INTEL_VGA_DEVICE(0x016a, &intel_ivybridge_d_info), /* GT2 server */
INTEL_VGA_DEVICE(0x0402, &intel_haswell_d_info), /* GT1 desktop */
INTEL_VGA_DEVICE(0x0412, &intel_haswell_d_info), /* GT2 desktop */
+ INTEL_VGA_DEVICE(0x0422, &intel_haswell_d_info), /* GT2 desktop */
INTEL_VGA_DEVICE(0x040a, &intel_haswell_d_info), /* GT1 server */
INTEL_VGA_DEVICE(0x041a, &intel_haswell_d_info), /* GT2 server */
+ INTEL_VGA_DEVICE(0x042a, &intel_haswell_d_info), /* GT2 server */
INTEL_VGA_DEVICE(0x0406, &intel_haswell_m_info), /* GT1 mobile */
INTEL_VGA_DEVICE(0x0416, &intel_haswell_m_info), /* GT2 mobile */
- INTEL_VGA_DEVICE(0x0c16, &intel_haswell_d_info), /* SDV */
+ INTEL_VGA_DEVICE(0x0426, &intel_haswell_m_info), /* GT2 mobile */
+ INTEL_VGA_DEVICE(0x0C02, &intel_haswell_d_info), /* SDV GT1 desktop */
+ INTEL_VGA_DEVICE(0x0C12, &intel_haswell_d_info), /* SDV GT2 desktop */
+ INTEL_VGA_DEVICE(0x0C22, &intel_haswell_d_info), /* SDV GT2 desktop */
+ INTEL_VGA_DEVICE(0x0C0A, &intel_haswell_d_info), /* SDV GT1 server */
+ INTEL_VGA_DEVICE(0x0C1A, &intel_haswell_d_info), /* SDV GT2 server */
+ INTEL_VGA_DEVICE(0x0C2A, &intel_haswell_d_info), /* SDV GT2 server */
+ INTEL_VGA_DEVICE(0x0C06, &intel_haswell_m_info), /* SDV GT1 mobile */
+ INTEL_VGA_DEVICE(0x0C16, &intel_haswell_m_info), /* SDV GT2 mobile */
+ INTEL_VGA_DEVICE(0x0C26, &intel_haswell_m_info), /* SDV GT2 mobile */
+ INTEL_VGA_DEVICE(0x0A02, &intel_haswell_d_info), /* ULT GT1 desktop */
+ INTEL_VGA_DEVICE(0x0A12, &intel_haswell_d_info), /* ULT GT2 desktop */
+ INTEL_VGA_DEVICE(0x0A22, &intel_haswell_d_info), /* ULT GT2 desktop */
+ INTEL_VGA_DEVICE(0x0A0A, &intel_haswell_d_info), /* ULT GT1 server */
+ INTEL_VGA_DEVICE(0x0A1A, &intel_haswell_d_info), /* ULT GT2 server */
+ INTEL_VGA_DEVICE(0x0A2A, &intel_haswell_d_info), /* ULT GT2 server */
+ INTEL_VGA_DEVICE(0x0A06, &intel_haswell_m_info), /* ULT GT1 mobile */
+ INTEL_VGA_DEVICE(0x0A16, &intel_haswell_m_info), /* ULT GT2 mobile */
+ INTEL_VGA_DEVICE(0x0A26, &intel_haswell_m_info), /* ULT GT2 mobile */
+ INTEL_VGA_DEVICE(0x0D12, &intel_haswell_d_info), /* CRW GT1 desktop */
+ INTEL_VGA_DEVICE(0x0D22, &intel_haswell_d_info), /* CRW GT2 desktop */
+ INTEL_VGA_DEVICE(0x0D32, &intel_haswell_d_info), /* CRW GT2 desktop */
+ INTEL_VGA_DEVICE(0x0D1A, &intel_haswell_d_info), /* CRW GT1 server */
+ INTEL_VGA_DEVICE(0x0D2A, &intel_haswell_d_info), /* CRW GT2 server */
+ INTEL_VGA_DEVICE(0x0D3A, &intel_haswell_d_info), /* CRW GT2 server */
+ INTEL_VGA_DEVICE(0x0D16, &intel_haswell_m_info), /* CRW GT1 mobile */
+ INTEL_VGA_DEVICE(0x0D26, &intel_haswell_m_info), /* CRW GT2 mobile */
+ INTEL_VGA_DEVICE(0x0D36, &intel_haswell_m_info), /* CRW GT2 mobile */
INTEL_VGA_DEVICE(0x0f30, &intel_valleyview_m_info),
INTEL_VGA_DEVICE(0x0157, &intel_valleyview_m_info),
INTEL_VGA_DEVICE(0x0155, &intel_valleyview_d_info),
* This should make it easier to transition modules over to the
* new register block scheme, since we can do it incrementally.
*/
- if (reg >= 0x180000)
+ if (reg >= VLV_DISPLAY_BASE)
return false;
if (reg >= RENDER_RING_BASE &&
__i915_write(32, l)
__i915_write(64, q)
#undef __i915_write
+
+static const struct register_whitelist {
+ uint64_t offset;
+ uint32_t size;
+ uint32_t gen_bitmask; /* support gens, 0x10 for 4, 0x30 for 4 and 5, etc. */
+} whitelist[] = {
+ { RING_TIMESTAMP(RENDER_RING_BASE), 8, 0xF0 },
+};
+
+int i915_reg_read_ioctl(struct drm_device *dev,
+ void *data, struct drm_file *file)
+{
+ struct drm_i915_private *dev_priv = dev->dev_private;
+ struct drm_i915_reg_read *reg = data;
+ struct register_whitelist const *entry = whitelist;
+ int i;
+
+ for (i = 0; i < ARRAY_SIZE(whitelist); i++, entry++) {
+ if (entry->offset == reg->offset &&
+ (1 << INTEL_INFO(dev)->gen & entry->gen_bitmask))
+ break;
+ }
+
+ if (i == ARRAY_SIZE(whitelist))
+ return -EINVAL;
+
+ switch (entry->size) {
+ case 8:
+ reg->val = I915_READ64(reg->offset);
+ break;
+ case 4:
+ reg->val = I915_READ(reg->offset);
+ break;
+ case 2:
+ reg->val = I915_READ16(reg->offset);
+ break;
+ case 1:
+ reg->val = I915_READ8(reg->offset);
+ break;
+ default:
+ WARN_ON(1);
+ return -EINVAL;
+ }
+
+ return 0;
+}
#include "intel_drv.h"
#include <linux/dma_remapping.h>
-struct change_domains {
- uint32_t invalidate_domains;
- uint32_t flush_domains;
- uint32_t flush_rings;
- uint32_t flips;
-};
-
-/*
- * Set the next domain for the specified object. This
- * may not actually perform the necessary flushing/invaliding though,
- * as that may want to be batched with other set_domain operations
- *
- * This is (we hope) the only really tricky part of gem. The goal
- * is fairly simple -- track which caches hold bits of the object
- * and make sure they remain coherent. A few concrete examples may
- * help to explain how it works. For shorthand, we use the notation
- * (read_domains, write_domain), e.g. (CPU, CPU) to indicate the
- * a pair of read and write domain masks.
- *
- * Case 1: the batch buffer
- *
- * 1. Allocated
- * 2. Written by CPU
- * 3. Mapped to GTT
- * 4. Read by GPU
- * 5. Unmapped from GTT
- * 6. Freed
- *
- * Let's take these a step at a time
- *
- * 1. Allocated
- * Pages allocated from the kernel may still have
- * cache contents, so we set them to (CPU, CPU) always.
- * 2. Written by CPU (using pwrite)
- * The pwrite function calls set_domain (CPU, CPU) and
- * this function does nothing (as nothing changes)
- * 3. Mapped by GTT
- * This function asserts that the object is not
- * currently in any GPU-based read or write domains
- * 4. Read by GPU
- * i915_gem_execbuffer calls set_domain (COMMAND, 0).
- * As write_domain is zero, this function adds in the
- * current read domains (CPU+COMMAND, 0).
- * flush_domains is set to CPU.
- * invalidate_domains is set to COMMAND
- * clflush is run to get data out of the CPU caches
- * then i915_dev_set_domain calls i915_gem_flush to
- * emit an MI_FLUSH and drm_agp_chipset_flush
- * 5. Unmapped from GTT
- * i915_gem_object_unbind calls set_domain (CPU, CPU)
- * flush_domains and invalidate_domains end up both zero
- * so no flushing/invalidating happens
- * 6. Freed
- * yay, done
- *
- * Case 2: The shared render buffer
- *
- * 1. Allocated
- * 2. Mapped to GTT
- * 3. Read/written by GPU
- * 4. set_domain to (CPU,CPU)
- * 5. Read/written by CPU
- * 6. Read/written by GPU
- *
- * 1. Allocated
- * Same as last example, (CPU, CPU)
- * 2. Mapped to GTT
- * Nothing changes (assertions find that it is not in the GPU)
- * 3. Read/written by GPU
- * execbuffer calls set_domain (RENDER, RENDER)
- * flush_domains gets CPU
- * invalidate_domains gets GPU
- * clflush (obj)
- * MI_FLUSH and drm_agp_chipset_flush
- * 4. set_domain (CPU, CPU)
- * flush_domains gets GPU
- * invalidate_domains gets CPU
- * wait_rendering (obj) to make sure all drawing is complete.
- * This will include an MI_FLUSH to get the data from GPU
- * to memory
- * clflush (obj) to invalidate the CPU cache
- * Another MI_FLUSH in i915_gem_flush (eliminate this somehow?)
- * 5. Read/written by CPU
- * cache lines are loaded and dirtied
- * 6. Read written by GPU
- * Same as last GPU access
- *
- * Case 3: The constant buffer
- *
- * 1. Allocated
- * 2. Written by CPU
- * 3. Read by GPU
- * 4. Updated (written) by CPU again
- * 5. Read by GPU
- *
- * 1. Allocated
- * (CPU, CPU)
- * 2. Written by CPU
- * (CPU, CPU)
- * 3. Read by GPU
- * (CPU+RENDER, 0)
- * flush_domains = CPU
- * invalidate_domains = RENDER
- * clflush (obj)
- * MI_FLUSH
- * drm_agp_chipset_flush
- * 4. Updated (written) by CPU again
- * (CPU, CPU)
- * flush_domains = 0 (no previous write domain)
- * invalidate_domains = 0 (no new read domains)
- * 5. Read by GPU
- * (CPU+RENDER, 0)
- * flush_domains = CPU
- * invalidate_domains = RENDER
- * clflush (obj)
- * MI_FLUSH
- * drm_agp_chipset_flush
- */
-static void
-i915_gem_object_set_to_gpu_domain(struct drm_i915_gem_object *obj,
- struct intel_ring_buffer *ring,
- struct change_domains *cd)
-{
- uint32_t invalidate_domains = 0, flush_domains = 0;
-
- /*
- * If the object isn't moving to a new write domain,
- * let the object stay in multiple read domains
- */
- if (obj->base.pending_write_domain == 0)
- obj->base.pending_read_domains |= obj->base.read_domains;
-
- /*
- * Flush the current write domain if
- * the new read domains don't match. Invalidate
- * any read domains which differ from the old
- * write domain
- */
- if (obj->base.write_domain &&
- (((obj->base.write_domain != obj->base.pending_read_domains ||
- obj->ring != ring)) ||
- (obj->fenced_gpu_access && !obj->pending_fenced_gpu_access))) {
- flush_domains |= obj->base.write_domain;
- invalidate_domains |=
- obj->base.pending_read_domains & ~obj->base.write_domain;
- }
- /*
- * Invalidate any read caches which may have
- * stale data. That is, any new read domains.
