- 10 Jul, 2018 5 commits
-
-
Chris Wilson authored
In the next patch, we will make a fairly minor change to flush outstanding resets before suspend. In order to keep churn to a minimum in that functional patch, we fix up the comments and coding style now. Signed-off-by:
Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by:
Mika Kuoppala <mika.kuoppala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180709130208.11730-7-chris@chris-wilson.co.uk
-
Chris Wilson authored
Across a reset, the seqno (and thus hangcheck) should restart and the hangcheck naturally progress, for when it does not, we want to declare an emergency. Currently, we only detect if reset and reinit fails, but we do not detect if the call to reinit succeeds but the HW is fried - as we are resetting hangcheck on initialisation the engine. Remove that and rely on the natural progress to reset the hangcheck timer. References: e21b1413 ("drm/i915: Mark the hangcheck as idle when unparking the engines") References: 1fd00c0f ("drm/i915: Declare the driver wedged if hangcheck makes no progress") Signed-off-by:
-
Chris Wilson authored
In our swizzling selftests, we cannot predict the physical address of the target page (at least not simply!) and so skip bit17 swizzles. However, there are two bit17 swizzle modes and we only skipped one, with the second being observed on the lab gdg causing the test to fail, as soon as we hit a page with bit17 set in its address. Testcase: igt/drv_selftest/live_objects #gdg Signed-off-by:
Matthew Auld <matthew.auld@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180709194915.5789-1-chris@chris-wilson.co.uk
-
Chris Wilson authored
Be pessimistic and presume that we actually allocate every page we exercise via the mock_gtt (e.g. for gvt). In which case we have to keep our working set under the available physical memory to prevent oom. Signed-off-by:
-
Chris Wilson authored
Error messages are intended to be addressed to the user; be clear, succinct, instructive and unambiguous. Adding the function name to that message does not add any information the user requires and in the process makes the message less clear. E.g. [ 245.539711] i915 0000:00:02.0: [drm:i915_gem_init [i915]] Failed to initialize GPU, declaring it wedged! becomes [ 245.539711] i915 0000:00:02.0: Failed to initialize GPU, declaring it wedged! Signed-off-by:
Rodrigo Vivi <rodrigo.vivi@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180709134858.12446-1-chris@chris-wilson.co.uk
-
- 09 Jul, 2018 6 commits
-
-
Rodrigo Vivi authored
Signed-off-by: -
Chris Wilson authored
igt_mmap_offset_exhaustion() wants to test what happens when the mmap space is filled with zombie objects, objects discarded by userspace but still active on the GPU. As they are only protected by the active reference, we have to be certain that active reference is kept while we peek into our dangling pointer. That active reference should not be freed until we retire, but we do that retirement from a background thread. This leaves us with a subtle timing problem, exacerbated and highlighted by KASAN: <3>[ 132.380399] BUG: KASAN: use-after-free in drm_gem_create_mmap_offset+0x8c/0xd0 <3>[ 132.380430] Read of size 8 at addr ffff8801e13245f8 by task drv_selftest/5822 <4>[ 132.380470] CPU: 0 PID: 5822 Comm: drv_selftest Tainted: G U 4.18.0-rc3-g7ae7763aa2be-kasan_48+ #1 <4>[ 132.380473] Hardware name: Dell Inc. XPS 8300 /0Y2MRG, BIOS A06 10/17/2011 <4>[ 132.380475] Call Trace: <4>[ 132.380481] dump_stack+0x7c/0xbb <4>[ 132.380487] print_address_description+0x65/0x270 <4>[ 132.380493] kasan_report+0x25b/0x380 <4>[ 132.380497] ? drm_gem_create_mmap_offset+0x8c/0xd0 <4>[ 132.380503] drm_gem_create_mmap_offset+0x8c/0xd0 <4>[ 132.380584] i915_gem_object_create_mmap_offset+0x6d/0x100 [i915] <4>[ 132.380650] igt_mmap_offset_exhaustion+0x462/0x940 [i915] <4>[ 132.380714] ? i915_gem_close_object+0x740/0x740 [i915] <4>[ 132.380784] ? igt_gem_huge+0x269/0x3d0 [i915] <4>[ 132.380865] __i915_subtests+0x5a/0x160 [i915] <4>[ 132.380936] __run_selftests+0x1a2/0x2f0 [i915] <4>[ 132.381008] i915_live_selftests+0x4e/0x80 [i915] <4>[ 132.381071] i915_pci_probe+0xd8/0x1b0 [i915] <4>[ 132.381077] pci_device_probe+0x1c5/0x3a0 <4>[ 132.381087] driver_probe_device+0x6b6/0xcb0 <4>[ 132.381094] __driver_attach+0x22d/0x2c0 <4>[ 132.381100] ? driver_probe_device+0xcb0/0xcb0 <4>[ 132.381103] bus_for_each_dev+0x113/0x1a0 <4>[ 132.381108] ? check_flags.part.24+0x450/0x450 <4>[ 132.381112] ? subsys_dev_iter_exit+0x10/0x10 <4>[ 132.381123] bus_add_driver+0x38b/0x6e0 <4>[ 132.381131] driver_register+0x189/0x400 <4>[ 132.381136] ? 