summaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/nouveau/nvkm/engine/fifo
AgeCommit message (Collapse)AuthorFilesLines
2017-02-17drm/nouveau/fifo/gk104-: trigger mmu fault before attempting engine recoveryBen Skeggs1-0/+41
Greatly improves the chances of recovering the GPU from a CTXSW_TIMEOUT. Tested with piglit's arb_shader_image_load_store-atomicity, which causes GR to hang in such a way that recovery failed (CTXSW_TIMEOUT continually re-triggers). Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2017-02-17drm/nouveau/fifo/gk104-: ACK SCHED_ERROR before attempting CTXSW_TIMEOUT ↵Ben Skeggs1-0/+8
recovery Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2017-02-17drm/nouveau/fifo/gk104-: directly use new recovery code for ctxsw timeoutBen Skeggs1-24/+6
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2017-02-17drm/nouveau/fifo/gk104-: directly use new recovery code for mmu faultsBen Skeggs3-8/+37
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2017-02-17drm/nouveau/fifo/gk104-: reset all engines a killed channel is still active onBen Skeggs1-0/+13
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2017-02-17drm/nouveau/fifo/gk104-: refactor recovery codeBen Skeggs1-14/+81
This will serve as a basis for implementing some improvements to how we recover the GPU from channel errors. Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2017-02-17drm/nouveau/fifo/gk104-: better detection of chid when parsing engine statusBen Skeggs1-1/+19
The previous commit simply changes the interface, but should result in the same behaviour as previously. This commit has been split out from it as it can result in a different channel being selected. Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2017-02-17drm/nouveau/fifo/gk104-: separate out engine status parsingBen Skeggs1-13/+46
We'll be wanting to reuse this logic in more places. Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2017-02-17drm/nouveau/fifo: add an api for initiating channel recoveryBen Skeggs2-0/+13
This will be used by callers outside of fifo interrupt handlers. Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2017-02-17drm/nouveau/fifo/gf100-: provide notification to user if channel is killedBen Skeggs8-5/+55
There are instances (such as non-recoverable GPU page faults) where NVKM decides that a channel's context is no longer viable, and will be removed from the runlist. This commit notifies the owner of the channel when this happens, so it has the opportunity to take some kind of recovery action instead of hanging. Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2017-02-17drm/nouveau/fifo/g84-: rename non-stall interrupt eventBen Skeggs1-1/+1
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2017-02-17drm/nouveau/fifo: tidy up channel creation event codeBen Skeggs3-6/+13
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2017-02-17drm/nouveau/core/memory: distinguish between coherent/non-coherent targetsBen Skeggs2-5/+15
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2017-02-17drm/nouveau/dma: lookup objects with nvkm_object_search()Ben Skeggs1-3/+3
Custom code is no longer needed here. Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-12-13drm/nouveau/fifo/gf100-: recover from host mmu faultsBen Skeggs4-28/+29
This has been on the TODO list for a while now, recovering from things such as attempting to execute a push buffer or touch a semaphore in an unmapped memory area. The only thing required on the HW side here is that the offending channel is removed from the runlist, and *not* a full reset of PFIFO. This used to be a bit messier to handle before the rework to make use of engine topology info, but is apparently now trivial. Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-11-07drm/nouveau/fifo/gf100-: protect channel preempt with subdev mutexBen Skeggs2-6/+11
This avoids an issue that occurs when we're attempting to preempt multiple channels simultaneously. HW seems to ignore preempt requests while it's still processing a previous one, which, well, makes sense. Fixes random "fifo: SCHED_ERROR 0d []" + GPCCS page faults during parallel piglit runs on (at least) GM107. Signed-off-by: Ben Skeggs <bskeggs@redhat.com> Cc: stable@vger.kernel.org
2016-11-07drm/nouveau: mark symbols static where possibleBaoyou Xie1-2/+2
We get a few warnings when building kernel with W=1: drivers/gpu/drm/nouveau/nvkm/subdev/bios/fan.c:29:1: warning: no previous prototype for 'nvbios_fan_table' [-Wmissing-prototypes] drivers/gpu/drm/nouveau/nvkm/subdev/bios/fan.c:56:1: warning: no previous prototype for 'nvbios_fan_entry' [-Wmissing-prototypes] drivers/gpu/drm/nouveau/nvkm/subdev/clk/gt215.c:184:1: warning: no previous prototype for 'gt215_clk_info' [-Wmissing-prototypes] drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramgt215.c:99:1: warning: no previous prototype for 'gt215_link_train_calc' [-Wmissing-prototypes] drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramgt215.c:153:1: warning: no previous prototype for 'gt215_link_train' [-Wmissing-prototypes] drivers/gpu/drm/nouveau/nvkm/subdev/fb/ramgt215.c:271:1: warning: no previous prototype for 'gt215_link_train_init' [-Wmissing-prototypes] .... In fact, both functions are only used in the file in which they are declared and don't need a declaration, but can be made static. So this patch marks these functions with 'static'. Signed-off-by: Baoyou Xie <baoyou.xie@linaro.org> Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-09-22drm/nouveau/fifo/nv04: avoid ramht race against cookie insertionIlia Mirkin1-0/+3
