, duration %zd sec)\n",
++ percent, vmar->clusters_read*VMA_CLUSTER_SIZE,
++ duration);
++ fflush(stdout);
++ vmar->clusters_read_per = percent;
++ }
++ }
++
++ /* try to write whole clusters to speedup restore */
++ if (mask == 0xffff) {
++ if ((start + VMA_CLUSTER_SIZE) > extent_size) {
++ error_setg(errp, "short vma extent - too many blocks");
++ return -1;
++ }
++ int64_t sector_num = (cluster_num * VMA_CLUSTER_SIZE) /
++ BDRV_SECTOR_SIZE;
++ int64_t end_sector = sector_num +
++ VMA_CLUSTER_SIZE/BDRV_SECTOR_SIZE;
++
++ if (end_sector > max_sector) {
++ end_sector = max_sector;
++ }
++
++ if (end_sector <= sector_num) {
++ error_setg(errp, "got wrong block address - write beyond end");
++ return -1;
++ }
++
++ if (!verify) {
++ int nb_sectors = end_sector - sector_num;
++ if (restore_write_data(vmar, dev_id, target, vmstate_fd,
++ buf + start, sector_num, nb_sectors,
++ errp) < 0) {
++ return -1;
++ }
++ }
++
++ start += VMA_CLUSTER_SIZE;
++ } else {
++ int j;
++ int bit = 1;
++
++ for (j = 0; j < 16; j++) {
++ int64_t sector_num = (cluster_num*VMA_CLUSTER_SIZE +
++ j*VMA_BLOCK_SIZE)/BDRV_SECTOR_SIZE;
++
++ int64_t end_sector = sector_num +
++ VMA_BLOCK_SIZE/BDRV_SECTOR_SIZE;
++ if (end_sector > max_sector) {
++ end_sector = max_sector;
++ }
++
++ if (mask & bit) {
++ if ((start + VMA_BLOCK_SIZE) > extent_size) {
++ error_setg(errp, "short vma extent - too many blocks");
++ return -1;
++ }
++
++ if (end_sector <= sector_num) {
++ error_setg(errp, "got wrong block address - "
++ "write beyond end");
++ return -1;
++ }
++
++ if (!verify) {
++ int nb_sectors = end_sector - sector_num;
++ if (restore_write_data(vmar, dev_id, target, vmstate_fd,
++ buf + start, sector_num,
++ nb_sectors, errp) < 0) {
++ return -1;
++ }
++ }
++
++ start += VMA_BLOCK_SIZE;
++
++ } else {
++
++
++ if (end_sector > sector_num) {
++ /* Todo: use bdrv_co_write_zeroes (but that need to
++ * be run inside coroutine?)
++ */
++ int nb_sectors = end_sector - sector_num;
++ int zero_size = BDRV_SECTOR_SIZE*nb_sectors;
++ vmar->zero_cluster_data += zero_size;
++ if (mask != 0) {
++ vmar->partial_zero_cluster_data += zero_size;
++ }
++
++ if (rstate->write_zeroes && !verify) {
++ if (restore_write_data(vmar, dev_id, target, vmstate_fd,
++ zero_vma_block, sector_num,
++ nb_sectors, errp) < 0) {
++ return -1;
++ }
++ }
++ }
++ }
++
++ bit = bit << 1;
++ }
++ }
++ }
++
++ if (start != extent_size) {
++ error_setg(errp, "vma extent error - missing blocks");
++ return -1;
++ }
++
++ return 0;
++}
++
++static int vma_reader_restore_full(VmaReader *vmar, int vmstate_fd,
++ bool verbose, bool verify,
++ Error **errp)
++{
++ assert(vmar);
++ assert(vmar->head_data);
++
++ int ret = 0;
++ unsigned char buf[VMA_MAX_EXTENT_SIZE];
++ int buf_pos = 0;
++ unsigned char md5sum[16];
++ VmaHeader *h = (VmaHeader *)vmar->head_data;
++
++ vmar->start_time = time(NULL);
++
++ while (1) {
++ int bytes = full_read(vmar->fd, buf + buf_pos, sizeof(buf) - buf_pos);
++ if (bytes < 0) {
++ error_setg(errp, "read failed - %s", g_strerror(errno));
++ return -1;
++ }
++
++ buf_pos += bytes;
++
++ if (!buf_pos) {
++ break; /* EOF */
++ }
++
++ if (buf_pos < VMA_EXTENT_HEADER_SIZE) {
++ error_setg(errp, "read short extent (%d bytes)", buf_pos);
++ return -1;
++ }
++
++ VmaExtentHeader *ehead = (VmaExtentHeader *)buf;
++
++ /* extract md5sum */
++ memcpy(md5sum, ehead->md5sum, sizeof(ehead->md5sum));
++ memset(ehead->md5sum, 0, sizeof(ehead->md5sum));
++
++ g_checksum_reset(vmar->md5csum);
++ g_checksum_update(vmar->md5csum, buf, VMA_EXTENT_HEADER_SIZE);
++ gsize csize = 16;
++ g_checksum_get_digest(vmar->md5csum, ehead->md5sum, &csize);
++
++ if (memcmp(md5sum, ehead->md5sum, 16) != 0) {
++ error_setg(errp, "wrong vma extent header chechsum");
++ return -1;
++ }
++
++ if (memcmp(h->uuid, ehead->uuid, sizeof(ehead->uuid)) != 0) {
++ error_setg(errp, "wrong vma extent uuid");
++ return -1;
++ }
++
++ if (ehead->magic != VMA_EXTENT_MAGIC || ehead->reserved1 != 0) {
++ error_setg(errp, "wrong vma extent header magic");
++ return -1;
++ }
++
++ int block_count = GUINT16_FROM_BE(ehead->block_count);
++ int extent_size = VMA_EXTENT_HEADER_SIZE + block_count*VMA_BLOCK_SIZE;
++
++ if (buf_pos < extent_size) {
++ error_setg(errp, "short vma extent (%d < %d)", buf_pos,
++ extent_size);
++ return -1;
++ }
++
++ if (restore_extent(vmar, buf, extent_size, vmstate_fd, verbose,
++ verify, errp) < 0) {
++ return -1;
++ }
++
++ if (buf_pos > extent_size) {
++ memmove(buf, buf + extent_size, buf_pos - extent_size);
++ buf_pos = buf_pos - extent_size;
++ } else {
++ buf_pos = 0;
++ }
++ }
++
++ bdrv_drain_all();
++
++ int i;
++ for (i = 1; i < 256; i++) {
++ VmaRestoreState *rstate = &vmar->rstate[i];
++ if (!rstate->target) {
++ continue;
++ }
++
++ if (blk_flush(rstate->target) < 0) {
++ error_setg(errp, "vma blk_flush %s failed",
++ vmar->devinfo[i].devname);
++ return -1;
++ }
++
++ if (vmar->devinfo[i].size &&
++ (strcmp(vmar->devinfo[i].devname, "vmstate") != 0)) {
++ assert(rstate->bitmap);
++
++ int64_t cluster_num, end;
++
++ end = (vmar->devinfo[i].size + VMA_CLUSTER_SIZE - 1) /
++ VMA_CLUSTER_SIZE;
++
++ for (cluster_num = 0; cluster_num < end; cluster_num++) {
++ if (!vma_reader_get_bitmap(rstate, cluster_num)) {
++ error_setg(errp, "detected missing cluster %zd "
++ "for stream %s", cluster_num,
++ vmar->devinfo[i].devname);
++ return -1;
++ }
++ }
++ }
++ }
++
++ if (verbose) {
++ if (vmar->clusters_read) {
++ printf("total bytes read %zd, sparse bytes %zd (%.3g%%)\n",
++ vmar->clusters_read*VMA_CLUSTER_SIZE,
++ vmar->zero_cluster_data,
++ (double)(100.0*vmar->zero_cluster_data)/
++ (vmar->clusters_read*VMA_CLUSTER_SIZE));
++
++ int64_t datasize = vmar->clusters_read*VMA_CLUSTER_SIZE-vmar->zero_cluster_data;
++ if (datasize) { // this does not make sense for empty files
++ printf("space reduction due to 4K zero blocks %.3g%%\n",
++ (double)(100.0*vmar->partial_zero_cluster_data) / datasize);
++ }
++ } else {
++ printf("vma archive contains no image data\n");
++ }
++ }
++ return ret;
++}
++
++int vma_reader_restore(VmaReader *vmar, int vmstate_fd, bool verbose,
++ Error **errp)
++{
++ return vma_reader_restore_full(vmar, vmstate_fd, verbose, false, errp);
++}
++
++int vma_reader_verify(VmaReader *vmar, bool verbose, Error **errp)
++{
++ guint8 dev_id;
++
++ for (dev_id = 1; dev_id < 255; dev_id++) {
++ if (vma_reader_get_device_info(vmar, dev_id)) {
++ allocate_rstate(vmar, dev_id, NULL, false);
++ }
++ }
++
++ return vma_reader_restore_full(vmar, -1, verbose, true, errp);
++}
++
+diff --git a/vma-writer.c b/vma-writer.c
+new file mode 100644
+index 0000000000..fe86b18a60
+--- /dev/null
++++ b/vma-writer.c
+@@ -0,0 +1,771 @@
++/*
++ * VMA: Virtual Machine Archive
++ *
++ * Copyright (C) 2012 Proxmox Server Solutions
++ *
++ * Authors:
++ * Dietmar Maurer (dietmar@proxmox.com)
++ *
++ * This work is licensed under the terms of the GNU GPL, version 2 or later.
++ * See the COPYING file in the top-level directory.
++ *
++ */
++
++#include "qemu/osdep.h"
++#include <glib.h>
++#include <uuid/uuid.h>
++
++#include "vma.h"
++#include "block/block.h"
++#include "monitor/monitor.h"
++#include "qemu/main-loop.h"
++#include "qemu/coroutine.h"
++#include "qemu/cutils.h"
++
++#define DEBUG_VMA 0
++
++#define DPRINTF(fmt, ...)\
++ do { if (DEBUG_VMA) { printf("vma: " fmt, ## __VA_ARGS__); } } while (0)
++
++#define WRITE_BUFFERS 5
++#define HEADER_CLUSTERS 8
++#define HEADERBUF_SIZE (VMA_CLUSTER_SIZE*HEADER_CLUSTERS)
++
++struct VmaWriter {
++ int fd;
++ FILE *cmd;
++ int status;
++ char errmsg[8192];
++ uuid_t uuid;
++ bool header_written;
++ bool closed;
++
++ /* we always write extents */
++ unsigned char *outbuf;
++ int outbuf_pos; /* in bytes */
++ int outbuf_count; /* in VMA_BLOCKS */
++ uint64_t outbuf_block_info[VMA_BLOCKS_PER_EXTENT];
++
++ unsigned char *headerbuf;
++
++ GChecksum *md5csum;
++ CoMutex flush_lock;
++ Coroutine *co_writer;
++
++ /* drive informations */
++ VmaStreamInfo stream_info[256];
++ guint stream_count;
++
++ guint8 vmstate_stream;
++ uint32_t vmstate_clusters;
++
++ /* header blob table */
++ char *header_blob_table;
++ uint32_t header_blob_table_size;
++ uint32_t header_blob_table_pos;
++
++ /* store for config blobs */
++ uint32_t config_names[VMA_MAX_CONFIGS]; /* offset into blob_buffer table */
++ uint32_t config_data[VMA_MAX_CONFIGS]; /* offset into blob_buffer table */
++ uint32_t config_count;
++};
++
++void vma_writer_set_error(VmaWriter *vmaw, const char *fmt, ...)
++{
++ va_list ap;
++
++ if (vmaw->status < 0) {
++ return;
++ }
++
++ vmaw->status = -1;
++
++ va_start(ap, fmt);
++ g_vsnprintf(vmaw->errmsg, sizeof(vmaw->errmsg), fmt, ap);
++ va_end(ap);
++
++ DPRINTF("vma_writer_set_error: %s\n", vmaw->errmsg);
++}
++
++static uint32_t allocate_header_blob(VmaWriter *vmaw, const char *data,
++ size_t len)
++{
++ if (len > 65535) {
++ return 0;
++ }
++
++ if (!vmaw->header_blob_table ||
++ (vmaw->header_blob_table_size <
++ (vmaw->header_blob_table_pos + len + 2))) {
++ int newsize = vmaw->header_blob_table_size + ((len + 2 + 511)/512)*512;
++
++ vmaw->header_blob_table = g_realloc(vmaw->header_blob_table, newsize);
++ memset(vmaw->header_blob_table + vmaw->header_blob_table_size,
++ 0, newsize - vmaw->header_blob_table_size);
++ vmaw->header_blob_table_size = newsize;
++ }
++
++ uint32_t cpos = vmaw->header_blob_table_pos;
++ vmaw->header_blob_table[cpos] = len & 255;
++ vmaw->header_blob_table[cpos+1] = (len >> 8) & 255;
++ memcpy(vmaw->header_blob_table + cpos + 2, data, len);
++ vmaw->header_blob_table_pos += len + 2;
++ return cpos;
++}
++
++static uint32_t allocate_header_string(VmaWriter *vmaw, const char *str)
++{
++ assert(vmaw);
++
++ size_t len = strlen(str) + 1;
++
++ return allocate_header_blob(vmaw, str, len);
++}
++
++int vma_writer_add_config(VmaWriter *vmaw, const char *name, gpointer data,
++ gsize len)
++{
++ assert(vmaw);
++ assert(!vmaw->header_written);
++ assert(vmaw->config_count < VMA_MAX_CONFIGS);
++ assert(name);
++ assert(data);
++
++ gchar *basename = g_path_get_basename(name);
++ uint32_t name_ptr = allocate_header_string(vmaw, basename);
++ g_free(basename);
++
++ if (!name_ptr) {
++ return -1;
++ }
++
++ uint32_t data_ptr = allocate_header_blob(vmaw, data, len);
++ if (!data_ptr) {
++ return -1;
++ }
++
++ vmaw->config_names[vmaw->config_count] = name_ptr;
++ vmaw->config_data[vmaw->config_count] = data_ptr;
++
++ vmaw->config_count++;
++
++ return 0;
++}
++
++int vma_writer_register_stream(VmaWriter *vmaw, const char *devname,
++ size_t size)
++{
++ assert(vmaw);
++ assert(devname);
++ assert(!vmaw->status);
++
++ if (vmaw->header_written) {
++ vma_writer_set_error(vmaw, "vma_writer_register_stream: header "
++ "already written");
++ return -1;
++ }
++
++ guint n = vmaw->stream_count + 1;
++
++ /* we can have dev_ids form 1 to 255 (0 reserved)
++ * 255(-1) reseverd for safety
++ */
++ if (n > 254) {
++ vma_writer_set_error(vmaw, "vma_writer_register_stream: "
++ "too many drives");
++ return -1;
++ }
++
++ if (size <= 0) {
++ vma_writer_set_error(vmaw, "vma_writer_register_stream: "
++ "got strange size %zd", size);
++ return -1;
++ }
++
++ DPRINTF("vma_writer_register_stream %s %zu %d\n", devname, size, n);
++
++ vmaw->stream_info[n].devname = g_strdup(devname);
++ vmaw->stream_info[n].size = size;
++
++ vmaw->stream_info[n].cluster_count = (size + VMA_CLUSTER_SIZE - 1) /
++ VMA_CLUSTER_SIZE;
++
++ vmaw->stream_count = n;
++
++ if (strcmp(devname, "vmstate") == 0) {
++ vmaw->vmstate_stream = n;
++ }
++
++ return n;
++}
++
++static void coroutine_fn yield_until_fd_writable(int fd)
++{
++ assert(qemu_in_coroutine());
++ AioContext *ctx = qemu_get_current_aio_context();
++ aio_set_fd_handler(ctx, fd, false, NULL, (IOHandler *)qemu_coroutine_enter,
++ NULL, qemu_coroutine_self());
++ qemu_coroutine_yield();
++ aio_set_fd_handler(ctx, fd, false, NULL, NULL, NULL, NULL);
++}
++
++static ssize_t coroutine_fn
++vma_queue_write(VmaWriter *vmaw, const void *buf, size_t bytes)
++{
++ DPRINTF("vma_queue_write enter %zd\n", bytes);
++
++ assert(vmaw);
++ assert(buf);
++ assert(bytes <= VMA_MAX_EXTENT_SIZE);
++
++ size_t done = 0;
++ ssize_t ret;
++
++ assert(vmaw->co_writer == NULL);
++
++ vmaw->co_writer = qemu_coroutine_self();
++
++ while (done < bytes) {
++ if (vmaw->status < 0) {
++ DPRINTF("vma_queue_write detected canceled backup\n");
++ done = -1;
++ break;
++ }
++ yield_until_fd_writable(vmaw->fd);
++ ret = write(vmaw->fd, buf + done, bytes - done);
++ if (ret > 0) {
++ done += ret;
++ DPRINTF("vma_queue_write written %zd %zd\n", done, ret);
++ } else if (ret < 0) {
++ if (errno == EAGAIN || errno == EWOULDBLOCK) {
++ /* try again */
++ } else {
++ vma_writer_set_error(vmaw, "vma_queue_write: write error - %s",
++ g_strerror(errno));
++ done = -1; /* always return failure for partial writes */
++ break;
++ }
++ } else if (ret == 0) {
++ /* should not happen - simply try again */
++ }
++ }
++
++ vmaw->co_writer = NULL;
++
++ return (done == bytes) ? bytes : -1;
++}
++
++VmaWriter *vma_writer_create(const char *filename, uuid_t uuid, Error **errp)
++{
++ const char *p;
++
++ assert(sizeof(VmaHeader) == (4096 + 8192));
++ assert(G_STRUCT_OFFSET(VmaHeader, config_names) == 2044);
++ assert(G_STRUCT_OFFSET(VmaHeader, config_data) == 3068);
++ assert(G_STRUCT_OFFSET(VmaHeader, dev_info) == 4096);
++ assert(sizeof(VmaExtentHeader) == 512);
++
++ VmaWriter *vmaw = g_new0(VmaWriter, 1);
++ vmaw->fd = -1;
++
++ vmaw->md5csum = g_checksum_new(G_CHECKSUM_MD5);
++ if (!vmaw->md5csum) {
++ error_setg(errp, "can't allocate cmsum\n");
++ goto err;
++ }
++
++ if (strstart(filename, "exec:", &p)) {
++ vmaw->cmd = popen(p, "w");
++ if (vmaw->cmd == NULL) {
++ error_setg(errp, "can't popen command '%s' - %s\n", p,
++ g_strerror(errno));
++ goto err;
++ }
++ vmaw->fd = fileno(vmaw->cmd);
++
++ /* try to use O_NONBLOCK */
++ fcntl(vmaw->fd, F_SETFL, fcntl(vmaw->fd, F_GETFL)|O_NONBLOCK);
++
++ } else {
++ struct stat st;
++ int oflags;
++ const char *tmp_id_str;
++
++ if ((stat(filename, &st) == 0) && S_ISFIFO(st.st_mode)) {
++ oflags = O_NONBLOCK|O_WRONLY;
++ vmaw->fd = qemu_open(filename, oflags, 0644);
++ } else if (strstart(filename, "/dev/fdset/", &tmp_id_str)) {
++ oflags = O_NONBLOCK|O_WRONLY;
++ vmaw->fd = qemu_open(filename, oflags, 0644);
++ } else if (strstart(filename, "/dev/fdname/", &tmp_id_str)) {
++ vmaw->fd = monitor_get_fd(cur_mon, tmp_id_str, errp);
++ if (vmaw->fd < 0) {
++ goto err;
++ }
++ /* try to use O_NONBLOCK */
++ fcntl(vmaw->fd, F_SETFL, fcntl(vmaw->fd, F_GETFL)|O_NONBLOCK);
++ } else {
++ oflags = O_NONBLOCK|O_DIRECT|O_WRONLY|O_CREAT|O_EXCL;
++ vmaw->fd = qemu_open(filename, oflags, 0644);
++ }
++
++ if (vmaw->fd < 0) {
++ error_setg(errp, "can't open file %s - %s\n", filename,
++ g_strerror(errno));
++ goto err;
++ }
++ }
++
++ /* we use O_DIRECT, so we need to align IO buffers */
++
++ vmaw->outbuf = qemu_memalign(512, VMA_MAX_EXTENT_SIZE);
++ vmaw->headerbuf = qemu_memalign(512, HEADERBUF_SIZE);
++
++ vmaw->outbuf_count = 0;
++ vmaw->outbuf_pos = VMA_EXTENT_HEADER_SIZE;
++
++ vmaw->header_blob_table_pos = 1; /* start at pos 1 */
++
++ qemu_co_mutex_init(&vmaw->flush_lock);
++
++ uuid_copy(vmaw->uuid, uuid);
++
++ return vmaw;
++
++err:
++ if (vmaw) {
++ if (vmaw->cmd) {
++ pclose(vmaw->cmd);
++ } else if (vmaw->fd >= 0) {
++ close(vmaw->fd);
++ }
++
++ if (vmaw->md5csum) {
++ g_checksum_free(vmaw->md5csum);
++ }
++
++ g_free(vmaw);
++ }
++
++ return NULL;
++}
++
++static int coroutine_fn vma_write_header(VmaWriter *vmaw)
++{
++ assert(vmaw);
++ unsigned char *buf = vmaw->headerbuf;
++ VmaHeader *head = (VmaHeader *)buf;
++
++ int i;
++
++ DPRINTF("VMA WRITE HEADER\n");
++
++ if (vmaw->status < 0) {
++ return vmaw->status;
++ }
++
++ memset(buf, 0, HEADERBUF_SIZE);
++
++ head->magic = VMA_MAGIC;
++ head->version = GUINT32_TO_BE(1); /* v1 */
++ memcpy(head->uuid, vmaw->uuid, 16);
++
++ time_t ctime = time(NULL);
++ head->ctime = GUINT64_TO_BE(ctime);
++
++ for (i = 0; i < VMA_MAX_CONFIGS; i++) {
++ head->config_names[i] = GUINT32_TO_BE(vmaw->config_names[i]);
++ head->config_data[i] = GUINT32_TO_BE(vmaw->config_data[i]);
++ }
++
++ /* 32 bytes per device (12 used currently) = 8192 bytes max */
++ for (i = 1; i <= 254; i++) {
++ VmaStreamInfo *si = &vmaw->stream_info[i];
++ if (si->size) {
++ assert(si->devname);
++ uint32_t devname_ptr = allocate_header_string(vmaw, si->devname);
++ if (!devname_ptr) {
++ return -1;
++ }
++ head->dev_info[i].devname_ptr = GUINT32_TO_BE(devname_ptr);
++ head->dev_info[i].size = GUINT64_TO_BE(si->size);
++ }
++ }
++
++ uint32_t header_size = sizeof(VmaHeader) + vmaw->header_blob_table_size;
++ head->header_size = GUINT32_TO_BE(header_size);
++
++ if (header_size > HEADERBUF_SIZE) {
++ return -1; /* just to be sure */
++ }
++
++ uint32_t blob_buffer_offset = sizeof(VmaHeader);
++ memcpy(buf + blob_buffer_offset, vmaw->header_blob_table,
++ vmaw->header_blob_table_size);
++ head->blob_buffer_offset = GUINT32_TO_BE(blob_buffer_offset);
++ head->blob_buffer_size = GUINT32_TO_BE(vmaw->header_blob_table_pos);
++
++ g_checksum_reset(vmaw->md5csum);
++ g_checksum_update(vmaw->md5csum, (const guchar *)buf, header_size);
++ gsize csize = 16;
++ g_checksum_get_digest(vmaw->md5csum, (guint8 *)(head->md5sum), &csize);
++
++ return vma_queue_write(vmaw, buf, header_size);
++}
++
++static int coroutine_fn vma_writer_flush(VmaWriter *vmaw)
++{
++ assert(vmaw);
++
++ int ret;
++ int i;
++
++ if (vmaw->status < 0) {
++ return vmaw->status;
++ }
++
++ if (!vmaw->header_written) {
++ vmaw->header_written = true;
++ ret = vma_write_header(vmaw);
++ if (ret < 0) {
++ vma_writer_set_error(vmaw, "vma_writer_flush: write header failed");
++ return ret;
++ }
++ }
++
++ DPRINTF("VMA WRITE FLUSH %d %d\n", vmaw->outbuf_count, vmaw->outbuf_pos);
++
++
++ VmaExtentHeader *ehead = (VmaExtentHeader *)vmaw->outbuf;
++
++ ehead->magic = VMA_EXTENT_MAGIC;
++ ehead->reserved1 = 0;
++
++ for (i = 0; i < VMA_BLOCKS_PER_EXTENT; i++) {
++ ehead->blockinfo[i] = GUINT64_TO_BE(vmaw->outbuf_block_info[i]);
++ }
++
++ guint16 block_count = (vmaw->outbuf_pos - VMA_EXTENT_HEADER_SIZE) /
++ VMA_BLOCK_SIZE;
++
++ ehead->block_count = GUINT16_TO_BE(block_count);
++
++ memcpy(ehead->uuid, vmaw->uuid, sizeof(ehead->uuid));
++ memset(ehead->md5sum, 0, sizeof(ehead->md5sum));
++
++ g_checksum_reset(vmaw->md5csum);
++ g_checksum_update(vmaw->md5csum, vmaw->outbuf, VMA_EXTENT_HEADER_SIZE);
++ gsize csize = 16;
++ g_checksum_get_digest(vmaw->md5csum, ehead->md5sum, &csize);
++
++ int bytes = vmaw->outbuf_pos;
++ ret = vma_queue_write(vmaw, vmaw->outbuf, bytes);
++ if (ret != bytes) {
++ vma_writer_set_error(vmaw, "vma_writer_flush: failed write");
++ }
++
++ vmaw->outbuf_count = 0;
++ vmaw->outbuf_pos = VMA_EXTENT_HEADER_SIZE;
++
++ for (i = 0; i < VMA_BLOCKS_PER_EXTENT; i++) {
++ vmaw->outbuf_block_info[i] = 0;
++ }
++
++ return vmaw->status;
++}
++
++static int vma_count_open_streams(VmaWriter *vmaw)
++{
++ g_assert(vmaw != NULL);
++
++ int i;
++ int open_drives = 0;
++ for (i = 0; i <= 255; i++) {
++ if (vmaw->stream_info[i].size && !vmaw->stream_info[i].finished) {
++ open_drives++;
++ }
++ }
++
++ return open_drives;
++}
++
++
++/**
++ * You need to call this if the vma archive does not contain
++ * any data stream.
