- 23 Apr, 2020 4 commits
-
-
Chris Wilson authored
Under ideal circumstances, the driver should be able to keep the GPU fully saturated with work. Measure how close to ideal we get under the harshest of conditions with no user payload. v2: Also measure throughput using only one thread. Signed-off-by:
Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by:
Andi Shyti <andi.shyti@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20200422074203.9799-1-chris@chris-wilson.co.uk
-
Chris Wilson authored
intel_gt_wait_for_idle() tries to wait until all the outstanding requests are retired and the GPU is idle. As a side effect of retiring requests, we may submit more work to flush any pm barriers, and so the wait-for-idle tries to flush the background pm work and catch the new requests. However, if the work completed in the background before we were able to flush, it would queue the extra barrier request without us noticing -- and so we would return from wait-for-idle with one request remaining. (This breaks e.g. record_default_state where we need to wait until that barrier is retired, and it may slow suspend down by causing us to wait on the background retirement worker as opposed to immediately retiring the barrier.) However, since we track if there has been a submission since the engine pm barrier, we can very quickly detect if the idle barrier is still outstanding. Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/1763Signed-off-by:
Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20200423085940.28168-1-chris@chris-wilson.co.uk
-
Chris Wilson authored
During the virtual engine's submission tasklet, we take the request and insert into the submission queue on each of our siblings. This seems quite simply, and so no problems with ordering. However, the sibling execlists' submission tasklets may run concurrently with the virtual engine's tasklet, submitting the request to HW before the virtual finishes its task of telling all the siblings. If this happens, the sibling tasklet may *reorder* the ve->sibling[] array that the virtual engine tasklet is processing. This can *only* reorder within the elements already processed by the virtual engine, nevertheless the race is detected by KCSAN: [ 185.580014] BUG: KCSAN: data-race in execlists_dequeue [i915] / virtual_submission_tasklet [i915] [ 185.580054] [ 185.580076] write to 0xffff8881f1919860 of 8 bytes by interrupt on cpu 2: [ 185.580553] execlists_dequeue+0x6ad/0x1600 [i915] [ 185.581044] __execlists_submission_tasklet+0x48/0x60 [i915] [ 185.581517] execlists_submission_tasklet+0xd3/0x170 [i915] [ 185.581554] tasklet_action_common.isra.0+0x42/0x90 [ 185.581585] __do_softirq+0xc8/0x206 [ 185.581613] run_ksoftirqd+0x15/0x20 [ 185.581641] smpboot_thread_fn+0x15a/0x270 [ 185.581669] kthread+0x19a/0x1e0 [ 185.581695] ret_from_fork+0x1f/0x30 [ 185.581717] [ 185.581736] read to 0xffff8881f1919860 of 8 bytes by interrupt on cpu 0: [ 185.582231] virtual_submission_tasklet+0x10e/0x5c0 [i915] [ 185.582265] tasklet_action_common.isra.0+0x42/0x90 [ 185.582291] __do_softirq+0xc8/0x206 [ 185.582315] run_ksoftirqd+0x15/0x20 [ 185.582340] smpboot_thread_fn+0x15a/0x270 [ 185.582368] kthread+0x19a/0x1e0 [ 185.582395] ret_from_fork+0x1f/0x30 [ 185.582417] We can prevent this race by checking for the ve->request after looking up the sibling array. Signed-off-by:
-
Imre Deak authored
Fix the check for when an AUX power well enabling timeout is expected on a legacy TypeC port. Fixes: 89e01caa ("drm/i915: Use single set of AUX powerwell ops for gen11+") Cc: Matt Roper <matthew.d.roper@intel.com> Cc: José Roberto de Souza <jose.souza@intel.com> Signed-off-by:
Imre Deak <imre.deak@intel.com> Reviewed-by:
José Roberto de Souza <jose.souza@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20200422123440.19522-1-imre.deak@intel.com
-
- 22 Apr, 2020 4 commits
-
-
Chris Wilson authored