- */
- invalidate_domains |= obj->base.pending_read_domains & ~obj->base.read_domains;
- if ((flush_domains | invalidate_domains) & I915_GEM_DOMAIN_CPU)
- i915_gem_clflush_object(obj);
-
- if (obj->base.pending_write_domain)
- cd->flips |= atomic_read(&obj->pending_flip);
-
- /* The actual obj->write_domain will be updated with
- * pending_write_domain after we emit the accumulated flush for all
- * of our domain changes in execbuffers (which clears objects'
- * write_domains). So if we have a current write domain that we
- * aren't changing, set pending_write_domain to that.
- */
- if (flush_domains == 0 && obj->base.pending_write_domain == 0)
- obj->base.pending_write_domain = obj->base.write_domain;
-
- cd->invalidate_domains |= invalidate_domains;
- cd->flush_domains |= flush_domains;
- if (flush_domains & I915_GEM_GPU_DOMAINS)
- cd->flush_rings |= intel_ring_flag(obj->ring);
- if (invalidate_domains & I915_GEM_GPU_DOMAINS)
- cd->flush_rings |= intel_ring_flag(ring);
-}
-
struct eb_objects {
int and;
struct hlist_head buckets[0];
target_i915_obj = to_intel_bo(target_obj);
target_offset = target_i915_obj->gtt_offset;
+ /* Sandybridge PPGTT errata: We need a global gtt mapping for MI and
+ * pipe_control writes because the gpu doesn't properly redirect them
+ * through the ppgtt for non_secure batchbuffers. */
+ if (unlikely(IS_GEN6(dev) &&
+ reloc->write_domain == I915_GEM_DOMAIN_INSTRUCTION &&
+ !target_i915_obj->has_global_gtt_mapping)) {
+ i915_gem_gtt_bind_object(target_i915_obj,
+ target_i915_obj->cache_level);
+ }
+
/* The target buffer should have appeared before us in the
* exec_object list, so it should have a GTT space bound by now.
*/
io_mapping_unmap_atomic(reloc_page);
}
- /* Sandybridge PPGTT errata: We need a global gtt mapping for MI and
- * pipe_control writes because the gpu doesn't properly redirect them
- * through the ppgtt for non_secure batchbuffers. */
- if (unlikely(IS_GEN6(dev) &&
- reloc->write_domain == I915_GEM_DOMAIN_INSTRUCTION &&
- !target_i915_obj->has_global_gtt_mapping)) {
- i915_gem_gtt_bind_object(target_i915_obj,
- target_i915_obj->cache_level);
- }
-
/* and update the user's relocation entry */
reloc->presumed_offset = target_offset;
obj->base.pending_read_domains = 0;
obj->base.pending_write_domain = 0;
+ obj->pending_fenced_gpu_access = false;
}
list_splice(&ordered_objects, objects);
return ret;
}
-static void
-i915_gem_execbuffer_flush(struct drm_device *dev,
- uint32_t invalidate_domains,
- uint32_t flush_domains)
-{
- if (flush_domains & I915_GEM_DOMAIN_CPU)
- intel_gtt_chipset_flush();
-
- if (flush_domains & I915_GEM_DOMAIN_GTT)
- wmb();
-}
-
static int
i915_gem_execbuffer_wait_for_flips(struct intel_ring_buffer *ring, u32 flips)
{
return 0;
}
-
static int
i915_gem_execbuffer_move_to_gpu(struct intel_ring_buffer *ring,
struct list_head *objects)
{
struct drm_i915_gem_object *obj;
- struct change_domains cd;
+ uint32_t flush_domains = 0;
+ uint32_t flips = 0;
int ret;
- memset(&cd, 0, sizeof(cd));
- list_for_each_entry(obj, objects, exec_list)
- i915_gem_object_set_to_gpu_domain(obj, ring, &cd);
-
- if (cd.invalidate_domains | cd.flush_domains) {
- i915_gem_execbuffer_flush(ring->dev,
- cd.invalidate_domains,
- cd.flush_domains);
- }
-
- if (cd.flips) {
- ret = i915_gem_execbuffer_wait_for_flips(ring, cd.flips);
+ list_for_each_entry(obj, objects, exec_list) {
+ ret = i915_gem_object_sync(obj, ring);
if (ret)
return ret;
+
+ if (obj->base.write_domain & I915_GEM_DOMAIN_CPU)
+ i915_gem_clflush_object(obj);
+
+ if (obj->base.pending_write_domain)
+ flips |= atomic_read(&obj->pending_flip);
+
+ flush_domains |= obj->base.write_domain;
}
- list_for_each_entry(obj, objects, exec_list) {
- ret = i915_gem_object_sync(obj, ring);
+ if (flips) {
+ ret = i915_gem_execbuffer_wait_for_flips(ring, flips);
if (ret)
return ret;
}
+ if (flush_domains & I915_GEM_DOMAIN_CPU)
+ intel_gtt_chipset_flush();
+
+ if (flush_domains & I915_GEM_DOMAIN_GTT)
+ wmb();
+
/* Unconditionally invalidate gpu caches and ensure that we do flush
* any residual writes from the previous batch.
*/
- ret = i915_gem_flush_ring(ring,
- I915_GEM_GPU_DOMAINS,
- ring->gpu_caches_dirty ? I915_GEM_GPU_DOMAINS : 0);
- if (ret)
- return ret;
-
- ring->gpu_caches_dirty = false;
- return 0;
+ return intel_ring_invalidate_all_caches(ring);
}
static bool
struct drm_i915_gem_object *obj;
list_for_each_entry(obj, objects, exec_list) {
- u32 old_read = obj->base.read_domains;
- u32 old_write = obj->base.write_domain;
-
+ u32 old_read = obj->base.read_domains;
+ u32 old_write = obj->base.write_domain;
obj->base.read_domains = obj->base.pending_read_domains;
obj->base.write_domain = obj->base.pending_write_domain;
i915_gem_object_move_to_active(obj, ring, seqno);
if (obj->base.write_domain) {
obj->dirty = 1;
- obj->pending_gpu_write = true;
- list_move_tail(&obj->gpu_write_list,
- &ring->gpu_write_list);
+ obj->last_write_seqno = seqno;
if (obj->pin_count) /* check for potential scanout */
- intel_mark_busy(ring->dev, obj);
+ intel_mark_fb_busy(obj);
}
trace_i915_gem_object_change_domain(obj, old_read, old_write);
}
-
- intel_mark_busy(ring->dev, NULL);
}
static void
struct drm_file *file,
struct intel_ring_buffer *ring)
{
- struct drm_i915_gem_request *request;
-
/* Unconditionally force add_request to emit a full flush. */
ring->gpu_caches_dirty = true;
/* Add a breadcrumb for the completion of the batch buffer */
- request = kzalloc(sizeof(*request), GFP_KERNEL);
- if (request == NULL || i915_add_request(ring, file, request)) {
- kfree(request);
- }
+ (void)i915_add_request(ring, file, NULL);
}
static int
struct drm_device *dev = obj->base.dev;
struct drm_i915_private *dev_priv = dev->dev_private;
- if (dev_priv->mm.gtt->needs_dmar)
+ /* don't map imported dma buf objects */
+ if (dev_priv->mm.gtt->needs_dmar && !obj->sg_table)
return intel_gtt_map_memory(obj->pages,
obj->base.size >> PAGE_SHIFT,
&obj->sg_list,
undo_idling(dev_priv, interruptible);
}
+static void i915_gtt_color_adjust(struct drm_mm_node *node,
+ unsigned long color,
+ unsigned long *start,
+ unsigned long *end)
+{
+ if (node->color != color)
+ *start += 4096;
+
+ if (!list_empty(&node->node_list)) {
+ node = list_entry(node->node_list.next,
+ struct drm_mm_node,
+ node_list);
+ if (node->allocated && node->color != color)
+ *end -= 4096;
+ }
+}
+
void i915_gem_init_global_gtt(struct drm_device *dev,
unsigned long start,
unsigned long mappable_end,
/* Substract the guard page ... */
drm_mm_init(&dev_priv->mm.gtt_space, start, end - start - PAGE_SIZE);
+ if (!HAS_LLC(dev))
+ dev_priv->mm.gtt_space.color_adjust = i915_gtt_color_adjust;
dev_priv->mm.gtt_start = start;
dev_priv->mm.gtt_mappable_end = mappable_end;
drm_helper_hpd_irq_event(dev);
}
-static void i915_handle_rps_change(struct drm_device *dev)
+/* defined intel_pm.c */
+extern spinlock_t mchdev_lock;
+
+static void ironlake_handle_rps_change(struct drm_device *dev)
{
drm_i915_private_t *dev_priv = dev->dev_private;
u32 busy_up, busy_down, max_avg, min_avg;
- u8 new_delay = dev_priv->cur_delay;
+ u8 new_delay;
+ unsigned long flags;
+
+ spin_lock_irqsave(&mchdev_lock, flags);
+
+ I915_WRITE16(MEMINTRSTS, I915_READ(MEMINTRSTS));
+
+ new_delay = dev_priv->cur_delay;
I915_WRITE16(MEMINTRSTS, MEMINT_EVAL_CHG);
busy_up = I915_READ(RCPREVBSYTUPAVG);
if (ironlake_set_drps(dev, new_delay))
dev_priv->cur_delay = new_delay;
+ spin_unlock_irqrestore(&mchdev_lock, flags);
+
return;
}
if (ring->obj == NULL)
return;
- trace_i915_gem_request_complete(ring, ring->get_seqno(ring));
+ trace_i915_gem_request_complete(ring, ring->get_seqno(ring, false));
wake_up_all(&ring->irq_queue);
if (i915_enable_hangcheck) {
static void gen6_pm_rps_work(struct work_struct *work)
{
drm_i915_private_t *dev_priv = container_of(work, drm_i915_private_t,
- rps_work);
+ rps.work);
u32 pm_iir, pm_imr;
u8 new_delay;
- spin_lock_irq(&dev_priv->rps_lock);
- pm_iir = dev_priv->pm_iir;
- dev_priv->pm_iir = 0;
+ spin_lock_irq(&dev_priv->rps.lock);
+ pm_iir = dev_priv->rps.pm_iir;
+ dev_priv->rps.pm_iir = 0;
pm_imr = I915_READ(GEN6_PMIMR);
I915_WRITE(GEN6_PMIMR, 0);
- spin_unlock_irq(&dev_priv->rps_lock);
+ spin_unlock_irq(&dev_priv->rps.lock);
if ((pm_iir & GEN6_PM_DEFERRED_EVENTS) == 0)
return;
mutex_lock(&dev_priv->dev->struct_mutex);
if (pm_iir & GEN6_PM_RP_UP_THRESHOLD)
- new_delay = dev_priv->cur_delay + 1;
+ new_delay = dev_priv->rps.cur_delay + 1;
else
- new_delay = dev_priv->cur_delay - 1;
+ new_delay = dev_priv->rps.cur_delay - 1;
gen6_set_rps(dev_priv->dev, new_delay);
drm_i915_private_t *dev_priv = (drm_i915_private_t *) dev->dev_private;
unsigned long flags;
- if (!IS_IVYBRIDGE(dev))
+ if (!HAS_L3_GPU_CACHE(dev))
return;
spin_lock_irqsave(&dev_priv->irq_lock, flags);
* IIR bits should never already be set because IMR should
* prevent an interrupt from being shown in IIR. The warning
* displays a case where we've unsafely cleared
- * dev_priv->pm_iir. Although missing an interrupt of the same
+ * dev_priv->rps.pm_iir. Although missing an interrupt of the same
* type is not a problem, it displays a problem in the logic.
*
- * The mask bit in IMR is cleared by rps_work.
+ * The mask bit in IMR is cleared by dev_priv->rps.work.