++ */
++int coroutine_fn
++vma_writer_flush_output(VmaWriter *vmaw)
++{
++ qemu_co_mutex_lock(&vmaw->flush_lock);
++ int ret = vma_writer_flush(vmaw);
++ qemu_co_mutex_unlock(&vmaw->flush_lock);
++ if (ret < 0) {
++ vma_writer_set_error(vmaw, "vma_writer_flush_header failed");
++ }
++ return ret;
++}
++
++/**
++ * all jobs should call this when there is no more data
++ * Returns: number of remaining stream (0 ==> finished)
++ */
++int coroutine_fn
++vma_writer_close_stream(VmaWriter *vmaw, uint8_t dev_id)
++{
++ g_assert(vmaw != NULL);
++
++ DPRINTF("vma_writer_set_status %d\n", dev_id);
++ if (!vmaw->stream_info[dev_id].size) {
++ vma_writer_set_error(vmaw, "vma_writer_close_stream: "
++ "no such stream %d", dev_id);
++ return -1;
++ }
++ if (vmaw->stream_info[dev_id].finished) {
++ vma_writer_set_error(vmaw, "vma_writer_close_stream: "
++ "stream already closed %d", dev_id);
++ return -1;
++ }
++
++ vmaw->stream_info[dev_id].finished = true;
++
++ int open_drives = vma_count_open_streams(vmaw);
++
++ if (open_drives <= 0) {
++ DPRINTF("vma_writer_set_status all drives completed\n");
++ vma_writer_flush_output(vmaw);
++ }
++
++ return open_drives;
++}
++
++int vma_writer_get_status(VmaWriter *vmaw, VmaStatus *status)
++{
++ int i;
++
++ g_assert(vmaw != NULL);
++
++ if (status) {
++ status->status = vmaw->status;
++ g_strlcpy(status->errmsg, vmaw->errmsg, sizeof(status->errmsg));
++ for (i = 0; i <= 255; i++) {
++ status->stream_info[i] = vmaw->stream_info[i];
++ }
++
++ uuid_unparse_lower(vmaw->uuid, status->uuid_str);
++ }
++
++ status->closed = vmaw->closed;
++
++ return vmaw->status;
++}
++
++static int vma_writer_get_buffer(VmaWriter *vmaw)
++{
++ int ret = 0;
++
++ qemu_co_mutex_lock(&vmaw->flush_lock);
++
++ /* wait until buffer is available */
++ while (vmaw->outbuf_count >= (VMA_BLOCKS_PER_EXTENT - 1)) {
++ ret = vma_writer_flush(vmaw);
++ if (ret < 0) {
++ vma_writer_set_error(vmaw, "vma_writer_get_buffer: flush failed");
++ break;
++ }
++ }
++
++ qemu_co_mutex_unlock(&vmaw->flush_lock);
++
++ return ret;
++}
++
++
++int64_t coroutine_fn
++vma_writer_write(VmaWriter *vmaw, uint8_t dev_id, int64_t cluster_num,
++ const unsigned char *buf, size_t *zero_bytes)
++{
++ g_assert(vmaw != NULL);
++ g_assert(zero_bytes != NULL);
++
++ *zero_bytes = 0;
++
++ if (vmaw->status < 0) {
++ return vmaw->status;
++ }
++
++ if (!dev_id || !vmaw->stream_info[dev_id].size) {
++ vma_writer_set_error(vmaw, "vma_writer_write: "
++ "no such stream %d", dev_id);
++ return -1;
++ }
++
++ if (vmaw->stream_info[dev_id].finished) {
++ vma_writer_set_error(vmaw, "vma_writer_write: "
++ "stream already closed %d", dev_id);
++ return -1;
++ }
++
++
++ if (cluster_num >= (((uint64_t)1)<<32)) {
++ vma_writer_set_error(vmaw, "vma_writer_write: "
++ "cluster number out of range");
++ return -1;
++ }
++
++ if (dev_id == vmaw->vmstate_stream) {
++ if (cluster_num != vmaw->vmstate_clusters) {
++ vma_writer_set_error(vmaw, "vma_writer_write: "
++ "non sequential vmstate write");
++ }
++ vmaw->vmstate_clusters++;
++ } else if (cluster_num >= vmaw->stream_info[dev_id].cluster_count) {
++ vma_writer_set_error(vmaw, "vma_writer_write: cluster number too big");
++ return -1;
++ }
++
++ /* wait until buffer is available */
++ if (vma_writer_get_buffer(vmaw) < 0) {
++ vma_writer_set_error(vmaw, "vma_writer_write: "
++ "vma_writer_get_buffer failed");
++ return -1;
++ }
++
++ DPRINTF("VMA WRITE %d %zd\n", dev_id, cluster_num);
++
++ uint16_t mask = 0;
++
++ if (buf) {
++ int i;
++ int bit = 1;
++ for (i = 0; i < 16; i++) {
++ const unsigned char *vmablock = buf + (i*VMA_BLOCK_SIZE);
++ if (!buffer_is_zero(vmablock, VMA_BLOCK_SIZE)) {
++ mask |= bit;
++ memcpy(vmaw->outbuf + vmaw->outbuf_pos, vmablock,
++ VMA_BLOCK_SIZE);
++ vmaw->outbuf_pos += VMA_BLOCK_SIZE;
++ } else {
++ DPRINTF("VMA WRITE %zd ZERO BLOCK %d\n", cluster_num, i);
++ vmaw->stream_info[dev_id].zero_bytes += VMA_BLOCK_SIZE;
++ *zero_bytes += VMA_BLOCK_SIZE;
++ }
++
++ bit = bit << 1;
++ }
++ } else {
++ DPRINTF("VMA WRITE %zd ZERO CLUSTER\n", cluster_num);
++ vmaw->stream_info[dev_id].zero_bytes += VMA_CLUSTER_SIZE;
++ *zero_bytes += VMA_CLUSTER_SIZE;
++ }
++
++ uint64_t block_info = ((uint64_t)mask) << (32+16);
++ block_info |= ((uint64_t)dev_id) << 32;
++ block_info |= (cluster_num & 0xffffffff);
++ vmaw->outbuf_block_info[vmaw->outbuf_count] = block_info;
++
++ DPRINTF("VMA WRITE MASK %zd %zx\n", cluster_num, block_info);
++
++ vmaw->outbuf_count++;
++
++ /** NOTE: We allways write whole clusters, but we correctly set
++ * transferred bytes. So transferred == size when when everything
++ * went OK.
++ */
++ size_t transferred = VMA_CLUSTER_SIZE;
++
++ if (dev_id != vmaw->vmstate_stream) {
++ uint64_t last = (cluster_num + 1) * VMA_CLUSTER_SIZE;
++ if (last > vmaw->stream_info[dev_id].size) {
++ uint64_t diff = last - vmaw->stream_info[dev_id].size;
++ if (diff >= VMA_CLUSTER_SIZE) {
++ vma_writer_set_error(vmaw, "vma_writer_write: "
++ "read after last cluster");
++ return -1;
++ }
++ transferred -= diff;
++ }
++ }
++
++ vmaw->stream_info[dev_id].transferred += transferred;
++
++ return transferred;
++}
++
++void vma_writer_error_propagate(VmaWriter *vmaw, Error **errp)
++{
++ if (vmaw->status < 0 && *errp == NULL) {
++ error_setg(errp, "%s", vmaw->errmsg);
++ }
++}
++
++int vma_writer_close(VmaWriter *vmaw, Error **errp)
++{
++ g_assert(vmaw != NULL);
++
++ int i;
++
++ qemu_co_mutex_lock(&vmaw->flush_lock); // wait for pending writes
++
++ assert(vmaw->co_writer == NULL);
++
++ if (vmaw->cmd) {
++ if (pclose(vmaw->cmd) < 0) {
++ vma_writer_set_error(vmaw, "vma_writer_close: "
++ "pclose failed - %s", g_strerror(errno));
++ }
++ } else {
++ if (close(vmaw->fd) < 0) {
++ vma_writer_set_error(vmaw, "vma_writer_close: "
++ "close failed - %s", g_strerror(errno));
++ }
++ }
++
++ for (i = 0; i <= 255; i++) {
++ VmaStreamInfo *si = &vmaw->stream_info[i];
++ if (si->size) {
++ if (!si->finished) {
++ vma_writer_set_error(vmaw, "vma_writer_close: "
++ "detected open stream '%s'", si->devname);
++ } else if ((si->transferred != si->size) &&
++ (i != vmaw->vmstate_stream)) {
++ vma_writer_set_error(vmaw, "vma_writer_close: "
++ "incomplete stream '%s' (%zd != %zd)",
++ si->devname, si->transferred, si->size);
++ }
++ }
++ }
++
++ for (i = 0; i <= 255; i++) {
++ vmaw->stream_info[i].finished = 1; /* mark as closed */
++ }
++
++ vmaw->closed = 1;
++
++ if (vmaw->status < 0 && *errp == NULL) {
++ error_setg(errp, "%s", vmaw->errmsg);
++ }
++
++ qemu_co_mutex_unlock(&vmaw->flush_lock);
++
++ return vmaw->status;
++}
++
++void vma_writer_destroy(VmaWriter *vmaw)
++{
++ assert(vmaw);
++
++ int i;
++
++ for (i = 0; i <= 255; i++) {
++ if (vmaw->stream_info[i].devname) {
++ g_free(vmaw->stream_info[i].devname);
++ }
++ }
++
++ if (vmaw->md5csum) {
++ g_checksum_free(vmaw->md5csum);
++ }
++
++ g_free(vmaw);
++}
+diff --git a/vma.c b/vma.c
+new file mode 100644
+index 0000000000..a82752448a
+--- /dev/null
++++ b/vma.c
+@@ -0,0 +1,837 @@
++/*
++ * VMA: Virtual Machine Archive
++ *
++ * Copyright (C) 2012-2013 Proxmox Server Solutions
++ *
++ * Authors:
++ * Dietmar Maurer (dietmar@proxmox.com)
++ *
++ * This work is licensed under the terms of the GNU GPL, version 2 or later.
++ * See the COPYING file in the top-level directory.
++ *
++ */
++
++#include "qemu/osdep.h"
++#include <glib.h>
++
++#include "vma.h"
++#include "qemu-common.h"
++#include "qemu/module.h"
++#include "qemu/error-report.h"
++#include "qemu/main-loop.h"
++#include "qemu/cutils.h"
++#include "qapi/qmp/qdict.h"
++#include "sysemu/block-backend.h"
++
++static void help(void)
++{
++ const char *help_msg =
++ "usage: vma command [command options]\n"
++ "\n"
++ "vma list <filename>\n"
++ "vma config <filename> [-c config]\n"
++ "vma create <filename> [-c config] pathname ...\n"
++ "vma extract <filename> [-r <fifo>] <targetdir>\n"
++ "vma verify <filename> [-v]\n"
++ ;
++
++ printf("%s", help_msg);
++ exit(1);
++}
++
++static const char *extract_devname(const char *path, char **devname, int index)
++{
++ assert(path);
++
++ const char *sep = strchr(path, '=');
++
++ if (sep) {
++ *devname = g_strndup(path, sep - path);
++ path = sep + 1;
++ } else {
++ if (index >= 0) {
++ *devname = g_strdup_printf("disk%d", index);
++ } else {
++ *devname = NULL;
++ }
++ }
++
++ return path;
++}
++
++static void print_content(VmaReader *vmar)
++{
++ assert(vmar);
++
++ VmaHeader *head = vma_reader_get_header(vmar);
++
++ GList *l = vma_reader_get_config_data(vmar);
++ while (l && l->data) {
++ VmaConfigData *cdata = (VmaConfigData *)l->data;
++ l = g_list_next(l);
++ printf("CFG: size: %d name: %s\n", cdata->len, cdata->name);
++ }
++
++ int i;
++ VmaDeviceInfo *di;
++ for (i = 1; i < 255; i++) {
++ di = vma_reader_get_device_info(vmar, i);
++ if (di) {
++ if (strcmp(di->devname, "vmstate") == 0) {
++ printf("VMSTATE: dev_id=%d memory: %zd\n", i, di->size);
++ } else {
++ printf("DEV: dev_id=%d size: %zd devname: %s\n",
++ i, di->size, di->devname);
++ }
++ }
++ }
++ /* ctime is the last entry we print */
++ printf("CTIME: %s", ctime(&head->ctime));
++ fflush(stdout);
++}
++
++static int list_content(int argc, char **argv)
++{
++ int c, ret = 0;
++ const char *filename;
++
++ for (;;) {
++ c = getopt(argc, argv, "h");
++ if (c == -1) {
++ break;
++ }
++ switch (c) {
++ case '?':
++ case 'h':
++ help();
++ break;
++ default:
++ g_assert_not_reached();
++ }
++ }
++
++ /* Get the filename */
++ if ((optind + 1) != argc) {
++ help();
++ }
++ filename = argv[optind++];
++
++ Error *errp = NULL;
++ VmaReader *vmar = vma_reader_create(filename, &errp);
++
++ if (!vmar) {
++ g_error("%s", error_get_pretty(errp));
++ }
++
++ print_content(vmar);
++
++ vma_reader_destroy(vmar);
++
++ return ret;
++}
++
++typedef struct RestoreMap {
++ char *devname;
++ char *path;
++ char *format;
++ uint64_t throttling_bps;
++ char *throttling_group;
++ char *cache;
++ bool write_zero;
++} RestoreMap;
++
++static bool try_parse_option(char **line, const char *optname, char **out, const char *inbuf) {
++ size_t optlen = strlen(optname);
++ if (strncmp(*line, optname, optlen) != 0 || (*line)[optlen] != '=') {
++ return false;
++ }
++ if (*out) {
++ g_error("read map failed - duplicate value for option '%s'", optname);
++ }
++ char *value = (*line) + optlen + 1; /* including a '=' */
++ char *colon = strchr(value, ':');
++ if (!colon) {
++ g_error("read map failed - option '%s' not terminated ('%s')",
++ optname, inbuf);
++ }
++ *line = colon+1;
++ *out = g_strndup(value, colon - value);
++ return true;
++}
++
++static uint64_t verify_u64(const char *text) {
++ uint64_t value;
++ const char *endptr = NULL;
++ if (qemu_strtou64(text, &endptr, 0, &value) != 0 || !endptr || *endptr) {
++ g_error("read map failed - not a number: %s", text);
++ }
++ return value;
++}
++
++static int extract_content(int argc, char **argv)
++{
++ int c, ret = 0;
++ int verbose = 0;
++ const char *filename;
++ const char *dirname;
++ const char *readmap = NULL;
++
++ for (;;) {
++ c = getopt(argc, argv, "hvr:");
++ if (c == -1) {
++ break;
++ }
++ switch (c) {
++ case '?':
++ case 'h':
++ help();
++ break;
++ case 'r':
++ readmap = optarg;
++ break;
++ case 'v':
++ verbose = 1;
++ break;
++ default:
++ help();
++ }
++ }
++
++ /* Get the filename */
++ if ((optind + 2) != argc) {
++ help();
++ }
++ filename = argv[optind++];
++ dirname = argv[optind++];
++
++ Error *errp = NULL;
++ VmaReader *vmar = vma_reader_create(filename, &errp);
++
++ if (!vmar) {
++ g_error("%s", error_get_pretty(errp));
++ }
++
++ if (mkdir(dirname, 0777) < 0) {
++ g_error("unable to create target directory %s - %s",
++ dirname, g_strerror(errno));
++ }
++
++ GList *l = vma_reader_get_config_data(vmar);
++ while (l && l->data) {
++ VmaConfigData *cdata = (VmaConfigData *)l->data;
++ l = g_list_next(l);
++ char *cfgfn = g_strdup_printf("%s/%s", dirname, cdata->name);
++ GError *err = NULL;
++ if (!g_file_set_contents(cfgfn, (gchar *)cdata->data, cdata->len,
++ &err)) {
++ g_error("unable to write file: %s", err->message);
++ }
++ }
++
++ GHashTable *devmap = g_hash_table_new(g_str_hash, g_str_equal);
++
++ if (readmap) {
++ print_content(vmar);
++
++ FILE *map = fopen(readmap, "r");
++ if (!map) {
++ g_error("unable to open fifo %s - %s", readmap, g_strerror(errno));
++ }
++
++ while (1) {
++ char inbuf[8192];
++ char *line = fgets(inbuf, sizeof(inbuf), map);
++ char *format = NULL;
++ char *bps = NULL;
++ char *group = NULL;
++ char *cache = NULL;
++ if (!line || line[0] == '\0' || !strcmp(line, "done\n")) {
++ break;
++ }
++ int len = strlen(line);
++ if (line[len - 1] == '\n') {
++ line[len - 1] = '\0';
++ if (len == 1) {
++ break;
++ }
++ }
++
++ while (1) {
++ if (!try_parse_option(&line, "format", &format, inbuf) &&
++ !try_parse_option(&line, "throttling.bps", &bps, inbuf) &&
++ !try_parse_option(&line, "throttling.group", &group, inbuf) &&
++ !try_parse_option(&line, "cache", &cache, inbuf))
++ {
++ break;
++ }
++ }
++
++ uint64_t bps_value = 0;
++ if (bps) {
++ bps_value = verify_u64(bps);
++ g_free(bps);
++ }
++
++ const char *path;
++ bool write_zero;
++ if (line[0] == '0' && line[1] == ':') {
++ path = line + 2;
++ write_zero = false;
++ } else if (line[0] == '1' && line[1] == ':') {
++ path = line + 2;
++ write_zero = true;
++ } else {
++ g_error("read map failed - parse error ('%s')", inbuf);
++ }
++
++ char *devname = NULL;
++ path = extract_devname(path, &devname, -1);
++ if (!devname) {
++ g_error("read map failed - no dev name specified ('%s')",
++ inbuf);
++ }
++
++ RestoreMap *map = g_new0(RestoreMap, 1);
++ map->devname = g_strdup(devname);
++ map->path = g_strdup(path);
++ map->format = format;
++ map->throttling_bps = bps_value;
++ map->throttling_group = group;
++ map->cache = cache;
++ map->write_zero = write_zero;
++
++ g_hash_table_insert(devmap, map->devname, map);
++
++ };
++ }
++
++ int i;
++ int vmstate_fd = -1;
++ guint8 vmstate_stream = 0;
++
++ BlockBackend *blk = NULL;
++
++ for (i = 1; i < 255; i++) {
++ VmaDeviceInfo *di = vma_reader_get_device_info(vmar, i);
++ if (di && (strcmp(di->devname, "vmstate") == 0)) {
++ vmstate_stream = i;
++ char *statefn = g_strdup_printf("%s/vmstate.bin", dirname);
++ vmstate_fd = open(statefn, O_WRONLY|O_CREAT|O_EXCL, 0644);
++ if (vmstate_fd < 0) {
++ g_error("create vmstate file '%s' failed - %s", statefn,
++ g_strerror(errno));
++ }
++ g_free(statefn);
++ } else if (di) {
++ char *devfn = NULL;
++ const char *format = NULL;
++ uint64_t throttling_bps = 0;
++ const char *throttling_group = NULL;
++ const char *cache = NULL;
++ int flags = BDRV_O_RDWR;
++ bool write_zero = true;
++
++ if (readmap) {
++ RestoreMap *map;
++ map = (RestoreMap *)g_hash_table_lookup(devmap, di->devname);
++ if (map == NULL) {
++ g_error("no device name mapping for %s", di->devname);
++ }
++ devfn = map->path;
++ format = map->format;
++ throttling_bps = map->throttling_bps;
++ throttling_group = map->throttling_group;
++ cache = map->cache;
++ write_zero = map->write_zero;
++ } else {
++ devfn = g_strdup_printf("%s/tmp-disk-%s.raw",
++ dirname, di->devname);
++ printf("DEVINFO %s %zd\n", devfn, di->size);
++
++ bdrv_img_create(devfn, "raw", NULL, NULL, NULL, di->size,
++ flags, true, &errp);
++ if (errp) {
++ g_error("can't create file %s: %s", devfn,
++ error_get_pretty(errp));
++ }
++
++ /* Note: we created an empty file above, so there is no
++ * need to write zeroes (so we generate a sparse file)
++ */
++ write_zero = false;
++ }
++
++ size_t devlen = strlen(devfn);
++ QDict *options = NULL;
++ bool writethrough;
++ if (format) {
++ /* explicit format from commandline */
++ options = qdict_new();
++ qdict_put_str(options, "driver", format);
++ } else if ((devlen > 4 && strcmp(devfn+devlen-4, ".raw") == 0) ||
++ strncmp(devfn, "/dev/", 5) == 0)
++ {
++ /* This part is now deprecated for PVE as well (just as qemu
++ * deprecated not specifying an explicit raw format, too.
++ */
++ /* explicit raw format */
++ options = qdict_new();
++ qdict_put_str(options, "driver", "raw");
++ }
++ if (cache && bdrv_parse_cache_mode(cache, &flags, &writethrough)) {
++ g_error("invalid cache option: %s\n", cache);
++ }
++
++ if (errp || !(blk = blk_new_open(devfn, NULL, options, flags, &errp))) {
++ g_error("can't open file %s - %s", devfn,
++ error_get_pretty(errp));
++ }
++
++ if (cache) {
++ blk_set_enable_write_cache(blk, !writethrough);
++ }
++
++ if (throttling_group) {
++ blk_io_limits_enable(blk, throttling_group);
++ }
++
++ if (throttling_bps) {
++ if (!throttling_group) {
++ blk_io_limits_enable(blk, devfn);
++ }
++
++ ThrottleConfig cfg;
++ throttle_config_init(&cfg);
++ cfg.buckets[THROTTLE_BPS_WRITE].avg = throttling_bps;
++ Error *err = NULL;
++ if (!throttle_is_valid(&cfg, &err)) {
++ error_report_err(err);
++ g_error("failed to apply throttling");
++ }
++ blk_set_io_limits(blk, &cfg);
++ }
++
++ if (vma_reader_register_bs(vmar, i, blk, write_zero, &errp) < 0) {
++ g_error("%s", error_get_pretty(errp));
++ }
++
++ if (!readmap) {
++ g_free(devfn);
++ }
++ }
++ }
++
++ if (vma_reader_restore(vmar, vmstate_fd, verbose, &errp) < 0) {
++ g_error("restore failed - %s", error_get_pretty(errp));
++ }
++
++ if (!readmap) {
++ for (i = 1; i < 255; i++) {
++ VmaDeviceInfo *di = vma_reader_get_device_info(vmar, i);
++ if (di && (i != vmstate_stream)) {
++ char *tmpfn = g_strdup_printf("%s/tmp-disk-%s.raw",
++ dirname, di->devname);
++ char *fn = g_strdup_printf("%s/disk-%s.raw",
++ dirname, di->devname);
++ if (rename(tmpfn, fn) != 0) {
++ g_error("rename %s to %s failed - %s",
++ tmpfn, fn, g_strerror(errno));
++ }
++ }
++ }
++ }
++
++ vma_reader_destroy(vmar);
++
++ blk_unref(blk);
++
++ bdrv_close_all();
++
++ return ret;
++}
++
++static int verify_content(int argc, char **argv)
++{
++ int c, ret = 0;
++ int verbose = 0;
++ const char *filename;
++
++ for (;;) {
++ c = getopt(argc, argv, "hv");
++ if (c == -1) {
++ break;
++ }
++ switch (c) {
++ case '?':
++ case 'h':
++ help();
++ break;
++ case 'v':
++ verbose = 1;
++ break;
++ default:
++ help();
++ }
++ }
++
++ /* Get the filename */
++ if ((optind + 1) != argc) {
++ help();
++ }
++ filename = argv[optind++];
++
++ Error *errp = NULL;
++ VmaReader *vmar = vma_reader_create(filename, &errp);
++
++ if (!vmar) {
++ g_error("%s", error_get_pretty(errp));
++ }
++
++ if (verbose) {
++ print_content(vmar);
++ }
++
++ if (vma_reader_verify(vmar, verbose, &errp) < 0) {
++ g_error("verify failed - %s", error_get_pretty(errp));
++ }
++
++ vma_reader_destroy(vmar);
++
++ bdrv_close_all();
++
++ return ret;
++}
++
++typedef struct BackupJob {
++ BlockBackend *target;
++ int64_t len;
++ VmaWriter *vmaw;
++ uint8_t dev_id;
++} BackupJob;
++
++#define BACKUP_SECTORS_PER_CLUSTER (VMA_CLUSTER_SIZE / BDRV_SECTOR_SIZE)
++
++static void coroutine_fn backup_run_empty(void *opaque)
++{
++ VmaWriter *vmaw = (VmaWriter *)opaque;
++
++ vma_writer_flush_output(vmaw);
++
++ Error *err = NULL;
++ if (vma_writer_close(vmaw, &err) != 0) {
++ g_warning("vma_writer_close failed %s", error_get_pretty(err));
++ }
++}
++
++static void coroutine_fn backup_run(void *opaque)
++{
++ BackupJob *job = (BackupJob *)opaque;
++ struct iovec iov;
++ QEMUIOVector qiov;
++
++ int64_t start, end;
++ int ret = 0;
++
++ unsigned char *buf = blk_blockalign(job->target, VMA_CLUSTER_SIZE);
++
++ start = 0;
++ end = DIV_ROUND_UP(job->len / BDRV_SECTOR_SIZE,
++ BACKUP_SECTORS_PER_CLUSTER);
++
++ for (; start < end; start++) {
++ iov.iov_base = buf;
++ iov.iov_len = VMA_CLUSTER_SIZE;
++ qemu_iovec_init_external(&qiov, &iov, 1);
++
++ ret = blk_co_preadv(job->target, start * VMA_CLUSTER_SIZE,
++ VMA_CLUSTER_SIZE, &qiov, 0);
++ if (ret < 0) {
++ vma_writer_set_error(job->vmaw, "read error", -1);
++ goto out;
++ }
++
++ size_t zb = 0;
++ if (vma_writer_write(job->vmaw, job->dev_id, start, buf, &zb) < 0) {
++ vma_writer_set_error(job->vmaw, "backup_dump_cb vma_writer_write failed", -1);
++ goto out;
++ }
++ }
++
++
++out:
++ if (vma_writer_close_stream(job->vmaw, job->dev_id) <= 0) {
++ Error *err = NULL;
++ if (vma_writer_close(job->vmaw, &err) != 0) {
++ g_warning("vma_writer_close failed %s", error_get_pretty(err));
++ }
++ }
++}
++
++static int create_archive(int argc, char **argv)
++{
++ int i, c;
++ int verbose = 0;
++ const char *archivename;
++ GList *config_files = NULL;
++
++ for (;;) {
++ c = getopt(argc, argv, "hvc:");
++ if (c == -1) {
++ break;
++ }
++ switch (c) {
++ case '?':
++ case 'h':
++ help();
++ break;
++ case 'c':
++ config_files = g_list_append(config_files, optarg);
++ break;
++ case 'v':
++ verbose = 1;
++ break;
++ default:
++ g_assert_not_reached();
++ }
++ }
++
++
++ /* make sure we an archive name */
++ if ((optind + 1) > argc) {
++ help();
++ }
++
++ archivename = argv[optind++];
++
++ uuid_t uuid;
++ uuid_generate(uuid);
++
++ Error *local_err = NULL;
++ VmaWriter *vmaw = vma_writer_create(archivename, uuid, &local_err);
++
++ if (vmaw == NULL) {
++ g_error("%s", error_get_pretty(local_err));
++ }
++
++ GList *l = config_files;
++ while (l && l->data) {
++ char *name = l->data;
++ char *cdata = NULL;
++ gsize clen = 0;
++ GError *err = NULL;
++ if (!g_file_get_contents(name, &cdata, &clen, &err)) {
++ unlink(archivename);
++ g_error("Unable to read file: %s", err->message);
++ }
++
++ if (vma_writer_add_config(vmaw, name, cdata, clen) != 0) {
++ unlink(archivename);
++ g_error("Unable to append config data %s (len = %zd)",
++ name, clen);
++ }
++ l = g_list_next(l);
++ }
++
++ int devcount = 0;
++ while (optind < argc) {
++ const char *path = argv[optind++];
++ char *devname = NULL;
++ path = extract_devname(path, &devname, devcount++);
++
++ Error *errp = NULL;
++ BlockBackend *target;
++
++ target = blk_new_open(path, NULL, NULL, 0, &errp);
++ if (!target) {
++ unlink(archivename);
++ g_error("bdrv_open '%s' failed - %s", path, error_get_pretty(errp));
++ }
++ int64_t size = blk_getlength(target);
++ int dev_id = vma_writer_register_stream(vmaw, devname, size);
++ if (dev_id <= 0) {
++ unlink(archivename);
++ g_error("vma_writer_register_stream '%s' failed", devname);
++ }
++
++ BackupJob *job = g_new0(BackupJob, 1);
++ job->len = size;
++ job->target = target;
++ job->vmaw = vmaw;
++ job->dev_id = dev_id;
++
++ Coroutine *co = qemu_coroutine_create(backup_run, job);
++ qemu_coroutine_enter(co);
++ }
++
++ VmaStatus vmastat;
++ int percent = 0;
++ int last_percent = -1;
++
++ if (devcount) {
++ while (1) {
++ main_loop_wait(false);
++ vma_writer_get_status(vmaw, &vmastat);
++
++ if (verbose) {
++
++ uint64_t total = 0;
++ uint64_t transferred = 0;
++ uint64_t zero_bytes = 0;
++
++ int i;
++ for (i = 0; i < 256; i++) {
++ if (vmastat.stream_info[i].size) {
++ total += vmastat.stream_info[i].size;
++ transferred += vmastat.stream_info[i].transferred;
++ zero_bytes += vmastat.stream_info[i].zero_bytes;
++ }
++ }
++ percent = (transferred*100)/total;
++ if (percent != last_percent) {
++ fprintf(stderr, "progress %d%% %zd/%zd %zd\n", percent,
++ transferred, total, zero_bytes);
++ fflush(stderr);
++
++ last_percent = percent;
++ }
++ }
++
++ if (vmastat.closed) {
++ break;
++ }
++ }
++ } else {
++ Coroutine *co = qemu_coroutine_create(backup_run_empty, vmaw);
++ qemu_coroutine_enter(co);
++ while (1) {
++ main_loop_wait(false);
++ vma_writer_get_status(vmaw, &vmastat);
++ if (vmastat.closed) {
++ break;
++ }
++ }
++ }
++
++ bdrv_drain_all();
++
++ vma_writer_get_status(vmaw, &vmastat);
++
++ if (verbose) {
++ for (i = 0; i < 256; i++) {
++ VmaStreamInfo *si = &vmastat.stream_info[i];
++ if (si->size) {
++ fprintf(stderr, "image %s: size=%zd zeros=%zd saved=%zd\n",
++ si->devname, si->size, si->zero_bytes,
++ si->size - si->zero_bytes);
++ }
++ }
++ }
++
++ if (vmastat.status < 0) {
++ unlink(archivename);
++ g_error("creating vma archive failed");
++ }
++
++ return 0;
++}
++
++static int dump_config(int argc, char **argv)
++{
++ int c, ret = 0;
++ const char *filename;
++ const char *config_name = "qemu-server.conf";
++
++ for (;;) {
++ c = getopt(argc, argv, "hc:");
++ if (c == -1) {
++ break;
++ }
++ switch (c) {
++ case '?':
++ case 'h':
++ help();
++ break;
++ case 'c':
++ config_name = optarg;
++ break;
++ default:
++ help();
++ }
++ }
++
++ /* Get the filename */
++ if ((optind + 1) != argc) {
++ help();
++ }
++ filename = argv[optind++];
++
++ Error *errp = NULL;
++ VmaReader *vmar = vma_reader_create(filename, &errp);
++
++ if (!vmar) {
++ g_error("%s", error_get_pretty(errp));
++ }
++
++ int found = 0;
++ GList *l = vma_reader_get_config_data(vmar);
++ while (l && l->data) {
++ VmaConfigData *cdata = (VmaConfigData *)l->data;
++ l = g_list_next(l);
++ if (strcmp(cdata->name, config_name) == 0) {
++ found = 1;
++ fwrite(cdata->data, cdata->len, 1, stdout);
++ break;
++ }
++ }
++
++ vma_reader_destroy(vmar);
++
++ bdrv_close_all();
++
++ if (!found) {
++ fprintf(stderr, "unable to find configuration data '%s'\n", config_name);
++ return -1;
++ }
++
++ return ret;
++}
++
++int main(int argc, char **argv)
++{
++ const char *cmdname;
++ Error *main_loop_err = NULL;
++
++ error_init(argv[0]);
++ module_call_init(MODULE_INIT_TRACE);
++ qemu_init_exec_dir(argv[0]);
++
++ if (qemu_init_main_loop(&main_loop_err)) {
++ g_error("%s", error_get_pretty(main_loop_err));
++ }
++
++ bdrv_init();
++ module_call_init(MODULE_INIT_QOM);
++
++ if (argc < 2) {
++ help();
++ }
++
++ cmdname = argv[1];
++ argc--; argv++;
++
++
++ if (!strcmp(cmdname, "list")) {
++ return list_content(argc, argv);
++ } else if (!strcmp(cmdname, "create")) {
++ return create_archive(argc, argv);
++ } else if (!strcmp(cmdname, "extract")) {
++ return extract_content(argc, argv);
++ } else if (!strcmp(cmdname, "verify")) {
++ return verify_content(argc, argv);
++ } else if (!strcmp(cmdname, "config")) {
++ return dump_config(argc, argv);
++ }
++
++ help();
++ return 0;
++}
+diff --git a/vma.h b/vma.h
+new file mode 100644
+index 0000000000..c895c97f6d
+--- /dev/null
++++ b/vma.h
+@@ -0,0 +1,150 @@
++/*
++ * VMA: Virtual Machine Archive
++ *
++ * Copyright (C) Proxmox Server Solutions
++ *
++ * Authors:
++ * Dietmar Maurer (dietmar@proxmox.com)
++ *
++ * This work is licensed under the terms of the GNU GPL, version 2 or later.