When we migrated to execlists, one of the conditions we wanted to test for was whether the breadcrumb seqno was being written before the breadcumb interrupt was delivered. This was following on from issues observed on previous generations which were not so strongly ordered. With the removal of the missed interrupt detection, we have not reliable means of detecting the out-of-order seqno/interrupt but instead tried to assert that the relationship between the CS event interrupt and the breadwrite should be strongly ordered. However, Icelake proves it is possible for the HW implementation to forget about minor little details such as write ordering and so the order between *processing* the CS event and the breadcrumb is unreliable. Remove the unreliable assertion, but leave a debug telltale in case we have reason to suspect. Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/1658Signed-off-by:
Mika Kuoppala <mika.kuoppala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20200422141749.28709-1-chris@chris-wilson.co.uk
-
Chris Wilson authored
While the ggtt vma are protected by their object lifetime, the list continues until it hits a non-ggtt vma, and that vma is not protected and may be freed as we inspect it. Hence, we require the obj->vma.lock to protect the list as we iterate. An example of forgetting to hold the obj->vma.lock is [1642834.464973] general protection fault, probably for non-canonical address 0xdead000000000122: 0000 [#1] SMP PTI [1642834.464977] CPU: 3 PID: 1954 Comm: Xorg Not tainted 5.6.0-300.fc32.x86_64 #1 [1642834.464979] Hardware name: LENOVO 20ARS25701/20ARS25701, BIOS GJET94WW (2.44 ) 09/14/2017 [1642834.465021] RIP: 0010:i915_gem_object_set_tiling+0x2c0/0x3e0 [i915] [1642834.465024] Code: 8b 84 24 18 01 00 00 f6 c4 80 74 59 49 8b 94 24 a0 00 00 00 49 8b 84 24 e0 00 00 00 49 8b 74 24 10 48 8b 92 30 01 00 00 89 c7 <80> ba 0a 06 00 00 03 0f 87 86 00 00 00 ba 00 00 08 00 b9 00 00 10 [1642834.465025] RSP: 0018:ffffa98780c77d60 EFLAGS: 00010282 [1642834.465028] RAX: ffff8d232bfb2578 RBX: 0000000000000002 RCX: ffff8d25873a0000 [1642834.465029] RDX: dead000000000122 RSI: fffff0af8ac6e408 RDI: 000000002bfb2578 [1642834.465030] RBP: ffff8d25873a0000 R08: ffff8d252bfb5638 R09: 0000000000000000 [1642834.465031] R10: 0000000000000000 R11: ffff8d252bfb5640 R12: ffffa987801cb8f8 [1642834.465032] R13: 0000000000001000 R14: ffff8d233e972e50 R15: ffff8d233e972d00 [1642834.465034] FS: 00007f6a3d327f00(0000) GS:ffff8d25926c0000(0000) knlGS:0000000000000000 [1642834.465036] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [1642834.465037] CR2: 00007f6a2064d000 CR3: 00000002fb57c001 CR4: 00000000001606e0 [1642834.465038] Call Trace: [1642834.465083] i915_gem_set_tiling_ioctl+0x122/0x230 [i915] [1642834.465121] ? i915_gem_object_set_tiling+0x3e0/0x3e0 [i915] [1642834.465151] drm_ioctl_kernel+0x86/0xd0 [drm] [1642834.465156] ? avc_has_perm+0x3b/0x160 [1642834.465178] drm_ioctl+0x206/0x390 [drm] [1642834.465216] ? i915_gem_object_set_tiling+0x3e0/0x3e0 [i915] [1642834.465221] ? selinux_file_ioctl+0x122/0x1c0 [1642834.465226] ? __do_munmap+0x24b/0x4d0 [1642834.465231] ksys_ioctl+0x82/0xc0 [1642834.465235] __x64_sys_ioctl+0x16/0x20 [1642834.465238] do_syscall_64+0x5b/0xf0 [1642834.465243] entry_SYSCALL_64_after_hwframe+0x44/0xa9 [1642834.465245] RIP: 0033:0x7f6a3d7b047b [1642834.465247] Code: 0f 1e fa 48 8b 05 1d aa 0c 00 64 c7 00 26 00 00 00 48 c7 c0 ff ff ff ff c3 66 0f 1f 44 00 00 f3 0f 1e fa b8 10 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d ed a9 0c 00 f7 d8 64 89 01 48 [1642834.465249] RSP: 002b:00007ffe71adba28 EFLAGS: 00000246 ORIG_RAX: 0000000000000010 [1642834.465251] RAX: ffffffffffffffda RBX: 000055f99048fa40 RCX: 00007f6a3d7b047b [1642834.465253] RDX: 00007ffe71adba30 RSI: 00000000c0106461 RDI: 000000000000000e [1642834.465254] RBP: 0000000000000002 R08: 000055f98f3f1798 R09: 0000000000000002 [1642834.465255] R10: 0000000000001000 R11: 0000000000000246 R12: 0000000000000080 [1642834.465257] R13: 000055f98f3f1690 R14: 00000000c0106461 R15: 00007ffe71adba30 Now to take the spinlock during the list iteration, we need to break it down into two phases. In the first phase under the lock, we cannot sleep and so must defer the actual work to a second list, protected by the ggtt->mutex. We also need to hold the spinlock during creation of a new vma to serialise with updates of the tiling on the object. Reported-by:
Dave Airlie <airlied@redhat.com> Fixes: 2850748e ("drm/i915: Pull i915_vma_pin under the vm->mutex") Signed-off-by:
-
Chris Wilson authored