*/
- spin_lock_irqsave(&dev_priv->rps_lock, flags);
- dev_priv->pm_iir |= pm_iir;
- I915_WRITE(GEN6_PMIMR, dev_priv->pm_iir);
+ spin_lock_irqsave(&dev_priv->rps.lock, flags);
- WARN(dev_priv->rps.pm_iir & pm_iir, "Missed a PM interrupt\n");
+ dev_priv->rps.pm_iir |= pm_iir;
+ I915_WRITE(GEN6_PMIMR, dev_priv->rps.pm_iir);
POSTING_READ(GEN6_PMIMR);
- spin_unlock_irqrestore(&dev_priv->rps_lock, flags);
+ spin_unlock_irqrestore(&dev_priv->rps.lock, flags);
- queue_work(dev_priv->wq, &dev_priv->rps_work);
+ queue_work(dev_priv->wq, &dev_priv->rps.work);
}
static irqreturn_t valleyview_irq_handler(DRM_IRQ_ARGS)
ibx_irq_handler(dev, pch_iir);
}
- if (de_iir & DE_PCU_EVENT) {
- I915_WRITE16(MEMINTRSTS, I915_READ(MEMINTRSTS));
- i915_handle_rps_change(dev);
- }
+ if (IS_GEN5(dev) && de_iir & DE_PCU_EVENT)
+ ironlake_handle_rps_change(dev);
if (IS_GEN6(dev) && pm_iir & GEN6_PM_DEFERRED_EVENTS)
gen6_queue_rps_work(dev_priv, pm_iir);
{
err->size = obj->base.size;
err->name = obj->base.name;
- err->seqno = obj->last_rendering_seqno;
+ err->rseqno = obj->last_read_seqno;
+ err->wseqno = obj->last_write_seqno;
err->gtt_offset = obj->gtt_offset;
err->read_domains = obj->base.read_domains;
err->write_domain = obj->base.write_domain;
if (!ring->get_seqno)
return NULL;
- seqno = ring->get_seqno(ring);
+ seqno = ring->get_seqno(ring, false);
list_for_each_entry(obj, &dev_priv->mm.active_list, mm_list) {
if (obj->ring != ring)
continue;
- if (i915_seqno_passed(seqno, obj->last_rendering_seqno))
+ if (i915_seqno_passed(seqno, obj->last_read_seqno))
continue;
if ((obj->base.read_domains & I915_GEM_DOMAIN_COMMAND) == 0)
error->waiting[ring->id] = waitqueue_active(&ring->irq_queue);
error->instpm[ring->id] = I915_READ(RING_INSTPM(ring->mmio_base));
- error->seqno[ring->id] = ring->get_seqno(ring);
+ error->seqno[ring->id] = ring->get_seqno(ring, false);
error->acthd[ring->id] = intel_ring_get_active_head(ring);
error->head[ring->id] = I915_READ_HEAD(ring);
error->tail[ring->id] = I915_READ_TAIL(ring);
static bool i915_hangcheck_ring_idle(struct intel_ring_buffer *ring, bool *err)
{
if (list_empty(&ring->request_list) ||
- i915_seqno_passed(ring->get_seqno(ring), ring_last_seqno(ring))) {
+ i915_seqno_passed(ring->get_seqno(ring, false),
+ ring_last_seqno(ring))) {
/* Issue a wake-up to catch stuck h/w. */
if (waitqueue_active(&ring->irq_queue)) {
DRM_ERROR("Hangcheck timer elapsed... %s idle\n",
INIT_WORK(&dev_priv->hotplug_work, i915_hotplug_work_func);
INIT_WORK(&dev_priv->error_work, i915_error_work_func);
- INIT_WORK(&dev_priv->rps_work, gen6_pm_rps_work);
+ INIT_WORK(&dev_priv->rps.work, gen6_pm_rps_work);
INIT_WORK(&dev_priv->parity_error_work, ivybridge_parity_work);
dev->driver->get_vblank_counter = i915_get_vblank_counter;
#include "intel_drv.h"
#include "i915_drv.h"
+ #ifdef CONFIG_PM
static u32 calc_residency(struct drm_device *dev, const u32 reg)
{
struct drm_i915_private *dev_priv = dev->dev_private;
DRM_ERROR("RC6 residency sysfs setup failed\n");
}
- if (IS_IVYBRIDGE(dev)) {
+ if (HAS_L3_GPU_CACHE(dev)) {
ret = device_create_bin_file(&dev->primary->kdev, &dpf_attrs);
if (ret)
DRM_ERROR("l3 parity sysfs setup failed\n");
device_remove_bin_file(&dev->primary->kdev, &dpf_attrs);
sysfs_unmerge_group(&dev->primary->kdev.kobj, &rc6_attr_group);
}
+ #else
+ void i915_setup_sysfs(struct drm_device *dev)
+ {
+ return;
+ }
+
+ void i915_teardown_sysfs(struct drm_device *dev)
+ {
+ return;
+ }
+ #endif /* CONFIG_PM */
unsigned long bestppm, ppm, absppm;
int dotclk, flag;
+ flag = 0;
dotclk = target * 1000;
bestppm = 1000000;
ppm = absppm = 0;
* protect mechanism may be enabled.
*
* Note! This is for pre-ILK only.
+ *
+ * Unfortunately needed by dvo_ns2501 since the dvo depends on it running.
*/
-static void intel_enable_pll(struct drm_i915_private *dev_priv, enum pipe pipe)
+void intel_enable_pll(struct drm_i915_private *dev_priv, enum pipe pipe)
{
int reg;
u32 val;
static bool intel_crtc_driving_pch(struct drm_crtc *crtc)
{
struct drm_device *dev = crtc->dev;
- struct intel_encoder *encoder;
+ struct intel_encoder *intel_encoder;
/*
* If there's a non-PCH eDP on this crtc, it must be DP_A, and that
* must be driven by its own crtc; no sharing is possible.
*/
- for_each_encoder_on_crtc(dev, crtc, encoder) {
+ for_each_encoder_on_crtc(dev, crtc, intel_encoder) {
/* On Haswell, LPT PCH handles the VGA connection via FDI, and Haswell
* CPU handles all others */
/* It is still unclear how this will work on PPT, so throw up a warning */
WARN_ON(!HAS_PCH_LPT(dev));
- if (encoder->type == DRM_MODE_ENCODER_DAC) {
+ if (intel_encoder->type == INTEL_OUTPUT_ANALOG) {
DRM_DEBUG_KMS("Haswell detected DAC encoder, assuming is PCH\n");
return true;
} else {
DRM_DEBUG_KMS("Haswell detected encoder %d, assuming is CPU\n",
- encoder->type);
+ intel_encoder->type);
return false;
}
}
- switch (encoder->type) {
+ switch (intel_encoder->type) {
case INTEL_OUTPUT_EDP:
- if (!intel_encoder_is_pch_edp(&encoder->base))
+ if (!intel_encoder_is_pch_edp(&intel_encoder->base))
return false;
continue;
}
continue;
}
- if (intel_encoder->type == INTEL_OUTPUT_EDP) {
- /* Use VBT settings if we have an eDP panel */
- unsigned int edp_bpc = dev_priv->edp.bpp / 3;
-
- if (edp_bpc < display_bpc) {
- DRM_DEBUG_KMS("clamping display bpc (was %d) to eDP (%d)\n", display_bpc, edp_bpc);
- display_bpc = edp_bpc;
- }
- continue;
- }
-
/* Not one of the known troublemakers, check the EDID */
list_for_each_entry(connector, &dev->mode_config.connector_list,
head) {
return mode;
}
-#define GPU_IDLE_TIMEOUT 500 /* ms */
-
-/* When this timer fires, we've been idle for awhile */
-static void intel_gpu_idle_timer(unsigned long arg)
-{
- struct drm_device *dev = (struct drm_device *)arg;
- drm_i915_private_t *dev_priv = dev->dev_private;
-
- if (!list_empty(&dev_priv->mm.active_list)) {
- /* Still processing requests, so just re-arm the timer. */
- mod_timer(&dev_priv->idle_timer, jiffies +
- msecs_to_jiffies(GPU_IDLE_TIMEOUT));
- return;
- }
-
- dev_priv->busy = false;
- queue_work(dev_priv->wq, &dev_priv->idle_work);
-}
-
-#define CRTC_IDLE_TIMEOUT 1000 /* ms */
-
-static void intel_crtc_idle_timer(unsigned long arg)
-{
- struct intel_crtc *intel_crtc = (struct intel_crtc *)arg;
- struct drm_crtc *crtc = &intel_crtc->base;
- drm_i915_private_t *dev_priv = crtc->dev->dev_private;
- struct intel_framebuffer *intel_fb;
-
- intel_fb = to_intel_framebuffer(crtc->fb);
- if (intel_fb && intel_fb->obj->active) {
- /* The framebuffer is still being accessed by the GPU. */
- mod_timer(&intel_crtc->idle_timer, jiffies +
- msecs_to_jiffies(CRTC_IDLE_TIMEOUT));
- return;
- }
-
- intel_crtc->busy = false;
- queue_work(dev_priv->wq, &dev_priv->idle_work);
-}
-
static void intel_increase_pllclock(struct drm_crtc *crtc)
{
struct drm_device *dev = crtc->dev;
if (dpll & DISPLAY_RATE_SELECT_FPA1)
DRM_DEBUG_DRIVER("failed to upclock LVDS!\n");
}
-
- /* Schedule downclock */
- mod_timer(&intel_crtc->idle_timer, jiffies +
- msecs_to_jiffies(CRTC_IDLE_TIMEOUT));
}
static void intel_decrease_pllclock(struct drm_crtc *crtc)
}
-/**
- * intel_idle_update - adjust clocks for idleness
- * @work: work struct
- *
- * Either the GPU or display (or both) went idle. Check the busy status
- * here and adjust the CRTC and GPU clocks as necessary.
- */
-static void intel_idle_update(struct work_struct *work)
+void intel_mark_busy(struct drm_device *dev)
+{
+ i915_update_gfx_val(dev->dev_private);
+}
+
+void intel_mark_idle(struct drm_device *dev)
{
- drm_i915_private_t *dev_priv = container_of(work, drm_i915_private_t,
- idle_work);
- struct drm_device *dev = dev_priv->dev;
+}
+
+void intel_mark_fb_busy(struct drm_i915_gem_object *obj)
+{
+ struct drm_device *dev = obj->base.dev;
struct drm_crtc *crtc;
- struct intel_crtc *intel_crtc;
if (!i915_powersave)
return;
- mutex_lock(&dev->struct_mutex);
-
- i915_update_gfx_val(dev_priv);
-
list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
- /* Skip inactive CRTCs */
if (!crtc->fb)
continue;
- intel_crtc = to_intel_crtc(crtc);
- if (!intel_crtc->busy)
- intel_decrease_pllclock(crtc);
+ if (to_intel_framebuffer(crtc->fb)->obj == obj)
+ intel_increase_pllclock(crtc);
}
-
-
- mutex_unlock(&dev->struct_mutex);
}
-/**
- * intel_mark_busy - mark the GPU and possibly the display busy
- * @dev: drm device
- * @obj: object we're operating on
- *
- * Callers can use this function to indicate that the GPU is busy processing
- * commands. If @obj matches one of the CRTC objects (i.e. it's a scanout
- * buffer), we'll also mark the display as busy, so we know to increase its
- * clock frequency.