++ * See the COPYING file in the top-level directory.
++ *
++ */
++
++#ifndef BACKUP_VMA_H
++#define BACKUP_VMA_H
++
++#include <uuid/uuid.h>
++#include "qapi/error.h"
++#include "block/block.h"
++
++#define VMA_BLOCK_BITS 12
++#define VMA_BLOCK_SIZE (1<<VMA_BLOCK_BITS)
++#define VMA_CLUSTER_BITS (VMA_BLOCK_BITS+4)
++#define VMA_CLUSTER_SIZE (1<<VMA_CLUSTER_BITS)
++
++#if VMA_CLUSTER_SIZE != 65536
++#error unexpected cluster size
++#endif
++
++#define VMA_EXTENT_HEADER_SIZE 512
++#define VMA_BLOCKS_PER_EXTENT 59
++#define VMA_MAX_CONFIGS 256
++
++#define VMA_MAX_EXTENT_SIZE \
++ (VMA_EXTENT_HEADER_SIZE+VMA_CLUSTER_SIZE*VMA_BLOCKS_PER_EXTENT)
++#if VMA_MAX_EXTENT_SIZE != 3867136
++#error unexpected VMA_EXTENT_SIZE
++#endif
++
++/* File Format Definitions */
++
++#define VMA_MAGIC (GUINT32_TO_BE(('V'<<24)|('M'<<16)|('A'<<8)|0x00))
++#define VMA_EXTENT_MAGIC (GUINT32_TO_BE(('V'<<24)|('M'<<16)|('A'<<8)|'E'))
++
++typedef struct VmaDeviceInfoHeader {
++ uint32_t devname_ptr; /* offset into blob_buffer table */
++ uint32_t reserved0;
++ uint64_t size; /* device size in bytes */
++ uint64_t reserved1;
++ uint64_t reserved2;
++} VmaDeviceInfoHeader;
++
++typedef struct VmaHeader {
++ uint32_t magic;
++ uint32_t version;
++ unsigned char uuid[16];
++ int64_t ctime;
++ unsigned char md5sum[16];
++
++ uint32_t blob_buffer_offset;
++ uint32_t blob_buffer_size;
++ uint32_t header_size;
++
++ unsigned char reserved[1984];
++
++ uint32_t config_names[VMA_MAX_CONFIGS]; /* offset into blob_buffer table */
++ uint32_t config_data[VMA_MAX_CONFIGS]; /* offset into blob_buffer table */
++
++ uint32_t reserved1;
++
++ VmaDeviceInfoHeader dev_info[256];
++} VmaHeader;
++
++typedef struct VmaExtentHeader {
++ uint32_t magic;
++ uint16_t reserved1;
++ uint16_t block_count;
++ unsigned char uuid[16];
++ unsigned char md5sum[16];
++ uint64_t blockinfo[VMA_BLOCKS_PER_EXTENT];
++} VmaExtentHeader;
++
++/* functions/definitions to read/write vma files */
++
++typedef struct VmaReader VmaReader;
++
++typedef struct VmaWriter VmaWriter;
++
++typedef struct VmaConfigData {
++ const char *name;
++ const void *data;
++ uint32_t len;
++} VmaConfigData;
++
++typedef struct VmaStreamInfo {
++ uint64_t size;
++ uint64_t cluster_count;
++ uint64_t transferred;
++ uint64_t zero_bytes;
++ int finished;
++ char *devname;
++} VmaStreamInfo;
++
++typedef struct VmaStatus {
++ int status;
++ bool closed;
++ char errmsg[8192];
++ char uuid_str[37];
++ VmaStreamInfo stream_info[256];
++} VmaStatus;
++
++typedef struct VmaDeviceInfo {
++ uint64_t size; /* device size in bytes */
++ const char *devname;
++} VmaDeviceInfo;
++
++VmaWriter *vma_writer_create(const char *filename, uuid_t uuid, Error **errp);
++int vma_writer_close(VmaWriter *vmaw, Error **errp);
++void vma_writer_error_propagate(VmaWriter *vmaw, Error **errp);
++void vma_writer_destroy(VmaWriter *vmaw);
++int vma_writer_add_config(VmaWriter *vmaw, const char *name, gpointer data,
++ size_t len);
++int vma_writer_register_stream(VmaWriter *vmaw, const char *devname,
++ size_t size);
++
++int64_t coroutine_fn vma_writer_write(VmaWriter *vmaw, uint8_t dev_id,
++ int64_t cluster_num,
++ const unsigned char *buf,
++ size_t *zero_bytes);
++
++int coroutine_fn vma_writer_close_stream(VmaWriter *vmaw, uint8_t dev_id);
++int coroutine_fn vma_writer_flush_output(VmaWriter *vmaw);
++
++int vma_writer_get_status(VmaWriter *vmaw, VmaStatus *status);
++void vma_writer_set_error(VmaWriter *vmaw, const char *fmt, ...);
++
++
++VmaReader *vma_reader_create(const char *filename, Error **errp);
++void vma_reader_destroy(VmaReader *vmar);
++VmaHeader *vma_reader_get_header(VmaReader *vmar);
++GList *vma_reader_get_config_data(VmaReader *vmar);
++VmaDeviceInfo *vma_reader_get_device_info(VmaReader *vmar, guint8 dev_id);
++int vma_reader_register_bs(VmaReader *vmar, guint8 dev_id,
++ BlockBackend *target, bool write_zeroes,
++ Error **errp);
++int vma_reader_restore(VmaReader *vmar, int vmstate_fd, bool verbose,
++ Error **errp);
++int vma_reader_verify(VmaReader *vmar, bool verbose, Error **errp);
++
++#endif /* BACKUP_VMA_H */
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Mon, 24 Jun 2019 10:19:50 +0200
-Subject: [PATCH] PVE: move snapshot cleanup into bottom half
-
-as per:
- (0ceccd858a8d) migration: qemu_savevm_state_cleanup() in cleanup
-
-may affect held locks and therefore change assumptions made
-by that function!
-
-Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- savevm-async.c | 3 ++-
- 1 file changed, 2 insertions(+), 1 deletion(-)
-
-diff --git a/savevm-async.c b/savevm-async.c
-index 0bbbbf51ba..f9355c5036 100644
---- a/savevm-async.c
-+++ b/savevm-async.c
-@@ -197,6 +197,8 @@ static void process_savevm_cleanup(void *opaque)
- int ret;
- qemu_bh_delete(snap_state.cleanup_bh);
- snap_state.cleanup_bh = NULL;
-+ qemu_savevm_state_cleanup();
-+
- qemu_mutex_unlock_iothread();
- qemu_thread_join(&snap_state.thread);
- qemu_mutex_lock_iothread();
-@@ -273,7 +275,6 @@ static void *process_savevm_thread(void *opaque)
- save_snapshot_error("qemu_savevm_state_iterate error %d", ret);
- break;
- }
-- qemu_savevm_state_cleanup();
- DPRINTF("save complete\n");
- break;
- }
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Dietmar Maurer <dietmar@proxmox.com>
+Date: Tue, 10 Mar 2020 12:55:27 +0100
+Subject: [PATCH 30/32] PVE-Backup: add backup-dump block driver
+
+- add backup-dump block driver block/backup-dump.c
+- move BackupBlockJob declaration from block/backup.c to include/block/block_int.h
+- block/backup.c - backup-job-create: also consider source cluster size
+- block/io.c - bdrv_do_drained_begin_quiesce: check for coroutine
+- job.c: make job_should_pause non-static
+---
+ block/Makefile.objs | 1 +
+ block/backup-dump.c | 169 ++++++++++++++++++++++++++++++++++++++
+ block/backup.c | 23 ++----
+ block/io.c | 8 +-
+ include/block/block_int.h | 30 +++++++
+ job.c | 3 +-
+ 6 files changed, 214 insertions(+), 20 deletions(-)
+ create mode 100644 block/backup-dump.c
+
+diff --git a/block/Makefile.objs b/block/Makefile.objs
+index a10ceabf5b..5cd9e40d8d 100644
+--- a/block/Makefile.objs
++++ b/block/Makefile.objs
+@@ -33,6 +33,7 @@ block-obj-$(CONFIG_RBD) += rbd.o
+ block-obj-$(CONFIG_GLUSTERFS) += gluster.o
+ block-obj-$(CONFIG_VXHS) += vxhs.o
+ block-obj-$(CONFIG_LIBSSH) += ssh.o
++block-obj-y += backup-dump.o
+ block-obj-y += accounting.o dirty-bitmap.o
+ block-obj-y += write-threshold.o
+ block-obj-y += backup.o
+diff --git a/block/backup-dump.c b/block/backup-dump.c
+new file mode 100644
+index 0000000000..3066ab0698
+--- /dev/null
++++ b/block/backup-dump.c
+@@ -0,0 +1,169 @@
++/*
++ * BlockDriver to send backup data stream to a callback function
++ *
++ * Copyright (C) 2020 Proxmox Server Solutions GmbH
++ *
++ * This work is licensed under the terms of the GNU GPL, version 2 or later.
++ * See the COPYING file in the top-level directory.
++ *
++ */
++
++#include "qemu/osdep.h"
++#include "qemu-common.h"
++#include "qom/object_interfaces.h"
++#include "block/block_int.h"
++
++typedef struct {
++ int dump_cb_block_size;
++ uint64_t byte_size;
++ BackupDumpFunc *dump_cb;
++ void *dump_cb_data;
++} BDRVBackupDumpState;
++
++static int qemu_backup_dump_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
++{
++ BDRVBackupDumpState *s = bs->opaque;
++
++ bdi->cluster_size = s->dump_cb_block_size;
++ bdi->unallocated_blocks_are_zero = true;
++ return 0;
++}
++
++static int qemu_backup_dump_check_perm(
++ BlockDriverState *bs,
++ uint64_t perm,
++ uint64_t shared,
++ Error **errp)
++{
++ /* Nothing to do. */
++ return 0;
++}
++
++static void qemu_backup_dump_set_perm(
++ BlockDriverState *bs,
++ uint64_t perm,
++ uint64_t shared)
++{
++ /* Nothing to do. */
++}
++
++static void qemu_backup_dump_abort_perm_update(BlockDriverState *bs)
++{
++ /* Nothing to do. */
++}
++
++static void qemu_backup_dump_refresh_limits(BlockDriverState *bs, Error **errp)
++{
++ bs->bl.request_alignment = BDRV_SECTOR_SIZE; /* No sub-sector I/O */
++}
++
++static void qemu_backup_dump_close(BlockDriverState *bs)
++{
++ /* Nothing to do. */
++}
++
++static int64_t qemu_backup_dump_getlength(BlockDriverState *bs)
++{
++ BDRVBackupDumpState *s = bs->opaque;
++
++ return s->byte_size;
++}
++
++static coroutine_fn int qemu_backup_dump_co_writev(
++ BlockDriverState *bs,
++ int64_t sector_num,
++ int nb_sectors,
++ QEMUIOVector *qiov,
++ int flags)
++{
++ /* flags can be only values we set in supported_write_flags */
++ assert(flags == 0);
++
++ BDRVBackupDumpState *s = bs->opaque;
++ off_t offset = sector_num * BDRV_SECTOR_SIZE;
++
++ uint64_t written = 0;
++
++ for (int i = 0; i < qiov->niov; ++i) {
++ const struct iovec *v = &qiov->iov[i];
++
++ int rc = s->dump_cb(s->dump_cb_data, offset, v->iov_len, v->iov_base);
++ if (rc < 0) {
++ return rc;
++ }
++
++ if (rc != v->iov_len) {
++ return -EIO;
++ }
++
++ written += v->iov_len;
++ offset += v->iov_len;
++ }
++
++ return written;
++}
++
++static void qemu_backup_dump_child_perm(
++ BlockDriverState *bs,
++ BdrvChild *c,
++ const BdrvChildRole *role,
++ BlockReopenQueue *reopen_queue,
++ uint64_t perm, uint64_t shared,
++ uint64_t *nperm, uint64_t *nshared)
++{
++ *nperm = BLK_PERM_ALL;
++ *nshared = BLK_PERM_ALL;
++}
++
++static BlockDriver bdrv_backup_dump_drive = {
++ .format_name = "backup-dump-drive",
++ .protocol_name = "backup-dump",
++ .instance_size = sizeof(BDRVBackupDumpState),
++
++ .bdrv_close = qemu_backup_dump_close,
++ .bdrv_has_zero_init = bdrv_has_zero_init_1,
++ .bdrv_getlength = qemu_backup_dump_getlength,
++ .bdrv_get_info = qemu_backup_dump_get_info,
++
++ .bdrv_co_writev = qemu_backup_dump_co_writev,
++
++ .bdrv_refresh_limits = qemu_backup_dump_refresh_limits,
++ .bdrv_check_perm = qemu_backup_dump_check_perm,
++ .bdrv_set_perm = qemu_backup_dump_set_perm,
++ .bdrv_abort_perm_update = qemu_backup_dump_abort_perm_update,
++ .bdrv_child_perm = qemu_backup_dump_child_perm,
++};
++
++static void bdrv_backup_dump_init(void)
++{
++ bdrv_register(&bdrv_backup_dump_drive);
++}
++
++block_init(bdrv_backup_dump_init);
++
++
++BlockDriverState *bdrv_backup_dump_create(
++ int dump_cb_block_size,
++ uint64_t byte_size,
++ BackupDumpFunc *dump_cb,
++ void *dump_cb_data,
++ Error **errp)
++{
++ BDRVBackupDumpState *state;
++ BlockDriverState *bs = bdrv_new_open_driver(
++ &bdrv_backup_dump_drive, NULL, BDRV_O_RDWR, errp);
++
++ if (!bs) {
++ return NULL;
++ }
++
++ bs->total_sectors = byte_size / BDRV_SECTOR_SIZE;
++ bs->opaque = state = g_new0(BDRVBackupDumpState, 1);
++
++ state->dump_cb_block_size = dump_cb_block_size;
++ state->byte_size = byte_size;
++ state->dump_cb = dump_cb;
++ state->dump_cb_data = dump_cb_data;
++
++ return bs;
++}
+diff --git a/block/backup.c b/block/backup.c
+index c155081de2..9d23da027f 100644
+--- a/block/backup.c
++++ b/block/backup.c
+@@ -32,24 +32,6 @@
+
+ #define BACKUP_CLUSTER_SIZE_DEFAULT (1 << 16)
+
+-typedef struct BackupBlockJob {
+- BlockJob common;
+- BlockDriverState *backup_top;
+- BlockDriverState *source_bs;
+-
+- BdrvDirtyBitmap *sync_bitmap;
+-
+- MirrorSyncMode sync_mode;
+- BitmapSyncMode bitmap_mode;
+- BlockdevOnError on_source_error;
+- BlockdevOnError on_target_error;
+- uint64_t len;
+- uint64_t bytes_read;
+- int64_t cluster_size;
+-
+- BlockCopyState *bcs;
+-} BackupBlockJob;
+-
+ static const BlockJobDriver backup_job_driver;
+
+ static void backup_progress_bytes_callback(int64_t bytes, void *opaque)
+@@ -420,6 +402,11 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
+ goto error;
+ }
+
++ BlockDriverInfo bdi;
++ if (bdrv_get_info(bs, &bdi) == 0) {
++ cluster_size = MAX(cluster_size, bdi.cluster_size);
++ }
++
+ /*
+ * If source is in backing chain of target assume that target is going to be
+ * used for "image fleecing", i.e. it should represent a kind of snapshot of
+diff --git a/block/io.c b/block/io.c
+index f75777f5ea..75dea5ae33 100644
+--- a/block/io.c
++++ b/block/io.c
+@@ -381,7 +381,13 @@ static void coroutine_fn bdrv_co_yield_to_drain(BlockDriverState *bs,
+ void bdrv_do_drained_begin_quiesce(BlockDriverState *bs,
+ BdrvChild *parent, bool ignore_bds_parents)
+ {
+- assert(!qemu_in_coroutine());
++ // AFAICT this function is just an optimization, but sadly it doesn't play
++ // nice with the PVE backup code (when we're in a coroutine, even in
++ // pvebackup_co_start), so just call the full-blown drain begin instead
++ if (qemu_in_coroutine()) {
++ bdrv_do_drained_begin(bs, false, parent, ignore_bds_parents, false);
++ return;
++ }
+
+ /* Stop things in parent-to-child order */
+ if (atomic_fetch_inc(&bs->quiesce_counter) == 0) {
+diff --git a/include/block/block_int.h b/include/block/block_int.h
+index b0d5eb9485..105bffe0f7 100644
+--- a/include/block/block_int.h
++++ b/include/block/block_int.h
+@@ -60,6 +60,36 @@
+
+ #define BLOCK_PROBE_BUF_SIZE 512
+
++typedef int BackupDumpFunc(void *opaque, uint64_t offset, uint64_t bytes, const void *buf);
++
++BlockDriverState *bdrv_backuo_dump_create(
++ int dump_cb_block_size,
++ uint64_t byte_size,
++ BackupDumpFunc *dump_cb,
++ void *dump_cb_data,
++ Error **errp);
++
++// Needs to be defined here, since it's used in blockdev.c to detect PVE backup
++// jobs with source_bs
++typedef struct BlockCopyState BlockCopyState;
++typedef struct BackupBlockJob {
++ BlockJob common;
++ BlockDriverState *backup_top;
++ BlockDriverState *source_bs;
++
++ BdrvDirtyBitmap *sync_bitmap;
++
++ MirrorSyncMode sync_mode;
++ BitmapSyncMode bitmap_mode;
++ BlockdevOnError on_source_error;
++ BlockdevOnError on_target_error;
++ uint64_t len;
++ uint64_t bytes_read;
++ int64_t cluster_size;
++
++ BlockCopyState *bcs;
++} BackupBlockJob;
++
+ enum BdrvTrackedRequestType {
+ BDRV_TRACKED_READ,
+ BDRV_TRACKED_WRITE,
+diff --git a/job.c b/job.c
+index 7554f735e3..b03ef989e2 100644
+--- a/job.c
++++ b/job.c
+@@ -248,7 +248,8 @@ static bool job_started(Job *job)
+ return job->co;
+ }
+
+-static bool job_should_pause(Job *job)
++bool job_should_pause(Job *job);
++bool job_should_pause(Job *job)
+ {
+ return job->pause_count > 0;
+ }
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Thu, 14 Nov 2019 14:38:02 +0100
-Subject: [PATCH] PVE: monitor: disable oob capability
-
-A bisect revealed that commit 8258292e18c3
-("monitor: Remove "x-oob", offer capability "oob" unconditionally")
-causes unexpected hangs when restoring live snapshots from some
-types of block devices (particularly RBD).
-We need to figure out what's happnening there. For now, since we
-had this disabled before and probably don't need it now either,
-disable oob, so we can get a functioning qemu out...
-
-Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- monitor/qmp.c | 3 +--
- 1 file changed, 1 insertion(+), 2 deletions(-)
-
-diff --git a/monitor/qmp.c b/monitor/qmp.c
-index fb3e66c62a..ad6703dc52 100644
---- a/monitor/qmp.c
-+++ b/monitor/qmp.c
-@@ -379,8 +379,7 @@ void monitor_init_qmp(Chardev *chr, bool pretty)
- MonitorQMP *mon = g_new0(MonitorQMP, 1);
-
- /* Note: we run QMP monitor in I/O thread when @chr supports that */
-- monitor_data_init(&mon->common, true, false,
-- qemu_chr_has_feature(chr, QEMU_CHAR_FEATURE_GCONTEXT));
-+ monitor_data_init(&mon->common, true, false, false);
-
- mon->pretty = pretty;
-
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Dietmar Maurer <dietmar@proxmox.com>
+Date: Tue, 10 Mar 2020 12:55:28 +0100
+Subject: [PATCH 31/32] PVE-Backup: proxmox backup patches for qemu
+
+---
+ Makefile.objs | 1 +
+ Makefile.target | 2 +-
+ blockdev.c | 1 +
+ hmp-commands-info.hx | 13 +
+ hmp-commands.hx | 31 ++
+ include/block/block_int.h | 2 +-
+ include/monitor/hmp.h | 3 +
+ monitor/hmp-cmds.c | 77 +++
+ proxmox-backup-client.c | 182 ++++++++
+ proxmox-backup-client.h | 52 +++
+ pve-backup.c | 959 ++++++++++++++++++++++++++++++++++++++
+ qapi/block-core.json | 109 +++++
+ qapi/common.json | 13 +
+ qapi/misc.json | 13 -
+ 14 files changed, 1443 insertions(+), 15 deletions(-)
+ create mode 100644 proxmox-backup-client.c
+ create mode 100644 proxmox-backup-client.h
+ create mode 100644 pve-backup.c
+
+diff --git a/Makefile.objs b/Makefile.objs
+index db7fbbe73b..4406dcfeba 100644
+--- a/Makefile.objs
++++ b/Makefile.objs
+@@ -46,6 +46,7 @@ io-obj-y = io/
+ ifeq ($(CONFIG_SOFTMMU),y)
+ common-obj-y = blockdev.o blockdev-nbd.o block/
+ common-obj-y += bootdevice.o iothread.o
++common-obj-y += proxmox-backup-client.o pve-backup.o
+ common-obj-y += dump/
+ common-obj-y += job-qmp.o
+ common-obj-y += monitor/
+diff --git a/Makefile.target b/Makefile.target
+index 24d79d26eb..d7d91085a4 100644
+--- a/Makefile.target
++++ b/Makefile.target
+@@ -160,7 +160,7 @@ obj-y += qapi/
+ obj-y += memory.o
+ obj-y += memory_mapping.o
+ obj-y += migration/ram.o
+-LIBS := $(libs_softmmu) $(LIBS)
++LIBS := $(libs_softmmu) $(LIBS) -lproxmox_backup_qemu
+
+ # Hardware support
+ ifeq ($(TARGET_NAME), sparc64)
+diff --git a/blockdev.c b/blockdev.c
+index c7fa663ebf..e5310cb939 100644
+--- a/blockdev.c
++++ b/blockdev.c
+@@ -36,6 +36,7 @@
+ #include "hw/block/block.h"
+ #include "block/blockjob.h"
+ #include "block/qdict.h"
++#include "block/blockjob_int.h"
+ #include "block/throttle-groups.h"
+ #include "monitor/monitor.h"
+ #include "qemu/error-report.h"
+diff --git a/hmp-commands-info.hx b/hmp-commands-info.hx
+index 139e673bea..8db5ce03a7 100644
+--- a/hmp-commands-info.hx
++++ b/hmp-commands-info.hx
+@@ -536,6 +536,19 @@ STEXI
+ @item info cpustats
+ @findex info cpustats
+ Show CPU statistics.
++ETEXI
++
++ {
++ .name = "backup",
++ .args_type = "",
++ .params = "",
++ .help = "show backup status",
++ .cmd = hmp_info_backup,
++ },
++
++STEXI
++@item info backup
++show backup status
+ ETEXI
+
+ #if defined(CONFIG_SLIRP)
+diff --git a/hmp-commands.hx b/hmp-commands.hx
+index 104288322d..29d11dd321 100644
+--- a/hmp-commands.hx
++++ b/hmp-commands.hx
+@@ -105,6 +105,37 @@ STEXI
+ @item block_stream
+ @findex block_stream
+ Copy data from a backing file into a block device.
++ETEXI
++
++ {
++ .name = "backup",
++ .args_type = "directory:-d,backupfile:s,speed:o?,devlist:s?",
++ .params = "[-d] backupfile [speed [devlist]]",
++ .help = "create a VM Backup."
++ "\n\t\t\t Use -d to dump data into a directory instead"
++ "\n\t\t\t of using VMA format.",
++ .cmd = hmp_backup,
++ },
++
++STEXI
++@item backup
++@findex backup
++Create a VM backup.
++ETEXI
++
++ {
++ .name = "backup_cancel",
++ .args_type = "",
++ .params = "",
++ .help = "cancel the current VM backup",
++ .cmd = hmp_backup_cancel,
++ },
++
++STEXI
++@item backup_cancel
++@findex backup_cancel
++Cancel the current VM backup.
++
+ ETEXI
+
+ {
+diff --git a/include/block/block_int.h b/include/block/block_int.h
+index 105bffe0f7..43b00c15ac 100644
+--- a/include/block/block_int.h
++++ b/include/block/block_int.h
+@@ -62,7 +62,7 @@
+
+ typedef int BackupDumpFunc(void *opaque, uint64_t offset, uint64_t bytes, const void *buf);
+
+-BlockDriverState *bdrv_backuo_dump_create(
++BlockDriverState *bdrv_backup_dump_create(
+ int dump_cb_block_size,
+ uint64_t byte_size,
+ BackupDumpFunc *dump_cb,
+diff --git a/include/monitor/hmp.h b/include/monitor/hmp.h
+index c6ee8295f0..0f2f96c4af 100644
+--- a/include/monitor/hmp.h
++++ b/include/monitor/hmp.h
+@@ -30,6 +30,7 @@ void hmp_info_migrate(Monitor *mon, const QDict *qdict);
+ void hmp_info_migrate_capabilities(Monitor *mon, const QDict *qdict);
+ void hmp_info_migrate_parameters(Monitor *mon, const QDict *qdict);
+ void hmp_info_migrate_cache_size(Monitor *mon, const QDict *qdict);
++void hmp_info_backup(Monitor *mon, const QDict *qdict);
+ void hmp_info_cpus(Monitor *mon, const QDict *qdict);
+ void hmp_info_block(Monitor *mon, const QDict *qdict);
+ void hmp_info_blockstats(Monitor *mon, const QDict *qdict);
+@@ -90,6 +91,8 @@ void hmp_eject(Monitor *mon, const QDict *qdict);
+ void hmp_change(Monitor *mon, const QDict *qdict);
+ void hmp_block_set_io_throttle(Monitor *mon, const QDict *qdict);
+ void hmp_block_stream(Monitor *mon, const QDict *qdict);
++void hmp_backup(Monitor *mon, const QDict *qdict);
++void hmp_backup_cancel(Monitor *mon, const QDict *qdict);
+ void hmp_block_job_set_speed(Monitor *mon, const QDict *qdict);
+ void hmp_block_job_cancel(Monitor *mon, const QDict *qdict);
+ void hmp_block_job_pause(Monitor *mon, const QDict *qdict);
+diff --git a/monitor/hmp-cmds.c b/monitor/hmp-cmds.c
+index 90aa34be25..1e33c61a2c 100644
+--- a/monitor/hmp-cmds.c
++++ b/monitor/hmp-cmds.c
+@@ -194,6 +194,50 @@ void hmp_info_mice(Monitor *mon, const QDict *qdict)
+ qapi_free_MouseInfoList(mice_list);
+ }
+
++void hmp_info_backup(Monitor *mon, const QDict *qdict)
++{
++ BackupStatus *info;
++
++ info = qmp_query_backup(NULL);
++
++ if (!info) {
++ monitor_printf(mon, "Backup status: not initialized\n");
++ return;
++ }
++
++ if (info->has_status) {
++ if (info->has_errmsg) {
++ monitor_printf(mon, "Backup status: %s - %s\n",
++ info->status, info->errmsg);
++ } else {
++ monitor_printf(mon, "Backup status: %s\n", info->status);
++ }
++ }
++
++ if (info->has_backup_file) {
++ monitor_printf(mon, "Start time: %s", ctime(&info->start_time));
++ if (info->end_time) {
++ monitor_printf(mon, "End time: %s", ctime(&info->end_time));
++ }
++
++ int per = (info->has_total && info->total &&
++ info->has_transferred && info->transferred) ?