0xffffffffc12d8000 <4>[ 132.381140] do_one_initcall+0xa0/0x4c0 <4>[ 132.381145] ? initcall_blacklisted+0x180/0x180 <4>[ 132.381152] ? do_init_module+0x4a/0x54c <4>[ 132.381156] ? rcu_lockdep_current_cpu_online+0xdc/0x130 <4>[ 132.381161] ? kasan_unpoison_shadow+0x30/0x40 <4>[ 132.381169] do_init_module+0x1b5/0x54c <4>[ 132.381177] load_module+0x619e/0x9b70 <4>[ 132.381202] ? module_frob_arch_sections+0x20/0x20 <4>[ 132.381211] ? vfs_read+0x257/0x2f0 <4>[ 132.381214] ? vfs_read+0x257/0x2f0 <4>[ 132.381221] ? kernel_read+0x8b/0x130 <4>[ 132.381231] ? copy_strings_kernel+0x120/0x120 <4>[ 132.381244] ? __se_sys_finit_module+0x17c/0x1a0 <4>[ 132.381248] __se_sys_finit_module+0x17c/0x1a0 <4>[ 132.381252] ? __ia32_sys_init_module+0xa0/0xa0 <4>[ 132.381261] ? __se_sys_newstat+0x77/0xd0 <4>[ 132.381265] ? cp_new_stat+0x590/0x590 <4>[ 132.381269] ? kmem_cache_free+0x2f0/0x340 <4>[ 132.381285] do_syscall_64+0x97/0x400 <4>[ 132.381292] entry_SYSCALL_64_after_hwframe+0x49/0xbe <4>[ 132.381295] RIP: 0033:0x7eff4af46839 <4>[ 132.381297] Code: 00 f3 c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 1f f6 2c 00 f7 d8 64 89 01 48 <4>[ 132.381426] RSP: 002b:00007ffcd84f4cf8 EFLAGS: 00000246 ORIG_RAX: 0000000000000139 <4>[ 132.381432] RAX: ffffffffffffffda RBX: 000055dfdeb429a0 RCX: 00007eff4af46839 <4>[ 132.381435] RDX: 0000000000000000 RSI: 000055dfdeb43670 RDI: 0000000000000004 <4>[ 132.381437] RBP: 000055dfdeb43670 R08: 0000000000000004 R09: 0000000000000000 <4>[ 132.381440] R10: 00007ffcd84f4e60 R11: 0000000000000246 R12: 0000000000000000 <4>[ 132.381442] R13: 000055dfdeb3bec0 R14: 0000000000000000 R15: 000000000000003b <3>[ 132.381466] Allocated by task 5822: <4>[ 132.381485] kmem_cache_alloc+0xdf/0x2e0 <4>[ 132.381546] i915_gem_object_create_internal+0x24/0x1e0 [i915] <4>[ 132.381609] igt_mmap_offset_exhaustion+0x257/0x940 [i915] <4>[ 132.381677] __i915_subtests+0x5a/0x160 [i915] <4>[ 132.381742] __run_selftests+0x1a2/0x2f0 [i915] <4>[ 132.381806] i915_live_selftests+0x4e/0x80 [i915] <4>[ 132.381865] i915_pci_probe+0xd8/0x1b0 [i915] <4>[ 132.381868] pci_device_probe+0x1c5/0x3a0 <4>[ 132.381871] driver_probe_device+0x6b6/0xcb0 <4>[ 132.381874] __driver_attach+0x22d/0x2c0 <4>[ 132.381877] bus_for_each_dev+0x113/0x1a0 <4>[ 132.381880] bus_add_driver+0x38b/0x6e0 <4>[ 132.381884] driver_register+0x189/0x400 <4>[ 132.381886] do_one_initcall+0xa0/0x4c0 <4>[ 132.381889] do_init_module+0x1b5/0x54c <4>[ 132.381892] load_module+0x619e/0x9b70 <4>[ 132.381895] __se_sys_finit_module+0x17c/0x1a0 <4>[ 132.381898] do_syscall_64+0x97/0x400 <4>[ 132.381901] entry_SYSCALL_64_after_hwframe+0x49/0xbe <3>[ 132.381914] Freed by task 150: <4>[ 132.381931] kmem_cache_free+0xb7/0x340 <4>[ 132.381995] __i915_gem_free_objects+0x875/0xf50 [i915] <4>[ 132.382054] __i915_gem_free_work+0x69/0xb0 [i915] <4>[ 132.382058] process_one_work+0x78b/0x1740 <4>[ 132.382061] worker_thread+0x82/0xb80 <4>[ 132.382064] kthread+0x30c/0x3d0 <4>[ 132.382067] ret_from_fork+0x3a/0x50 <3>[ 132.382081] The buggy address belongs to the object at ffff8801e1324500 which belongs to the cache drm_i915_gem_object of size 1168 <3>[ 132.382133] The buggy address is located 248 bytes inside of 1168-byte region [ffff8801e1324500, ffff8801e1324990) <3>[ 132.382179] The buggy address belongs to the page: <0>[ 132.382202] page:ffffea000784c800 count:1 mapcount:0 mapping:ffff8801dedf6500 index:0xffff8801e1323ec0 compound_mapcount: 0 <0>[ 132.382251] flags: 0x8000000000008100(slab|head) <1>[ 132.382274] raw: 8000000000008100 ffff8801d6317440 ffff8801d6317440 ffff8801dedf6500 <1>[ 132.382307] raw: ffff8801e1323ec0 0000000000140013 00000001ffffffff 0000000000000000 <1>[ 132.382339] page dumped because: kasan: bad access