Paolo Bonzini [Fri, 15 Jan 2010 08:42:08 +0000 (09:42 +0100)]
add assertions about env->current_tb
By virtue of the previous patch env->current_tb will always be NULL at
the top of cpu_exec's outermost for loop, and at the end of the innermost
while loop.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Stefan Weil [Sun, 20 Dec 2009 15:52:22 +0000 (16:52 +0100)]
eepro100: Restructure code (new function tx_command)
Handling of transmit commands is rather complex,
so about 80 lines of code were moved from function
action_command to the new function tx_command.
The two new values "tx" and "cb_address" in the
eepro100 status structure made this possible without
passing too many parameters.
In addition, the moved code was cleaned a little bit:
old comments marked with //~ were removed, C++ style
comments were replaced by C style comments, C++ like
variable declarations after code were reordered.
Simplified mode is still broken. Nor did I fix
endianess issues. Both problems will be fixed in
additional patches (which need this one).
Signed-off-by: Stefan Weil <weil@mail.berlios.de> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Blue Swirl [Sun, 17 Jan 2010 16:51:57 +0000 (16:51 +0000)]
Sparc: improve CPU register dump
Common:
* Remove unnecessary 0x prefix
* Print %y
* Fix NZVC flag print order to match CPU bit order
Sparc64 specific:
* Print registers without line wrapping
* Print %f40-%f63
* Pretty print CCR flags
* Print %fsr and %fprs in full precision
* More consistent formatting
The commit was badly broken, Gentoo has sdl as the default driver,
consequently 5 gentoo users have hit the breakage and were kind enough
to report, so thank you:
Claes Gyllenswrd
vekin
Chris
But above all thanks to Toralf Foerster who actually provied enough
information to pinpoint the breakage to sdlaudio.
Artyom Tarasenko [Sat, 16 Jan 2010 09:06:32 +0000 (09:06 +0000)]
sparc32 do not clear interrupts when masking
Don't clear interrupts on disabling, because
* Sun4M_SystemArchitecture_edited2.pdf doesn't describe
that masking or un-masking IRQ shall clear pending ones.
* Field tests also show that SPARCstation-20 doesn't
clear them.
* The patch makes Solaris 2.5.1/2.6 boot ~1500 times
faster (~20 seconds instead of ~8 hours)
Signed-off-by: Artyom Tarasenko <atar4qemu@gmail.com> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Artyom Tarasenko [Fri, 15 Jan 2010 21:28:56 +0000 (22:28 +0100)]
sparc32 do_unassigned_access overhaul v2
According to pages 9-31 - 9-34 of "SuperSPARC & MultiCache Controller
User's Manual":
1. "A lower priority fault may not overwrite the
MFSR status of a higher priority fault."
2. The MFAR is overwritten according to the policy defined for the MFSR
3. The overwrite bit is asserted if the fault status register (MFSR)
has been written more than once by faults of the same class
4. SuperSPARC will never place instruction fault addresses in the MFAR.
Implementation of points 1-3 allows booting Solaris 2.6 and 2.5.1.
v2: CODING_STYLE fixes
Signed-off-by: Artyom Tarasenko <atar4qemu@gmail.com> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
The existing P_REXB internal opcode flag unconditionally emits
the REX prefix. Technically it's not needed if the register in
question is %al, %bl, %cl, %dl.
Eliding the prefix requires splitting the P_REXB flag into two,
in order to indicate whether the byte register in question is
in the REG or the R/M field. Within TCG, the byte register is
in the REG field only for stores.
Signed-off-by: Richard Henderson <rth@twiddle.net> Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
This avoids an unnecessary REX.W prefix when dealing with AND
operands that fit into a 32-bit quantity. The most common change
actually seen is movz[wb]q -> movz[wb]l.
Similarly, avoid REXW in ext{8,16}u_i64 tcg opcodes.
Signed-off-by: Richard Henderson <rth@twiddle.net> Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
Vince Weaver [Tue, 29 Dec 2009 05:01:22 +0000 (00:01 -0500)]
alpha: fix stat64 issue
The stat64/fstat64 syscalls are broken for alpha linux-user.
This is because Alpha, even though it is native 64-bits, has a stat64
syscall that is different than regular stat. This means that the
"TARGET_LONG_BITS==64" check in syscall.c isn't enough. Below is
a patch that fixes things for me, although it might not be the cleanest
fix.
This issue keeps sixtrack and fma3d spec2k benchmarks from running.