++ (info->transferred * 100)/info->total : 0;
++ int zero_per = (info->has_total && info->total &&
++ info->has_zero_bytes && info->zero_bytes) ?
++ (info->zero_bytes * 100)/info->total : 0;
++ monitor_printf(mon, "Backup file: %s\n", info->backup_file);
++ monitor_printf(mon, "Backup uuid: %s\n", info->uuid);
++ monitor_printf(mon, "Total size: %zd\n", info->total);
++ monitor_printf(mon, "Transferred bytes: %zd (%d%%)\n",
++ info->transferred, per);
++ monitor_printf(mon, "Zero bytes: %zd (%d%%)\n",
++ info->zero_bytes, zero_per);
++ }
++
++ qapi_free_BackupStatus(info);
++}
++
+ static char *SocketAddress_to_str(SocketAddress *addr)
+ {
+ switch (addr->type) {
+@@ -2062,6 +2106,39 @@ void hmp_block_stream(Monitor *mon, const QDict *qdict)
+ hmp_handle_error(mon, &error);
+ }
+
++void hmp_backup_cancel(Monitor *mon, const QDict *qdict)
++{
++ Error *error = NULL;
++
++ qmp_backup_cancel(&error);
++
++ hmp_handle_error(mon, &error);
++}
++
++void hmp_backup(Monitor *mon, const QDict *qdict)
++{
++ Error *error = NULL;
++
++ int dir = qdict_get_try_bool(qdict, "directory", 0);
++ const char *backup_file = qdict_get_str(qdict, "backupfile");
++ const char *devlist = qdict_get_try_str(qdict, "devlist");
++ int64_t speed = qdict_get_try_int(qdict, "speed", 0);
++
++ qmp_backup(
++ backup_file,
++ false, NULL, // BPS password
++ false, NULL, // BPS keyfile
++ false, NULL, // BPS key_password
++ false, NULL, // BPS fingerprint
++ false, NULL, // BPS backup-id
++ false, 0, // BPS backup-time
++ true, dir ? BACKUP_FORMAT_DIR : BACKUP_FORMAT_VMA,
++ false, NULL, false, NULL, !!devlist,
++ devlist, qdict_haskey(qdict, "speed"), speed, &error);
++
++ hmp_handle_error(mon, &error);
++}
++
+ void hmp_block_job_set_speed(Monitor *mon, const QDict *qdict)
+ {
+ Error *error = NULL;
+diff --git a/proxmox-backup-client.c b/proxmox-backup-client.c
+new file mode 100644
+index 0000000000..b7bc7f2574
+--- /dev/null
++++ b/proxmox-backup-client.c
+@@ -0,0 +1,182 @@
++#include "proxmox-backup-client.h"
++#include "qemu/main-loop.h"
++#include "block/aio-wait.h"
++#include "qapi/error.h"
++
++/* Proxmox Backup Server client bindings using coroutines */
++
++typedef struct BlockOnCoroutineWrapper {
++ AioContext *ctx;
++ CoroutineEntry *entry;
++ void *entry_arg;
++ bool finished;
++} BlockOnCoroutineWrapper;
++
++// Waker implementaion to syncronice with proxmox backup rust code
++typedef struct ProxmoxBackupWaker {
++ Coroutine *co;
++ AioContext *ctx;
++} ProxmoxBackupWaker;
++
++static void coroutine_fn block_on_coroutine_wrapper(void *opaque)
++{
++ BlockOnCoroutineWrapper *wrapper = opaque;
++ wrapper->entry(wrapper->entry_arg);
++ wrapper->finished = true;
++ aio_wait_kick();
++}
++
++void block_on_coroutine_fn(CoroutineEntry *entry, void *entry_arg)
++{
++ assert(!qemu_in_coroutine());
++
++ AioContext *ctx = qemu_get_current_aio_context();
++ BlockOnCoroutineWrapper wrapper = {
++ .finished = false,
++ .entry = entry,
++ .entry_arg = entry_arg,
++ .ctx = ctx,
++ };
++ Coroutine *wrapper_co = qemu_coroutine_create(block_on_coroutine_wrapper, &wrapper);
++ aio_co_enter(ctx, wrapper_co);
++ AIO_WAIT_WHILE(ctx, !wrapper.finished);
++}
++
++// This is called from another thread, so we use aio_co_schedule()
++static void proxmox_backup_schedule_wake(void *data) {
++ ProxmoxBackupWaker *waker = (ProxmoxBackupWaker *)data;
++ aio_co_schedule(waker->ctx, waker->co);
++}
++
++int coroutine_fn
++proxmox_backup_co_connect(ProxmoxBackupHandle *pbs, Error **errp)
++{
++ Coroutine *co = qemu_coroutine_self();
++ AioContext *ctx = qemu_get_current_aio_context();
++ ProxmoxBackupWaker waker = { .co = co, .ctx = ctx };
++ char *pbs_err = NULL;
++ int pbs_res = -1;
++
++ proxmox_backup_connect_async(pbs, proxmox_backup_schedule_wake, &waker, &pbs_res, &pbs_err);
++ qemu_coroutine_yield();
++ if (pbs_res < 0) {
++ if (errp) error_setg(errp, "backup connect failed: %s", pbs_err ? pbs_err : "unknown error");
++ if (pbs_err) proxmox_backup_free_error(pbs_err);
++ }
++ return pbs_res;
++}
++
++int coroutine_fn
++proxmox_backup_co_add_config(
++ ProxmoxBackupHandle *pbs,
++ const char *name,
++ const uint8_t *data,
++ uint64_t size,
++ Error **errp)
++{
++ Coroutine *co = qemu_coroutine_self();
++ AioContext *ctx = qemu_get_current_aio_context();
++ ProxmoxBackupWaker waker = { .co = co, .ctx = ctx };
++ char *pbs_err = NULL;
++ int pbs_res = -1;
++
++ proxmox_backup_add_config_async(
++ pbs, name, data, size ,proxmox_backup_schedule_wake, &waker, &pbs_res, &pbs_err);
++ qemu_coroutine_yield();
++ if (pbs_res < 0) {
++ if (errp) error_setg(errp, "backup add_config %s failed: %s", name, pbs_err ? pbs_err : "unknown error");
++ if (pbs_err) proxmox_backup_free_error(pbs_err);
++ }
++ return pbs_res;
++}
++
++int coroutine_fn
++proxmox_backup_co_register_image(
++ ProxmoxBackupHandle *pbs,
++ const char *device_name,
++ uint64_t size,
++ Error **errp)
++{
++ Coroutine *co = qemu_coroutine_self();
++ AioContext *ctx = qemu_get_current_aio_context();
++ ProxmoxBackupWaker waker = { .co = co, .ctx = ctx };
++ char *pbs_err = NULL;
++ int pbs_res = -1;
++
++ proxmox_backup_register_image_async(
++ pbs, device_name, size ,proxmox_backup_schedule_wake, &waker, &pbs_res, &pbs_err);
++ qemu_coroutine_yield();
++ if (pbs_res < 0) {
++ if (errp) error_setg(errp, "backup register image failed: %s", pbs_err ? pbs_err : "unknown error");
++ if (pbs_err) proxmox_backup_free_error(pbs_err);
++ }
++ return pbs_res;
++}
++
++int coroutine_fn
++proxmox_backup_co_finish(
++ ProxmoxBackupHandle *pbs,
++ Error **errp)
++{
++ Coroutine *co = qemu_coroutine_self();
++ AioContext *ctx = qemu_get_current_aio_context();
++ ProxmoxBackupWaker waker = { .co = co, .ctx = ctx };
++ char *pbs_err = NULL;
++ int pbs_res = -1;
++
++ proxmox_backup_finish_async(
++ pbs, proxmox_backup_schedule_wake, &waker, &pbs_res, &pbs_err);
++ qemu_coroutine_yield();
++ if (pbs_res < 0) {
++ if (errp) error_setg(errp, "backup finish failed: %s", pbs_err ? pbs_err : "unknown error");
++ if (pbs_err) proxmox_backup_free_error(pbs_err);
++ }
++ return pbs_res;
++}
++
++int coroutine_fn
++proxmox_backup_co_close_image(
++ ProxmoxBackupHandle *pbs,
++ uint8_t dev_id,
++ Error **errp)
++{
++ Coroutine *co = qemu_coroutine_self();
++ AioContext *ctx = qemu_get_current_aio_context();
++ ProxmoxBackupWaker waker = { .co = co, .ctx = ctx };
++ char *pbs_err = NULL;
++ int pbs_res = -1;
++
++ proxmox_backup_close_image_async(
++ pbs, dev_id, proxmox_backup_schedule_wake, &waker, &pbs_res, &pbs_err);
++ qemu_coroutine_yield();
++ if (pbs_res < 0) {
++ if (errp) error_setg(errp, "backup close image failed: %s", pbs_err ? pbs_err : "unknown error");
++ if (pbs_err) proxmox_backup_free_error(pbs_err);
++ }
++ return pbs_res;
++}
++
++int coroutine_fn
++proxmox_backup_co_write_data(
++ ProxmoxBackupHandle *pbs,
++ uint8_t dev_id,
++ const uint8_t *data,
++ uint64_t offset,
++ uint64_t size,
++ Error **errp)
++{
++ Coroutine *co = qemu_coroutine_self();
++ AioContext *ctx = qemu_get_current_aio_context();
++ ProxmoxBackupWaker waker = { .co = co, .ctx = ctx };
++ char *pbs_err = NULL;
++ int pbs_res = -1;
++
++ proxmox_backup_write_data_async(
++ pbs, dev_id, data, offset, size, proxmox_backup_schedule_wake, &waker, &pbs_res, &pbs_err);
++ qemu_coroutine_yield();
++ if (pbs_res < 0) {
++ if (errp) error_setg(errp, "backup write data failed: %s", pbs_err ? pbs_err : "unknown error");
++ if (pbs_err) proxmox_backup_free_error(pbs_err);
++ }
++ return pbs_res;
++}
+diff --git a/proxmox-backup-client.h b/proxmox-backup-client.h
+new file mode 100644
+index 0000000000..b311bf8de8
+--- /dev/null
++++ b/proxmox-backup-client.h
+@@ -0,0 +1,52 @@
++#ifndef PROXMOX_BACKUP_CLIENT_H
++#define PROXMOX_BACKUP_CLIENT_H
++
++#include "qemu/osdep.h"
++#include "qemu/coroutine.h"
++#include "proxmox-backup-qemu.h"
++
++void block_on_coroutine_fn(CoroutineEntry *entry, void *entry_arg);
++
++int coroutine_fn
++proxmox_backup_co_connect(
++ ProxmoxBackupHandle *pbs,
++ Error **errp);
++
++int coroutine_fn
++proxmox_backup_co_add_config(
++ ProxmoxBackupHandle *pbs,
++ const char *name,
++ const uint8_t *data,
++ uint64_t size,
++ Error **errp);
++
++int coroutine_fn
++proxmox_backup_co_register_image(
++ ProxmoxBackupHandle *pbs,
++ const char *device_name,
++ uint64_t size,
++ Error **errp);
++
++
++int coroutine_fn
++proxmox_backup_co_finish(
++ ProxmoxBackupHandle *pbs,
++ Error **errp);
++
++int coroutine_fn
++proxmox_backup_co_close_image(
++ ProxmoxBackupHandle *pbs,
++ uint8_t dev_id,
++ Error **errp);
++
++int coroutine_fn
++proxmox_backup_co_write_data(
++ ProxmoxBackupHandle *pbs,
++ uint8_t dev_id,
++ const uint8_t *data,
++ uint64_t offset,
++ uint64_t size,
++ Error **errp);
++
++
++#endif /* PROXMOX_BACKUP_CLIENT_H */
+diff --git a/pve-backup.c b/pve-backup.c
+new file mode 100644
+index 0000000000..9ae89fb679
+--- /dev/null
++++ b/pve-backup.c
+@@ -0,0 +1,959 @@
++#include "proxmox-backup-client.h"
++#include "vma.h"
++
++#include "qemu/osdep.h"
++#include "qemu/module.h"
++#include "sysemu/block-backend.h"
++#include "sysemu/blockdev.h"
++#include "block/blockjob.h"
++#include "qapi/qapi-commands-block.h"
++#include "qapi/qmp/qerror.h"
++
++/* PVE backup state and related function */
++
++
++static struct PVEBackupState {
++ struct {
++ // Everithing accessed from qmp command, protected using rwlock
++ CoRwlock rwlock;
++ Error *error;
++ time_t start_time;
++ time_t end_time;
++ char *backup_file;
++ uuid_t uuid;
++ char uuid_str[37];
++ size_t total;
++ size_t transferred;
++ size_t zero_bytes;
++ bool cancel;
++ } stat;
++ int64_t speed;
++ VmaWriter *vmaw;
++ ProxmoxBackupHandle *pbs;
++ GList *di_list;
++ CoMutex backup_mutex;
++} backup_state;
++
++static void pvebackup_init(void)
++{
++ qemu_co_rwlock_init(&backup_state.stat.rwlock);
++ qemu_co_mutex_init(&backup_state.backup_mutex);
++}
++
++// initialize PVEBackupState at startup
++opts_init(pvebackup_init);
++
++typedef struct PVEBackupDevInfo {
++ BlockDriverState *bs;
++ size_t size;
++ uint8_t dev_id;
++ bool completed;
++ char targetfile[PATH_MAX];
++ BlockDriverState *target;
++} PVEBackupDevInfo;
++
++static void pvebackup_co_run_next_job(void);
++
++static int coroutine_fn
++pvebackup_co_dump_cb(
++ void *opaque,
++ uint64_t start,
++ uint64_t bytes,
++ const void *pbuf)
++{
++ assert(qemu_in_coroutine());
++
++ const uint64_t size = bytes;
++ const unsigned char *buf = pbuf;
++ PVEBackupDevInfo *di = opaque;
++
++ qemu_co_rwlock_rdlock(&backup_state.stat.rwlock);
++ bool cancel = backup_state.stat.cancel;
++ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
++
++ if (cancel) {
++ return size; // return success
++ }
++
++ qemu_co_mutex_lock(&backup_state.backup_mutex);
++
++ int ret = -1;
++
++ if (backup_state.vmaw) {
++ size_t zero_bytes = 0;
++ uint64_t remaining = size;
++
++ uint64_t cluster_num = start / VMA_CLUSTER_SIZE;
++ if ((cluster_num * VMA_CLUSTER_SIZE) != start) {
++ qemu_co_rwlock_rdlock(&backup_state.stat.rwlock);
++ if (!backup_state.stat.error) {
++ qemu_co_rwlock_upgrade(&backup_state.stat.rwlock);
++ error_setg(&backup_state.stat.error,
++ "got unaligned write inside backup dump "
++ "callback (sector %ld)", start);
++ }
++ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
++ qemu_co_mutex_unlock(&backup_state.backup_mutex);
++ return -1; // not aligned to cluster size
++ }
++
++ while (remaining > 0) {
++ ret = vma_writer_write(backup_state.vmaw, di->dev_id, cluster_num,
++ buf, &zero_bytes);
++ ++cluster_num;
++ if (buf) {
++ buf += VMA_CLUSTER_SIZE;
++ }
++ if (ret < 0) {
++ qemu_co_rwlock_rdlock(&backup_state.stat.rwlock);
++ if (!backup_state.stat.error) {
++ qemu_co_rwlock_upgrade(&backup_state.stat.rwlock);
++ vma_writer_error_propagate(backup_state.vmaw, &backup_state.stat.error);
++ }
++ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
++
++ qemu_co_mutex_unlock(&backup_state.backup_mutex);
++ return ret;
++ } else {
++ qemu_co_rwlock_wrlock(&backup_state.stat.rwlock);
++ backup_state.stat.zero_bytes += zero_bytes;
++ if (remaining >= VMA_CLUSTER_SIZE) {
++ backup_state.stat.transferred += VMA_CLUSTER_SIZE;
++ remaining -= VMA_CLUSTER_SIZE;
++ } else {
++ backup_state.stat.transferred += remaining;
++ remaining = 0;
++ }
++ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
++ }
++ }
++ } else if (backup_state.pbs) {
++ Error *local_err = NULL;
++ int pbs_res = -1;
++
++ pbs_res = proxmox_backup_co_write_data(backup_state.pbs, di->dev_id, buf, start, size, &local_err);
++
++ qemu_co_rwlock_wrlock(&backup_state.stat.rwlock);
++
++ if (pbs_res < 0) {
++ error_propagate(&backup_state.stat.error, local_err);
++ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
++ qemu_co_mutex_unlock(&backup_state.backup_mutex);
++ return pbs_res;
++ } else {
++ if (!buf) {
++ backup_state.stat.zero_bytes += size;
++ }
++ backup_state.stat.transferred += size;
++ }
++
++ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
++
++ } else {
++ qemu_co_rwlock_wrlock(&backup_state.stat.rwlock);
++ if (!buf) {
++ backup_state.stat.zero_bytes += size;
++ }
++ backup_state.stat.transferred += size;
++ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
++ }
++
++ qemu_co_mutex_unlock(&backup_state.backup_mutex);
++
++ return size;
++}
++
++static void coroutine_fn pvebackup_co_cleanup(void)
++{
++ assert(qemu_in_coroutine());
++
++ qemu_co_mutex_lock(&backup_state.backup_mutex);
++
++ qemu_co_rwlock_wrlock(&backup_state.stat.rwlock);
++ backup_state.stat.end_time = time(NULL);
++ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
++
++ if (backup_state.vmaw) {
++ Error *local_err = NULL;
++ vma_writer_close(backup_state.vmaw, &local_err);
++
++ if (local_err != NULL) {
++ qemu_co_rwlock_wrlock(&backup_state.stat.rwlock);
++ error_propagate(&backup_state.stat.error, local_err);
++ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
++ }
++
++ backup_state.vmaw = NULL;
++ }
++
++ if (backup_state.pbs) {
++ qemu_co_rwlock_rdlock(&backup_state.stat.rwlock);
++ bool error_or_canceled = backup_state.stat.error || backup_state.stat.cancel;
++ if (!error_or_canceled) {
++ Error *local_err = NULL;
++ proxmox_backup_co_finish(backup_state.pbs, &local_err);
++ if (local_err != NULL) {
++ qemu_co_rwlock_upgrade(&backup_state.stat.rwlock);
++ error_propagate(&backup_state.stat.error, local_err);
++ }
++ }
++ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
++
++ proxmox_backup_disconnect(backup_state.pbs);
++ backup_state.pbs = NULL;
++ }
++
++ g_list_free(backup_state.di_list);
++ backup_state.di_list = NULL;
++ qemu_co_mutex_unlock(&backup_state.backup_mutex);
++}
++
++typedef struct PVEBackupCompeteCallbackData {
++ PVEBackupDevInfo *di;
++ int result;
++} PVEBackupCompeteCallbackData;
++
++static void coroutine_fn pvebackup_co_complete_cb(void *opaque)
++{
++ assert(qemu_in_coroutine());
++
++ PVEBackupCompeteCallbackData *cb_data = opaque;
++
++ qemu_co_mutex_lock(&backup_state.backup_mutex);
++
++ PVEBackupDevInfo *di = cb_data->di;
++ int ret = cb_data->result;
++
++ di->completed = true;
++
++ qemu_co_rwlock_rdlock(&backup_state.stat.rwlock);
++ bool error_or_canceled = backup_state.stat.error || backup_state.stat.cancel;
++
++ if (ret < 0 && !backup_state.stat.error) {
++ qemu_co_rwlock_upgrade(&backup_state.stat.rwlock);
++ error_setg(&backup_state.stat.error, "job failed with err %d - %s",
++ ret, strerror(-ret));
++ }
++ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
++
++ di->bs = NULL;
++
++ if (di->target) {
++ bdrv_unref(di->target);
++ di->target = NULL;
++ }
++
++ if (backup_state.vmaw) {
++ vma_writer_close_stream(backup_state.vmaw, di->dev_id);
++ }
++
++ if (backup_state.pbs && !error_or_canceled) {
++ Error *local_err = NULL;
++ proxmox_backup_co_close_image(backup_state.pbs, di->dev_id, &local_err);
++ if (local_err != NULL) {
++ qemu_co_rwlock_wrlock(&backup_state.stat.rwlock);
++ error_propagate(&backup_state.stat.error, local_err);
++ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
++ }
++ }
++
++ // remove self from job queue
++ backup_state.di_list = g_list_remove(backup_state.di_list, di);
++ g_free(di);
++
++ int pending_jobs = g_list_length(backup_state.di_list);
++
++ qemu_co_mutex_unlock(&backup_state.backup_mutex);
++
++ if (pending_jobs > 0) {
++ pvebackup_co_run_next_job();
++ } else {
++ pvebackup_co_cleanup();
++ }
++}
++
++static void pvebackup_complete_cb(void *opaque, int ret)
++{
++ // This can be called from the main loop, or from a coroutine
++ PVEBackupCompeteCallbackData cb_data = {
++ .di = opaque,
++ .result = ret,
++ };
++
++ if (qemu_in_coroutine()) {
++ pvebackup_co_complete_cb(&cb_data);
++ } else {
++ block_on_coroutine_fn(pvebackup_co_complete_cb, &cb_data);
++ }
++}
++
++static void coroutine_fn pvebackup_co_cancel(void *opaque)
++{
++ assert(qemu_in_coroutine());
++
++ qemu_co_rwlock_wrlock(&backup_state.stat.rwlock);
++ backup_state.stat.cancel = true;
++ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
++
++ qemu_co_mutex_lock(&backup_state.backup_mutex);
++
++ // Avoid race between block jobs and backup-cancel command:
++ if (!(backup_state.vmaw || backup_state.pbs)) {
++ qemu_co_mutex_unlock(&backup_state.backup_mutex);
++ return;
++ }
++
++ qemu_co_rwlock_rdlock(&backup_state.stat.rwlock);
++ if (!backup_state.stat.error) {
++ qemu_co_rwlock_upgrade(&backup_state.stat.rwlock);
++ error_setg(&backup_state.stat.error, "backup cancelled");
++ }
++ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
++
++ if (backup_state.vmaw) {
++ /* make sure vma writer does not block anymore */
++ vma_writer_set_error(backup_state.vmaw, "backup cancelled");
++ }
++
++ if (backup_state.pbs) {
++ proxmox_backup_abort(backup_state.pbs, "backup cancelled");
++ }
++
++ bool running_jobs = 0;
++ GList *l = backup_state.di_list;
++ while (l) {
++ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
++ l = g_list_next(l);
++ if (!di->completed && di->bs) {
++ for (BlockJob *job = block_job_next(NULL); job; job = block_job_next(job)) {
++ if (job->job.driver->job_type != JOB_TYPE_BACKUP) {
++ continue;
++ }
++
++ BackupBlockJob *bjob = container_of(job, BackupBlockJob, common);
++ if (bjob && bjob->source_bs == di->bs) {
++ AioContext *aio_context = job->job.aio_context;
++ aio_context_acquire(aio_context);
++
++ if (!di->completed) {
++ running_jobs += 1;
++ job_cancel(&job->job, false);
++ }
++ aio_context_release(aio_context);
++ }
++ }
++ }
++ }
++
++ qemu_co_mutex_unlock(&backup_state.backup_mutex);
++
++ if (running_jobs == 0) pvebackup_co_cleanup(); // else job will call completion handler
++}
++
++void qmp_backup_cancel(Error **errp)
++{
++ block_on_coroutine_fn(pvebackup_co_cancel, NULL);
++}
++
++static int coroutine_fn pvebackup_co_add_config(
++ const char *file,
++ const char *name,
++ BackupFormat format,
++ const char *backup_dir,
++ VmaWriter *vmaw,
++ ProxmoxBackupHandle *pbs,
++ Error **errp)
++{
++ int res = 0;
++
++ char *cdata = NULL;
++ gsize clen = 0;
++ GError *err = NULL;
++ if (!g_file_get_contents(file, &cdata, &clen, &err)) {
++ error_setg(errp, "unable to read file '%s'", file);
++ return 1;
++ }
++
++ char *basename = g_path_get_basename(file);
++ if (name == NULL) name = basename;
++
++ if (format == BACKUP_FORMAT_VMA) {
++ if (vma_writer_add_config(vmaw, name, cdata, clen) != 0) {
++ error_setg(errp, "unable to add %s config data to vma archive", file);
++ goto err;
++ }
++ } else if (format == BACKUP_FORMAT_PBS) {
++ if (proxmox_backup_co_add_config(pbs, name, (unsigned char *)cdata, clen, errp) < 0)
++ goto err;
++ } else if (format == BACKUP_FORMAT_DIR) {
++ char config_path[PATH_MAX];
++ snprintf(config_path, PATH_MAX, "%s/%s", backup_dir, name);
++ if (!g_file_set_contents(config_path, cdata, clen, &err)) {
++ error_setg(errp, "unable to write config file '%s'", config_path);
++ goto err;
++ }
++ }
++
++ out:
++ g_free(basename);
++ g_free(cdata);
++ return res;
++
++ err:
++ res = -1;
++ goto out;
++}
++
++bool job_should_pause(Job *job);
++
++static void coroutine_fn pvebackup_co_run_next_job(void)
++{
++ assert(qemu_in_coroutine());
++
++ qemu_co_mutex_lock(&backup_state.backup_mutex);
++
++ GList *l = backup_state.di_list;
++ while (l) {
++ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
++ l = g_list_next(l);
++ if (!di->completed && di->bs) {
++ for (BlockJob *job = block_job_next(NULL); job; job = block_job_next(job)) {
++ if (job->job.driver->job_type != JOB_TYPE_BACKUP) {
++ continue;
++ }
++
++ BackupBlockJob *bjob = container_of(job, BackupBlockJob, common);
++ if (bjob && bjob->source_bs == di->bs) {
++ AioContext *aio_context = job->job.aio_context;
++ qemu_co_mutex_unlock(&backup_state.backup_mutex);
++ aio_context_acquire(aio_context);
++
++ if (job_should_pause(&job->job)) {
++ qemu_co_rwlock_rdlock(&backup_state.stat.rwlock);
++ bool error_or_canceled = backup_state.stat.error || backup_state.stat.cancel;
++ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
++
++ if (error_or_canceled) {
++ job_cancel(&job->job, false);
++ } else {
++ job_resume(&job->job);
++ }
++ }
++ aio_context_release(aio_context);
++ return;
++ }
++ }
++ }
++ }
++ qemu_co_mutex_unlock(&backup_state.backup_mutex);
++}
++
++typedef struct QmpBackupTask {
++ const char *backup_file;
++ bool has_password;
++ const char *password;
++ bool has_keyfile;
++ const char *keyfile;
++ bool has_key_password;
++ const char *key_password;
++ bool has_backup_id;
++ const char *backup_id;
++ bool has_backup_time;
++ const char *fingerprint;
++ bool has_fingerprint;
++ int64_t backup_time;
++ bool has_format;
++ BackupFormat format;
++ bool has_config_file;
++ const char *config_file;
++ bool has_firewall_file;
++ const char *firewall_file;
++ bool has_devlist;
++ const char *devlist;
++ bool has_speed;
++ int64_t speed;
++ Error **errp;
++ UuidInfo *result;
++} QmpBackupTask;
++
++static void coroutine_fn pvebackup_co_start(void *opaque)
++{
++ assert(qemu_in_coroutine());
++
++ QmpBackupTask *task = opaque;
++
++ task->result = NULL; // just to be sure
++
++ BlockBackend *blk;
++ BlockDriverState *bs = NULL;
++ const char *backup_dir = NULL;
++ Error *local_err = NULL;
++ uuid_t uuid;
++ VmaWriter *vmaw = NULL;
++ ProxmoxBackupHandle *pbs = NULL;
++ gchar **devs = NULL;
++ GList *di_list = NULL;
++ GList *l;
++ UuidInfo *uuid_info;
++ BlockJob *job;
++
++ const char *config_name = "qemu-server.conf";
++ const char *firewall_name = "qemu-server.fw";
++
++ qemu_co_mutex_lock(&backup_state.backup_mutex);
++
++ if (backup_state.di_list) {
++ qemu_co_mutex_unlock(&backup_state.backup_mutex);
++ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
++ "previous backup not finished");
++ return;
++ }
++
++ /* Todo: try to auto-detect format based on file name */
++ BackupFormat format = task->has_format ? task->format : BACKUP_FORMAT_VMA;
++
++ if (task->has_devlist) {
++ devs = g_strsplit_set(task->devlist, ",;:", -1);
++
++ gchar **d = devs;
++ while (d && *d) {
++ blk = blk_by_name(*d);
++ if (blk) {
++ bs = blk_bs(blk);
++ if (bdrv_is_read_only(bs)) {
++ error_setg(task->errp, "Node '%s' is read only", *d);
++ goto err;
++ }
++ if (!bdrv_is_inserted(bs)) {
++ error_setg(task->errp, QERR_DEVICE_HAS_NO_MEDIUM, *d);
++ goto err;
++ }
++ PVEBackupDevInfo *di = g_new0(PVEBackupDevInfo, 1);
++ di->bs = bs;
++ di_list = g_list_append(di_list, di);
++ } else {
++ error_set(task->errp, ERROR_CLASS_DEVICE_NOT_FOUND,
++ "Device '%s' not found", *d);
++ goto err;
++ }
++ d++;
++ }
++
++ } else {
++ BdrvNextIterator it;
++
++ bs = NULL;
++ for (bs = bdrv_first(&it); bs; bs = bdrv_next(&it)) {
++ if (!bdrv_is_inserted(bs) || bdrv_is_read_only(bs)) {
++ continue;
++ }
++
++ PVEBackupDevInfo *di = g_new0(PVEBackupDevInfo, 1);