detected <3>[ 132.382373] Memory state around the buggy address: <3>[ 132.382395] ffff8801e1324480: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc <3>[ 132.382426] ffff8801e1324500: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb <3>[ 132.382457] >ffff8801e1324580: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb <3>[ 132.382488] ^ <3>[ 132.382517] ffff8801e1324600: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb <3>[ 132.382548] ffff8801e1324680: fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb This patch tricks the system into running without the background retire thread, until after we finish the test. The only reaping should then be performed by the mmap offset routine to reclaim the space as required. Signed-off-by: -
Chris Wilson authored
In igt_flush_test() we install a background timer in order to ensure that the wait completes within a certain time. We can now tell the wait that it has to complete within a timeout, and so no longer need the background timer. Signed-off-by:
-
Chris Wilson authored
With a broken GPU we expect it to fail during the initial GPU setup where do a couple of context switches to record the defaults. This is a task that takes a few milliseconds even on the slowest of devices, but we may have to wait 60s for hangcheck to give in and declare the machine inoperable. In this a case where any gpu hang is unacceptable, both from a timeliness and practical standpoint. We can therefore set a timeout on our wait-for-idle that is shorter than the hangcheck (which may be up to 60s for a declaring a wedged driver) and so detect the broken GPU much more quickly during driver load (and so prevent stalling userspace for ages). Signed-off-by:
-
Chris Wilson authored
Usually we have no idea about the upper bound we need to wait to catch up with userspace when idling the device, but in a few situations we know the system was idle beforehand and can provide a short timeout in order to very quickly catch a failure, long before hangcheck kicks in. In the following patches, we will use the timeout to curtain two overly long waits, where we know we can expect the GPU to complete within a reasonable time or declare it broken. In particular, with a broken GPU we expect it to fail during the initial GPU setup where do a couple of context switches to record the defaults. This is a task that takes a few milliseconds even on the slowest of devices, but we may have to wait 60s for hangcheck to give in and declare the machine inoperable. In this a case where any gpu hang is unacceptable, both from a timeliness and practical standpoint. The other improvement is that in selftests, we do not need to arm an independent timer to inject a wedge, as we can just limit the timeout on the wait directly. v2: Include the timeout parameter in the trace. Signed-off-by:
-
Chris Wilson authored
i915g has a slightly different tiling layout, and so requires a different reference swizzle pattern. Testcase: igt/drv_selftests/live_objects #gdg Signed-off-by:
-
- 07 Jul, 2018 1 commit
-
-
Chris Wilson authored
In the next patch, we will want a third distinct class of timeline that may overlap with the current pair of client and engine timeline classes. Rather than use the ad hoc markup of SINGLE_DEPTH_NESTING, initialise the different timeline classes with an explicit subclass. Signed-off-by:
Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180706210710.16251-1-chris@chris-wilson.co.uk
-
- 06 Jul, 2018 28 commits
-
-
Chris Wilson authored
Inside the mock GEM device, we try to grab the runtime pm for the fake device to prevent it from ever suspending. However, if CONFIG_PM is not set, trying to obtain the wakref returns an error which we WARN about. Suppress the expected warning. Signed-off-by:
-
Chris Wilson authored
clflush is an unserialised instruction and the IA manual strongly advises you to serialise it with a mb. To be cautious, apply one before and one after, so that it is serialised with both writes and reads without worrying too much about the required direction. Signed-off-by:
-
Chris Wilson authored