Milan Plzik [Thu, 7 Jan 2010 12:39:43 +0000 (13:39 +0100)]
Qemu's internal TFTP server breaks lock-step-iness of TFTP
According to RFC 1350 and RFC 2347, TFTP server should answer RRQ by
either OACK or DATA packet. Qemu's internal TFTP server answers RRQ with
additional options by sending both OACK and DATA packet, thus breaking
the "lock-step" feature of the protocol, and also confuses client.
Proposed solution would be to, in case of OACK packet, wait for ACK
from client and just then start sending data. Attached patch implements
this.
Signed-off-by: Thomas Horsten <thomas@horsten.com> Signed-off-by: Milan Plzik <milan.plzik@gmail.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
The backing device is only modified from bdrv_commit. So instead of
flushing it every time bdrv_flush is called for the front-end device
only flush it after we're written data to it in bdrv_commit.
Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Kevin Wolf [Wed, 13 Jan 2010 15:20:56 +0000 (16:20 +0100)]
osdep.c: Fix accept4 fallback
Commit 3a03bfa5 added a fallback in case the Linux kernel running qemu is older
than the kernel of the build system. Unfortunately, v1 was committed instead of
v2, so the code has a bug that was revealed in the review (checking for the
wrong error code).
Signed-off-by: Kevin Wolf <kwolf@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Kevin Wolf [Tue, 12 Jan 2010 11:55:18 +0000 (12:55 +0100)]
qemu-img rebase
This adds a rebase subcommand to qemu-img which allows to change the backing
file of an image.
In default mode, both the current and the new backing file need to exist, and
after the rebase, the COW image is guaranteed to have the same guest visible
content as before. To achieve this, old and new backing file are compared and,
if necessary, data is copied from the old backing file into the COW image.
With -u an unsafe mode is enabled that doesn't require the backing files to
exist. It merely changes the backing file reference in the COW image. This is
useful for renaming or moving the backing file. The user is responsible to make
sure that the new backing file has no changes compared to the old one, or
corruption may occur.
Signed-off-by: Kevin Wolf <kwolf@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Kevin Wolf [Tue, 12 Jan 2010 11:55:16 +0000 (12:55 +0100)]
block: Introduce BDRV_O_NO_BACKING
If an image references a backing file that doesn't exist, qemu-img info fails
to open this image. Exactly in this case the info would be valuable, though:
the user might want to find out which file is missing.
This patch introduces a BDRV_O_NO_BACKING flag to ignore the backing file when
opening the image. qemu-img info is the first user and provides info now even
if the backing file is invalid.
Signed-off-by: Kevin Wolf <kwolf@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Paolo Bonzini [Wed, 13 Jan 2010 08:52:54 +0000 (09:52 +0100)]
fixes to the static compilation case for sdl
After the next commit, pkg-config could be used for the shared library
configuration case and sdl-config for static libraries. So I prepare
the test here by doing two changes:
at the same time I remove useless backslashes from the invocation of
grep;
1) fixing a typo ($sd_cflags). The typo has been there since commit 1ac88f2 (remove sdl_static. Just do the right thing if static is yes,
2009-07-27).
2) fixing an erroneous "test `... | grep > /dev/null`" idiom that would
never succeed since grep's output would be empty;
3) checking the status code after executing sdl-config --static --libs;
this is needed for the next patch only.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Blue Swirl [Wed, 13 Jan 2010 18:54:54 +0000 (18:54 +0000)]
Sparc: comment out unused variable, spotted by clang
The variable is_annulled is referenced in a FIXME comment, so instead
of removing the variable, comment out the lines so that the FIXME
can be understood.
Stefan Weil [Thu, 7 Jan 2010 16:13:30 +0000 (17:13 +0100)]
eepro100: Update ROM file support
Use new way to associate ROM files to devices.
Currently, there is only a ROM file for i82559er
included in QEMU, so the patch does not add
.romfile for the other devices.
When flexible mode is fixed in eepro100, adding
more ROM files will be possible. It should be
possible to create them from pxe-i82559er.bin,
because etherboot uses the same driver for all
eepro100 devices (only PCI ids differ).
Maybe it is even possible to create a single
pxe-i8255x.bin which supports all eepro100 devices
(not supported with current etherboot).
Signed-off-by: Stefan Weil <weil@mail.berlios.de> Acked-by: Gerd Hoffmann <kraxel@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Juergen Lock [Tue, 12 Jan 2010 22:48:04 +0000 (23:48 +0100)]
Workaround for broken OSS_GETVERSION on FreeBSD, part two
Turns out on those versions of FreeBSD (>= 7.x) that know OSS_GETVERSION
the ioctl doesn't actually work yet (except in the Linuxolator), so if
building on FreeBSD assume the sound drivers are new enough if the ioctl
returns the errno it does currently on FreeBSD.