++ di->bs = bs;
++ di_list = g_list_append(di_list, di);
++ }
++ }
++
++ if (!di_list) {
++ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "empty device list");
++ goto err;
++ }
++
++ size_t total = 0;
++
++ l = di_list;
++ while (l) {
++ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
++ l = g_list_next(l);
++ if (bdrv_op_is_blocked(di->bs, BLOCK_OP_TYPE_BACKUP_SOURCE, task->errp)) {
++ goto err;
++ }
++
++ ssize_t size = bdrv_getlength(di->bs);
++ if (size < 0) {
++ error_setg_errno(task->errp, -di->size, "bdrv_getlength failed");
++ goto err;
++ }
++ di->size = size;
++ total += size;
++ }
++
++ uuid_generate(uuid);
++
++ if (format == BACKUP_FORMAT_PBS) {
++ if (!task->has_password) {
++ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "missing parameter 'password'");
++ goto err;
++ }
++ if (!task->has_backup_id) {
++ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "missing parameter 'backup-id'");
++ goto err;
++ }
++ if (!task->has_backup_time) {
++ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "missing parameter 'backup-time'");
++ goto err;
++ }
++
++ int dump_cb_block_size = PROXMOX_BACKUP_DEFAULT_CHUNK_SIZE; // Hardcoded (4M)
++ firewall_name = "fw.conf";
++
++ char *pbs_err = NULL;
++ pbs = proxmox_backup_new(
++ task->backup_file,
++ task->backup_id,
++ task->backup_time,
++ dump_cb_block_size,
++ task->has_password ? task->password : NULL,
++ task->has_keyfile ? task->keyfile : NULL,
++ task->has_key_password ? task->key_password : NULL,
++ task->has_fingerprint ? task->fingerprint : NULL,
++ &pbs_err);
++
++ if (!pbs) {
++ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
++ "proxmox_backup_new failed: %s", pbs_err);
++ proxmox_backup_free_error(pbs_err);
++ goto err;
++ }
++
++ if (proxmox_backup_co_connect(pbs, task->errp) < 0)
++ goto err;
++
++ /* register all devices */
++ l = di_list;
++ while (l) {
++ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
++ l = g_list_next(l);
++
++ const char *devname = bdrv_get_device_name(di->bs);
++
++ int dev_id = proxmox_backup_co_register_image(pbs, devname, di->size, task->errp);
++ if (dev_id < 0)
++ goto err;
++
++ if (!(di->target = bdrv_backup_dump_create(dump_cb_block_size, di->size, pvebackup_co_dump_cb, di, task->errp))) {
++ goto err;
++ }
++
++ di->dev_id = dev_id;
++ }
++ } else if (format == BACKUP_FORMAT_VMA) {
++ vmaw = vma_writer_create(task->backup_file, uuid, &local_err);
++ if (!vmaw) {
++ if (local_err) {
++ error_propagate(task->errp, local_err);
++ }
++ goto err;
++ }
++
++ /* register all devices for vma writer */
++ l = di_list;
++ while (l) {
++ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
++ l = g_list_next(l);
++
++ if (!(di->target = bdrv_backup_dump_create(VMA_CLUSTER_SIZE, di->size, pvebackup_co_dump_cb, di, task->errp))) {
++ goto err;
++ }
++
++ const char *devname = bdrv_get_device_name(di->bs);
++ di->dev_id = vma_writer_register_stream(vmaw, devname, di->size);
++ if (di->dev_id <= 0) {
++ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
++ "register_stream failed");
++ goto err;
++ }
++ }
++ } else if (format == BACKUP_FORMAT_DIR) {
++ if (mkdir(task->backup_file, 0640) != 0) {
++ error_setg_errno(task->errp, errno, "can't create directory '%s'\n",
++ task->backup_file);
++ goto err;
++ }
++ backup_dir = task->backup_file;
++
++ l = di_list;
++ while (l) {
++ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
++ l = g_list_next(l);
++
++ const char *devname = bdrv_get_device_name(di->bs);
++ snprintf(di->targetfile, PATH_MAX, "%s/%s.raw", backup_dir, devname);
++
++ int flags = BDRV_O_RDWR;
++ bdrv_img_create(di->targetfile, "raw", NULL, NULL, NULL,
++ di->size, flags, false, &local_err);
++ if (local_err) {
++ error_propagate(task->errp, local_err);
++ goto err;
++ }
++
++ di->target = bdrv_open(di->targetfile, NULL, NULL, flags, &local_err);
++ if (!di->target) {
++ error_propagate(task->errp, local_err);
++ goto err;
++ }
++ }
++ } else {
++ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "unknown backup format");
++ goto err;
++ }
++
++
++ /* add configuration file to archive */
++ if (task->has_config_file) {
++ if (pvebackup_co_add_config(task->config_file, config_name, format, backup_dir,
++ vmaw, pbs, task->errp) != 0) {
++ goto err;
++ }
++ }
++
++ /* add firewall file to archive */
++ if (task->has_firewall_file) {
++ if (pvebackup_co_add_config(task->firewall_file, firewall_name, format, backup_dir,
++ vmaw, pbs, task->errp) != 0) {
++ goto err;
++ }
++ }
++ /* initialize global backup_state now */
++
++ qemu_co_rwlock_wrlock(&backup_state.stat.rwlock);
++
++ backup_state.stat.cancel = false;
++
++ if (backup_state.stat.error) {
++ error_free(backup_state.stat.error);
++ backup_state.stat.error = NULL;
++ }
++
++ backup_state.stat.start_time = time(NULL);
++ backup_state.stat.end_time = 0;
++
++ if (backup_state.stat.backup_file) {
++ g_free(backup_state.stat.backup_file);
++ }
++ backup_state.stat.backup_file = g_strdup(task->backup_file);
++
++ uuid_copy(backup_state.stat.uuid, uuid);
++ uuid_unparse_lower(uuid, backup_state.stat.uuid_str);
++ char *uuid_str = g_strdup(backup_state.stat.uuid_str);
++
++ backup_state.stat.total = total;
++ backup_state.stat.transferred = 0;
++ backup_state.stat.zero_bytes = 0;
++
++ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
++
++ backup_state.speed = (task->has_speed && task->speed > 0) ? task->speed : 0;
++
++ backup_state.vmaw = vmaw;
++ backup_state.pbs = pbs;
++
++ backup_state.di_list = di_list;
++
++ /* start all jobs (paused state) */
++ l = di_list;
++ while (l) {
++ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
++ l = g_list_next(l);
++
++ // make sure target runs in same aoi_context as source
++ AioContext *aio_context = bdrv_get_aio_context(di->bs);
++ aio_context_acquire(aio_context);
++ GSList *ignore = NULL;
++ bdrv_set_aio_context_ignore(di->target, aio_context, &ignore);
++ g_slist_free(ignore);
++ aio_context_release(aio_context);
++
++ job = backup_job_create(NULL, di->bs, di->target, backup_state.speed, MIRROR_SYNC_MODE_FULL, NULL,
++ BITMAP_SYNC_MODE_NEVER, false, NULL, BLOCKDEV_ON_ERROR_REPORT, BLOCKDEV_ON_ERROR_REPORT,
++ JOB_DEFAULT, pvebackup_complete_cb, di, 1, NULL, &local_err);
++ if (!job || local_err != NULL) {
++ qemu_co_rwlock_wrlock(&backup_state.stat.rwlock);
++ error_setg(&backup_state.stat.error, "backup_job_create failed");
++ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
++ break;
++ }
++ job_start(&job->job);
++ if (di->target) {
++ bdrv_unref(di->target);
++ di->target = NULL;
++ }
++ }
++
++ qemu_co_mutex_unlock(&backup_state.backup_mutex);
++
++ qemu_co_rwlock_rdlock(&backup_state.stat.rwlock);
++ bool no_errors = !backup_state.stat.error;
++ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
++
++ if (no_errors) {
++ pvebackup_co_run_next_job(); // run one job
++ } else {
++ pvebackup_co_cancel(NULL);
++ }
++
++ uuid_info = g_malloc0(sizeof(*uuid_info));
++ uuid_info->UUID = uuid_str;
++
++ task->result = uuid_info;
++ return;
++
++err:
++
++ l = di_list;
++ while (l) {
++ PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
++ l = g_list_next(l);
++
++ if (di->target) {
++ bdrv_unref(di->target);
++ }
++
++ if (di->targetfile[0]) {
++ unlink(di->targetfile);
++ }
++ g_free(di);
++ }
++ g_list_free(di_list);
++
++ if (devs) {
++ g_strfreev(devs);
++ }
++
++ if (vmaw) {
++ Error *err = NULL;
++ vma_writer_close(vmaw, &err);
++ unlink(task->backup_file);
++ }
++
++ if (pbs) {
++ proxmox_backup_disconnect(pbs);
++ }
++
++ if (backup_dir) {
++ rmdir(backup_dir);
++ }
++
++ qemu_co_mutex_unlock(&backup_state.backup_mutex);
++
++ task->result = NULL;
++ return;
++}
++
++UuidInfo *qmp_backup(
++ const char *backup_file,
++ bool has_password, const char *password,
++ bool has_keyfile, const char *keyfile,
++ bool has_key_password, const char *key_password,
++ bool has_fingerprint, const char *fingerprint,
++ bool has_backup_id, const char *backup_id,
++ bool has_backup_time, int64_t backup_time,
++ bool has_format, BackupFormat format,
++ bool has_config_file, const char *config_file,
++ bool has_firewall_file, const char *firewall_file,
++ bool has_devlist, const char *devlist,
++ bool has_speed, int64_t speed, Error **errp)
++{
++ QmpBackupTask task = {
++ .backup_file = backup_file,
++ .has_password = has_password,
++ .password = password,
++ .has_key_password = has_key_password,
++ .key_password = key_password,
++ .has_fingerprint = has_fingerprint,
++ .fingerprint = fingerprint,
++ .has_backup_id = has_backup_id,
++ .backup_id = backup_id,
++ .has_backup_time = has_backup_time,
++ .backup_time = backup_time,
++ .has_format = has_format,
++ .format = format,
++ .has_config_file = has_config_file,
++ .config_file = config_file,
++ .has_firewall_file = has_firewall_file,
++ .firewall_file = firewall_file,
++ .has_devlist = has_devlist,
++ .devlist = devlist,
++ .has_speed = has_speed,
++ .speed = speed,
++ .errp = errp,
++ };
++
++ block_on_coroutine_fn(pvebackup_co_start, &task);
++
++ return task.result;
++}
++
++
++typedef struct QmpQueryBackupTask {
++ Error **errp;
++ BackupStatus *result;
++} QmpQueryBackupTask;
++
++static void coroutine_fn pvebackup_co_query(void *opaque)
++{
++ assert(qemu_in_coroutine());
++
++ QmpQueryBackupTask *task = opaque;
++
++ BackupStatus *info = g_malloc0(sizeof(*info));
++
++ qemu_co_rwlock_rdlock(&backup_state.stat.rwlock);
++
++ if (!backup_state.stat.start_time) {
++ /* not started, return {} */
++ task->result = info;
++ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
++ return;
++ }
++
++ info->has_status = true;
++ info->has_start_time = true;
++ info->start_time = backup_state.stat.start_time;
++
++ if (backup_state.stat.backup_file) {
++ info->has_backup_file = true;
++ info->backup_file = g_strdup(backup_state.stat.backup_file);
++ }
++
++ info->has_uuid = true;
++ info->uuid = g_strdup(backup_state.stat.uuid_str);
++
++ if (backup_state.stat.end_time) {
++ if (backup_state.stat.error) {
++ info->status = g_strdup("error");
++ info->has_errmsg = true;
++ info->errmsg = g_strdup(error_get_pretty(backup_state.stat.error));
++ } else {
++ info->status = g_strdup("done");
++ }
++ info->has_end_time = true;
++ info->end_time = backup_state.stat.end_time;
++ } else {
++ info->status = g_strdup("active");
++ }
++
++ info->has_total = true;
++ info->total = backup_state.stat.total;
++ info->has_zero_bytes = true;
++ info->zero_bytes = backup_state.stat.zero_bytes;
++ info->has_transferred = true;
++ info->transferred = backup_state.stat.transferred;
++
++ task->result = info;
++
++ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
++}
++
++BackupStatus *qmp_query_backup(Error **errp)
++{
++ QmpQueryBackupTask task = {
++ .errp = errp,
++ .result = NULL,
++ };
++
++ block_on_coroutine_fn(pvebackup_co_query, &task);
++
++ return task.result;
++}
+diff --git a/qapi/block-core.json b/qapi/block-core.json
+index 783a868eb2..0b987ad6e3 100644
+--- a/qapi/block-core.json
++++ b/qapi/block-core.json
+@@ -800,6 +800,115 @@
+ { 'command': 'query-block', 'returns': ['BlockInfo'] }
+
+
++##
++# @BackupStatus:
++#
++# Detailed backup status.
++#
++# @status: string describing the current backup status.
++# This can be 'active', 'done', 'error'. If this field is not
++# returned, no backup process has been initiated
++#
++# @errmsg: error message (only returned if status is 'error')
++#
++# @total: total amount of bytes involved in the backup process
++#
++# @transferred: amount of bytes already backed up.
++#
++# @zero-bytes: amount of 'zero' bytes detected.
++#
++# @start-time: time (epoch) when backup job started.
++#
++# @end-time: time (epoch) when backup job finished.
++#
++# @backup-file: backup file name
++#
++# @uuid: uuid for this backup job
++#
++##
++{ 'struct': 'BackupStatus',
++ 'data': {'*status': 'str', '*errmsg': 'str', '*total': 'int',
++ '*transferred': 'int', '*zero-bytes': 'int',
++ '*start-time': 'int', '*end-time': 'int',
++ '*backup-file': 'str', '*uuid': 'str' } }
++
++##
++# @BackupFormat:
++#
++# An enumeration of supported backup formats.
++#
++# @vma: Proxmox vma backup format
++##
++{ 'enum': 'BackupFormat',
++ 'data': [ 'vma', 'dir', 'pbs' ] }
++
++##
++# @backup:
++#
++# Starts a VM backup.
++#
++# @backup-file: the backup file name
++#
++# @format: format of the backup file
++#
++# @config-file: a configuration file to include into
++# the backup archive.
++#
++# @speed: the maximum speed, in bytes per second
++#
++# @devlist: list of block device names (separated by ',', ';'
++# or ':'). By default the backup includes all writable block devices.
++#
++# @password: backup server passsword (required for format 'pbs')
++#
++# @keyfile: keyfile used for encryption (optional for format 'pbs')
++#
++# @key-password: password for keyfile (optional for format 'pbs')
++#
++# @fingerprint: server cert fingerprint (optional for format 'pbs')
++#
++# @backup-id: backup ID (required for format 'pbs')
++#
++# @backup-time: backup timestamp (Unix epoch, required for format 'pbs')
++#
++# Returns: the uuid of the backup job
++#
++##
++{ 'command': 'backup', 'data': { 'backup-file': 'str',
++ '*password': 'str',
++ '*keyfile': 'str',
++ '*key-password': 'str',
++ '*fingerprint': 'str',
++ '*backup-id': 'str',
++ '*backup-time': 'int',
++ '*format': 'BackupFormat',
++ '*config-file': 'str',
++ '*firewall-file': 'str',
++ '*devlist': 'str', '*speed': 'int' },
++ 'returns': 'UuidInfo' }
++
++##
++# @query-backup:
++#
++# Returns information about current/last backup task.
++#
++# Returns: @BackupStatus
++#
++##
++{ 'command': 'query-backup', 'returns': 'BackupStatus' }
++
++##
++# @backup-cancel:
++#
++# Cancel the current executing backup process.
++#
++# Returns: nothing on success
++#
++# Notes: This command succeeds even if there is no backup process running.
++#
++##
++{ 'command': 'backup-cancel' }
++
+ ##
+ # @BlockDeviceTimedStats:
+ #
+diff --git a/qapi/common.json b/qapi/common.json
+index 7b9cbcd97b..c3b8bb7b48 100644
+--- a/qapi/common.json
++++ b/qapi/common.json
+@@ -144,3 +144,16 @@
+ ##
+ { 'enum': 'PCIELinkWidth',
+ 'data': [ '1', '2', '4', '8', '12', '16', '32' ] }
++
++##
++# @UuidInfo:
++#
++# Guest UUID information (Universally Unique Identifier).
++#
++# @UUID: the UUID of the guest
++#
++# Since: 0.14.0
++#
++# Notes: If no UUID was specified for the guest, a null UUID is returned.
++##
++{ 'struct': 'UuidInfo', 'data': {'UUID': 'str'} }
+diff --git a/qapi/misc.json b/qapi/misc.json
+index 4c4618a574..7d506b5300 100644
+--- a/qapi/misc.json
++++ b/qapi/misc.json
+@@ -270,19 +270,6 @@
+ ##
+ { 'command': 'query-kvm', 'returns': 'KvmInfo' }
+
+-##
+-# @UuidInfo:
+-#
+-# Guest UUID information (Universally Unique Identifier).
+-#
+-# @UUID: the UUID of the guest
+-#
+-# Since: 0.14.0
+-#
+-# Notes: If no UUID was specified for the guest, a null UUID is returned.
+-##
+-{ 'struct': 'UuidInfo', 'data': {'UUID': 'str'} }
+-
+ ##
+ # @query-uuid:
+ #
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Dietmar Maurer <dietmar@proxmox.com>
-Date: Mon, 21 Oct 2019 11:51:57 +0200
-Subject: [PATCH] PVE bug fix #1071 - vma-writer.c: use correct AioContext
-
-Signed-off-by: Dietmar Maurer <dietmar@proxmox.com>
----
- vma-writer.c | 16 ++++++++--------
- 1 file changed, 8 insertions(+), 8 deletions(-)
-
-diff --git a/vma-writer.c b/vma-writer.c
-index fd9567634d..b163fa2d3a 100644
---- a/vma-writer.c
-+++ b/vma-writer.c
-@@ -199,12 +199,14 @@ int vma_writer_register_stream(VmaWriter *vmaw, const char *devname,
- return n;
- }
-
--static void vma_co_continue_write(void *opaque)
-+static void coroutine_fn yield_until_fd_writable(int fd)
- {
-- VmaWriter *vmaw = opaque;
--
-- DPRINTF("vma_co_continue_write\n");
-- qemu_coroutine_enter(vmaw->co_writer);
-+ assert(qemu_in_coroutine());
-+ AioContext *ctx = qemu_get_current_aio_context();
-+ aio_set_fd_handler(ctx, fd, false, NULL, (IOHandler *)qemu_coroutine_enter,
-+ NULL, qemu_coroutine_self());
-+ qemu_coroutine_yield();
-+ aio_set_fd_handler(ctx, fd, false, NULL, NULL, NULL, NULL);
- }
-
- static ssize_t coroutine_fn
-@@ -224,14 +226,12 @@ vma_queue_write(VmaWriter *vmaw, const void *buf, size_t bytes)
- vmaw->co_writer = qemu_coroutine_self();
-
- while (done < bytes) {
-- aio_set_fd_handler(qemu_get_aio_context(), vmaw->fd, false, NULL, vma_co_continue_write, NULL, vmaw);
-- qemu_coroutine_yield();
-- aio_set_fd_handler(qemu_get_aio_context(), vmaw->fd, false, NULL, NULL, NULL, NULL);
- if (vmaw->status < 0) {
- DPRINTF("vma_queue_write detected canceled backup\n");
- done = -1;
- break;
- }
-+ yield_until_fd_writable(vmaw->fd);
- ret = write(vmaw->fd, buf + done, bytes - done);
- if (ret > 0) {
- done += ret;
--- /dev/null
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Dietmar Maurer <dietmar@proxmox.com>
+Date: Tue, 10 Mar 2020 12:55:29 +0100
+Subject: [PATCH 32/32] PVE-Backup: pbs-restore - new command to restore from
+ proxmox backup server
+
+---
+ Makefile | 4 +-
+ pbs-restore.c | 208 ++++++++++++++++++++++++++++++++++++++++++++++++++
+ 2 files changed, 211 insertions(+), 1 deletion(-)
+ create mode 100644 pbs-restore.c
+
+diff --git a/Makefile b/Makefile
+index 18d2dba2e4..469a1e3666 100644
+--- a/Makefile
++++ b/Makefile
+@@ -453,7 +453,7 @@ dummy := $(call unnest-vars,, \
+
+ include $(SRC_PATH)/tests/Makefile.include
+
+-all: $(DOCS) $(if $(BUILD_DOCS),sphinxdocs) $(TOOLS) vma$(EXESUF) $(HELPERS-y) recurse-all modules $(vhost-user-json-y)
++all: $(DOCS) $(if $(BUILD_DOCS),sphinxdocs) $(TOOLS) vma$(EXESUF) pbs-restore$(EXESUF) $(HELPERS-y) recurse-all modules $(vhost-user-json-y)
+
+ qemu-version.h: FORCE
+ $(call quiet-command, \
+@@ -568,6 +568,8 @@ qemu-img$(EXESUF): qemu-img.o $(authz-obj-y) $(block-obj-y) $(crypto-obj-y) $(io
+ qemu-nbd$(EXESUF): qemu-nbd.o $(authz-obj-y) $(block-obj-y) $(crypto-obj-y) $(io-obj-y) $(qom-obj-y) $(COMMON_LDADDS)
+ qemu-io$(EXESUF): qemu-io.o $(authz-obj-y) $(block-obj-y) $(crypto-obj-y) $(io-obj-y) $(qom-obj-y) $(COMMON_LDADDS)
+ vma$(EXESUF): vma.o vma-reader.o $(authz-obj-y) $(block-obj-y) $(crypto-obj-y) $(io-obj-y) $(qom-obj-y) $(COMMON_LDADDS)
++pbs-restore$(EXESUF): pbs-restore.o $(authz-obj-y) $(block-obj-y) $(crypto-obj-y) $(io-obj-y) $(qom-obj-y) $(COMMON_LDADDS)
++pbs-restore$(EXESUF): LIBS += -lproxmox_backup_qemu
+
+ qemu-bridge-helper$(EXESUF): qemu-bridge-helper.o $(COMMON_LDADDS)
+
+diff --git a/pbs-restore.c b/pbs-restore.c
+new file mode 100644
+index 0000000000..f65de8b890
+--- /dev/null
++++ b/pbs-restore.c
+@@ -0,0 +1,208 @@
++/*
++ * Qemu image restore helper for Proxmox Backup
++ *
++ * Copyright (C) 2019 Proxmox Server Solutions
++ *
++ * Authors:
++ * Dietmar Maurer (dietmar@proxmox.com)
++ *
++ * This work is licensed under the terms of the GNU GPL, version 2 or later.
++ * See the COPYING file in the top-level directory.
++ *
++ */
++
++#include "qemu/osdep.h"
++#include <glib.h>
++#include <getopt.h>
++#include <string.h>
++
++#include "qemu-common.h"
++#include "qemu/module.h"
++#include "qemu/error-report.h"
++#include "qemu/main-loop.h"
++#include "qemu/cutils.h"
++#include "qapi/error.h"
++#include "qapi/qmp/qdict.h"
++#include "sysemu/block-backend.h"
++
++#include <proxmox-backup-qemu.h>
++
++static void help(void)
++{
++ const char *help_msg =
++ "usage: pbs-restore [--repository <repo>] snapshot archive-name target [command options]\n"
++ ;
++
++ printf("%s", help_msg);
++ exit(1);
++}
++
++typedef struct CallbackData {
++ BlockBackend *target;
++ uint64_t last_offset;
++ bool skip_zero;
++} CallbackData;
++
++static int write_callback(
++ void *callback_data_ptr,
++ uint64_t offset,
++ const unsigned char *data,
++ uint64_t data_len)
++{
++ int res = -1;
++
++ CallbackData *callback_data = (CallbackData *)callback_data_ptr;
++
++ uint64_t last_offset = callback_data->last_offset;
++ if (offset > last_offset) callback_data->last_offset = offset;
++
++ if (data == NULL) {
++ if (callback_data->skip_zero && offset > last_offset) {
++ return 0;
++ }
++ res = blk_pwrite_zeroes(callback_data->target, offset, data_len, 0);
++ } else {
++ res = blk_pwrite(callback_data->target, offset, data, data_len, 0);
++ }
++
++ if (res < 0) {
++ fprintf(stderr, "blk_pwrite failed at offset %ld length %ld (%d) - %s\n", offset, data_len, res, strerror(-res));
++ return res;
++ }
++
++ return 0;
++}
++
++int main(int argc, char **argv)
++{
++ Error *main_loop_err = NULL;
++ const char *format = "raw";
++ const char *repository = NULL;
++ const char *keyfile = NULL;
++ int verbose = false;
++ bool skip_zero = false;
++
++ error_init(argv[0]);
++
++ for(;;) {
++ static const struct option long_options[] = {
++ {"help", no_argument, 0, 'h'},
++ {"skip-zero", no_argument, 0, 'S'},
++ {"verbose", no_argument, 0, 'v'},
++ {"format", required_argument, 0, 'f'},
++ {"repository", required_argument, 0, 'r'},
++ {"keyfile", required_argument, 0, 'k'},
++ {0, 0, 0, 0}
++ };
++ int c = getopt_long(argc, argv, "hvf:r:k:", long_options, NULL);
++ if (c == -1) {
++ break;
++ }
++ switch(c) {
++ case ':':
++ fprintf(stderr, "missing argument for option '%s'", argv[optind - 1]);
++ return -1;
++ case '?':
++ fprintf(stderr, "unrecognized option '%s'", argv[optind - 1]);
++ return -1;
++ case 'f':
++ format = g_strdup(argv[optind - 1]);
++ break;
++ case 'r':
++ repository = g_strdup(argv[optind - 1]);
++ break;
++ case 'k':
++ keyfile = g_strdup(argv[optind - 1]);
++ break;
++ case 'v':
++ verbose = true;
++ break;
++ case 'S':
++ skip_zero = true;
++ break;
++ case 'h':
++ help();
++ return 0;
++ }
++ }
++
++ if (optind >= argc - 2) {
++ fprintf(stderr, "missing arguments\n");
++ help();
++ return -1;
++ }
++
++ if (repository == NULL) {
++ repository = getenv("PBS_REPOSITORY");
++ }
++
++ if (repository == NULL) {
++ fprintf(stderr, "no repository specified\n");
++ help();
++ return -1;
++ }
++
++ char *snapshot = argv[optind++];
++ char *archive_name = argv[optind++];
++ char *target = argv[optind++];
++
++ const char *password = getenv("PBS_PASSWORD");
++ const char *fingerprint = getenv("PBS_FINGERPRINT");
++ const char *key_password = getenv("PBS_ENCRYPTION_PASSWORD");
++
++ if (qemu_init_main_loop(&main_loop_err)) {
++ g_error("%s", error_get_pretty(main_loop_err));
++ }
++
++ bdrv_init();
++ module_call_init(MODULE_INIT_QOM);
++
++ char *pbs_error = NULL;
++ ProxmoxRestoreHandle *conn = proxmox_restore_connect(
++ repository, snapshot, password, keyfile, key_password, fingerprint, &pbs_error);
++ if (conn == NULL) {
++ fprintf(stderr, "restore failed: %s\n", pbs_error);
++ return -1;
++ }
++
++ QDict *options = qdict_new();
++ qdict_put_str(options, "driver", format);
++
++ if (format) {
++ qdict_put_str(options, "driver", format);
++ }
++
++ Error *local_err = NULL;
++ int flags = BDRV_O_RDWR;
++
++ BlockBackend *blk = blk_new_open(target, NULL, options, flags, &local_err);
++ if (!blk) {
++ fprintf(stderr, "%s\n", error_get_pretty(local_err));
++ return -1;
++ }
++
++ CallbackData *callback_data = calloc(sizeof( CallbackData), 1);
++
++ callback_data->target = blk;
++ callback_data->skip_zero = skip_zero;
++ callback_data->last_offset = 0;
++
++ // blk_set_enable_write_cache(blk, !writethrough);
++
++ int res = proxmox_restore_image(
++ conn,
++ archive_name,
++ write_callback,
++ callback_data,
++ &pbs_error,
++ verbose);
++
++ proxmox_restore_disconnect(conn);
++
++ if (res < 0) {
++ fprintf(stderr, "restore failed: %s\n", pbs_error);
++ return -1;
++ }
++
++ return 0;
++}
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Dietmar Maurer <dietmar@proxmox.com>
-Date: Tue, 22 Oct 2019 12:48:17 +0200
-Subject: [PATCH] qmp_backup: run backup related code inside coroutines
-
-So that we can safely use coroutines/yield everywhere.