- */
-void intel_mark_busy(struct drm_device *dev, struct drm_i915_gem_object *obj)
+void intel_mark_fb_idle(struct drm_i915_gem_object *obj)
{
- drm_i915_private_t *dev_priv = dev->dev_private;
- struct drm_crtc *crtc = NULL;
- struct intel_framebuffer *intel_fb;
- struct intel_crtc *intel_crtc;
-
- if (!drm_core_check_feature(dev, DRIVER_MODESET))
- return;
-
- if (!dev_priv->busy) {
- intel_sanitize_pm(dev);
- dev_priv->busy = true;
- } else
- mod_timer(&dev_priv->idle_timer, jiffies +
- msecs_to_jiffies(GPU_IDLE_TIMEOUT));
+ struct drm_device *dev = obj->base.dev;
+ struct drm_crtc *crtc;
- if (obj == NULL)
+ if (!i915_powersave)
return;
list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
if (!crtc->fb)
continue;
- intel_crtc = to_intel_crtc(crtc);
- intel_fb = to_intel_framebuffer(crtc->fb);
- if (intel_fb->obj == obj) {
- if (!intel_crtc->busy) {
- /* Non-busy -> busy, upclock */
- intel_increase_pllclock(crtc);
- intel_crtc->busy = true;
- } else {
- /* Busy -> busy, put off timer */
- mod_timer(&intel_crtc->idle_timer, jiffies +
- msecs_to_jiffies(CRTC_IDLE_TIMEOUT));
- }
- }
+ if (to_intel_framebuffer(crtc->fb)->obj == obj)
+ intel_decrease_pllclock(crtc);
}
}
default:
WARN_ONCE(1, "unknown plane in flip command\n");
ret = -ENODEV;
- goto err;
+ goto err_unpin;
}
ret = intel_ring_begin(ring, 4);
goto cleanup_pending;
intel_disable_fbc(dev);
- intel_mark_busy(dev, obj);
+ intel_mark_fb_busy(obj);
mutex_unlock(&dev->struct_mutex);
trace_i915_flip_request(intel_crtc->plane, obj);
}
drm_crtc_helper_add(&intel_crtc->base, &intel_helper_funcs);
-
- intel_crtc->busy = false;
-
- setup_timer(&intel_crtc->idle_timer, intel_crtc_idle_timer,
- (unsigned long)intel_crtc);
}
int intel_get_pipe_from_crtc_id(struct drm_device *dev, void *data,
return 0;
}
-static int intel_encoder_clones(struct drm_device *dev, int type_mask)
+static int intel_encoder_clones(struct intel_encoder *encoder)
{
- struct intel_encoder *encoder;
+ struct drm_device *dev = encoder->base.dev;
+ struct intel_encoder *source_encoder;
int index_mask = 0;
int entry = 0;
- list_for_each_entry(encoder, &dev->mode_config.encoder_list, base.head) {
- if (type_mask & encoder->clone_mask)
+ list_for_each_entry(source_encoder,
+ &dev->mode_config.encoder_list, base.head) {
+
+ if (encoder == source_encoder)
index_mask |= (1 << entry);
+
+ /* Intel hw has only one MUX where enocoders could be cloned. */
+ if (encoder->cloneable && source_encoder->cloneable)
+ index_mask |= (1 << entry);
+
entry++;
}
dpd_is_edp = intel_dpd_is_edp(dev);
if (has_edp_a(dev))
- intel_dp_init(dev, DP_A);
+ intel_dp_init(dev, DP_A, PORT_A);
if (dpd_is_edp && (I915_READ(PCH_DP_D) & DP_DETECTED))
- intel_dp_init(dev, PCH_DP_D);
+ intel_dp_init(dev, PCH_DP_D, PORT_D);
}
intel_crt_init(dev);
/* PCH SDVOB multiplex with HDMIB */
found = intel_sdvo_init(dev, PCH_SDVOB, true);
if (!found)
- intel_hdmi_init(dev, HDMIB);
+ intel_hdmi_init(dev, HDMIB, PORT_B);
if (!found && (I915_READ(PCH_DP_B) & DP_DETECTED))
- intel_dp_init(dev, PCH_DP_B);
+ intel_dp_init(dev, PCH_DP_B, PORT_B);
}
if (I915_READ(HDMIC) & PORT_DETECTED)
- intel_hdmi_init(dev, HDMIC);
+ intel_hdmi_init(dev, HDMIC, PORT_C);
if (!dpd_is_edp && I915_READ(HDMID) & PORT_DETECTED)
- intel_hdmi_init(dev, HDMID);
+ intel_hdmi_init(dev, HDMID, PORT_D);
if (I915_READ(PCH_DP_C) & DP_DETECTED)
- intel_dp_init(dev, PCH_DP_C);
+ intel_dp_init(dev, PCH_DP_C, PORT_C);
if (!dpd_is_edp && (I915_READ(PCH_DP_D) & DP_DETECTED))
- intel_dp_init(dev, PCH_DP_D);
+ intel_dp_init(dev, PCH_DP_D, PORT_D);
} else if (IS_VALLEYVIEW(dev)) {
int found;
/* SDVOB multiplex with HDMIB */
found = intel_sdvo_init(dev, SDVOB, true);
if (!found)
- intel_hdmi_init(dev, SDVOB);
+ intel_hdmi_init(dev, SDVOB, PORT_B);
if (!found && (I915_READ(DP_B) & DP_DETECTED))
- intel_dp_init(dev, DP_B);
+ intel_dp_init(dev, DP_B, PORT_B);
}
if (I915_READ(SDVOC) & PORT_DETECTED)
- intel_hdmi_init(dev, SDVOC);
+ intel_hdmi_init(dev, SDVOC, PORT_C);
/* Shares lanes with HDMI on SDVOC */
if (I915_READ(DP_C) & DP_DETECTED)
- intel_dp_init(dev, DP_C);
+ intel_dp_init(dev, DP_C, PORT_C);
} else if (SUPPORTS_DIGITAL_OUTPUTS(dev)) {
bool found = false;
found = intel_sdvo_init(dev, SDVOB, true);
if (!found && SUPPORTS_INTEGRATED_HDMI(dev)) {
DRM_DEBUG_KMS("probing HDMI on SDVOB\n");
- intel_hdmi_init(dev, SDVOB);
+ intel_hdmi_init(dev, SDVOB, PORT_B);
}
if (!found && SUPPORTS_INTEGRATED_DP(dev)) {
DRM_DEBUG_KMS("probing DP_B\n");
- intel_dp_init(dev, DP_B);
+ intel_dp_init(dev, DP_B, PORT_B);
}
}
if (SUPPORTS_INTEGRATED_HDMI(dev)) {
DRM_DEBUG_KMS("probing HDMI on SDVOC\n");
- intel_hdmi_init(dev, SDVOC);
+ intel_hdmi_init(dev, SDVOC, PORT_C);
}
if (SUPPORTS_INTEGRATED_DP(dev)) {
DRM_DEBUG_KMS("probing DP_C\n");
- intel_dp_init(dev, DP_C);
+ intel_dp_init(dev, DP_C, PORT_C);
}
}
if (SUPPORTS_INTEGRATED_DP(dev) &&
(I915_READ(DP_D) & DP_DETECTED)) {
DRM_DEBUG_KMS("probing DP_D\n");
- intel_dp_init(dev, DP_D);
+ intel_dp_init(dev, DP_D, PORT_D);
}
} else if (IS_GEN2(dev))
intel_dvo_init(dev);
list_for_each_entry(encoder, &dev->mode_config.encoder_list, base.head) {
encoder->base.possible_crtcs = encoder->crtc_mask;
encoder->base.possible_clones =
- intel_encoder_clones(dev, encoder->clone_mask);
+ intel_encoder_clones(encoder);
}
/* disable all the possible outputs/crtcs before entering KMS mode */
POSTING_READ(vga_reg);
}
- static void ivb_pch_pwm_override(struct drm_device *dev)
- {
- struct drm_i915_private *dev_priv = dev->dev_private;
-
- /*
- * IVB has CPU eDP backlight regs too, set things up to let the
- * PCH regs control the backlight
- */
- I915_WRITE(BLC_PWM_CPU_CTL2, BLM_PWM_ENABLE);
- I915_WRITE(BLC_PWM_CPU_CTL, 0);
- I915_WRITE(BLC_PWM_PCH_CTL1, BLM_PCH_PWM_ENABLE | BLM_PCH_OVERRIDE_ENABLE);
- }
-
void intel_modeset_init_hw(struct drm_device *dev)
{
/* We attempt to init the necessary power wells early in the initialization
mutex_lock(&dev->struct_mutex);
intel_enable_gt_powersave(dev);
mutex_unlock(&dev->struct_mutex);
-
- if (IS_IVYBRIDGE(dev))
- ivb_pch_pwm_override(dev);
}
void intel_modeset_init(struct drm_device *dev)
/* Just disable it once at startup */
i915_disable_vga(dev);
intel_setup_outputs(dev);
-
- INIT_WORK(&dev_priv->idle_work, intel_idle_update);
- setup_timer(&dev_priv->idle_timer, intel_gpu_idle_timer,
- (unsigned long)dev);
}
void intel_modeset_gem_init(struct drm_device *dev)
* enqueue unpin/hotplug work. */
drm_irq_uninstall(dev);
cancel_work_sync(&dev_priv->hotplug_work);
- cancel_work_sync(&dev_priv->rps_work);
+ cancel_work_sync(&dev_priv->rps.work);
/* flush any delayed tasks or pending work */
flush_scheduled_work();
- /* Shut off idle work before the crtcs get freed. */
- list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
- intel_crtc = to_intel_crtc(crtc);
- del_timer_sync(&intel_crtc->idle_timer);
- }
- del_timer_sync(&dev_priv->idle_timer);
- cancel_work_sync(&dev_priv->idle_work);
-
drm_mode_config_cleanup(dev);
}
#include "intel_drv.h"
#include "i915_drm.h"
#include "i915_drv.h"
-#include "drm_dp_helper.h"
-#define DP_RECEIVER_CAP_SIZE 0xf
#define DP_LINK_STATUS_SIZE 6
#define DP_LINK_CHECK_TIMEOUT (10 * 1000)
-#define DP_LINK_CONFIGURATION_SIZE 9
-
-struct intel_dp {
- struct intel_encoder base;
- uint32_t output_reg;
- uint32_t DP;
- uint8_t link_configuration[DP_LINK_CONFIGURATION_SIZE];
- bool has_audio;
- enum hdmi_force_audio force_audio;
- uint32_t color_range;
- int dpms_mode;
- uint8_t link_bw;
- uint8_t lane_count;
- uint8_t dpcd[DP_RECEIVER_CAP_SIZE];
- struct i2c_adapter adapter;
- struct i2c_algo_dp_aux_data algo;
- bool is_pch_edp;
- uint8_t train_set[4];
- int panel_power_up_delay;
- int panel_power_down_delay;
- int panel_power_cycle_delay;
- int backlight_on_delay;
- int backlight_off_delay;
- struct drm_display_mode *panel_fixed_mode; /* for eDP */
- struct delayed_work panel_vdd_work;
- bool want_panel_vdd;
- struct edid *edid; /* cached EDID for eDP */
- int edid_mode_count;
-};
-
/**
* is_edp - is the given port attached to an eDP panel (either CPU or PCH)
* @intel_dp: DP struct
WARN(!intel_dp->want_panel_vdd, "Need VDD to turn off panel\n");
pp = ironlake_get_pp_control(dev_priv);
- pp &= ~(POWER_TARGET_ON | PANEL_POWER_RESET | EDP_BLC_ENABLE);
+ /* We need to switch off panel power _and_ force vdd, for otherwise some