Mac feature bit isn't going to work as all network cards already have a
'mac' property to set the mac address. Remove it from mask and add in
get_features.
Reported-by: Gerd Hoffmann <kraxel@redhat.com> Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Jan Kiszka [Tue, 12 Jan 2010 18:53:48 +0000 (19:53 +0100)]
kvm: Detect availability of kvm_para.h
Will be required for upcoming KVM cpuid leaf. Host kernels >= 2.6.32 as
well as future kvm-kmod releases (more recent than kvm-kmod-2.6.32.3)
do/will provide them.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
The 32-bit right-shift instructions is defined to extend the shifted
output to 64-bits. A shift count of zero therefore is a simple
extension without actually shifting.
Signed-off-by: Richard Henderson <rth@twiddle.net> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
The {div,divu}2 opcodes are intended for systems for which the
division instruction produces both quotient and remainder. Sparc
is not such a system. Indeed, the remainder must be computed as
quot = a / b
rem = a - (quot * b)
Split out a tcg_out_div32 function that properly initializes Y
with the extension of the input to 64-bits. Discard the code
that used the 64-bit DIVX on sparc9/sparcv8plus without extending
the inputs to 64-bits. Implement remainders in terms of division
followed by multiplication.
Signed-off-by: Richard Henderson <rth@twiddle.net>
[blauwirbel@gmail.com: applied rth's typo fix in tcg_out_div32] Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Add a function to handle the register-vs-immediate test for arithmetic.
Also, adjust the OP_32_64 macro so that it auto-indents properly.
Rename the gen_arith32 label to gen_arith, since it handles 64-bit
arithmetic as well.
Signed-off-by: Richard Henderson <rth@twiddle.net> Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Hi all,
calling vnc_update_client in vnc_refresh might have the unlikely side
effect of setting vd->timer = NULL, if the last vnc client disconnected.
In this case we have to return from vnc_refresh without updating the
timer, otherwise we cause a segfault.
Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Anthony Liguori [Mon, 11 Jan 2010 19:27:35 +0000 (13:27 -0600)]
Update to SeaBIOS 0.5.1
5da6833 Update version to 0.5.1 669c991 Fix sign error preventing incorrect memory over 4gig calculation. 7e6bd3e Minor - better indent assembler in int1587. 48cf232 Add comment explaining why mptable is in low memory. 643062f Add int1589 support. 085debd Set FDPT in irq table even for small drives. 7c1b186 Reduce #ifs by weeding out some cross-chunk function definitions. f9b25d3 Fix vgahook sign issue; add warning to build to catch future cases. 3862b2d vgabios: Fix compile error due to fixed prototypes. 1ca05b0 Be sure to add "void" to all function prototypes that take no args. b5bb9db mptable: Reset pinmask on new bus or device. 8918989 Detect latest FC12 gcc -combine breakage. c9d3c2d Minor vga binary cleanups. 9a8609f Make MTRR region 0xc0000-0x100000 be cached. fdca418 Force a link error if a function is used from the wrong code chunk. dad41d9 Add __noreturn define for __attribute__((noreturn)). c003148 Implement native 32bit APM support. 5c99b6c Commit compiled dsdt file; misc comment updates. 29f4b91 prevent acpi from rerouting SCI interrupt 4c94b7e enumerate all PCI buses in mptable 871e0a0 Add support for 32bit PCI BIOS entry. eda2c83 Only add "addr32" to memory accesses that require them. 52a300f Introduce MODESEGMENT define; rename VISIBLE32 to VISIBLE32FLAT. fe2c3ee Allocate smbios in temp space and copy into final location. b164d2c Clear user reserved interrupts (0x60-0x66). d9104ff Remove pci_bios_bigmem_addr; set pci_bios_mem_addr=0xe0000000 14021f2 Add initial support for ATA DMA. 8362699 Allocate mptable in temp space and copy into final location. 979862e Also report memory over 4G during init. 928d4df provide correct pci routing information in mptable afc02da Add symbolic definitions for MTRR code. fb214dc Fix yield() so it works from boot code. 2ceeec9 Fix potential build failure due to text16 section being too large. a2195e4 Increase version in preparation for next release.
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Gerd Hoffmann [Fri, 8 Jan 2010 14:25:41 +0000 (15:25 +0100)]
pci: allow loading roms via fw_cfg.