-
-Signed-off-by: Dietmar Maurer <dietmar@proxmox.com>
----
- blockdev.c | 250 ++++++++++++++++++++++++++++++++++++++---------------
- 1 file changed, 180 insertions(+), 70 deletions(-)
-
-diff --git a/blockdev.c b/blockdev.c
-index cee7952bbb..cec0f770e8 100644
---- a/blockdev.c
-+++ b/blockdev.c
-@@ -3173,6 +3173,34 @@ out:
-
- /* PVE backup related function */
-
-+typedef struct BlockOnCoroutineWrapper {
-+ AioContext *ctx;
-+ CoroutineEntry *entry;
-+ void *entry_arg;
-+ bool finished;
-+} BlockOnCoroutineWrapper;
-+
-+static void coroutine_fn block_on_coroutine_wrapper(void *opaque)
-+{
-+ BlockOnCoroutineWrapper *wrapper = opaque;
-+ wrapper->entry(wrapper->entry_arg);
-+ wrapper->finished = true;
-+}
-+
-+static void block_on_coroutine_fn(CoroutineEntry *entry, void *entry_arg)
-+{
-+ AioContext *ctx = qemu_get_current_aio_context();
-+ BlockOnCoroutineWrapper wrapper = {
-+ .finished = false,
-+ .entry = entry,
-+ .entry_arg = entry_arg,
-+ .ctx = ctx,
-+ };
-+ Coroutine *wrapper_co = qemu_coroutine_create(block_on_coroutine_wrapper, &wrapper);
-+ aio_co_enter(ctx, wrapper_co);
-+ AIO_WAIT_WHILE(ctx, !wrapper.finished);
-+}
-+
- static struct PVEBackupState {
- Error *error;
- bool cancel;
-@@ -3200,12 +3228,14 @@ typedef struct PVEBackupDevInfo {
- BlockDriverState *target;
- } PVEBackupDevInfo;
-
--static void pvebackup_run_next_job(void);
-+static void pvebackup_co_run_next_job(void);
-
--static int pvebackup_dump_cb(void *opaque, BlockBackend *target,
-+static int coroutine_fn pvebackup_co_dump_cb(void *opaque, BlockBackend *target,
- uint64_t start, uint64_t bytes,
- const void *pbuf)
- {
-+ assert(qemu_in_coroutine());
-+
- const uint64_t size = bytes;
- const unsigned char *buf = pbuf;
- PVEBackupDevInfo *di = opaque;
-@@ -3267,8 +3297,10 @@ static int pvebackup_dump_cb(void *opaque, BlockBackend *target,
- return size;
- }
-
--static void pvebackup_cleanup(void)
-+static void coroutine_fn pvebackup_co_cleanup(void)
- {
-+ assert(qemu_in_coroutine());
-+
- qemu_mutex_lock(&backup_state.backup_mutex);
- // Avoid race between block jobs and backup-cancel command:
- if (!backup_state.vmaw) {
-@@ -3290,18 +3322,19 @@ static void pvebackup_cleanup(void)
- qemu_mutex_unlock(&backup_state.backup_mutex);
- }
-
--static void coroutine_fn backup_close_vma_stream(void *opaque)
--{
-- PVEBackupDevInfo *di = opaque;
-+typedef struct PVEBackupCompeteCallbackData {
-+ PVEBackupDevInfo *di;
-+ int result;
-+} PVEBackupCompeteCallbackData;
-
-- vma_writer_close_stream(backup_state.vmaw, di->dev_id);
--}
--
--static void pvebackup_complete_cb(void *opaque, int ret)
-+static void coroutine_fn pvebackup_co_complete_cb(void *opaque)
- {
-- // This always runs in the main loop
-+ assert(qemu_in_coroutine());
-
-- PVEBackupDevInfo *di = opaque;
-+ PVEBackupCompeteCallbackData *cb_data = opaque;
-+
-+ PVEBackupDevInfo *di = cb_data->di;
-+ int ret = cb_data->result;
-
- di->completed = true;
-
-@@ -3314,8 +3347,7 @@ static void pvebackup_complete_cb(void *opaque, int ret)
- di->target = NULL;
-
- if (backup_state.vmaw) {
-- Coroutine *co = qemu_coroutine_create(backup_close_vma_stream, di);
-- qemu_coroutine_enter(co);
-+ vma_writer_close_stream(backup_state.vmaw, di->dev_id);
- }
-
- // remove self from job queue
-@@ -3325,12 +3357,25 @@ static void pvebackup_complete_cb(void *opaque, int ret)
- qemu_mutex_unlock(&backup_state.backup_mutex);
-
- if (!backup_state.cancel) {
-- pvebackup_run_next_job();
-+ pvebackup_co_run_next_job();
- }
- }
-
--static void pvebackup_cancel(void *opaque)
-+static void pvebackup_complete_cb(void *opaque, int ret)
- {
-+ // This always called from the main loop
-+ PVEBackupCompeteCallbackData cb_data = {
-+ .di = opaque,
-+ .result = ret,
-+ };
-+
-+ block_on_coroutine_fn(pvebackup_co_complete_cb, &cb_data);
-+}
-+
-+static void coroutine_fn pvebackup_co_cancel(void *opaque)
-+{
-+ assert(qemu_in_coroutine());
-+
- backup_state.cancel = true;
- qemu_mutex_lock(&backup_state.backup_mutex);
- // Avoid race between block jobs and backup-cancel command:
-@@ -3365,21 +3410,16 @@ static void pvebackup_cancel(void *opaque)
- }
- }
-
-- qemu_mutex_unlock(&backup_state.backup_mutex);
-- pvebackup_cleanup();
-+ qemu_co_mutex_unlock(&backup_state.backup_mutex);
-+ pvebackup_co_cleanup();
- }
-
- void qmp_backup_cancel(Error **errp)
- {
- if (!backup_state.backup_mutex_initialized)
- return;
-- Coroutine *co = qemu_coroutine_create(pvebackup_cancel, NULL);
-- qemu_coroutine_enter(co);
-
-- while (backup_state.vmaw) {
-- /* vma writer use main aio context */
-- aio_poll(qemu_get_aio_context(), true);
-- }
-+ block_on_coroutine_fn(pvebackup_co_cancel, NULL);
- }
-
- static int config_to_vma(const char *file, BackupFormat format,
-@@ -3420,8 +3460,11 @@ static int config_to_vma(const char *file, BackupFormat format,
- }
-
- bool job_should_pause(Job *job);
--static void pvebackup_run_next_job(void)
-+
-+static void coroutine_fn pvebackup_co_run_next_job(void)
- {
-+ assert(qemu_in_coroutine());
-+
- qemu_mutex_lock(&backup_state.backup_mutex);
-
- GList *l = backup_state.di_list;
-@@ -3447,16 +3490,33 @@ static void pvebackup_run_next_job(void)
- qemu_mutex_unlock(&backup_state.backup_mutex);
-
- // no more jobs, run the cleanup
-- pvebackup_cleanup();
-+ pvebackup_co_cleanup();
- }
-
--UuidInfo *qmp_backup(const char *backup_file, bool has_format,
-- BackupFormat format,
-- bool has_config_file, const char *config_file,
-- bool has_firewall_file, const char *firewall_file,
-- bool has_devlist, const char *devlist,
-- bool has_speed, int64_t speed, Error **errp)
-+typedef struct QmpBackupTask {
-+ const char *backup_file;
-+ bool has_format;
-+ BackupFormat format;
-+ bool has_config_file;
-+ const char *config_file;
-+ bool has_firewall_file;
-+ const char *firewall_file;
-+ bool has_devlist;
-+ const char *devlist;
-+ bool has_speed;
-+ int64_t speed;
-+ Error **errp;
-+ UuidInfo *result;
-+} QmpBackupTask;
-+
-+static void coroutine_fn pvebackup_co_start(void *opaque)
- {
-+ assert(qemu_in_coroutine());
-+
-+ QmpBackupTask *task = opaque;
-+
-+ task->result = NULL; // just to be sure
-+
- BlockBackend *blk;
- BlockDriverState *bs = NULL;
- const char *backup_dir = NULL;
-@@ -3475,16 +3535,16 @@ UuidInfo *qmp_backup(const char *backup_file, bool has_format,
- }
-
- if (backup_state.di_list) {
-- error_set(errp, ERROR_CLASS_GENERIC_ERROR,
-+ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
- "previous backup not finished");
-- return NULL;
-+ return;
- }
-
- /* Todo: try to auto-detect format based on file name */
-- format = has_format ? format : BACKUP_FORMAT_VMA;
-+ BackupFormat format = task->has_format ? task->format : BACKUP_FORMAT_VMA;
-
-- if (has_devlist) {
-- devs = g_strsplit_set(devlist, ",;:", -1);
-+ if (task->has_devlist) {
-+ devs = g_strsplit_set(task->devlist, ",;:", -1);
-
- gchar **d = devs;
- while (d && *d) {
-@@ -3492,18 +3552,18 @@ UuidInfo *qmp_backup(const char *backup_file, bool has_format,
- if (blk) {
- bs = blk_bs(blk);
- if (bdrv_is_read_only(bs)) {
-- error_setg(errp, "Node '%s' is read only", *d);
-+ error_setg(task->errp, "Node '%s' is read only", *d);
- goto err;
- }
- if (!bdrv_is_inserted(bs)) {
-- error_setg(errp, QERR_DEVICE_HAS_NO_MEDIUM, *d);
-+ error_setg(task->errp, QERR_DEVICE_HAS_NO_MEDIUM, *d);
- goto err;
- }
- PVEBackupDevInfo *di = g_new0(PVEBackupDevInfo, 1);
- di->bs = bs;
- di_list = g_list_append(di_list, di);
- } else {
-- error_set(errp, ERROR_CLASS_DEVICE_NOT_FOUND,
-+ error_set(task->errp, ERROR_CLASS_DEVICE_NOT_FOUND,
- "Device '%s' not found", *d);
- goto err;
- }
-@@ -3526,7 +3586,7 @@ UuidInfo *qmp_backup(const char *backup_file, bool has_format,
- }
-
- if (!di_list) {
-- error_set(errp, ERROR_CLASS_GENERIC_ERROR, "empty device list");
-+ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "empty device list");
- goto err;
- }
-
-@@ -3536,13 +3596,13 @@ UuidInfo *qmp_backup(const char *backup_file, bool has_format,
- while (l) {
- PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
- l = g_list_next(l);
-- if (bdrv_op_is_blocked(di->bs, BLOCK_OP_TYPE_BACKUP_SOURCE, errp)) {
-+ if (bdrv_op_is_blocked(di->bs, BLOCK_OP_TYPE_BACKUP_SOURCE, task->errp)) {
- goto err;
- }
-
- ssize_t size = bdrv_getlength(di->bs);
- if (size < 0) {
-- error_setg_errno(errp, -di->size, "bdrv_getlength failed");
-+ error_setg_errno(task->errp, -di->size, "bdrv_getlength failed");
- goto err;
- }
- di->size = size;
-@@ -3552,10 +3612,10 @@ UuidInfo *qmp_backup(const char *backup_file, bool has_format,
- uuid_generate(uuid);
-
- if (format == BACKUP_FORMAT_VMA) {
-- vmaw = vma_writer_create(backup_file, uuid, &local_err);
-+ vmaw = vma_writer_create(task->backup_file, uuid, &local_err);
- if (!vmaw) {
- if (local_err) {
-- error_propagate(errp, local_err);
-+ error_propagate(task->errp, local_err);
- }
- goto err;
- }
-@@ -3569,18 +3629,18 @@ UuidInfo *qmp_backup(const char *backup_file, bool has_format,
- const char *devname = bdrv_get_device_name(di->bs);
- di->dev_id = vma_writer_register_stream(vmaw, devname, di->size);
- if (di->dev_id <= 0) {
-- error_set(errp, ERROR_CLASS_GENERIC_ERROR,
-+ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
- "register_stream failed");
- goto err;
- }
- }
- } else if (format == BACKUP_FORMAT_DIR) {
-- if (mkdir(backup_file, 0640) != 0) {
-- error_setg_errno(errp, errno, "can't create directory '%s'\n",
-- backup_file);
-+ if (mkdir(task->backup_file, 0640) != 0) {
-+ error_setg_errno(task->errp, errno, "can't create directory '%s'\n",
-+ task->backup_file);
- goto err;
- }
-- backup_dir = backup_file;
-+ backup_dir = task->backup_file;
-
- l = di_list;
- while (l) {
-@@ -3594,31 +3654,31 @@ UuidInfo *qmp_backup(const char *backup_file, bool has_format,
- bdrv_img_create(di->targetfile, "raw", NULL, NULL, NULL,
- di->size, flags, false, &local_err);
- if (local_err) {
-- error_propagate(errp, local_err);
-+ error_propagate(task->errp, local_err);
- goto err;
- }
-
- di->target = bdrv_open(di->targetfile, NULL, NULL, flags, &local_err);
- if (!di->target) {
-- error_propagate(errp, local_err);
-+ error_propagate(task->errp, local_err);
- goto err;
- }
- }
- } else {
-- error_set(errp, ERROR_CLASS_GENERIC_ERROR, "unknown backup format");
-+ error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "unknown backup format");
- goto err;
- }
-
- /* add configuration file to archive */
-- if (has_config_file) {
-- if (config_to_vma(config_file, format, backup_dir, vmaw, errp) != 0) {
-+ if (task->has_config_file) {
-+ if (config_to_vma(task->config_file, format, backup_dir, vmaw, task->errp) != 0) {
- goto err;
- }
- }
-
- /* add firewall file to archive */
-- if (has_firewall_file) {
-- if (config_to_vma(firewall_file, format, backup_dir, vmaw, errp) != 0) {
-+ if (task->has_firewall_file) {
-+ if (config_to_vma(task->firewall_file, format, backup_dir, vmaw, task->errp) != 0) {
- goto err;
- }
- }
-@@ -3631,7 +3691,7 @@ UuidInfo *qmp_backup(const char *backup_file, bool has_format,
- backup_state.error = NULL;
- }
-
-- backup_state.speed = (has_speed && speed > 0) ? speed : 0;
-+ backup_state.speed = (task->has_speed && task->speed > 0) ? task->speed : 0;
-
- backup_state.start_time = time(NULL);
- backup_state.end_time = 0;
-@@ -3639,7 +3699,7 @@ UuidInfo *qmp_backup(const char *backup_file, bool has_format,
- if (backup_state.backup_file) {
- g_free(backup_state.backup_file);
- }
-- backup_state.backup_file = g_strdup(backup_file);
-+ backup_state.backup_file = g_strdup(task->backup_file);
-
- backup_state.vmaw = vmaw;
-
-@@ -3658,14 +3718,13 @@ UuidInfo *qmp_backup(const char *backup_file, bool has_format,
- while (l) {
- PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
- l = g_list_next(l);
-- job = backup_job_create(NULL, di->bs, di->target, speed, MIRROR_SYNC_MODE_FULL, NULL,
-+ job = backup_job_create(NULL, di->bs, di->target, backup_state.speed, MIRROR_SYNC_MODE_FULL, NULL,
- false, BLOCKDEV_ON_ERROR_REPORT, BLOCKDEV_ON_ERROR_REPORT,
-- JOB_DEFAULT,
-- pvebackup_dump_cb, pvebackup_complete_cb, di,
-+ JOB_DEFAULT, pvebackup_co_dump_cb, pvebackup_complete_cb, di,
- 1, NULL, &local_err);
- if (!job || local_err != NULL) {
- error_setg(&backup_state.error, "backup_job_create failed");
-- pvebackup_cancel(NULL);
-+ pvebackup_co_cancel(NULL);
- } else {
- job_start(&job->job);
- }
-@@ -3678,13 +3737,14 @@ UuidInfo *qmp_backup(const char *backup_file, bool has_format,
- qemu_mutex_unlock(&backup_state.backup_mutex);
-
- if (!backup_state.error) {
-- pvebackup_run_next_job(); // run one job
-+ pvebackup_co_run_next_job(); // run one job
- }
-
- uuid_info = g_malloc0(sizeof(*uuid_info));
- uuid_info->UUID = g_strdup(backup_state.uuid_str);
-
-- return uuid_info;
-+ task->result = uuid_info;
-+ return;
-
- err:
-
-@@ -3711,23 +3771,61 @@ err:
- if (vmaw) {
- Error *err = NULL;
- vma_writer_close(vmaw, &err);
-- unlink(backup_file);
-+ unlink(task->backup_file);
- }
-
- if (backup_dir) {
- rmdir(backup_dir);
- }
-
-- return NULL;
-+ task->result = NULL;
-+ return;
- }
-
--BackupStatus *qmp_query_backup(Error **errp)
-+UuidInfo *qmp_backup(const char *backup_file, bool has_format,
-+ BackupFormat format,
-+ bool has_config_file, const char *config_file,
-+ bool has_firewall_file, const char *firewall_file,
-+ bool has_devlist, const char *devlist,
-+ bool has_speed, int64_t speed, Error **errp)
-+{
-+ QmpBackupTask task = {
-+ .backup_file = backup_file,
-+ .has_format = has_format,
-+ .format = format,
-+ .has_config_file = has_config_file,
-+ .config_file = config_file,
-+ .has_firewall_file = has_firewall_file,
-+ .firewall_file = firewall_file,
-+ .has_devlist = has_devlist,
-+ .devlist = devlist,
-+ .has_speed = has_speed,
-+ .speed = speed,
-+ .errp = errp,
-+ };
-+
-+ block_on_coroutine_fn(pvebackup_co_start, &task);
-+ return task.result;
-+}
-+
-+
-+typedef struct QmpQueryBackupTask {
-+ Error **errp;
-+ BackupStatus *result;
-+} QmpQueryBackupTask;
-+
-+static void coroutine_fn pvebackup_co_query(void *opaque)
- {
-+ assert(qemu_in_coroutine());
-+
-+ QmpQueryBackupTask *task = opaque;
-+
- BackupStatus *info = g_malloc0(sizeof(*info));
-
- if (!backup_state.start_time) {
- /* not started, return {} */
-- return info;
-+ task->result = info;
-+ return;
- }
-
- info->has_status = true;
-@@ -3763,7 +3861,19 @@ BackupStatus *qmp_query_backup(Error **errp)
- info->has_transferred = true;
- info->transferred = backup_state.transferred;
-
-- return info;
-+ task->result = info;
-+}
-+
-+BackupStatus *qmp_query_backup(Error **errp)
-+{
-+ QmpQueryBackupTask task = {
-+ .errp = errp,
-+ .result = NULL,
-+ };
-+
-+ block_on_coroutine_fn(pvebackup_co_query, &task);
-+
-+ return task.result;
- }
-
- void qmp_block_stream(bool has_job_id, const char *job_id, const char *device,
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Dietmar Maurer <dietmar@proxmox.com>
-Date: Tue, 22 Oct 2019 12:48:18 +0200
-Subject: [PATCH] qmp_backup: use a CoMutex to protect access to backup_state
-
-And use job_cancel instead of job_cancel_sync (which hangs sometimes)
-Also avoid calling pvebackup_co_cleanup if not necessary. If there are
-running jobs, this is called from the block job completion callbacks
-anyways.
-
-Signed-off-by: Dietmar Maurer <dietmar@proxmox.com>
----
- blockdev.c | 74 ++++++++++++++++++++++++++++++++++++++----------------
- 1 file changed, 52 insertions(+), 22 deletions(-)
-
-diff --git a/blockdev.c b/blockdev.c
-index cec0f770e8..29196c18d8 100644
---- a/blockdev.c
-+++ b/blockdev.c
-@@ -3215,7 +3215,7 @@ static struct PVEBackupState {
- size_t total;
- size_t transferred;
- size_t zero_bytes;
-- QemuMutex backup_mutex;
-+ CoMutex backup_mutex;
- bool backup_mutex_initialized;
- } backup_state;
-
-@@ -3240,7 +3240,10 @@ static int coroutine_fn pvebackup_co_dump_cb(void *opaque, BlockBackend *target,
- const unsigned char *buf = pbuf;
- PVEBackupDevInfo *di = opaque;
-
-+ qemu_co_mutex_lock(&backup_state.backup_mutex);
-+
- if (backup_state.cancel) {
-+ qemu_co_mutex_unlock(&backup_state.backup_mutex);
- return size; // return success
- }
-
-@@ -3251,6 +3254,7 @@ static int coroutine_fn pvebackup_co_dump_cb(void *opaque, BlockBackend *target,
- "got unaligned write inside backup dump "
- "callback (sector %ld)", start);
- }
-+ qemu_co_mutex_unlock(&backup_state.backup_mutex);
- return -1; // not aligned to cluster size
- }
-
-@@ -3292,6 +3296,8 @@ static int coroutine_fn pvebackup_co_dump_cb(void *opaque, BlockBackend *target,
- backup_state.transferred += size;
- }
-
-+ qemu_co_mutex_unlock(&backup_state.backup_mutex);
-+
- // Note: always return success, because we want that writes succeed anyways.