+ * panels get very unhappy and cease to work. */
+ pp &= ~(POWER_TARGET_ON | EDP_FORCE_VDD | PANEL_POWER_RESET | EDP_BLC_ENABLE);
I915_WRITE(PCH_PP_CONTROL, pp);
POSTING_READ(PCH_PP_CONTROL);
+ intel_dp->want_panel_vdd = false;
+
ironlake_wait_panel_off(intel_dp);
}
* ensure that we have vdd while we switch off the panel. */
ironlake_edp_panel_vdd_on(intel_dp);
ironlake_edp_backlight_off(intel_dp);
- ironlake_edp_panel_off(intel_dp);
-
intel_dp_sink_dpms(intel_dp, DRM_MODE_DPMS_ON);
+ ironlake_edp_panel_off(intel_dp);
intel_dp_link_down(intel_dp);
- ironlake_edp_panel_vdd_off(intel_dp, false);
}
static void intel_dp_commit(struct drm_encoder *encoder)
/* Switching the panel off requires vdd. */
ironlake_edp_panel_vdd_on(intel_dp);
ironlake_edp_backlight_off(intel_dp);
- ironlake_edp_panel_off(intel_dp);
-
intel_dp_sink_dpms(intel_dp, mode);
+ ironlake_edp_panel_off(intel_dp);
intel_dp_link_down(intel_dp);
- ironlake_edp_panel_vdd_off(intel_dp, false);
if (is_cpu_edp(intel_dp))
ironlake_edp_pll_off(encoder);
struct drm_i915_private *dev_priv = dev->dev_private;
int ret;
+ if (HAS_PCH_CPT(dev) && (IS_GEN7(dev) || !is_cpu_edp(intel_dp))) {
+ dp_reg_value &= ~DP_LINK_TRAIN_MASK_CPT;
+
+ switch (dp_train_pat & DP_TRAINING_PATTERN_MASK) {
+ case DP_TRAINING_PATTERN_DISABLE:
+ dp_reg_value |= DP_LINK_TRAIN_OFF_CPT;
+ break;
+ case DP_TRAINING_PATTERN_1:
+ dp_reg_value |= DP_LINK_TRAIN_PAT_1_CPT;
+ break;
+ case DP_TRAINING_PATTERN_2:
+ dp_reg_value |= DP_LINK_TRAIN_PAT_2_CPT;
+ break;
+ case DP_TRAINING_PATTERN_3:
+ DRM_ERROR("DP training pattern 3 not supported\n");
+ dp_reg_value |= DP_LINK_TRAIN_PAT_2_CPT;
+ break;
+ }
+
+ } else {
+ dp_reg_value &= ~DP_LINK_TRAIN_MASK;
+
+ switch (dp_train_pat & DP_TRAINING_PATTERN_MASK) {
+ case DP_TRAINING_PATTERN_DISABLE:
+ dp_reg_value |= DP_LINK_TRAIN_OFF;
+ break;
+ case DP_TRAINING_PATTERN_1:
+ dp_reg_value |= DP_LINK_TRAIN_PAT_1;
+ break;
+ case DP_TRAINING_PATTERN_2:
+ dp_reg_value |= DP_LINK_TRAIN_PAT_2;
+ break;
+ case DP_TRAINING_PATTERN_3:
+ DRM_ERROR("DP training pattern 3 not supported\n");
+ dp_reg_value |= DP_LINK_TRAIN_PAT_2;
+ break;
+ }
+ }
+
I915_WRITE(intel_dp->output_reg, dp_reg_value);
POSTING_READ(intel_dp->output_reg);
DP_TRAINING_PATTERN_SET,
dp_train_pat);
- ret = intel_dp_aux_native_write(intel_dp,
- DP_TRAINING_LANE0_SET,
- intel_dp->train_set,
- intel_dp->lane_count);
- if (ret != intel_dp->lane_count)
- return false;
+ if ((dp_train_pat & DP_TRAINING_PATTERN_MASK) !=
+ DP_TRAINING_PATTERN_DISABLE) {
+ ret = intel_dp_aux_native_write(intel_dp,
+ DP_TRAINING_LANE0_SET,
+ intel_dp->train_set,
+ intel_dp->lane_count);
+ if (ret != intel_dp->lane_count)
+ return false;
+ }
return true;
}
uint8_t voltage;
bool clock_recovery = false;
int voltage_tries, loop_tries;
- u32 reg;
uint32_t DP = intel_dp->DP;
/*
DP |= DP_PORT_EN;
- if (HAS_PCH_CPT(dev) && (IS_GEN7(dev) || !is_cpu_edp(intel_dp)))
- DP &= ~DP_LINK_TRAIN_MASK_CPT;
- else
- DP &= ~DP_LINK_TRAIN_MASK;
memset(intel_dp->train_set, 0, 4);
voltage = 0xff;
voltage_tries = 0;
DP = (DP & ~(DP_VOLTAGE_MASK|DP_PRE_EMPHASIS_MASK)) | signal_levels;
}
- if (HAS_PCH_CPT(dev) && (IS_GEN7(dev) || !is_cpu_edp(intel_dp)))
- reg = DP | DP_LINK_TRAIN_PAT_1_CPT;
- else
- reg = DP | DP_LINK_TRAIN_PAT_1;
-
- if (!intel_dp_set_link_train(intel_dp, reg,
+ if (!intel_dp_set_link_train(intel_dp, DP,
DP_TRAINING_PATTERN_1 |
DP_LINK_SCRAMBLING_DISABLE))
break;
intel_dp_complete_link_train(struct intel_dp *intel_dp)
{
struct drm_device *dev = intel_dp->base.base.dev;
- struct drm_i915_private *dev_priv = dev->dev_private;
bool channel_eq = false;
int tries, cr_tries;
- u32 reg;
uint32_t DP = intel_dp->DP;
/* channel equalization */
DP = (DP & ~(DP_VOLTAGE_MASK|DP_PRE_EMPHASIS_MASK)) | signal_levels;
}
- if (HAS_PCH_CPT(dev) && (IS_GEN7(dev) || !is_cpu_edp(intel_dp)))
- reg = DP | DP_LINK_TRAIN_PAT_2_CPT;
- else
- reg = DP | DP_LINK_TRAIN_PAT_2;
-
/* channel eq pattern */
- if (!intel_dp_set_link_train(intel_dp, reg,
+ if (!intel_dp_set_link_train(intel_dp, DP,
DP_TRAINING_PATTERN_2 |
DP_LINK_SCRAMBLING_DISABLE))
break;
++tries;
}
- if (HAS_PCH_CPT(dev) && (IS_GEN7(dev) || !is_cpu_edp(intel_dp)))
- reg = DP | DP_LINK_TRAIN_OFF_CPT;
- else
- reg = DP | DP_LINK_TRAIN_OFF;
-
- I915_WRITE(intel_dp->output_reg, reg);
- POSTING_READ(intel_dp->output_reg);
- intel_dp_aux_native_write_1(intel_dp,
- DP_TRAINING_PATTERN_SET, DP_TRAINING_PATTERN_DISABLE);
+ intel_dp_set_link_train(intel_dp, DP, DP_TRAINING_PATTERN_DISABLE);
}
static void
}
void
-intel_dp_init(struct drm_device *dev, int output_reg)
+intel_dp_init(struct drm_device *dev, int output_reg, enum port port)
{
struct drm_i915_private *dev_priv = dev->dev_private;
struct drm_connector *connector;
return;
intel_dp->output_reg = output_reg;
+ intel_dp->port = port;
intel_dp->dpms_mode = -1;
intel_connector = kzalloc(sizeof(struct intel_connector), GFP_KERNEL);
connector->polled = DRM_CONNECTOR_POLL_HPD;
- if (output_reg == DP_B || output_reg == PCH_DP_B)
- intel_encoder->clone_mask = (1 << INTEL_DP_B_CLONE_BIT);
- else if (output_reg == DP_C || output_reg == PCH_DP_C)
- intel_encoder->clone_mask = (1 << INTEL_DP_C_CLONE_BIT);
- else if (output_reg == DP_D || output_reg == PCH_DP_D)
- intel_encoder->clone_mask = (1 << INTEL_DP_D_CLONE_BIT);
+ intel_encoder->cloneable = false;
- if (is_edp(intel_dp)) {
- intel_encoder->clone_mask = (1 << INTEL_EDP_CLONE_BIT);
- INIT_DELAYED_WORK(&intel_dp->panel_vdd_work,
- ironlake_panel_vdd_work);
- }
+ INIT_DELAYED_WORK(&intel_dp->panel_vdd_work,
+ ironlake_panel_vdd_work);
intel_encoder->crtc_mask = (1 << 0) | (1 << 1) | (1 << 2);
drm_sysfs_connector_add(connector);
/* Set up the DDC bus. */
- switch (output_reg) {
- case DP_A:
- name = "DPDDC-A";
- break;
- case DP_B:
- case PCH_DP_B:
- dev_priv->hotplug_supported_mask |=
- DPB_HOTPLUG_INT_STATUS;
- name = "DPDDC-B";
- break;
- case DP_C:
- case PCH_DP_C:
- dev_priv->hotplug_supported_mask |=
- DPC_HOTPLUG_INT_STATUS;
- name = "DPDDC-C";
- break;
- case DP_D:
- case PCH_DP_D:
- dev_priv->hotplug_supported_mask |=
- DPD_HOTPLUG_INT_STATUS;
- name = "DPDDC-D";
- break;
+ switch (port) {
+ case PORT_A:
+ name = "DPDDC-A";
+ break;
+ case PORT_B:
+ dev_priv->hotplug_supported_mask |= DPB_HOTPLUG_INT_STATUS;
+ name = "DPDDC-B";
+ break;
+ case PORT_C:
+ dev_priv->hotplug_supported_mask |= DPC_HOTPLUG_INT_STATUS;
+ name = "DPDDC-C";
+ break;
+ case PORT_D:
+ dev_priv->hotplug_supported_mask |= DPD_HOTPLUG_INT_STATUS;
+ name = "DPDDC-D";
+ break;
+ default:
+ WARN(1, "Invalid port %c\n", port_name(port));
+ break;
}
intel_dp_i2c_init(intel_dp, intel_connector, name);
#include "drm_crtc.h"
#include "drm_crtc_helper.h"
#include "drm_fb_helper.h"
+#include "drm_dp_helper.h"
#define _wait_for(COND, MS, W) ({ \
unsigned long timeout__ = jiffies + msecs_to_jiffies(MS); \
})
#define wait_for_atomic_us(COND, US) ({ \
- int i, ret__ = -ETIMEDOUT; \
- for (i = 0; i < (US); i++) { \
- if ((COND)) { \
- ret__ = 0; \
- break; \
- } \
- udelay(1); \
- } \
- ret__; \
+ unsigned long timeout__ = jiffies + usecs_to_jiffies(US); \
+ int ret__ = 0; \
+ while (!(COND)) { \
+ if (time_after(jiffies, timeout__)) { \
+ ret__ = -ETIMEDOUT; \
+ break; \
+ } \
+ cpu_relax(); \
+ } \
+ ret__; \
})
#define wait_for(COND, MS) _wait_for(COND, MS, 1)
#define INTEL_OUTPUT_DISPLAYPORT 7
#define INTEL_OUTPUT_EDP 8
-/* Intel Pipe Clone Bit */
-#define INTEL_HDMIB_CLONE_BIT 1
-#define INTEL_HDMIC_CLONE_BIT 2
-#define INTEL_HDMID_CLONE_BIT 3
-#define INTEL_HDMIE_CLONE_BIT 4
-#define INTEL_HDMIF_CLONE_BIT 5
-#define INTEL_SDVO_NON_TV_CLONE_BIT 6
-#define INTEL_SDVO_TV_CLONE_BIT 7
-#define INTEL_SDVO_LVDS_CLONE_BIT 8
-#define INTEL_ANALOG_CLONE_BIT 9
-#define INTEL_TV_CLONE_BIT 10
-#define INTEL_DP_B_CLONE_BIT 11
-#define INTEL_DP_C_CLONE_BIT 12
-#define INTEL_DP_D_CLONE_BIT 13
-#define INTEL_LVDS_CLONE_BIT 14
-#define INTEL_DVO_TMDS_CLONE_BIT 15
-#define INTEL_DVO_LVDS_CLONE_BIT 16
-#define INTEL_EDP_CLONE_BIT 17
-
#define INTEL_DVO_CHIP_NONE 0
#define INTEL_DVO_CHIP_LVDS 1
#define INTEL_DVO_CHIP_TMDS 2
struct drm_encoder base;
int type;
bool needs_tv_clock;
+ /*
+ * Intel hw has only one MUX where encoders could be clone, hence a
+ * simple flag is enough to compute the possible_clones mask.