Since batch buffers dominant execution time, most preemption requests should naturally occur during execution of a batch buffer. We wish to verify that should a preemption occur within a batch buffer, when we come to restart that batch buffer, it occurs at the interrupted instruction and most importantly does not rollback to an earlier point. v2: Do not clear the GPR at the start of the batch, but rely on them being clear for new contexts. Suggested-by:
-
Chris Wilson authored
For verifying reciving the EI interrupts, we need to hold the GPU in very precise conditions (in terms of C0 cycles during the EI). If we preempt the busy load to handle the heartbeat, this may perturb the busy load causing us to miss the interrupt. The other tests, while not as time sensitive, may also be slightly perturbed, so apply the heartbeat protection across all the measurements. Signed-off-by:
-
- 21 Apr, 2020 18 commits
-
-
Chris Wilson authored
Having noticed that MI_BB_START is incurring a memory stall (see the correlation with uncore frequency), we have to unroll the loop in order to diminish the impact of the MI_BB_START on the instruction throughput. Signed-off-by:
-
Chris Wilson authored
Since we may lose the content of any buffer when we relinquish control of the system (e.g. suspend/resume), we have to be careful not to rely on regaining control. A good method to detect when we might be using garbage is by always injecting that garbage prior to first use on load/resume/etc. v2: Drop sanitize callback on cleanup v3: Move seqno reset to timeline enter, so we reset all timelines. However, this is done on every activation during runtime and not reset. The similar level of paranoia we apply to correcting context state after a period of inactivity. Suggested-by:
-
Chris Wilson authored
Let's isolate the impact of cpu frequency selection on determing the GPU throughput in response to selection of RPS frequencies. For real systems, we do have to be concerned with the impact of integrating c-states, p-states and rp-states, but for the sake of proving whether or not RPS works, one baby step at a time. For the record, as one would hope, it does not seem to impact on the measured performance, but we do it anyway to reduce the number of variables. Later, we can extend the testing to encourage the the cpu/pkg to try and sleep while the GPU is busy. Signed-off-by:
-
Chris Wilson authored
If we detect that the RPS end points do not scale perfectly, take the time to measure all the in between values as well. We are aborting the test, so we might as well spend the available time gathering critical debug information instead. Signed-off-by:
-
Chris Wilson authored
We want to see the pstate limits whenever we fail to set the minimum frequency as that may help for debugging. Signed-off-by:
-
Pankaj Bharadiya authored
struct drm_device specific drm_WARN* macros include device information in the backtrace, so we know what device the warnings originate from. Prefer drm_WARN_ON over WARN_ON. Signed-off-by:
Pankaj Bharadiya <pankaj.laxminarayan.bharadiya@intel.com> Signed-off-by:
Jani Nikula <jani.nikula@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20200406112800.23762-14-pankaj.laxminarayan.bharadiya@intel.com
-
Pankaj Bharadiya authored
struct drm_device specific drm_WARN* macros include device information in the backtrace, so we know what device the warnings originate from. Prefer drm_WARN_ON over WARN_ON. Signed-off-by:
-
Pankaj Bharadiya authored
struct drm_device specific drm_WARN* macros include device information in the backtrace, so we know what device the warnings originate from. Prefer drm_WARN* over WARN* calls. Signed-off-by:
-
Pankaj Bharadiya authored
struct drm_device specific drm_WARN* macros include device information in the backtrace, so we know what device the warnings originate from. Prefer drm_WARN_ON over WARN_ON. Signed-off-by:
-
Pankaj Bharadiya authored
struct drm_device specific drm_WARN* macros include device information in the backtrace, so we know what device the warnings originate from. Prefer drm_WARN_ON over WARN_ON at places where struct drm_device pointer can be extracted. Signed-off-by:
-
Pankaj Bharadiya authored