-
- return size;
-@@ -3301,10 +3307,9 @@ static void coroutine_fn pvebackup_co_cleanup(void)
- {
- assert(qemu_in_coroutine());
-
-- qemu_mutex_lock(&backup_state.backup_mutex);
-- // Avoid race between block jobs and backup-cancel command:
-+ qemu_co_mutex_lock(&backup_state.backup_mutex);
- if (!backup_state.vmaw) {
-- qemu_mutex_unlock(&backup_state.backup_mutex);
-+ qemu_co_mutex_unlock(&backup_state.backup_mutex);
- return;
- }
-
-@@ -3319,7 +3324,7 @@ static void coroutine_fn pvebackup_co_cleanup(void)
-
- g_list_free(backup_state.di_list);
- backup_state.di_list = NULL;
-- qemu_mutex_unlock(&backup_state.backup_mutex);
-+ qemu_co_mutex_unlock(&backup_state.backup_mutex);
- }
-
- typedef struct PVEBackupCompeteCallbackData {
-@@ -3333,6 +3338,8 @@ static void coroutine_fn pvebackup_co_complete_cb(void *opaque)
-
- PVEBackupCompeteCallbackData *cb_data = opaque;
-
-+ qemu_co_mutex_lock(&backup_state.backup_mutex);
-+
- PVEBackupDevInfo *di = cb_data->di;
- int ret = cb_data->result;
-
-@@ -3351,12 +3358,14 @@ static void coroutine_fn pvebackup_co_complete_cb(void *opaque)
- }
-
- // remove self from job queue
-- qemu_mutex_lock(&backup_state.backup_mutex);
- backup_state.di_list = g_list_remove(backup_state.di_list, di);
- g_free(di);
-- qemu_mutex_unlock(&backup_state.backup_mutex);
-
-- if (!backup_state.cancel) {
-+ bool cancel = backup_state.cancel;
-+
-+ qemu_co_mutex_unlock(&backup_state.backup_mutex);
-+
-+ if (!cancel) {
- pvebackup_co_run_next_job();
- }
- }
-@@ -3376,11 +3385,13 @@ static void coroutine_fn pvebackup_co_cancel(void *opaque)
- {
- assert(qemu_in_coroutine());
-
-+ qemu_co_mutex_lock(&backup_state.backup_mutex);
-+
- backup_state.cancel = true;
-- qemu_mutex_lock(&backup_state.backup_mutex);
-+
- // Avoid race between block jobs and backup-cancel command:
- if (!backup_state.vmaw) {
-- qemu_mutex_unlock(&backup_state.backup_mutex);
-+ qemu_co_mutex_unlock(&backup_state.backup_mutex);
- return;
- }
-
-@@ -3393,6 +3404,7 @@ static void coroutine_fn pvebackup_co_cancel(void *opaque)
- vma_writer_set_error(backup_state.vmaw, "backup cancelled");
- }
-
-+ bool running_jobs = 0;
- GList *l = backup_state.di_list;
- while (l) {
- PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
-@@ -3403,6 +3415,7 @@ static void coroutine_fn pvebackup_co_cancel(void *opaque)
- AioContext *aio_context = blk_get_aio_context(job->blk);
- aio_context_acquire(aio_context);
- if (!di->completed) {
-+ running_jobs += 1;
- job_cancel(&job->job, false);
- }
- aio_context_release(aio_context);
-@@ -3411,7 +3424,8 @@ static void coroutine_fn pvebackup_co_cancel(void *opaque)
- }
-
- qemu_co_mutex_unlock(&backup_state.backup_mutex);
-- pvebackup_co_cleanup();
-+
-+ if (running_jobs == 0) pvebackup_co_cleanup(); // else job will call completion handler
- }
-
- void qmp_backup_cancel(Error **errp)
-@@ -3465,7 +3479,7 @@ static void coroutine_fn pvebackup_co_run_next_job(void)
- {
- assert(qemu_in_coroutine());
-
-- qemu_mutex_lock(&backup_state.backup_mutex);
-+ qemu_co_mutex_lock(&backup_state.backup_mutex);
-
- GList *l = backup_state.di_list;
- while (l) {
-@@ -3474,11 +3488,13 @@ static void coroutine_fn pvebackup_co_run_next_job(void)
- if (!di->completed && di->bs && di->bs->job) {
- BlockJob *job = di->bs->job;
- AioContext *aio_context = blk_get_aio_context(job->blk);
-+ bool cancel_job = backup_state.error || backup_state.cancel;
-+ qemu_co_mutex_unlock(&backup_state.backup_mutex);
-+
- aio_context_acquire(aio_context);
-- qemu_mutex_unlock(&backup_state.backup_mutex);
- if (job_should_pause(&job->job)) {
-- if (backup_state.error || backup_state.cancel) {
-- job_cancel_sync(&job->job);
-+ if (cancel_job) {
-+ job_cancel(&job->job, false);
- } else {
- job_resume(&job->job);
- }
-@@ -3487,7 +3503,7 @@ static void coroutine_fn pvebackup_co_run_next_job(void)
- return;
- }
- }
-- qemu_mutex_unlock(&backup_state.backup_mutex);
-+ qemu_co_mutex_unlock(&backup_state.backup_mutex);
-
- // no more jobs, run the cleanup
- pvebackup_co_cleanup();
-@@ -3530,11 +3546,14 @@ static void coroutine_fn pvebackup_co_start(void *opaque)
- BlockJob *job;
-
- if (!backup_state.backup_mutex_initialized) {
-- qemu_mutex_init(&backup_state.backup_mutex);
-+ qemu_co_mutex_init(&backup_state.backup_mutex);
- backup_state.backup_mutex_initialized = true;
- }
-
-+ qemu_co_mutex_lock(&backup_state.backup_mutex);
-+
- if (backup_state.di_list) {
-+ qemu_co_mutex_unlock(&backup_state.backup_mutex);
- error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
- "previous backup not finished");
- return;
-@@ -3706,7 +3725,6 @@ static void coroutine_fn pvebackup_co_start(void *opaque)
- uuid_copy(backup_state.uuid, uuid);
- uuid_unparse_lower(uuid, backup_state.uuid_str);
-
-- qemu_mutex_lock(&backup_state.backup_mutex);
- backup_state.di_list = di_list;
-
- backup_state.total = total;
-@@ -3724,20 +3742,21 @@ static void coroutine_fn pvebackup_co_start(void *opaque)
- 1, NULL, &local_err);
- if (!job || local_err != NULL) {
- error_setg(&backup_state.error, "backup_job_create failed");
-- pvebackup_co_cancel(NULL);
-- } else {
-- job_start(&job->job);
-+ break;
- }
-+ job_start(&job->job);
- if (di->target) {
- bdrv_unref(di->target);
- di->target = NULL;
- }
- }
-
-- qemu_mutex_unlock(&backup_state.backup_mutex);
-+ qemu_co_mutex_unlock(&backup_state.backup_mutex);
-
- if (!backup_state.error) {
- pvebackup_co_run_next_job(); // run one job
-+ } else {
-+ pvebackup_co_cancel(NULL);
- }
-
- uuid_info = g_malloc0(sizeof(*uuid_info));
-@@ -3778,6 +3797,8 @@ err:
- rmdir(backup_dir);
- }
-
-+ qemu_co_mutex_unlock(&backup_state.backup_mutex);
-+
- task->result = NULL;
- return;
- }
-@@ -3805,6 +3826,7 @@ UuidInfo *qmp_backup(const char *backup_file, bool has_format,
- };
-
- block_on_coroutine_fn(pvebackup_co_start, &task);
-+
- return task.result;
- }
-
-@@ -3822,9 +3844,15 @@ static void coroutine_fn pvebackup_co_query(void *opaque)
-
- BackupStatus *info = g_malloc0(sizeof(*info));
-
-+ if (!backup_state.backup_mutex_initialized)
-+ return;
-+
-+ qemu_co_mutex_lock(&backup_state.backup_mutex);
-+
- if (!backup_state.start_time) {
- /* not started, return {} */
- task->result = info;
-+ qemu_co_mutex_unlock(&backup_state.backup_mutex);
- return;
- }
-
-@@ -3862,6 +3890,8 @@ static void coroutine_fn pvebackup_co_query(void *opaque)
- info->transferred = backup_state.transferred;
-
- task->result = info;
-+
-+ qemu_co_mutex_unlock(&backup_state.backup_mutex);
- }
-
- BackupStatus *qmp_query_backup(Error **errp)
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Dietmar Maurer <dietmar@proxmox.com>
-Date: Tue, 22 Oct 2019 12:48:19 +0200
-Subject: [PATCH] vma_writer_close: avoid call to aio_poll (acquire flush_lock
- instead)
-
-Signed-off-by: Dietmar Maurer <dietmar@proxmox.com>
----
- vma-writer.c | 6 +++---
- 1 file changed, 3 insertions(+), 3 deletions(-)
-
-diff --git a/vma-writer.c b/vma-writer.c
-index b163fa2d3a..fe86b18a60 100644
---- a/vma-writer.c
-+++ b/vma-writer.c
-@@ -705,9 +705,7 @@ int vma_writer_close(VmaWriter *vmaw, Error **errp)
-
- int i;
-
-- while (vmaw->co_writer) {
-- aio_poll(qemu_get_aio_context(), true);
-- }
-+ qemu_co_mutex_lock(&vmaw->flush_lock); // wait for pending writes
-
- assert(vmaw->co_writer == NULL);
-
-@@ -748,6 +746,8 @@ int vma_writer_close(VmaWriter *vmaw, Error **errp)
- error_setg(errp, "%s", vmaw->errmsg);
- }
-
-+ qemu_co_mutex_unlock(&vmaw->flush_lock);
-+
- return vmaw->status;
- }
-
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Dietmar Maurer <dietmar@proxmox.com>
-Date: Thu, 24 Oct 2019 08:06:50 +0200
-Subject: [PATCH] backup_job_create: pass cluster size for dump
-
-Signed-off-by: Dietmar Maurer <dietmar@proxmox.com>
----
- block/backup.c | 8 +++++++-
- block/replication.c | 2 +-
- blockdev.c | 10 ++++++----
- include/block/block_int.h | 4 ++++
- 4 files changed, 18 insertions(+), 6 deletions(-)
-
-diff --git a/block/backup.c b/block/backup.c
-index fdcfb0529c..2398a4d602 100644
---- a/block/backup.c
-+++ b/block/backup.c
-@@ -566,6 +566,7 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
- BlockdevOnError on_target_error,
- int creation_flags,
- BackupDumpFunc *dump_cb,
-+ int dump_cb_block_size,
- BlockCompletionFunc *cb, void *opaque,
- int pause_count,
- JobTxn *txn, Error **errp)
-@@ -636,7 +637,12 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
- goto error;
- }
-
-- cluster_size = backup_calculate_cluster_size(target ? target : bs, errp);
-+ if (target) {
-+ cluster_size = backup_calculate_cluster_size(target, errp);
-+ } else {
-+ cluster_size = dump_cb_block_size;
-+ }
-+
- if (cluster_size < 0) {
- goto error;
- }
-diff --git a/block/replication.c b/block/replication.c
-index f060755713..b9465c3587 100644
---- a/block/replication.c
-+++ b/block/replication.c
-@@ -546,7 +546,7 @@ static void replication_start(ReplicationState *rs, ReplicationMode mode,
- 0, MIRROR_SYNC_MODE_NONE, NULL, false,
- BLOCKDEV_ON_ERROR_REPORT,
- BLOCKDEV_ON_ERROR_REPORT, JOB_INTERNAL,
-- NULL,
-+ NULL, 0,
- backup_job_completed, bs, 0, NULL, &local_err);
- if (local_err) {
- error_propagate(errp, local_err);
-diff --git a/blockdev.c b/blockdev.c
-index 29196c18d8..a95beb823e 100644
---- a/blockdev.c
-+++ b/blockdev.c
-@@ -3544,6 +3544,7 @@ static void coroutine_fn pvebackup_co_start(void *opaque)
- GList *l;
- UuidInfo *uuid_info;
- BlockJob *job;
-+ int dump_cb_block_size = -1;
-
- if (!backup_state.backup_mutex_initialized) {
- qemu_co_mutex_init(&backup_state.backup_mutex);
-@@ -3631,6 +3632,7 @@ static void coroutine_fn pvebackup_co_start(void *opaque)
- uuid_generate(uuid);
-
- if (format == BACKUP_FORMAT_VMA) {
-+ dump_cb_block_size = VMA_CLUSTER_SIZE;
- vmaw = vma_writer_create(task->backup_file, uuid, &local_err);
- if (!vmaw) {
- if (local_err) {
-@@ -3738,8 +3740,8 @@ static void coroutine_fn pvebackup_co_start(void *opaque)
- l = g_list_next(l);
- job = backup_job_create(NULL, di->bs, di->target, backup_state.speed, MIRROR_SYNC_MODE_FULL, NULL,
- false, BLOCKDEV_ON_ERROR_REPORT, BLOCKDEV_ON_ERROR_REPORT,
-- JOB_DEFAULT, pvebackup_co_dump_cb, pvebackup_complete_cb, di,
-- 1, NULL, &local_err);
-+ JOB_DEFAULT, pvebackup_co_dump_cb, dump_cb_block_size,
-+ pvebackup_complete_cb, di, 1, NULL, &local_err);
- if (!job || local_err != NULL) {
- error_setg(&backup_state.error, "backup_job_create failed");
- break;
-@@ -4312,7 +4314,7 @@ static BlockJob *do_drive_backup(DriveBackup *backup, JobTxn *txn,
- job = backup_job_create(backup->job_id, bs, target_bs, backup->speed,
- backup->sync, bmap, backup->compress,
- backup->on_source_error, backup->on_target_error,
-- job_flags, NULL, NULL, NULL, 0, txn, &local_err);
-+ job_flags, NULL, 0, NULL, NULL, 0, txn, &local_err);
- if (local_err != NULL) {
- error_propagate(errp, local_err);
- goto unref;
-@@ -4417,7 +4419,7 @@ BlockJob *do_blockdev_backup(BlockdevBackup *backup, JobTxn *txn,
- job = backup_job_create(backup->job_id, bs, target_bs, backup->speed,
- backup->sync, bmap, backup->compress,
- backup->on_source_error, backup->on_target_error,
-- job_flags, NULL, NULL, NULL, 0, txn, &local_err);
-+ job_flags, NULL, 0, NULL, NULL, 0, txn, &local_err);
- if (local_err != NULL) {
- error_propagate(errp, local_err);
- }
-diff --git a/include/block/block_int.h b/include/block/block_int.h
-index 5a8b2e06c1..e7ca823058 100644
---- a/include/block/block_int.h
-+++ b/include/block/block_int.h
-@@ -1160,6 +1160,9 @@ void mirror_start(const char *job_id, BlockDriverState *bs,
- * @on_target_error: The action to take upon error writing to the target.
- * @creation_flags: Flags that control the behavior of the Job lifetime.
- * See @BlockJobCreateFlags
-+ * @dump_cb: Callback for PVE backup code. Called for each data block when
-+ * target is NULL.
-+ * @dump_cb_block_size: The minimum block size expected by dump_cb.
- * @cb: Completion function for the job.
- * @opaque: Opaque pointer value passed to @cb.
- * @txn: Transaction that this job is part of (may be NULL).
-@@ -1176,6 +1179,7 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
- BlockdevOnError on_target_error,
- int creation_flags,
- BackupDumpFunc *dump_cb,
-+ int dump_cb_block_size,
- BlockCompletionFunc *cb, void *opaque,
- int pause_count,
- JobTxn *txn, Error **errp);
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Dietmar Maurer <dietmar@proxmox.com>
-Date: Thu, 24 Oct 2019 08:06:51 +0200
-Subject: [PATCH] avoid calling dump_cb with NULL data pointer for small/last
- cluster
-
-The last block of a backup may be smaller than cluster_size.
-
-Signed-off-by: Dietmar Maurer <dietmar@proxmox.com>
----
- block/backup.c | 7 ++++++-
- 1 file changed, 6 insertions(+), 1 deletion(-)
-
-diff --git a/block/backup.c b/block/backup.c
-index 2398a4d602..c682c99f7a 100644
---- a/block/backup.c
-+++ b/block/backup.c
-@@ -131,7 +131,12 @@ static int coroutine_fn backup_cow_with_bounce_buffer(BackupBlockJob *job,
-
- if (buffer_is_zero(*bounce_buffer, nbytes)) {
- if (job->dump_cb) {
-- ret = job->dump_cb(job->common.job.opaque, job->target, start, nbytes, NULL);
-+ if (nbytes == job->cluster_size) {
-+ // Note: pass NULL to indicate that we want to write [0u8; cluster_size]
-+ ret = job->dump_cb(job->common.job.opaque, job->target, start, nbytes, NULL);
-+ } else {
-+ ret = job->dump_cb(job->common.job.opaque, job->target, start, nbytes, *bounce_buffer);
-+ }
- } else {
- ret = blk_co_pwrite_zeroes(job->target, start,
- nbytes, write_flags | BDRV_REQ_MAY_UNMAP);
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Dietmar Maurer <dietmar@proxmox.com>
-Date: Thu, 24 Oct 2019 08:06:52 +0200
-Subject: [PATCH] rename config_to_vma into pvebackup_co_add_config
-
-- mark it with coroutine_fn
-- add an additional parameter 'name'
-- return -1 on error (instead of 1)
-- code cleanup
-
-Signed-off-by: Dietmar Maurer <dietmar@proxmox.com>
----
- blockdev.c | 40 ++++++++++++++++++++++++++--------------
- 1 file changed, 26 insertions(+), 14 deletions(-)
-
-diff --git a/blockdev.c b/blockdev.c
-index a95beb823e..530b76c82f 100644
---- a/blockdev.c
-+++ b/blockdev.c
-@@ -3436,10 +3436,16 @@ void qmp_backup_cancel(Error **errp)
- block_on_coroutine_fn(pvebackup_co_cancel, NULL);
- }
-
--static int config_to_vma(const char *file, BackupFormat format,
-- const char *backup_dir, VmaWriter *vmaw,
-- Error **errp)
-+static int coroutine_fn pvebackup_co_add_config(
-+ const char *file,
-+ const char *name,
-+ BackupFormat format,
-+ const char *backup_dir,
-+ VmaWriter *vmaw,
-+ Error **errp)
- {
-+ int res = 0;
-+
- char *cdata = NULL;
- gsize clen = 0;
- GError *err = NULL;
-@@ -3449,28 +3455,30 @@ static int config_to_vma(const char *file, BackupFormat format,
- }
-
- char *basename = g_path_get_basename(file);
-+ if (name == NULL) name = basename;
-
- if (format == BACKUP_FORMAT_VMA) {
-- if (vma_writer_add_config(vmaw, basename, cdata, clen) != 0) {
-+ if (vma_writer_add_config(vmaw, name, cdata, clen) != 0) {
- error_setg(errp, "unable to add %s config data to vma archive", file);
-- g_free(cdata);
-- g_free(basename);
-- return 1;
-+ goto err;
- }
- } else if (format == BACKUP_FORMAT_DIR) {
- char config_path[PATH_MAX];
-- snprintf(config_path, PATH_MAX, "%s/%s", backup_dir, basename);
-+ snprintf(config_path, PATH_MAX, "%s/%s", backup_dir, name);
- if (!g_file_set_contents(config_path, cdata, clen, &err)) {
- error_setg(errp, "unable to write config file '%s'", config_path);
-- g_free(cdata);
-- g_free(basename);
-- return 1;
-+ goto err;
- }
- }
-
-+ out:
- g_free(basename);
- g_free(cdata);
-- return 0;
-+ return res;
-+
-+ err:
-+ res = -1;
-+ goto out;
- }
-
- bool job_should_pause(Job *job);
-@@ -3546,6 +3554,9 @@ static void coroutine_fn pvebackup_co_start(void *opaque)
- BlockJob *job;
- int dump_cb_block_size = -1;
-
-+ const char *config_name = "qemu-server.conf";
-+ const char *firewall_name = "qemu-server.fw";
-+
- if (!backup_state.backup_mutex_initialized) {
- qemu_co_mutex_init(&backup_state.backup_mutex);
- backup_state.backup_mutex_initialized = true;
-@@ -3690,16 +3701,17 @@ static void coroutine_fn pvebackup_co_start(void *opaque)
- goto err;
- }
-
-+
- /* add configuration file to archive */
- if (task->has_config_file) {
-- if (config_to_vma(task->config_file, format, backup_dir, vmaw, task->errp) != 0) {
-+ if (pvebackup_co_add_config(task->config_file, config_name, format, backup_dir, vmaw, task->errp) != 0) {
- goto err;
- }
- }
-
- /* add firewall file to archive */
- if (task->has_firewall_file) {
-- if (config_to_vma(task->firewall_file, format, backup_dir, vmaw, task->errp) != 0) {
-+ if (pvebackup_co_add_config(task->firewall_file, firewall_name, format, backup_dir, vmaw, task->errp) != 0) {
- goto err;
- }
- }
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Dietmar Maurer <dietmar@proxmox.com>
-Date: Wed, 30 Oct 2019 12:15:44 +0100
-Subject: [PATCH] pvebackup_co_dump_cb: do not call job->cancel()
-
-The backup loop will automatically abort if we return an error.
----
- blockdev.c | 6 ++----
- 1 file changed, 2 insertions(+), 4 deletions(-)
-
-diff --git a/blockdev.c b/blockdev.c
-index 530b76c82f..568f71fdb4 100644
---- a/blockdev.c
-+++ b/blockdev.c
-@@ -3274,10 +3274,8 @@ static int coroutine_fn pvebackup_co_dump_cb(void *opaque, BlockBackend *target,
- if (!backup_state.error) {
- vma_writer_error_propagate(backup_state.vmaw, &backup_state.error);
- }
-- if (di->bs && di->bs->job) {
-- job_cancel(&di->bs->job->job, true);
-- }
-- break;
-+ qemu_co_mutex_unlock(&backup_state.backup_mutex);
-+ return ret;
- } else {
- backup_state.zero_bytes += zero_bytes;
- if (remaining >= VMA_CLUSTER_SIZE) {
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Dietmar Maurer <dietmar@proxmox.com>
-Date: Wed, 30 Oct 2019 12:15:45 +0100
-Subject: [PATCH] fix backup job completion
-
-With recent changes, pvebackup_co_run_next_job cancels the job async,
-so we need to run pvebackup_co_cleanup in the completion handler
-instead. We call pvebackup_co_run_next as long as there are
-jobs in the list.
----
- blockdev.c | 9 ++++-----
- 1 file changed, 4 insertions(+), 5 deletions(-)
-
-diff --git a/blockdev.c b/blockdev.c
-index 568f71fdb4..66f2711185 100644
---- a/blockdev.c
-+++ b/blockdev.c
-@@ -3359,12 +3359,14 @@ static void coroutine_fn pvebackup_co_complete_cb(void *opaque)
- backup_state.di_list = g_list_remove(backup_state.di_list, di);
- g_free(di);
-
-- bool cancel = backup_state.cancel;
-+ int pending_jobs = g_list_length(backup_state.di_list);
-
- qemu_co_mutex_unlock(&backup_state.backup_mutex);
-
-- if (!cancel) {
-+ if (pending_jobs > 0) {
- pvebackup_co_run_next_job();
-+ } else {
-+ pvebackup_co_cleanup();
- }
- }
-
-@@ -3510,9 +3512,6 @@ static void coroutine_fn pvebackup_co_run_next_job(void)
- }
- }
- qemu_co_mutex_unlock(&backup_state.backup_mutex);
--
-- // no more jobs, run the cleanup
-- pvebackup_co_cleanup();
- }
-
- typedef struct QmpBackupTask {
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Dietmar Maurer <dietmar@proxmox.com>
-Date: Wed, 30 Oct 2019 12:15:46 +0100
-Subject: [PATCH] pvebackup_complete_cb: avoid poll loop if already inside
- coroutine
-
----
- blockdev.c | 10 ++++++++--
- 1 file changed, 8 insertions(+), 2 deletions(-)
-
-diff --git a/blockdev.c b/blockdev.c
-index 66f2711185..083ada6c8e 100644
---- a/blockdev.c
-+++ b/blockdev.c
-@@ -3189,6 +3189,8 @@ static void coroutine_fn block_on_coroutine_wrapper(void *opaque)
-
- static void block_on_coroutine_fn(CoroutineEntry *entry, void *entry_arg)
- {
-+ assert(!qemu_in_coroutine());
-+
- AioContext *ctx = qemu_get_current_aio_context();
- BlockOnCoroutineWrapper wrapper = {
- .finished = false,
-@@ -3372,13 +3374,17 @@ static void coroutine_fn pvebackup_co_complete_cb(void *opaque)
-
- static void pvebackup_complete_cb(void *opaque, int ret)
- {
-- // This always called from the main loop
-+ // This can be called from the main loop, or from a coroutine
- PVEBackupCompeteCallbackData cb_data = {
- .di = opaque,
- .result = ret,
- };
-
-- block_on_coroutine_fn(pvebackup_co_complete_cb, &cb_data);
-+ if (qemu_in_coroutine()) {
-+ pvebackup_co_complete_cb(&cb_data);
-+ } else {
-+ block_on_coroutine_fn(pvebackup_co_complete_cb, &cb_data);
-+ }
- }
-
- static void coroutine_fn pvebackup_co_cancel(void *opaque)
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-Date: Mon, 4 Nov 2019 14:18:40 +0100
-Subject: [PATCH] PVE: backup: consider source cluster size as well
-
-Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
----
- block/backup.c | 7 ++-----
- 1 file changed, 2 insertions(+), 5 deletions(-)
-
-diff --git a/block/backup.c b/block/backup.c
-index c682c99f7a..556367f8ba 100644
---- a/block/backup.c
-+++ b/block/backup.c
-@@ -642,11 +642,8 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
- goto error;
- }
-
-- if (target) {
-- cluster_size = backup_calculate_cluster_size(target, errp);
-- } else {
-- cluster_size = dump_cb_block_size;
-- }
-+ cluster_size = backup_calculate_cluster_size(target ? target : bs, errp);
-+ cluster_size = MAX(cluster_size, dump_cb_block_size);
-
- if (cluster_size < 0) {
- goto error;
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Thomas Lamprecht <t.lamprecht@proxmox.com>
-Date: Mon, 26 Aug 2019 18:40:10 +0200
-Subject: [PATCH] PVE: fixup: vma tool
-
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- vma.c | 5 ++++-
- 1 file changed, 4 insertions(+), 1 deletion(-)
-
-diff --git a/vma.c b/vma.c
-index 3289fd722f..a82752448a 100644
---- a/vma.c
-+++ b/vma.c
-@@ -16,6 +16,7 @@
-
- #include "vma.h"
- #include "qemu-common.h"
-+#include "qemu/module.h"
- #include "qemu/error-report.h"
- #include "qemu/main-loop.h"
- #include "qemu/cutils.h"
-@@ -800,7 +801,9 @@ int main(int argc, char **argv)
- const char *cmdname;
- Error *main_loop_err = NULL;
-
-- error_set_progname(argv[0]);
-+ error_init(argv[0]);
-+ module_call_init(MODULE_INIT_TRACE);
-+ qemu_init_exec_dir(argv[0]);
-
- if (qemu_init_main_loop(&main_loop_err)) {
- g_error("%s", error_get_pretty(main_loop_err));
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Thomas Lamprecht <t.lamprecht@proxmox.com>
-Date: Mon, 26 Aug 2019 18:40:44 +0200
-Subject: [PATCH] PVE: fixup: blockdev pvebackup integration: fix blockjob
-
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
----
- blockdev.c | 49 +++++++++++++++++++++++++++----------------------
- 1 file changed, 27 insertions(+), 22 deletions(-)
-
-diff --git a/blockdev.c b/blockdev.c
-index 083ada6c8e..46e8a2780a 100644
---- a/blockdev.c
-+++ b/blockdev.c
-@@ -3416,15 +3416,17 @@ static void coroutine_fn pvebackup_co_cancel(void *opaque)
- PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
- l = g_list_next(l);
- if (!di->completed && di->bs) {
-- BlockJob *job = di->bs->job;
-- if (job) {
-- AioContext *aio_context = blk_get_aio_context(job->blk);
-- aio_context_acquire(aio_context);
-- if (!di->completed) {
-- running_jobs += 1;
-- job_cancel(&job->job, false);
-+ for (BlockJob *job = block_job_next(NULL); job; job = block_job_next(job)) {
-+ if (block_job_has_bdrv(job, di->bs)) {
-+ AioContext *aio_context = job->job.aio_context;
-+ aio_context_acquire(aio_context);
-+
-+ if (!di->completed) {
-+ running_jobs += 1;
-+ job_cancel(&job->job, false);
-+ }
-+ aio_context_release(aio_context);
- }
-- aio_context_release(aio_context);
- }
- }
- }
-@@ -3499,22 +3501,25 @@ static void coroutine_fn pvebackup_co_run_next_job(void)
- while (l) {
- PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
- l = g_list_next(l);
-- if (!di->completed && di->bs && di->bs->job) {
-- BlockJob *job = di->bs->job;
-- AioContext *aio_context = blk_get_aio_context(job->blk);
-- bool cancel_job = backup_state.error || backup_state.cancel;
-- qemu_co_mutex_unlock(&backup_state.backup_mutex);
--
-- aio_context_acquire(aio_context);
-- if (job_should_pause(&job->job)) {
-- if (cancel_job) {
-- job_cancel(&job->job, false);
-- } else {
-- job_resume(&job->job);
-+ if (!di->completed && di->bs) {
-+ for (BlockJob *job = block_job_next(NULL); job; job = block_job_next(job)) {
-+ if (block_job_has_bdrv(job, di->bs)) {
-+ AioContext *aio_context = job->job.aio_context;
-+ qemu_co_mutex_unlock(&backup_state.backup_mutex);
-+ aio_context_acquire(aio_context);
-+
-+
-+ if (job_should_pause(&job->job)) {
-+ if (backup_state.error || backup_state.cancel) {
-+ job_cancel(&job->job, false);
-+ } else {
-+ job_resume(&job->job);
-+ }
-+ }
-+ aio_context_release(aio_context);
-+ return;
- }
- }
-- aio_context_release(aio_context);
-- return;
- }
- }
- qemu_co_mutex_unlock(&backup_state.backup_mutex);
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Mon, 21 Oct 2019 17:32:22 +0200
-Subject: [PATCH] Acquire aio_context before calling block_job_add_bdrv
-
-Otherwise backups immediately fail with 'permission denied' since
-_add_bdrv tries to release a lock we don't own.
-
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
----
- blockjob.c | 10 ++++++++++
- 1 file changed, 10 insertions(+)
-
-diff --git a/blockjob.c b/blockjob.c
-index 74abb97bfd..75b9180a7f 100644
---- a/blockjob.c
-+++ b/blockjob.c
-@@ -448,10 +448,20 @@ void *block_job_create(const char *job_id, const BlockJobDriver *driver,
- notifier_list_add(&job->job.on_ready, &job->ready_notifier);
- notifier_list_add(&job->job.on_idle, &job->idle_notifier);
-
-+ /* block_job_add_bdrv expects us to hold the aio context lock, so acquire it
-+ * before calling if we're not in the main context anyway. */
-+ if (job->job.aio_context != qemu_get_aio_context()) {
-+ aio_context_acquire(job->job.aio_context);
-+ }
-+
- error_setg(&job->blocker, "block device is in use by block job: %s",
- job_type_str(&job->job));
- block_job_add_bdrv(job, "main node", bs, 0, BLK_PERM_ALL, &error_abort);
-
-+ if (job->job.aio_context != qemu_get_aio_context()) {
-+ aio_context_release(job->job.aio_context);
-+ }
-+
- bdrv_op_unblock(bs, BLOCK_OP_TYPE_DATAPLANE, job->blocker);
-
- blk_set_allow_aio_context_change(blk, true);
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Thomas Lamprecht <t.lamprecht@proxmox.com>
-Date: Fri, 22 Nov 2019 08:56:20 +0100
-Subject: [PATCH] PVE: [Compat]: 4.0 used balloon qemu-4-0-config-size false
- here
-
-The underlying issue why this change from upstream to us arised in
-the first place is that QEMU 4.0 was already released at the point we
-run into this migration issue, so we did the then obvious fallback to
-false for virtio-balloon-device qemu-4-0-config-size.
-
-QEMU made that switch back in 4.1, where it now uses a backward
-compatible mechanism to detect if the bigger CFG sizes should be
-used, i.e., checking the VIRTIO_BALLOON_F_PAGE_POISON or
-VIRTIO_BALLOON_F_FREE_PAGE_HINT balloon feature flags.
-As for them, upstream released version 4.0 had this to true they keep
-it to true in their compatibility record for the 4.0 machine, to
-allow live migrations from 4.0 to 4.1.
-As for us, downstream released version 4.0 (first public release of
-this QEMU) had this to false, we change it back to false again, for
-the same reason.
-
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- hw/core/machine.c | 3 ++-
- 1 file changed, 2 insertions(+), 1 deletion(-)
-
-diff --git a/hw/core/machine.c b/hw/core/machine.c
-index 32d1ca9abc..0ee68f1473 100644
---- a/hw/core/machine.c
-+++ b/hw/core/machine.c
-@@ -34,7 +34,8 @@ GlobalProperty hw_compat_4_0[] = {
- { "virtio-vga", "edid", "false" },
- { "virtio-gpu", "edid", "false" },
- { "virtio-device", "use-started", "false" },
-- { "virtio-balloon-device", "qemu-4-0-config-size", "true" },
-+ // PVE differed from upstream for 4.0 balloon cfg size
-+ { "virtio-balloon-device", "qemu-4-0-config-size", "false" },
- { "pl031", "migrate-tick-offset", "false" },
- };
- const size_t hw_compat_4_0_len = G_N_ELEMENTS(hw_compat_4_0);
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Thu, 14 Nov 2019 17:56:12 +0100
-Subject: [PATCH] PVE: Allow version code in machine type
-
-E.g. pc-i440fx-4.0+pve3 would print 'pve3' as version code while
-selecting pc-i440fx-4.0 as machine type.
-
-Version is made available as 'pve-version' in query-machines (same as,
-and only if 'is-current').