+ */
+ bool cloneable;
void (*hot_plug)(struct intel_encoder *);
int crtc_mask;
- int clone_mask;
};
struct intel_connector {
int dpms_mode;
bool active; /* is the crtc on? independent of the dpms mode */
bool primary_disabled; /* is the crtc obscured by a plane? */
- bool busy; /* is scanout buffer being updated frequently? */
- struct timer_list idle_timer;
bool lowfreq_avail;
struct intel_overlay *overlay;
struct intel_unpin_work *unpin_work;
struct drm_display_mode *adjusted_mode);
};
+#define DP_RECEIVER_CAP_SIZE 0xf
+#define DP_LINK_CONFIGURATION_SIZE 9
+
+struct intel_dp {
+ struct intel_encoder base;
+ uint32_t output_reg;
+ uint32_t DP;
+ uint8_t link_configuration[DP_LINK_CONFIGURATION_SIZE];
+ bool has_audio;
+ enum hdmi_force_audio force_audio;
+ enum port port;
+ uint32_t color_range;
+ int dpms_mode;
+ uint8_t link_bw;
+ uint8_t lane_count;
+ uint8_t dpcd[DP_RECEIVER_CAP_SIZE];
+ struct i2c_adapter adapter;
+ struct i2c_algo_dp_aux_data algo;
+ bool is_pch_edp;
+ uint8_t train_set[4];
+ int panel_power_up_delay;
+ int panel_power_down_delay;
+ int panel_power_cycle_delay;
+ int backlight_on_delay;
+ int backlight_off_delay;
+ struct drm_display_mode *panel_fixed_mode; /* for eDP */
+ struct delayed_work panel_vdd_work;
+ bool want_panel_vdd;
+ struct edid *edid; /* cached EDID for eDP */
+ int edid_mode_count;
+};
+
static inline struct drm_crtc *
intel_get_crtc_for_pipe(struct drm_device *dev, int pipe)
{
extern void intel_attach_broadcast_rgb_property(struct drm_connector *connector);
extern void intel_crt_init(struct drm_device *dev);
-extern void intel_hdmi_init(struct drm_device *dev, int sdvox_reg);
+extern void intel_hdmi_init(struct drm_device *dev,
+ int sdvox_reg, enum port port);
extern struct intel_hdmi *enc_to_intel_hdmi(struct drm_encoder *encoder);
extern void intel_dip_infoframe_csum(struct dip_infoframe *avi_if);
extern bool intel_sdvo_init(struct drm_device *dev, uint32_t sdvo_reg,
bool is_sdvob);
extern void intel_dvo_init(struct drm_device *dev);
extern void intel_tv_init(struct drm_device *dev);
-extern void intel_mark_busy(struct drm_device *dev,
- struct drm_i915_gem_object *obj);
+extern void intel_mark_busy(struct drm_device *dev);
+extern void intel_mark_idle(struct drm_device *dev);
+extern void intel_mark_fb_busy(struct drm_i915_gem_object *obj);
+extern void intel_mark_fb_idle(struct drm_i915_gem_object *obj);
extern bool intel_lvds_init(struct drm_device *dev);
-extern void intel_dp_init(struct drm_device *dev, int dp_reg);
+extern void intel_dp_init(struct drm_device *dev, int output_reg,
+ enum port port);
void
intel_dp_set_m_n(struct drm_crtc *crtc, struct drm_display_mode *mode,
struct drm_display_mode *adjusted_mode);
extern void intel_flush_display_plane(struct drm_i915_private *dev_priv,
enum plane plane);
-void intel_sanitize_pm(struct drm_device *dev);
-
/* intel_panel.c */
extern void intel_fixed_panel_mode(struct drm_display_mode *fixed_mode,
struct drm_display_mode *adjusted_mode);
const struct drm_display_mode *mode,
struct drm_display_mode *adjusted_mode);
extern u32 intel_panel_get_max_backlight(struct drm_device *dev);
- extern u32 intel_panel_get_backlight(struct drm_device *dev);
extern void intel_panel_set_backlight(struct drm_device *dev, u32 level);
extern int intel_panel_setup_backlight(struct drm_device *dev);
extern void intel_panel_enable_backlight(struct drm_device *dev,
return NULL;
}
+/**
+ * Lock protecting IPS related data structures
+ * - i915_mch_dev
+ * - dev_priv->max_delay
+ * - dev_priv->min_delay
+ * - dev_priv->fmax
+ * - dev_priv->gpu_busy
+ * - dev_priv->gfx_power
+ */
+DEFINE_SPINLOCK(mchdev_lock);
+
+/* Global for IPS driver to get at the current i915 device. Protected by
+ * mchdev_lock. */
+static struct drm_i915_private *i915_mch_dev;
+
bool ironlake_set_drps(struct drm_device *dev, u8 val)
{
struct drm_i915_private *dev_priv = dev->dev_private;
u16 rgvswctl;
+ assert_spin_locked(&mchdev_lock);
+
rgvswctl = I915_READ16(MEMSWCTL);
if (rgvswctl & MEMCTL_CMD_STS) {
DRM_DEBUG("gpu busy, RCS change rejected\n");
u32 rgvmodectl = I915_READ(MEMMODECTL);
u8 fmax, fmin, fstart, vstart;
+ spin_lock_irq(&mchdev_lock);
+
/* Enable temp reporting */
I915_WRITE16(PMMISC, I915_READ(PMMISC) | MCPPCE_EN);
I915_WRITE16(TSC1, I915_READ(TSC1) | TSE);
rgvmodectl |= MEMMODE_SWMODE_EN;
I915_WRITE(MEMMODECTL, rgvmodectl);
- if (wait_for((I915_READ(MEMSWCTL) & MEMCTL_CMD_STS) == 0, 10))
+ if (wait_for_atomic((I915_READ(MEMSWCTL) & MEMCTL_CMD_STS) == 0, 10))
DRM_ERROR("stuck trying to change perf mode\n");
- msleep(1);
+ mdelay(1);
ironlake_set_drps(dev, fstart);
dev_priv->last_time1 = jiffies_to_msecs(jiffies);
dev_priv->last_count2 = I915_READ(0x112f4);
getrawmonotonic(&dev_priv->last_time2);
+
+ spin_unlock_irq(&mchdev_lock);
}
static void ironlake_disable_drps(struct drm_device *dev)
{
struct drm_i915_private *dev_priv = dev->dev_private;
- u16 rgvswctl = I915_READ16(MEMSWCTL);
+ u16 rgvswctl;
+
+ spin_lock_irq(&mchdev_lock);
+
+ rgvswctl = I915_READ16(MEMSWCTL);
/* Ack interrupts, disable EFC interrupt */
I915_WRITE(MEMINTREN, I915_READ(MEMINTREN) & ~MEMINT_EVAL_CHG_EN);
/* Go back to the starting frequency */
ironlake_set_drps(dev, dev_priv->fstart);
- msleep(1);
+ mdelay(1);
rgvswctl |= MEMCTL_CMD_STS;
I915_WRITE(MEMSWCTL, rgvswctl);
- msleep(1);
+ mdelay(1);
+ spin_unlock_irq(&mchdev_lock);
}
-void gen6_set_rps(struct drm_device *dev, u8 val)
+/* There's a funny hw issue where the hw returns all 0 when reading from
+ * GEN6_RP_INTERRUPT_LIMITS. Hence we always need to compute the desired value
+ * ourselves, instead of doing a rmw cycle (which might result in us clearing
+ * all limits and the gpu stuck at whatever frequency it is at atm).
+ */
+static u32 gen6_rps_limits(struct drm_i915_private *dev_priv, u8 *val)
{
- struct drm_i915_private *dev_priv = dev->dev_private;
u32 limits;
limits = 0;
- if (val >= dev_priv->max_delay)
- val = dev_priv->max_delay;
- else
- limits |= dev_priv->max_delay << 24;
- if (val <= dev_priv->min_delay)
- val = dev_priv->min_delay;
- else
- limits |= dev_priv->min_delay << 16;
+ if (*val >= dev_priv->rps.max_delay)
+ *val = dev_priv->rps.max_delay;
+ limits |= dev_priv->rps.max_delay << 24;
- if (val == dev_priv->cur_delay)
+ /* Only set the down limit when we've reached the lowest level to avoid
+ * getting more interrupts, otherwise leave this clear. This prevents a
+ * race in the hw when coming out of rc6: There's a tiny window where
+ * the hw runs at the minimal clock before selecting the desired
+ * frequency, if the down threshold expires in that window we will not
+ * receive a down interrupt. */
+ if (*val <= dev_priv->rps.min_delay) {
+ *val = dev_priv->rps.min_delay;
+ limits |= dev_priv->rps.min_delay << 16;
+ }
+
+ return limits;
+}
+
+void gen6_set_rps(struct drm_device *dev, u8 val)
+{
+ struct drm_i915_private *dev_priv = dev->dev_private;
+ u32 limits = gen6_rps_limits(dev_priv, &val);
+
+ WARN_ON(!mutex_is_locked(&dev->struct_mutex));
+
+ if (val == dev_priv->rps.cur_delay)
return;
I915_WRITE(GEN6_RPNSWREQ,
*/
I915_WRITE(GEN6_RP_INTERRUPT_LIMITS, limits);
- dev_priv->cur_delay = val;
+ dev_priv->rps.cur_delay = val;
}
static void gen6_disable_rps(struct drm_device *dev)
* register (PMIMR) to mask PM interrupts. The only risk is in leaving
* stale bits in PMIIR and PMIMR which gen6_enable_rps will clean up. */
- spin_lock_irq(&dev_priv->rps_lock);
- dev_priv->pm_iir = 0;
- spin_unlock_irq(&dev_priv->rps_lock);
+ spin_lock_irq(&dev_priv->rps.lock);
+ dev_priv->rps.pm_iir = 0;
+ spin_unlock_irq(&dev_priv->rps.lock);
I915_WRITE(GEN6_PMIIR, I915_READ(GEN6_PMIIR));
}
int intel_enable_rc6(const struct drm_device *dev)
{
- /*
- * Respect the kernel parameter if it is set
- */
+ /* Respect the kernel parameter if it is set */
if (i915_enable_rc6 >= 0)
return i915_enable_rc6;
- /*
- * Disable RC6 on Ironlake
- */
- if (INTEL_INFO(dev)->gen == 5)
- return 0;
+ if (INTEL_INFO(dev)->gen == 5) {
+ DRM_DEBUG_DRIVER("Ironlake: only RC6 available\n");
+ return INTEL_RC6_ENABLE;
+ }
- /* On Haswell, only RC6 is available. So let's enable it by default to
- * provide better testing and coverage since the beginning.