struct drm_device specific drm_WARN* macros include device information in the backtrace, so we know what device the warnings originate from. Prefer drm_WARN_ON over WARN_ON at places where struct drm_device pointer can be extracted. Signed-off-by:
-
Pankaj Bharadiya authored
struct drm_device specific drm_WARN* macros include device information in the backtrace, so we know what device the warnings originate from. Prefer drm_WARN* over WARN* calls. Signed-off-by:
-
Pankaj Bharadiya authored
struct drm_device specific drm_WARN* macros include device information in the backtrace, so we know what device the warnings originate from. Prefer drm_WARN_ON over WARN_ON. Signed-off-by:
-
Pankaj Bharadiya authored
struct drm_device specific drm_WARN* macros include device information in the backtrace, so we know what device the warnings originate from. Prefer drm_WARN_ON over WARN_ON. Signed-off-by:
-
Jani Nikula authored
The early check for compressed_bpp being zero is too early, as it is hit also when DSC is not enabled. Move the checks down to where the values are actually needed. This is a paranoid check for a situation that should not happen, so we don't really care about handling it gracefully apart from not oopsing. Fixes: 48b8b04c ("drm/i915/display: Enable DP Display Audio WA") Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/1750 Cc: Anshuman Gupta <anshuman.gupta@intel.com> Cc: Uma Shankar <uma.shankar@intel.com> Reviewed-by:
Uma Shankar <uma.shankar@intel.com> Signed-off-by:
-
Jani Nikula authored
Remove a number of inlines from .c files, and let the compiler decide what's best. There's more to do, but need to start somewhere, and need to start setting the example. Acked-by:
Ville Syrjälä <ville.syrjala@linux.intel.com> Signed-off-by:
-
Jani Nikula authored
Unused, hiding from the compiler warnings behind the inline keyword. Reviewed-by:
-
Matt Roper authored
AUX power wells sometimes need additional handling besides just programming the specific power well registers: * Type-C PHY's also require additional Type-C register programming * ICL combo PHY's require additional workarounds * TGL & EHL combo PHY's can be treated like any other power well Today we have dedicated aux ops for the ICL combo PHY and Type-C cases. This works fine, but means that when a new platform shows up with identical general power well handling, but different types of PHYs on its outputs, we have to define an entire new power well table for that platform and can't just re-use the table from the earlier platform -- as an example, see ehl_power_wells[], which is a subset of icl_power_wells[], *except* that we need to specify different AUX ops for the third display. If we instead create a single set of top-level aux ops that will check the PHY type and then dispatch to the appropriate handlers, we can get more reuse out of our power well definitions. This allows us to immediately eliminate ehl_power_wells[] and simply reuse the ICL table; if future platforms follow the same general power well assignments as either ICL or TGL, we'll be able to re-use those tables in the same way. Note that I've only changed ICL+ platforms over to using the new icl_aux ops; at this point it's unlikely that we'll have any new platforms that re-use gen9 or earlier power well configurations. v2: - ICL_AUX_PW_TO_PHY() won't return the proper PHY for TBT AUX power wells. But we know those wells will only used on Type-C outputs anyway, so we can just check is is_tc_tbt flag in the condition. (Jose). Cc: José Roberto de Souza <jose.souza@intel.com> Signed-off-by:
Matt Roper <matthew.d.roper@intel.com> Reviewed-by:
-
- 20 Apr, 2020 14 commits
-
-
Xiyu Yang authored
igt_ppgtt_pin_update() invokes i915_gem_context_get_vm_rcu(), which returns a reference of the i915_address_space object to "vm" with increased refcount. When igt_ppgtt_pin_update() returns, "vm" becomes invalid, so the refcount should be decreased to keep refcount balanced. The reference counting issue happens in two exception handling paths of igt_ppgtt_pin_update(). When i915_gem_object_create_internal() returns IS_ERR, the refcnt increased by i915_gem_context_get_vm_rcu() is not decreased, causing a refcnt leak. Fix this issue by jumping to "out_vm" label when i915_gem_object_create_internal() returns IS_ERR. Fixes: a4e7ccda ("drm/i915: Move context management under GEM") Signed-off-by:
Xiyu Yang <xiyuyang19@fudan.edu.cn> Signed-off-by:
Xin Tan <tanxin.ctf@gmail.com> Reviewed-by:
-
Chris Wilson authored
After having testing all the RPS controls individually, we need to take a step back and check how our RPS worker integrates them to perform dynamic GPU reclocking. So do that by submitting a spinner and wait and see what happens. Signed-off-by:
-
Chris Wilson authored
If we encounter an error while scaling, read back the frequency tables from the pcu. Signed-off-by:
-
Chris Wilson authored
Split the frequency measurement into two modes, so that we can judge the impact of the llc setup on top of the pure CS frequency scaling. Signed-off-by:
-
Chris Wilson authored
Check that the GPU does respond to our RPS frequency requests by setting our desired frequency. Signed-off-by:
-
Chris Wilson authored
If we can not manipulate the frequency with RPS, then comparing min/max power consumption is pointless / misleading. We will leave the warning about not being able to control the frequency selection via RPS to other tests so as not to confuse this more specialised check. Signed-off-by:
-
Chris Wilson authored
One of the core tenents of reclocking the GPU is that its throughput scales with the clock frequency. We can observe this by incrementing a loop counter on the GPU, and compare the different execution rates at the notional RPS frequencies. Signed-off-by:
-
Ville Syrjälä authored
We shouldn't try to do link retraining from the short hpd handler. We can't take any modeset locks there so this is racy as hell. Push the whole thing into the hotplug work like we do with SST. We'll just have to adjust the SST retraining code to deal with the MST encoders and multiple pipes. TODO: I have a feeling we should just rip this all out and do a full modeset instead. Stuff like port sync and the tgl+ MST master transcoder stuff maybe doesn't work well if we try to retrain without following the proper modeset sequence. So far haven't done any actual tests to confirm that though. Cc: Lyude Paul <lyude@redhat.com> Signed-off-by:
Lyude Paul <lyude@redhat.com>
-
Ville Syrjälä authored
Make intel_dp_check_mst_status() somewhat legible by humans. Note that the return value of drm_dp_mst_hpd_irq() is always either 0 or -ENOMEM, and we never did anything with the latter so we can just ignore the whole thing. We can also get rid of the direct drm_dp_mst_topology_mgr_set_mst(false) call since returning -EINVAL causes the caller to do the very same call for us. Cc: Lyude Paul <lyude@redhat.com> Signed-off-by:
-
Ville Syrjälä authored
Pass the encoder all the way down to intel_ddi_transcoder_func_reg_val_get(). Allows us eliminate the intel_ddi_get_crtc_encoder() eyesore. Signed-off-by:
-
Ville Syrjälä authored
Push the TRANS_DDI_FUNC_CTL into the encoder enable hook. The disable is already there, and as a followup will enable us to pass the encoder all the way down. Signed-off-by:
-
Ville Syrjälä authored
No reason that I can see why we should enable TRANS_DDI_FUNC_CTL before we set up the watermarks of configure the mbus stuff. In fact reordering these seems to match the bspec sequence better, and crucially will allow us to push the TRANS_DDI_FUNC_CTL enable into the encoder enable hook as a followup. Signed-off-by:
-
Ville Syrjälä authored
Since intel_ddi_enable_pipe_clock() was pushed down into the encoder hooks we can pass on the encoder instead of having to use intel_ddi_get_crtc_encoder(). Signed-off-by:
-
Peter Jones authored
This was sort of annoying me: random:~$ dmesg | tail -1 [523884.039227] [drm] Reducing the compressed framebuffer size. This may lead to less power savings than a non-reduced-size. Try to increase stolen memory size if available in BIOS. random:~$ dmesg | grep -c "Reducing the compressed" 47 This patch makes it DRM_INFO_ONCE() just like the similar message farther down in that function is pr_info_once(). Cc: stable@vger.kernel.org Signed-off-by:
Peter Jones <pjones@redhat.com> Acked-by:
Rodrigo Vivi <rodrigo.vivi@intel.com> Closes: https://gitlab.freedesktop.org/drm/intel/-/issues/1745 Link: https://patchwork.freedesktop.org/patch/msgid/20180706190424.29194-1-pjones@redhat.com [vsyrjala: Rebase due to per-device logging] Signed-off-by:
-