Using a VMA on more than one timeline concurrently is the exception rather than the rule (using it concurrently on multiple engines). As we expect to only use one active tracker, store the most recently used tracker inside the i915_vma itself and only fallback to the rbtree if we need a second or more concurrent active trackers. v2: Comments on how we overwrite any existing last_active cache. v3: __list_del_entry() before list_replace_init() is confusing and, much more important, entirely redundant. v4: Note that both last_active and the rbtree may be simultaneously tracking this timeline, albeit with different requests, and so the vma may be retired twice for the same timeline. v5: No, that list_del is required! Signed-off-by:
-
Chris Wilson authored
In the next patch, we will want to be able to use more flexible request timelines that can hop between engines. From the vma pov, we can then not rely on the binding of this request to an engine and so can not ensure that different requests are ordered through a per-engine timeline, and so we must track activity of all timelines. (We track activity on the vma itself to prevent unbinding from HW before the HW has finished accessing it.) v2: Switch to a rbtree for 32b safety (since using u64 as a radixtree index is fraught with aliasing of unsigned longs). v3: s/lookup_active/active_instance/ because we can never agree on names Signed-off-by:
-
Chris Wilson authored
i915_vma_move_to_active() has grown beyond its execbuf origins, and should take its rightful place in i915_vma.c as a method for i915_vma! Signed-off-by:
-
Chris Wilson authored
Handling such a late error in request construction is tricky, but to accommodate future patches which may allocate here, we potentially could err. To handle the error after already adjusting global state to track the new request, we must finish and submit the request. But we don't want to use the request as not everything is being tracked by it, so we opt to cancel the commands inside the request. Signed-off-by:
-
Chris Wilson authored
In the next patch, we will want to start skipping requests on failing to complete their payloads. So export the utility function current used to make requests inoperable following a failed gpu reset. Signed-off-by:
-
Chris Wilson authored
Currently all callers are responsible for adding the vma to the active timeline and then exporting its fence. Combine the two operations into i915_vma_move_to_active() to move all the extra handling from the callers to the single site. Signed-off-by:
-
Chris Wilson authored
We always want to use a virtual address (i.e. use the GTT) for MI_STORE_DWORD_IMM, but forgot the ever so important flag in live_hangcheck for gen3. Signed-off-by:
Ville Syrjälä <ville.syrjala@linux.intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180706142323.25699-2-chris@chris-wilson.co.uk
-
Chris Wilson authored
Replace the magic bit with the proper symbolic name for instructing MI_STORE_DWORD_IMM to use a virtual address (on gen3) or the global GTT address (still virtual!) on gen4+. Signed-off-by:
-
Chris Wilson authored
Limit the GTT size we try and allocate to ensure that it fits within RAM and does not trigger the oomkiller indiscriminately. Signed-off-by:
-
Chris Wilson authored
We already maually control the CPU cache for our page table directories, so we can tell the dma mapper to skip doing it as well. Signed-off-by:
-
Chris Wilson authored
As we propagate back the error to the caller for them to handle, we do not need the lowest level spitting out a redundant warning upon an allocation failure inside dma_map_page(). Signed-off-by:
-
Chris Wilson authored
If we have just completed a WC write, we must ensure that the WCB (Write Combining Buffer) is flushed out to main memory before we can expect to see the results. This is especially important when mixing WC with GTT as the physical paths are different and cachelines are not naturally flushed. Testcase: igt/drv_selftests/live_coherency #gdg Signed-off-by:
-
Chris Wilson authored
If the GPU is irrecoverably wedged, we can not execute any requests making testing execlists (request execution) pointless. Skip! Signed-off-by:
-
Chris Wilson authored