- */
- if (IS_HASWELL(dev))
+ if (IS_HASWELL(dev)) {
+ DRM_DEBUG_DRIVER("Haswell: only RC6 available\n");
return INTEL_RC6_ENABLE;
+ }
- /*
- * Disable rc6 on Sandybridge
- */
+ /* snb/ivb have more than one rc6 state. */
if (INTEL_INFO(dev)->gen == 6) {
DRM_DEBUG_DRIVER("Sandybridge: deep RC6 disabled\n");
return INTEL_RC6_ENABLE;
}
+
DRM_DEBUG_DRIVER("RC6 and deep RC6 enabled\n");
return (INTEL_RC6_ENABLE | INTEL_RC6p_ENABLE);
}
gt_perf_status = I915_READ(GEN6_GT_PERF_STATUS);
/* In units of 100MHz */
- dev_priv->max_delay = rp_state_cap & 0xff;
- dev_priv->min_delay = (rp_state_cap & 0xff0000) >> 16;
- dev_priv->cur_delay = 0;
+ dev_priv->rps.max_delay = rp_state_cap & 0xff;
+ dev_priv->rps.min_delay = (rp_state_cap & 0xff0000) >> 16;
+ dev_priv->rps.cur_delay = 0;
/* disable the counters and set deterministic thresholds */
I915_WRITE(GEN6_RC_CONTROL, 0);
I915_WRITE(GEN6_RP_DOWN_TIMEOUT, 1000000);
I915_WRITE(GEN6_RP_INTERRUPT_LIMITS,
- dev_priv->max_delay << 24 |
- dev_priv->min_delay << 16);
+ dev_priv->rps.max_delay << 24 |
+ dev_priv->rps.min_delay << 16);
if (IS_HASWELL(dev)) {
I915_WRITE(GEN6_RP_UP_THRESHOLD, 59400);
500))
DRM_ERROR("timeout waiting for pcode mailbox to finish\n");
if (pcu_mbox & (1<<31)) { /* OC supported */
- dev_priv->max_delay = pcu_mbox & 0xff;
+ dev_priv->rps.max_delay = pcu_mbox & 0xff;
DRM_DEBUG_DRIVER("overclocking supported, adjusting frequency max to %dMHz\n", pcu_mbox * 50);
}
/* requires MSI enabled */
I915_WRITE(GEN6_PMIER, GEN6_PM_DEFERRED_EVENTS);
- spin_lock_irq(&dev_priv->rps_lock);
- WARN_ON(dev_priv->pm_iir != 0);
+ spin_lock_irq(&dev_priv->rps.lock);
+ WARN_ON(dev_priv->rps.pm_iir != 0);
I915_WRITE(GEN6_PMIMR, 0);
- spin_unlock_irq(&dev_priv->rps_lock);
+ spin_unlock_irq(&dev_priv->rps.lock);
/* enable all PM interrupts */
I915_WRITE(GEN6_PMINTRMSK, 0);
* to use for memory access. We do this by specifying the IA frequency
* the PCU should use as a reference to determine the ring frequency.
*/
- for (gpu_freq = dev_priv->max_delay; gpu_freq >= dev_priv->min_delay;
+ for (gpu_freq = dev_priv->rps.max_delay; gpu_freq >= dev_priv->rps.min_delay;
gpu_freq--) {
- int diff = dev_priv->max_delay - gpu_freq;
+ int diff = dev_priv->rps.max_delay - gpu_freq;
/*
* For GPU frequencies less than 750MHz, just use the lowest
unsigned long now = jiffies_to_msecs(jiffies), diff1;
int i;
+ assert_spin_locked(&mchdev_lock);
+
diff1 = now - dev_priv->last_time1;
/* Prevent division-by-zero if we are asking too fast.
return v_table[pxvid].vd;
}
-void i915_update_gfx_val(struct drm_i915_private *dev_priv)
+static void __i915_update_gfx_val(struct drm_i915_private *dev_priv)
{
struct timespec now, diff1;
u64 diff;
unsigned long diffms;
u32 count;
- if (dev_priv->info->gen != 5)
- return;
+ assert_spin_locked(&mchdev_lock);
getrawmonotonic(&now);
diff1 = timespec_sub(now, dev_priv->last_time2);
dev_priv->gfx_power = diff;
}
+void i915_update_gfx_val(struct drm_i915_private *dev_priv)
+{
+ if (dev_priv->info->gen != 5)
+ return;
+
+ spin_lock_irq(&mchdev_lock);
+
+ __i915_update_gfx_val(dev_priv);
+
+ spin_unlock_irq(&mchdev_lock);
+}
+
unsigned long i915_gfx_val(struct drm_i915_private *dev_priv)
{
unsigned long t, corr, state1, corr2, state2;
u32 pxvid, ext_v;
- pxvid = I915_READ(PXVFREQ_BASE + (dev_priv->cur_delay * 4));
+ assert_spin_locked(&mchdev_lock);
+
+ pxvid = I915_READ(PXVFREQ_BASE + (dev_priv->rps.cur_delay * 4));
pxvid = (pxvid >> 24) & 0x7f;
ext_v = pvid_to_extvid(dev_priv, pxvid);
state2 = (corr2 * state1) / 10000;
state2 /= 100; /* convert to mW */
- i915_update_gfx_val(dev_priv);
+ __i915_update_gfx_val(dev_priv);
return dev_priv->gfx_power + state2;
}
-/* Global for IPS driver to get at the current i915 device */
-static struct drm_i915_private *i915_mch_dev;
-/*
- * Lock protecting IPS related data structures
- * - i915_mch_dev
- * - dev_priv->max_delay
- * - dev_priv->min_delay
- * - dev_priv->fmax
- * - dev_priv->gpu_busy
- */
-static DEFINE_SPINLOCK(mchdev_lock);
-
/**
* i915_read_mch_val - return value for IPS use
*
struct drm_i915_private *dev_priv;
unsigned long chipset_val, graphics_val, ret = 0;
- spin_lock(&mchdev_lock);
+ spin_lock_irq(&mchdev_lock);
if (!i915_mch_dev)
goto out_unlock;
dev_priv = i915_mch_dev;
ret = chipset_val + graphics_val;
out_unlock:
- spin_unlock(&mchdev_lock);
+ spin_unlock_irq(&mchdev_lock);
return ret;
}
struct drm_i915_private *dev_priv;
bool ret = true;
- spin_lock(&mchdev_lock);
+ spin_lock_irq(&mchdev_lock);
if (!i915_mch_dev) {
ret = false;
goto out_unlock;
dev_priv->max_delay--;
out_unlock:
- spin_unlock(&mchdev_lock);
+ spin_unlock_irq(&mchdev_lock);
return ret;
}
struct drm_i915_private *dev_priv;
bool ret = true;
- spin_lock(&mchdev_lock);
+ spin_lock_irq(&mchdev_lock);
if (!i915_mch_dev) {
ret = false;
goto out_unlock;
dev_priv->max_delay++;
out_unlock:
- spin_unlock(&mchdev_lock);
+ spin_unlock_irq(&mchdev_lock);
return ret;
}
bool i915_gpu_busy(void)
{
struct drm_i915_private *dev_priv;
+ struct intel_ring_buffer *ring;
bool ret = false;
+ int i;
- spin_lock(&mchdev_lock);
+ spin_lock_irq(&mchdev_lock);
if (!i915_mch_dev)
goto out_unlock;
dev_priv = i915_mch_dev;
- ret = dev_priv->busy;
+ for_each_ring(ring, dev_priv, i)
+ ret |= !list_empty(&ring->request_list);
out_unlock:
- spin_unlock(&mchdev_lock);
+ spin_unlock_irq(&mchdev_lock);
return ret;
}
struct drm_i915_private *dev_priv;
bool ret = true;
- spin_lock(&mchdev_lock);
+ spin_lock_irq(&mchdev_lock);
if (!i915_mch_dev) {
ret = false;
goto out_unlock;
ret = false;
out_unlock:
- spin_unlock(&mchdev_lock);
+ spin_unlock_irq(&mchdev_lock);
return ret;
}
void intel_gpu_ips_init(struct drm_i915_private *dev_priv)
{
- spin_lock(&mchdev_lock);
+ /* We only register the i915 ips part with intel-ips once everything is
+ * set up, to avoid intel-ips sneaking in and reading bogus values. */
+ spin_lock_irq(&mchdev_lock);
i915_mch_dev = dev_priv;
- dev_priv->mchdev_lock = &mchdev_lock;
- spin_unlock(&mchdev_lock);
+ spin_unlock_irq(&mchdev_lock);
ips_ping_for_i915_load();
}
void intel_gpu_ips_teardown(void)
{
- spin_lock(&mchdev_lock);
+ spin_lock_irq(&mchdev_lock);
i915_mch_dev = NULL;
- spin_unlock(&mchdev_lock);
+ spin_unlock_irq(&mchdev_lock);
}
static void intel_init_emon(struct drm_device *dev)
{
dev_priv->display.init_pch_clock_gating(dev);
}
-static void gen6_sanitize_pm(struct drm_device *dev)
-{
- struct drm_i915_private *dev_priv = dev->dev_private;
- u32 limits, delay, old;
-
- gen6_gt_force_wake_get(dev_priv);
-
- old = limits = I915_READ(GEN6_RP_INTERRUPT_LIMITS);
- /* Make sure we continue to get interrupts
- * until we hit the minimum or maximum frequencies.
- */
- limits &= ~(0x3f << 16 | 0x3f << 24);
- delay = dev_priv->cur_delay;
- if (delay < dev_priv->max_delay)
- limits |= (dev_priv->max_delay & 0x3f) << 24;
- if (delay > dev_priv->min_delay)
- limits |= (dev_priv->min_delay & 0x3f) << 16;
-
- if (old != limits) {
- /* Note that the known failure case is to read back 0. */
- DRM_DEBUG_DRIVER("Power management discrepancy: GEN6_RP_INTERRUPT_LIMITS "
- "expected %08x, was %08x\n", limits, old);
- I915_WRITE(GEN6_RP_INTERRUPT_LIMITS, limits);
- }
-
- gen6_gt_force_wake_put(dev_priv);
-}
-
-void intel_sanitize_pm(struct drm_device *dev)
-{
- struct drm_i915_private *dev_priv = dev->dev_private;
-
- if (dev_priv->display.sanitize_pm)
- dev_priv->display.sanitize_pm(dev);
-}
-
/* Starting with Haswell, we have different power wells for
* different parts of the GPU. This attempts to enable them all.
*/
dev_priv->display.update_wm = NULL;
}
dev_priv->display.init_clock_gating = gen6_init_clock_gating;
- dev_priv->display.sanitize_pm = gen6_sanitize_pm;
} else if (IS_IVYBRIDGE(dev)) {
/* FIXME: detect B0+ stepping and use auto training */
if (SNB_READ_WM0_LATENCY()) {
dev_priv->display.update_wm = NULL;
}
dev_priv->display.init_clock_gating = ivybridge_init_clock_gating;
- dev_priv->display.sanitize_pm = gen6_sanitize_pm;
} else if (IS_HASWELL(dev)) {
if (SNB_READ_WM0_LATENCY()) {
dev_priv->display.update_wm = sandybridge_update_wm;
dev_priv->display.update_wm = NULL;
}
dev_priv->display.init_clock_gating = haswell_init_clock_gating;
- dev_priv->display.sanitize_pm = gen6_sanitize_pm;
} else
dev_priv->display.update_wm = NULL;
} else if (IS_VALLEYVIEW(dev)) {
DRM_ERROR("Force wake wait timed out\n");
I915_WRITE_NOTRACE(FORCEWAKE, 1);
+ POSTING_READ(FORCEWAKE);
if (wait_for_atomic_us((I915_READ_NOTRACE(forcewake_ack) & 1), 500))
DRM_ERROR("Force wake wait timed out\n");
DRM_ERROR("Force wake wait timed out\n");
I915_WRITE_NOTRACE(FORCEWAKE_MT, _MASKED_BIT_ENABLE(1));
+ POSTING_READ(FORCEWAKE_MT);
if (wait_for_atomic_us((I915_READ_NOTRACE(forcewake_ack) & 1), 500))
DRM_ERROR("Force wake wait timed out\n");
static void __gen6_gt_force_wake_put(struct drm_i915_private *dev_priv)
{
I915_WRITE_NOTRACE(FORCEWAKE, 0);
- /* The below doubles as a POSTING_READ */
+ POSTING_READ(FORCEWAKE);
gen6_gt_check_fifodbg(dev_priv);
}
static void __gen6_gt_force_wake_mt_put(struct drm_i915_private *dev_priv)
{
I915_WRITE_NOTRACE(FORCEWAKE_MT, _MASKED_BIT_DISABLE(1));
- /* The below doubles as a POSTING_READ */
+ POSTING_READ(FORCEWAKE_MT);
gen6_gt_check_fifodbg(dev_priv);
}
u32 invalidate_domains, u32 flush_domains)
{
u32 flags = 0;
+ struct pipe_control *pc = ring->private;
+ u32 scratch_addr = pc->gtt_offset + 128;
int ret;
- /* Force SNB workarounds for PIPE_CONTROL flushes */
- ret = intel_emit_post_sync_nonzero_flush(ring);
- if (ret)
- return ret;
-
/* Just flush everything. Experiments have shown that reducing the
* number of bits based on the write domains has little performance
* impact.