This patch adds a pci bus property 'rombar' which specifies whenever
the pci rom should be loaded via pci rom bar (default) or via fw_cfg.
The later can be used for compatibility with older qemu versions where
no pci rom bar is present.
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Gerd Hoffmann [Fri, 8 Jan 2010 14:25:40 +0000 (15:25 +0100)]
roms: rework rom loading via fw
This patch changes the way rom loading via fw_cfg is handled.
Instead of having pc_init1() call a function which passed all
roms to the firmware config we simply pass a pointer to fw_cfg
to the rom loader.
Advantage: loading roms via firmware works also for devices which
are initialized after pc_init1(), i.e. everyting added via -device.
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Thomas Horsten [Thu, 7 Jan 2010 17:01:28 +0000 (17:01 +0000)]
Handle TFTP ERROR from client
If a PXE client only wants to find out the size of a file, it will
open the file and then abort the transfer by sending a TFTP ERROR packet.
The ERROR packet should cause qemu to terminate the session. If not,
the sessions will soon run out and cause timeouts in the client.
Also, if a TFTP session already exists with same IP/UDP port, it
should be terminated when a new RRQ is received, instead of creating a
duplicate (which will never be used).
A patch for gPXE to send the ERROR packet is also being submitted to
gPXE. Together they resolve slowness/hanging when booting pxegrub from
qemu's internal TFTP server. The patch from Milan Plzik to return
after sending OACK is also required for a complete fix.
Signed-off-by: Thomas Horsten <thomas@horsten.com> Signed-off-by: Milan Plzik <milan.plzik@gmail.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Currently the dmg image format driver simply opens the images as raw
if any kind of failure happens. This is contrarty to the behaviour
of all other image formats which just return an error and let the
block core deal with it.
Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Since patch ed757e140c0ada220f213036e4497315d24ca8bct, virtio will
sometimes clear all status registers on bus master disable, which loses
information such as VIRTIO_CONFIG_S_FAILED bit. This is a result of
a patch being misapplied: code uses ! instead of ~ for bit
operations as in Yan's original patch. This obviously does not make
sense.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Add feature bits as properties to virtio. This makes it possible to e.g. define
machine without indirect buffer support, which is required for 0.10
compatibility, or without hardware checksum support, which is required for 0.11
compatibility. Since default values for optional features are now set by qdev,
get_features callback has been modified: it sets non-optional bits, and clears
bits not supported by host.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Acked-by: Gerd Hoffmann <kraxel@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
This adds "bit" property type, which is a boolean stored in a 32 bit
integer field, with legal values on and off. Will be used by virtio for
feature bits.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com> Acked-by: Gerd Hoffmann <kraxel@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Gerd Hoffmann [Wed, 6 Jan 2010 16:08:04 +0000 (17:08 +0100)]
lsi: pass lsi_request to lsi_reselect
All callers of lsi_reselect have a lsi_request struct at hand anyway.
So just pass it directly instead of having lsi_reselect search for it
using the tag.
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Gerd Hoffmann [Wed, 6 Jan 2010 16:08:01 +0000 (17:08 +0100)]
lsi: have lsi_request for the whole life time of the request.
Right now lsi_request is allocated when a request is queued and released
when a request is unqueued. With this patch applied the lsi_request is
kept for the whole lifetime of the scsi request.
Rationale: We can use it for per-request data then. The patch does that
already for the request tag.
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Roland Dreier [Wed, 6 Jan 2010 04:43:34 +0000 (20:43 -0800)]
vmware_vga: Check cursor dimensions passed from guest to avoid buffer overflow
Check that the cursor dimensions passed from the guest for the
DEFINE_CURSOR command don't overflow the available space in the
cursor.image[] or cursor.mask[] arrays before copying data from the
guest into those arrays.
Signed-off-by: Roland Dreier <rolandd@cisco.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Huang Ying [Tue, 5 Jan 2010 08:32:24 +0000 (16:32 +0800)]
MCE: Fix bug of IA32_MCG_STATUS after system reset
Now, if we inject a fatal MCE into guest OS, for example Linux, Linux
will go panic and then reboot. But if we inject another MCE now,
system will reset directly instead of go panic firstly, because
MCG_STATUS.MCIP is set to 1 and not cleared after reboot. This is does
not follow the behavior in real hardware.
This patch fixes this via set env->mcg_status to 0 during system reset.
Signed-off-by: Huang Ying <ying.huang@intel.com> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
projects / qemu.git / log