-
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
----
- hw/core/machine-qmp-cmds.c | 6 ++++++
- include/hw/boards.h | 2 ++
- qapi/machine.json | 2 +-
- vl.c | 15 ++++++++++++++-
- 4 files changed, 23 insertions(+), 2 deletions(-)
-
-diff --git a/hw/core/machine-qmp-cmds.c b/hw/core/machine-qmp-cmds.c
-index fd68f9baf8..61fcad138d 100644
---- a/hw/core/machine-qmp-cmds.c
-+++ b/hw/core/machine-qmp-cmds.c
-@@ -232,6 +232,12 @@ MachineInfoList *qmp_query_machines(Error **errp)
- if (strcmp(mc->name, MACHINE_GET_CLASS(current_machine)->name) == 0) {
- info->has_is_current = true;
- info->is_current = true;
-+
-+ // PVE version string only exists for current machine
-+ if (mc->pve_version) {
-+ info->has_pve_version = true;
-+ info->pve_version = g_strdup(mc->pve_version);
-+ }
- }
-
- entry = g_malloc0(sizeof(*entry));
-diff --git a/include/hw/boards.h b/include/hw/boards.h
-index a71d1a53a5..952d2add82 100644
---- a/include/hw/boards.h
-+++ b/include/hw/boards.h
-@@ -178,6 +178,8 @@ struct MachineClass {
- const char *desc;
- const char *deprecation_reason;
-
-+ const char *pve_version;
-+
- void (*init)(MachineState *state);
- void (*reset)(MachineState *state);
- void (*hot_add_cpu)(MachineState *state, const int64_t id, Error **errp);
-diff --git a/qapi/machine.json b/qapi/machine.json
-index 7b82c5f7f5..11fef6714e 100644
---- a/qapi/machine.json
-+++ b/qapi/machine.json
-@@ -333,7 +333,7 @@
- 'data': { 'name': 'str', '*alias': 'str',
- '*is-default': 'bool', '*is-current': 'bool', 'cpu-max': 'int',
- 'hotpluggable-cpus': 'bool', 'numa-mem-supported': 'bool',
-- 'deprecated': 'bool' } }
-+ 'deprecated': 'bool', '*pve-version': 'str' } }
-
- ##
- # @query-machines:
-diff --git a/vl.c b/vl.c
-index 7ffcd271f1..1305cdaee7 100644
---- a/vl.c
-+++ b/vl.c
-@@ -2454,6 +2454,8 @@ static MachineClass *machine_parse(const char *name, GSList *machines)
- {
- MachineClass *mc;
- GSList *el;
-+ size_t pvever_index = 0;
-+ gchar *name_clean;
-
- if (is_help_option(name)) {
- printf("Supported machines are:\n");
-@@ -2470,12 +2472,23 @@ static MachineClass *machine_parse(const char *name, GSList *machines)
- exit(0);
- }
-
-- mc = find_machine(name, machines);
-+ // PVE version is specified with '+' as seperator, e.g. pc-i440fx+pvever
-+ pvever_index = strcspn(name, "+");
-+
-+ name_clean = g_strndup(name, pvever_index);
-+ mc = find_machine(name_clean, machines);
-+ g_free(name_clean);
-+
- if (!mc) {
- error_report("unsupported machine type");
- error_printf("Use -machine help to list supported machines\n");
- exit(1);
- }
-+
-+ if (pvever_index < strlen(name)) {
-+ mc->pve_version = &name[pvever_index+1];
-+ }
-+
- return mc;
- }
-
+++ /dev/null
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Tim Marx <t.marx@proxmox.com>
-Date: Wed, 22 Jan 2020 16:22:14 +0100
-Subject: [PATCH] PVE: fix hmp info backup cmd for not initialized
- backup_state.backup_mutex
-
-Signed-off-by: Tim Marx <t.marx@proxmox.com>
----
- monitor/hmp-cmds.c | 6 ++++++
- 1 file changed, 6 insertions(+)
-
-diff --git a/monitor/hmp-cmds.c b/monitor/hmp-cmds.c
-index bc9ca346f7..fb820410de 100644
---- a/monitor/hmp-cmds.c
-+++ b/monitor/hmp-cmds.c
-@@ -201,6 +201,12 @@ void hmp_info_backup(Monitor *mon, const QDict *qdict)
- BackupStatus *info;
-
- info = qmp_query_backup(NULL);
-+
-+ if (!info) {
-+ monitor_printf(mon, "Backup status: not initialized\n");
-+ return;
-+ }
-+
- if (info->has_status) {
- if (info->has_errmsg) {
- monitor_printf(mon, "Backup status: %s - %s\n",
---
-2.20.1
-
+++ /dev/null
-From 4c2e4e498716b8a556fccec20fb1348555c2b189 Mon Sep 17 00:00:00 2001
-From: Dietmar Maurer <dietmar@proxmox.com>
-Date: Fri, 14 Feb 2020 10:21:37 +0100
-Subject: [PATCH qemu 1/7] PVE backup: use separate CoRwlock for data accessed
- by qmp query_backup
-
-This should minimize contention between query_backup and running backup jobs.
----
- blockdev.c | 178 +++++++++++++++++++++++++++++++++--------------------
- 1 file changed, 112 insertions(+), 66 deletions(-)
-
-diff --git a/blockdev.c b/blockdev.c
-index 46e8a2780a..4367eaf2a7 100644
---- a/blockdev.c
-+++ b/blockdev.c
-@@ -3203,22 +3203,26 @@ static void block_on_coroutine_fn(CoroutineEntry *entry, void *entry_arg)
- AIO_WAIT_WHILE(ctx, !wrapper.finished);
- }
-
-+
- static struct PVEBackupState {
-- Error *error;
-+ struct {
-+ CoRwlock rwlock;
-+ Error *error;
-+ time_t start_time;
-+ time_t end_time;
-+ char *backup_file;
-+ uuid_t uuid;
-+ char uuid_str[37];
-+ size_t total;
-+ size_t transferred;
-+ size_t zero_bytes;
-+ } stat;
- bool cancel;
-- uuid_t uuid;
-- char uuid_str[37];
- int64_t speed;
-- time_t start_time;
-- time_t end_time;
-- char *backup_file;
- VmaWriter *vmaw;
- GList *di_list;
-- size_t total;
-- size_t transferred;
-- size_t zero_bytes;
- CoMutex backup_mutex;
-- bool backup_mutex_initialized;
-+ bool mutex_initialized;
- } backup_state;
-
- typedef struct PVEBackupDevInfo {
-@@ -3251,11 +3255,14 @@ static int coroutine_fn pvebackup_co_dump_cb(void *opaque, BlockBackend *target,
-
- uint64_t cluster_num = start / VMA_CLUSTER_SIZE;
- if ((cluster_num * VMA_CLUSTER_SIZE) != start) {
-- if (!backup_state.error) {
-- error_setg(&backup_state.error,
-+ qemu_co_rwlock_rdlock(&backup_state.stat.rwlock);
-+ if (!backup_state.stat.error) {
-+ qemu_co_rwlock_upgrade(&backup_state.stat.rwlock);
-+ error_setg(&backup_state.stat.error,
- "got unaligned write inside backup dump "
- "callback (sector %ld)", start);
- }
-+ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
- qemu_co_mutex_unlock(&backup_state.backup_mutex);
- return -1; // not aligned to cluster size
- }
-@@ -3273,27 +3280,35 @@ static int coroutine_fn pvebackup_co_dump_cb(void *opaque, BlockBackend *target,
- buf += VMA_CLUSTER_SIZE;
- }
- if (ret < 0) {
-- if (!backup_state.error) {
-- vma_writer_error_propagate(backup_state.vmaw, &backup_state.error);
-+ qemu_co_rwlock_rdlock(&backup_state.stat.rwlock);
-+ if (!backup_state.stat.error) {
-+ qemu_co_rwlock_upgrade(&backup_state.stat.rwlock);
-+ vma_writer_error_propagate(backup_state.vmaw, &backup_state.stat.error);
- }
-+ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
-+
- qemu_co_mutex_unlock(&backup_state.backup_mutex);
- return ret;
- } else {
-- backup_state.zero_bytes += zero_bytes;
-+ qemu_co_rwlock_wrlock(&backup_state.stat.rwlock);
-+ backup_state.stat.zero_bytes += zero_bytes;
- if (remaining >= VMA_CLUSTER_SIZE) {
-- backup_state.transferred += VMA_CLUSTER_SIZE;
-+ backup_state.stat.transferred += VMA_CLUSTER_SIZE;
- remaining -= VMA_CLUSTER_SIZE;
- } else {
-- backup_state.transferred += remaining;
-+ backup_state.stat.transferred += remaining;
- remaining = 0;
- }
-+ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
- }
- }
- } else {
-+ qemu_co_rwlock_wrlock(&backup_state.stat.rwlock);
- if (!buf) {
-- backup_state.zero_bytes += size;
-+ backup_state.stat.zero_bytes += size;
- }
-- backup_state.transferred += size;
-+ backup_state.stat.transferred += size;
-+ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
- }
-
- qemu_co_mutex_unlock(&backup_state.backup_mutex);
-@@ -3313,12 +3328,20 @@ static void coroutine_fn pvebackup_co_cleanup(void)
- return;
- }
-
-- backup_state.end_time = time(NULL);
-+ qemu_co_rwlock_wrlock(&backup_state.stat.rwlock);
-+ backup_state.stat.end_time = time(NULL);
-+ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
-
- if (backup_state.vmaw) {
- Error *local_err = NULL;
- vma_writer_close(backup_state.vmaw, &local_err);
-- error_propagate(&backup_state.error, local_err);
-+
-+ if (local_err != NULL) {
-+ qemu_co_rwlock_wrlock(&backup_state.stat.rwlock);
-+ error_propagate(&backup_state.stat.error, local_err);
-+ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
-+ }
-+
- backup_state.vmaw = NULL;
- }
-
-@@ -3345,10 +3368,14 @@ static void coroutine_fn pvebackup_co_complete_cb(void *opaque)
-
- di->completed = true;
-
-- if (ret < 0 && !backup_state.error) {
-- error_setg(&backup_state.error, "job failed with err %d - %s",
-+ qemu_co_rwlock_rdlock(&backup_state.stat.rwlock);
-+
-+ if (ret < 0 && !backup_state.stat.error) {
-+ qemu_co_rwlock_upgrade(&backup_state.stat.rwlock);
-+ error_setg(&backup_state.stat.error, "job failed with err %d - %s",
- ret, strerror(-ret));
- }
-+ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
-
- di->bs = NULL;
- di->target = NULL;
-@@ -3401,9 +3428,12 @@ static void coroutine_fn pvebackup_co_cancel(void *opaque)
- return;
- }
-
-- if (!backup_state.error) {
-- error_setg(&backup_state.error, "backup cancelled");
-+ qemu_co_rwlock_rdlock(&backup_state.stat.rwlock);
-+ if (!backup_state.stat.error) {
-+ qemu_co_rwlock_upgrade(&backup_state.stat.rwlock);
-+ error_setg(&backup_state.stat.error, "backup cancelled");
- }
-+ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
-
- if (backup_state.vmaw) {
- /* make sure vma writer does not block anymore */
-@@ -3438,7 +3468,7 @@ static void coroutine_fn pvebackup_co_cancel(void *opaque)
-
- void qmp_backup_cancel(Error **errp)
- {
-- if (!backup_state.backup_mutex_initialized)
-+ if (!backup_state.mutex_initialized)
- return;
-
- block_on_coroutine_fn(pvebackup_co_cancel, NULL);
-@@ -3508,9 +3538,13 @@ static void coroutine_fn pvebackup_co_run_next_job(void)
- qemu_co_mutex_unlock(&backup_state.backup_mutex);
- aio_context_acquire(aio_context);
-
--
-+
- if (job_should_pause(&job->job)) {
-- if (backup_state.error || backup_state.cancel) {
-+ qemu_co_rwlock_rdlock(&backup_state.stat.rwlock);
-+ bool error_or_canceled = backup_state.stat.error || backup_state.cancel;
-+ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
-+
-+ if (error_or_canceled) {
- job_cancel(&job->job, false);
- } else {
- job_resume(&job->job);
-@@ -3565,9 +3599,10 @@ static void coroutine_fn pvebackup_co_start(void *opaque)
- const char *config_name = "qemu-server.conf";
- const char *firewall_name = "qemu-server.fw";
-
-- if (!backup_state.backup_mutex_initialized) {
-+ if (!backup_state.mutex_initialized) {
-+ qemu_co_rwlock_init(&backup_state.stat.rwlock);
- qemu_co_mutex_init(&backup_state.backup_mutex);
-- backup_state.backup_mutex_initialized = true;
-+ backup_state.mutex_initialized = true;
- }
-
- qemu_co_mutex_lock(&backup_state.backup_mutex);
-@@ -3727,31 +3762,36 @@ static void coroutine_fn pvebackup_co_start(void *opaque)
-
- backup_state.cancel = false;
-
-- if (backup_state.error) {
-- error_free(backup_state.error);
-- backup_state.error = NULL;
-- }
-+ qemu_co_rwlock_wrlock(&backup_state.stat.rwlock);
-
-- backup_state.speed = (task->has_speed && task->speed > 0) ? task->speed : 0;
-+ if (backup_state.stat.error) {
-+ error_free(backup_state.stat.error);
-+ backup_state.stat.error = NULL;
-+ }
-
-- backup_state.start_time = time(NULL);
-- backup_state.end_time = 0;
-+ backup_state.stat.start_time = time(NULL);
-+ backup_state.stat.end_time = 0;
-
-- if (backup_state.backup_file) {
-- g_free(backup_state.backup_file);
-+ if (backup_state.stat.backup_file) {
-+ g_free(backup_state.stat.backup_file);
- }
-- backup_state.backup_file = g_strdup(task->backup_file);
-+ backup_state.stat.backup_file = g_strdup(task->backup_file);
-
-- backup_state.vmaw = vmaw;
-+ uuid_copy(backup_state.stat.uuid, uuid);
-+ uuid_unparse_lower(uuid, backup_state.stat.uuid_str);
-+ char *uuid_str = g_strdup(backup_state.stat.uuid_str);
-
-- uuid_copy(backup_state.uuid, uuid);
-- uuid_unparse_lower(uuid, backup_state.uuid_str);
-+ backup_state.stat.total = total;
-+ backup_state.stat.transferred = 0;
-+ backup_state.stat.zero_bytes = 0;
-
-- backup_state.di_list = di_list;
-+ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
-+
-+ backup_state.speed = (task->has_speed && task->speed > 0) ? task->speed : 0;
-+
-+ backup_state.vmaw = vmaw;
-
-- backup_state.total = total;
-- backup_state.transferred = 0;
-- backup_state.zero_bytes = 0;
-+ backup_state.di_list = di_list;
-
- /* start all jobs (paused state) */
- l = di_list;
-@@ -3763,7 +3803,9 @@ static void coroutine_fn pvebackup_co_start(void *opaque)
- JOB_DEFAULT, pvebackup_co_dump_cb, dump_cb_block_size,
- pvebackup_complete_cb, di, 1, NULL, &local_err);
- if (!job || local_err != NULL) {
-- error_setg(&backup_state.error, "backup_job_create failed");
-+ qemu_co_rwlock_wrlock(&backup_state.stat.rwlock);
-+ error_setg(&backup_state.stat.error, "backup_job_create failed");
-+ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
- break;
- }
- job_start(&job->job);
-@@ -3775,14 +3817,18 @@ static void coroutine_fn pvebackup_co_start(void *opaque)
-
- qemu_co_mutex_unlock(&backup_state.backup_mutex);
-
-- if (!backup_state.error) {
-+ qemu_co_rwlock_rdlock(&backup_state.stat.rwlock);
-+ bool no_errors = !backup_state.stat.error;
-+ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
-+
-+ if (no_errors) {
- pvebackup_co_run_next_job(); // run one job
- } else {
- pvebackup_co_cancel(NULL);
- }
-
- uuid_info = g_malloc0(sizeof(*uuid_info));
-- uuid_info->UUID = g_strdup(backup_state.uuid_str);
-+ uuid_info->UUID = uuid_str;
-
- task->result = uuid_info;
- return;
-@@ -3866,54 +3912,54 @@ static void coroutine_fn pvebackup_co_query(void *opaque)
-
- BackupStatus *info = g_malloc0(sizeof(*info));
-
-- if (!backup_state.backup_mutex_initialized)
-+ if (!backup_state.mutex_initialized)
- return;
-
-- qemu_co_mutex_lock(&backup_state.backup_mutex);
-+ qemu_co_rwlock_rdlock(&backup_state.stat.rwlock);
-
-- if (!backup_state.start_time) {
-+ if (!backup_state.stat.start_time) {
- /* not started, return {} */
- task->result = info;
-- qemu_co_mutex_unlock(&backup_state.backup_mutex);
-+ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
- return;
- }
-
- info->has_status = true;
- info->has_start_time = true;
-- info->start_time = backup_state.start_time;
-+ info->start_time = backup_state.stat.start_time;
-
-- if (backup_state.backup_file) {
-+ if (backup_state.stat.backup_file) {
- info->has_backup_file = true;
-- info->backup_file = g_strdup(backup_state.backup_file);
-+ info->backup_file = g_strdup(backup_state.stat.backup_file);
- }
-
- info->has_uuid = true;
-- info->uuid = g_strdup(backup_state.uuid_str);
-+ info->uuid = g_strdup(backup_state.stat.uuid_str);
-
-- if (backup_state.end_time) {
-- if (backup_state.error) {
-+ if (backup_state.stat.end_time) {
-+ if (backup_state.stat.error) {
- info->status = g_strdup("error");
- info->has_errmsg = true;
-- info->errmsg = g_strdup(error_get_pretty(backup_state.error));
-+ info->errmsg = g_strdup(error_get_pretty(backup_state.stat.error));
- } else {
- info->status = g_strdup("done");
- }
- info->has_end_time = true;
-- info->end_time = backup_state.end_time;
-+ info->end_time = backup_state.stat.end_time;
- } else {
- info->status = g_strdup("active");
- }
-
- info->has_total = true;
-- info->total = backup_state.total;
-+ info->total = backup_state.stat.total;
- info->has_zero_bytes = true;
-- info->zero_bytes = backup_state.zero_bytes;
-+ info->zero_bytes = backup_state.stat.zero_bytes;
- info->has_transferred = true;
-- info->transferred = backup_state.transferred;
-+ info->transferred = backup_state.stat.transferred;
-
- task->result = info;
-
-- qemu_co_mutex_unlock(&backup_state.backup_mutex);
-+ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
- }
-
- BackupStatus *qmp_query_backup(Error **errp)
---
-2.20.1
-
+++ /dev/null
-From ca384caae9a0dbbbbab2174ec09935702ac7a067 Mon Sep 17 00:00:00 2001
-From: Dietmar Maurer <dietmar@proxmox.com>
-Date: Fri, 14 Feb 2020 11:51:02 +0100
-Subject: [PATCH qemu 2/7] PVE backup - block_on_coroutine_wrapper: call
- aio_wait_kick() as described in block/aio-wait.h
-
----
- blockdev.c | 1 +
- 1 file changed, 1 insertion(+)
-
-diff --git a/blockdev.c b/blockdev.c
-index 4367eaf2a7..2f84f8832d 100644
---- a/blockdev.c
-+++ b/blockdev.c
-@@ -3185,6 +3185,7 @@ static void coroutine_fn block_on_coroutine_wrapper(void *opaque)
- BlockOnCoroutineWrapper *wrapper = opaque;
- wrapper->entry(wrapper->entry_arg);
- wrapper->finished = true;
-+ aio_wait_kick();
- }
-
- static void block_on_coroutine_fn(CoroutineEntry *entry, void *entry_arg)
---
-2.20.1
-
+++ /dev/null
-From c8856d5cbcf3d427cd02c9556f845486ab5362bc Mon Sep 17 00:00:00 2001
-From: Dietmar Maurer <dietmar@proxmox.com>
-Date: Fri, 14 Feb 2020 12:37:27 +0100
-Subject: [PATCH qemu 3/7] PVE backup: move backup_state.cancel to
- backup_state.stat.cancel
-
-In order to avoid lock contention with qmp_backup_cancel.
----
- blockdev.c | 25 +++++++++++++++----------
- 1 file changed, 15 insertions(+), 10 deletions(-)
-
-diff --git a/blockdev.c b/blockdev.c
-index 2f84f8832d..a4b5b98224 100644
---- a/blockdev.c
-+++ b/blockdev.c
-@@ -3204,9 +3204,9 @@ static void block_on_coroutine_fn(CoroutineEntry *entry, void *entry_arg)
- AIO_WAIT_WHILE(ctx, !wrapper.finished);
- }
-
--
- static struct PVEBackupState {
- struct {
-+ // Everithing accessed from qmp command, protected using rwlock
- CoRwlock rwlock;
- Error *error;
- time_t start_time;
-@@ -3217,8 +3217,8 @@ static struct PVEBackupState {
- size_t total;
- size_t transferred;
- size_t zero_bytes;
-+ bool cancel;
- } stat;
-- bool cancel;
- int64_t speed;
- VmaWriter *vmaw;
- GList *di_list;
-@@ -3247,13 +3247,16 @@ static int coroutine_fn pvebackup_co_dump_cb(void *opaque, BlockBackend *target,
- const unsigned char *buf = pbuf;
- PVEBackupDevInfo *di = opaque;
-
-- qemu_co_mutex_lock(&backup_state.backup_mutex);
-+ qemu_co_rwlock_rdlock(&backup_state.stat.rwlock);
-+ bool cancel = backup_state.stat.cancel;
-+ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
-
-- if (backup_state.cancel) {
-- qemu_co_mutex_unlock(&backup_state.backup_mutex);
-+ if (cancel) {
- return size; // return success
- }
-
-+ qemu_co_mutex_lock(&backup_state.backup_mutex);
-+
- uint64_t cluster_num = start / VMA_CLUSTER_SIZE;
- if ((cluster_num * VMA_CLUSTER_SIZE) != start) {
- qemu_co_rwlock_rdlock(&backup_state.stat.rwlock);
-@@ -3419,9 +3422,11 @@ static void coroutine_fn pvebackup_co_cancel(void *opaque)
- {
- assert(qemu_in_coroutine());
-
-- qemu_co_mutex_lock(&backup_state.backup_mutex);
-+ qemu_co_rwlock_wrlock(&backup_state.stat.rwlock);
-+ backup_state.stat.cancel = true;
-+ qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
-
-- backup_state.cancel = true;
-+ qemu_co_mutex_lock(&backup_state.backup_mutex);
-
- // Avoid race between block jobs and backup-cancel command:
- if (!backup_state.vmaw) {
-@@ -3542,7 +3547,7 @@ static void coroutine_fn pvebackup_co_run_next_job(void)
-
- if (job_should_pause(&job->job)) {
- qemu_co_rwlock_rdlock(&backup_state.stat.rwlock);
-- bool error_or_canceled = backup_state.stat.error || backup_state.cancel;
-+ bool error_or_canceled = backup_state.stat.error || backup_state.stat.cancel;
- qemu_co_rwlock_unlock(&backup_state.stat.rwlock);
-
- if (error_or_canceled) {
-@@ -3761,10 +3766,10 @@ static void coroutine_fn pvebackup_co_start(void *opaque)
- }
- /* initialize global backup_state now */
-
-- backup_state.cancel = false;
--
- qemu_co_rwlock_wrlock(&backup_state.stat.rwlock);
-
-+ backup_state.stat.cancel = false;
-+
- if (backup_state.stat.error) {
- error_free(backup_state.stat.error);
- backup_state.stat.error = NULL;
---
-2.20.1
-
-extra/0001-monitor-qmp-resume-monitor-when-clearing-its-queue.patch
-extra/0002-virtio-blk-schedule-virtio_notify_config-to-run-on-m.patch
-extra/0003-vnc-fix-memory-leak-when-vnc-disconnect.patch
-extra/0004-util-add-slirp_fmt-helpers.patch
-extra/0005-tcp_emu-fix-unsafe-snprintf-usages.patch
pve/0001-PVE-Config-block-file-change-locking-default-to-off.patch
pve/0002-PVE-Config-Adjust-network-script-path-to-etc-kvm.patch
pve/0003-PVE-Config-set-the-CPU-model-to-kvm64-32-instead-of-.patch
pve/0017-PVE-internal-snapshot-async.patch
pve/0018-PVE-block-add-the-zeroinit-block-driver-filter.patch
pve/0019-PVE-backup-modify-job-api.patch
-pve/0020-PVE-backup-introduce-vma-archive-format.patch
-pve/0021-PVE-Deprecated-adding-old-vma-files.patch
-pve/0022-PVE-vma-add-throttling-options-to-drive-mapping-fifo.patch
-pve/0023-PVE-vma-add-cache-option-to-device-map.patch
-pve/0024-PVE-vma-remove-forced-NO_FLUSH-option.patch
-pve/0025-PVE-Add-dummy-id-command-line-parameter.patch
-pve/0026-PVE-Config-Revert-target-i386-disable-LINT0-after-re.patch
-pve/0027-PVE-Up-Config-file-posix-make-locking-optiono-on-cre.patch
-pve/0028-PVE-savevm-async-kick-AIO-wait-on-block-state-write.patch
-pve/0029-PVE-move-snapshot-cleanup-into-bottom-half.patch
-pve/0030-PVE-monitor-disable-oob-capability.patch
-pve/0031-PVE-bug-fix-1071-vma-writer.c-use-correct-AioContext.patch
-pve/0032-qmp_backup-run-backup-related-code-inside-coroutines.patch
-pve/0033-qmp_backup-use-a-CoMutex-to-protect-access-to-backup.patch
-pve/0034-vma_writer_close-avoid-call-to-aio_poll-acquire-flus.patch
-pve/0035-backup_job_create-pass-cluster-size-for-dump.patch
-pve/0036-avoid-calling-dump_cb-with-NULL-data-pointer-for-sma.patch
-pve/0037-rename-config_to_vma-into-pvebackup_co_add_config.patch
-pve/0038-pvebackup_co_dump_cb-do-not-call-job-cancel.patch
-pve/0039-fix-backup-job-completion.patch
-pve/0040-pvebackup_complete_cb-avoid-poll-loop-if-already-ins.patch
-pve/0041-PVE-backup-consider-source-cluster-size-as-well.patch
-pve/0042-PVE-fixup-vma-tool.patch
-pve/0043-PVE-fixup-blockdev-pvebackup-integration-fix-blockjo.patch
-pve/0044-Acquire-aio_context-before-calling-block_job_add_bdr.patch
-pve/0045-PVE-Compat-4.0-used-balloon-qemu-4-0-config-size-fal.patch
-pve/0046-PVE-Allow-version-code-in-machine-type.patch
-pve/0047-PVE-fix-hmp-info-backup-cmd-for-not-initialized-back.patch
-pve/0048-PVE-backup-use-separate-CoRwlock-for-data-accessed-b.patch
-pve/0049-PVE-backup-block_on_coroutine_wrapper-call-aio_wait.patch
-pve/0050-PVE-backup-move-backup_state.cancel-to-backup_state.patch
+pve/0020-PVE-Add-dummy-id-command-line-parameter.patch
+pve/0021-PVE-Config-Revert-target-i386-disable-LINT0-after-re.patch
+pve/0022-PVE-Up-Config-file-posix-make-locking-optiono-on-cre.patch
+pve/0023-PVE-savevm-async-kick-AIO-wait-on-block-state-write.patch
+pve/0024-PVE-move-snapshot-cleanup-into-bottom-half.patch
+pve/0025-PVE-monitor-disable-oob-capability.patch
+pve/0026-PVE-Acquire-aio_context-before-calling-block_job_add.patch
+pve/0027-PVE-Compat-4.0-used-balloon-qemu-4-0-config-size-fal.patch
+pve/0028-PVE-Allow-version-code-in-machine-type.patch
+pve/0029-PVE-Backup-add-vma-backup-format-code.patch
+pve/0030-PVE-Backup-add-backup-dump-block-driver.patch
+pve/0031-PVE-Backup-proxmox-backup-patches-for-qemu.patch
+pve/0032-PVE-Backup-pbs-restore-new-command-to-restore-from-p.patch
rm $(destdir)/usr/share/kvm/s390-netboot.img
rm $(destdir)/usr/share/kvm/qemu_vga.ndrv
rm $(destdir)/usr/share/kvm/slof.bin
- rm $(destdir)/usr/share/kvm/spapr-rtas.bin
rm $(destdir)/usr/share/kvm/u-boot.e500
# remove Aplha files
rm $(destdir)/usr/share/kvm/palcode-clipper
-Subproject commit 99c5874a9b6c9f70aef285d6eff85d4f46de3c52
+Subproject commit b0ca999a43a22b38158a222233d3f5881648bb4f
projects / pve-qemu.git / commitdiff