*/
- flags |= PIPE_CONTROL_RENDER_TARGET_CACHE_FLUSH;
- flags |= PIPE_CONTROL_TLB_INVALIDATE;
- flags |= PIPE_CONTROL_INSTRUCTION_CACHE_INVALIDATE;
- flags |= PIPE_CONTROL_TEXTURE_CACHE_INVALIDATE;
- flags |= PIPE_CONTROL_DEPTH_CACHE_FLUSH;
- flags |= PIPE_CONTROL_VF_CACHE_INVALIDATE;
- flags |= PIPE_CONTROL_CONST_CACHE_INVALIDATE;
- flags |= PIPE_CONTROL_STATE_CACHE_INVALIDATE;
- /*
- * Ensure that any following seqno writes only happen when the render
- * cache is indeed flushed (but only if the caller actually wants that).
- */
- if (flush_domains)
+ if (flush_domains) {
+ flags |= PIPE_CONTROL_RENDER_TARGET_CACHE_FLUSH;
+ flags |= PIPE_CONTROL_DEPTH_CACHE_FLUSH;
+ /*
+ * Ensure that any following seqno writes only happen
+ * when the render cache is indeed flushed.
+ */
flags |= PIPE_CONTROL_CS_STALL;
+ }
+ if (invalidate_domains) {
+ flags |= PIPE_CONTROL_TLB_INVALIDATE;
+ flags |= PIPE_CONTROL_INSTRUCTION_CACHE_INVALIDATE;
+ flags |= PIPE_CONTROL_TEXTURE_CACHE_INVALIDATE;
+ flags |= PIPE_CONTROL_VF_CACHE_INVALIDATE;
+ flags |= PIPE_CONTROL_CONST_CACHE_INVALIDATE;
+ flags |= PIPE_CONTROL_STATE_CACHE_INVALIDATE;
+ /*
+ * TLB invalidate requires a post-sync write.
+ */
+ flags |= PIPE_CONTROL_QW_WRITE;
+ }
ret = intel_ring_begin(ring, 4);
if (ret)
intel_ring_emit(ring, GFX_OP_PIPE_CONTROL(4));
intel_ring_emit(ring, flags);
- intel_ring_emit(ring, 0);
+ intel_ring_emit(ring, scratch_addr | PIPE_CONTROL_GLOBAL_GTT);
intel_ring_emit(ring, 0);
intel_ring_advance(ring);
return 0;
}
+static int
+gen6_render_ring_flush__wa(struct intel_ring_buffer *ring,
+ u32 invalidate_domains, u32 flush_domains)
+{
+ int ret;
+
+ /* Force SNB workarounds for PIPE_CONTROL flushes */
+ ret = intel_emit_post_sync_nonzero_flush(ring);
+ if (ret)
+ return ret;
+
+ return gen6_render_ring_flush(ring, invalidate_domains, flush_domains);
+}
+
static void ring_write_tail(struct intel_ring_buffer *ring,
u32 value)
{
I915_WRITE_HEAD(ring, 0);
ring->write_tail(ring, 0);
- /* Initialize the ring. */
- I915_WRITE_START(ring, obj->gtt_offset);
head = I915_READ_HEAD(ring) & HEAD_ADDR;
/* G45 ring initialization fails to reset head to zero */
}
}
+ /* Initialize the ring. This must happen _after_ we've cleared the ring
+ * registers with the above sequence (the readback of the HEAD registers
+ * also enforces ordering), otherwise the hw might lose the new ring
+ * register values. */
+ I915_WRITE_START(ring, obj->gtt_offset);
I915_WRITE_CTL(ring,
((ring->size - PAGE_SIZE) & RING_NR_PAGES)
| RING_VALID);
if (INTEL_INFO(dev)->gen >= 6)
I915_WRITE(INSTPM, _MASKED_BIT_ENABLE(INSTPM_FORCE_ORDERING));
- if (IS_IVYBRIDGE(dev))
+ if (HAS_L3_GPU_CACHE(dev))
I915_WRITE_IMR(ring, ~GEN6_RENDER_L3_PARITY_ERROR);
return ret;
}
static u32
-gen6_ring_get_seqno(struct intel_ring_buffer *ring)
+gen6_ring_get_seqno(struct intel_ring_buffer *ring, bool lazy_coherency)
{
- struct drm_device *dev = ring->dev;
-
/* Workaround to force correct ordering between irq and seqno writes on
* ivb (and maybe also on snb) by reading from a CS register (like
* ACTHD) before reading the status page. */
- if (IS_GEN6(dev) || IS_GEN7(dev))
+ if (!lazy_coherency)
intel_ring_get_active_head(ring);
return intel_read_status_page(ring, I915_GEM_HWS_INDEX);
}
static u32
-ring_get_seqno(struct intel_ring_buffer *ring)
+ring_get_seqno(struct intel_ring_buffer *ring, bool lazy_coherency)
{
return intel_read_status_page(ring, I915_GEM_HWS_INDEX);
}
static u32
-pc_render_get_seqno(struct intel_ring_buffer *ring)
+pc_render_get_seqno(struct intel_ring_buffer *ring, bool lazy_coherency)
{
struct pipe_control *pc = ring->private;
return pc->cpu_page[0];
spin_lock_irqsave(&dev_priv->irq_lock, flags);
if (ring->irq_refcount++ == 0) {
- if (IS_IVYBRIDGE(dev) && ring->id == RCS)
+ if (HAS_L3_GPU_CACHE(dev) && ring->id == RCS)
I915_WRITE_IMR(ring, ~(ring->irq_enable_mask |
GEN6_RENDER_L3_PARITY_ERROR));
else
spin_lock_irqsave(&dev_priv->irq_lock, flags);
if (--ring->irq_refcount == 0) {
- if (IS_IVYBRIDGE(dev) && ring->id == RCS)
+ if (HAS_L3_GPU_CACHE(dev) && ring->id == RCS)
I915_WRITE_IMR(ring, ~GEN6_RENDER_L3_PARITY_ERROR);
else
I915_WRITE_IMR(ring, ~0);
ring->dev = dev;
INIT_LIST_HEAD(&ring->active_list);
INIT_LIST_HEAD(&ring->request_list);
- INIT_LIST_HEAD(&ring->gpu_write_list);
ring->size = 32 * PAGE_SIZE;
init_waitqueue_head(&ring->irq_queue);
if (INTEL_INFO(dev)->gen >= 6) {
ring->add_request = gen6_add_request;
ring->flush = gen6_render_ring_flush;
+ if (INTEL_INFO(dev)->gen == 6)
+ ring->flush = gen6_render_ring_flush__wa;
ring->irq_get = gen6_ring_get_irq;
ring->irq_put = gen6_ring_put_irq;
ring->irq_enable_mask = GT_USER_INTERRUPT;
ring->dev = dev;
INIT_LIST_HEAD(&ring->active_list);
INIT_LIST_HEAD(&ring->request_list);
- INIT_LIST_HEAD(&ring->gpu_write_list);
ring->size = size;
ring->effective_size = ring->size;
return intel_init_ring_buffer(dev, ring);
}
+
+int
+intel_ring_flush_all_caches(struct intel_ring_buffer *ring)
+{
+ int ret;
+
+ if (!ring->gpu_caches_dirty)
+ return 0;
+
+ ret = ring->flush(ring, 0, I915_GEM_GPU_DOMAINS);
+ if (ret)
+ return ret;
+
+ trace_i915_gem_ring_flush(ring, 0, I915_GEM_GPU_DOMAINS);
+
+ ring->gpu_caches_dirty = false;
+ return 0;
+}
+
+int
+intel_ring_invalidate_all_caches(struct intel_ring_buffer *ring)
+{
+ uint32_t flush_domains;
+ int ret;
+
+ flush_domains = 0;
+ if (ring->gpu_caches_dirty)
+ flush_domains = I915_GEM_GPU_DOMAINS;
+
+ ret = ring->flush(ring, I915_GEM_GPU_DOMAINS, flush_domains);
+ if (ret)
+ return ret;
+
+ trace_i915_gem_ring_flush(ring, I915_GEM_GPU_DOMAINS, flush_domains);
+
+ ring->gpu_caches_dirty = false;
+ return 0;
+}
struct i2c_msg *msgs;
int i, ret = true;
+ /* Would be simpler to allocate both in one go ? */
buf = (u8 *)kzalloc(args_len * 2 + 2, GFP_KERNEL);
if (!buf)
return false;
msgs = kcalloc(args_len + 3, sizeof(*msgs), GFP_KERNEL);
- if (!msgs)
+ if (!msgs) {
+ kfree(buf);
return false;
+ }
intel_sdvo_debug_write(intel_sdvo, cmd, args, args_len);
connector->connector_type = DRM_MODE_CONNECTOR_HDMIA;
intel_sdvo->is_hdmi = true;
}
- intel_sdvo->base.clone_mask = ((1 << INTEL_SDVO_NON_TV_CLONE_BIT) |
- (1 << INTEL_ANALOG_CLONE_BIT));
+ intel_sdvo->base.cloneable = true;
intel_sdvo_connector_init(intel_sdvo_connector, intel_sdvo);
if (intel_sdvo->is_hdmi)
intel_sdvo->is_tv = true;
intel_sdvo->base.needs_tv_clock = true;
- intel_sdvo->base.clone_mask = 1 << INTEL_SDVO_TV_CLONE_BIT;
+ intel_sdvo->base.cloneable = false;
intel_sdvo_connector_init(intel_sdvo_connector, intel_sdvo);
intel_sdvo_connector->output_flag = SDVO_OUTPUT_RGB1;
}
- intel_sdvo->base.clone_mask = ((1 << INTEL_SDVO_NON_TV_CLONE_BIT) |
- (1 << INTEL_ANALOG_CLONE_BIT));
+ intel_sdvo->base.cloneable = true;
intel_sdvo_connector_init(intel_sdvo_connector,
intel_sdvo);
intel_sdvo_connector->output_flag = SDVO_OUTPUT_LVDS1;
}
- intel_sdvo->base.clone_mask = ((1 << INTEL_ANALOG_CLONE_BIT) |
- (1 << INTEL_SDVO_LVDS_CLONE_BIT));
+ /* SDVO LVDS is cloneable because the SDVO encoder does the upscaling,
+ * as opposed to native LVDS, where we upscale with the panel-fitter
+ * (and hence only the native LVDS resolution could be cloned). */
+ intel_sdvo->base.cloneable = true;
intel_sdvo_connector_init(intel_sdvo_connector, intel_sdvo);
if (!intel_sdvo_create_enhance_property(intel_sdvo, intel_sdvo_connector))
projects / mirror_ubuntu-zesty-kernel.git / commitdiff