If the HW (or driver) doesn't support logical contexts, don't pretend we gain anything from trying to execute GPU commands with them. At best it reports -ENODEV, which is an unhelpful failure that we should just skip. v2: Be more specific and check the driver/engine caps for logical (HW) context support. Signed-off-by:
-
Chris Wilson authored
Avoid looking at the magical engines[RCS] to decide if the HW and driver supports logical contexts, and instead record that knowledge during initialisation. Signed-off-by:
-
Imre Deak authored
We can simplify the encoder's get_power_domains() hook by calling it only if the encoder is active. That way the hook can return its power domains unconditionally without checking the active state by calling encoder::get_hw_state(). This get_hw_state() query is in fact redundant since it's already done by intel_modeset_readout_hw_state() setting the encoder's crtc or leaving it NULL accordingly. Let's use this fact to decide if the encoder is active. While at it clarify the comment in intel_ddi_get_power_domains() about primary vs. fake MST encoders and make sure we never do an incorrect encoder->dig_port cast for fake MST encoders. Suggested-by:
Imre Deak <imre.deak@intel.com> Reviewed-by:
-
Maarten Lankhorst authored
This interface is deprecated, and has been replaced by the upstream drm crc interface. Signed-off-by:
Maarten Lankhorst <maarten.lankhorst@linux.intel.com> Cc: Tomi Sarvela <tomi.p.sarvela@intel.com> Cc: Petri Latvala <petri.latvala@intel.com> Cc: Jani Nikula <jani.nikula@intel.com> Cc: Ville Syrjälä <ville.syrjala@linux.intel.com> Acked-by:
Jani Nikula <jani.nikula@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180628072303.14175-1-maarten.lankhorst@linux.intel.com
-
Chris Wilson authored
If the GPU is terminally wedged we cannot submit any requests into a context, completely unfulfilling our purpose of doing so. As this expectedly fails, skip over the test. Signed-off-by:
-
Chris Wilson authored
We test the GPU handling of huge pages by submitting requests that write into a huge page, but if the GPU is irrecoverably wedged we cannot submit any requests. As the test expectedly fails, skip over it. Signed-off-by:
-
Chris Wilson authored
If the GPU is irrecoverably wedged, we cannot submit any requests and so cannot make the GTT busy in order to test evicting active objects. As this expectedly fails, skip over the test. Signed-off-by:
-
Chris Wilson authored
If the GPU is irrecoverably wedged, we cannot submit any request and therefore cannot query the register state of the context (which is done using the GPU command stream). So skip over the test as it expectedly fails. Signed-off-by:
-
Chris Wilson authored
If the GPU is irrecoverably wedge, we cannot submit any request and so all of the request selftests will expectedly fail. Skip over them. Signed-off-by:
-
Chris Wilson authored
If the GPU is wedged, we cannot make the object busy as trying to submit a request will generate -EIO. Skip to the end of the test. Signed-off-by:
-
Chris Wilson authored
If the GPU is irrecoverably broken, we can not use it to dirty memory and check for cache coherency with the CPU. All we can do is simply skip over the GPU subtests and focus on the CPU domains (WC, WB) cache management. Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=107127Signed-off-by:
-
Chris Wilson authored
As we keep VMA around until the object is destroyed, when testing partial tiling we instantiate many, many VMA (as the object is huge allowing for many different partial regions). We test elsewhere our handling of populating large objects with a full set of VMA and checking we can retrieve them afterwards, but in this test we incur the cost of flushing all VMA after every GTT write, dramatically slowing down the test. References: https://bugs.freedesktop.org/show_bug.cgi?id=107130Signed-off-by:
-
Chris Wilson authored
Having found the error causing the IGT test to fail, downgrade the verbose logging so that we stop flooding the syslogs as we deliberately provoke it many thousands of time during selftests. References: 10195b1e ("drm/i915: Show vma allocator stack when in doubt") Signed-off-by:
-