Signed-off-by: Ilia Mirkin <imirkin@alum.mit.edu> Cc: stable@vger.kernel.org Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-07-14drm/nouveau/fifo/gp100: initial supportBen Skeggs4-0/+104
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-07-14drm/nouveau/fifo/gk104-: translate engidx into human-readable name in debug ↵Ben Skeggs1-2/+2
output Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-07-14drm/nouveau/top: take nvkm_device as argument to public functionsBen Skeggs1-3/+2
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-05-20drm/nouveau/fifo/gm107-: remove engines from mmu engine mapping arrayBen Skeggs4-3/+28
These are specified by PTOP on Maxwell GPUs. Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-05-20drm/nouveau/fifo/gk104-: identify mmu engine ids for host faultsBen Skeggs1-4/+14
It appears these don't map to PBDMAs (at least on Kepler, it may or may be valid for Fermi - this hasn't been checked), but to runlists. This drops the NVKM_ENGINE_FIFO data from the entries too, as resetting all of PFIFO is *not* the way to handle such faults. Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-05-20drm/nouveau/fifo/gk104-: implement support for PTOP fault infoBen Skeggs1-2/+16
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-05-20drm/nouveau/fifo/gk104-: abstract mmu fault data structuresBen Skeggs8-111/+150
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-05-20drm/nouveau/fifo/gk104-: subclass funcBen Skeggs10-73/+65
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-05-20drm/nouveau/fifo/gk104-: use device info from top subdevBen Skeggs1-73/+16
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-05-20drm/nouveau/core: remove pmc_enable argument from subdev ctorBen Skeggs1-2/+1
These are now specified directly in the MC subdev. Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-03-14drm/nouveau/fifo/gm107: KeplerChannelGpfifoB, and 2048 channelsBen Skeggs2-0/+47
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-03-14drm/nouveau/fifo/gk110: expose KeplerChannelGpfifoBBen Skeggs4-0/+83
This class supports a WFI method (0x0078) that's not present on the KeplerChannelGpfifoA class. The binary driver exposes both classes on these GPUs for some reason, though there doesn't appear to be any difference in the setup that's done for each (ie. even if you allocate GpfifoA, the WFI method will still work). We shall just expose GpfifoB, as I don't see a good reason to report the presence of both. Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-03-14drm/nouveau/fifo/gk104: submit NOP after all PBDMA_INTR_0, not just DEVICEBen Skeggs1-1/+2
Prevents the same interrupt from re-triggering forever. Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-03-14drm/nouveau/fifo/gk104: add vic plumbingBen Skeggs1-0/+2
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-03-14drm/nouveau/fifo/gk104: add sec plumbingBen Skeggs1-0/+2
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-03-14drm/nouveau/fifo/gk104: add nvdec plumbingBen Skeggs2-0/+3
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-03-14drm/nouveau/fifo/gk104: add nvenc plumbingBen Skeggs2-7/+23
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-03-14drm/nouveau/fifo/gk104: add msenc plumbingBen Skeggs1-0/+1
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-03-14drm/nouveau/fifo/gk104: make use of topology info during gpfifo constructionBen Skeggs1-47/+97
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-03-14drm/nouveau/fifo/gk104: make use of topology info during fault recoveryBen Skeggs2-34/+26
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-03-14drm/nouveau/fifo/gk104: make use of topology info when handling ctxsw timeoutBen Skeggs1-20/+9
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-03-14drm/nouveau/fifo/gk104: read device topology information from hwBen Skeggs3-5/+103
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-03-14drm/nouveau/fifo/gk104: cosmetic engine->runlist changesBen Skeggs4-48/+50
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-03-14drm/nouveau/fifo/gk104: don't attempt recovery of unknown mmu enginesBen Skeggs1-1/+1
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-03-14drm/nouveau/fifo/gk104: identify fault-recovery members more clearlyBen Skeggs2-10/+12
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-03-14drm/nouveau/fifo/gk104: rename spoon to pbdma, and move detection to oneinitBen Skeggs2-10/+14
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-03-14drm/nouveau/fifo/gf100: fix certain engines not being recovered after a faultBen Skeggs1-2/+2
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-03-14drm/nouveau/fifo/gf100: don't attempt recovery of unknown mmu enginesBen Skeggs1-1/+1
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-03-14drm/nouveau/fifo/gf100: identify fault-recovery members more clearlyBen Skeggs2-10/+12
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-03-14drm/nouveau/fifo/gf100: rename spooon to pbdma, and move detection to oneinitBen Skeggs2-12/+17
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-03-14drm/nouveau/fifo/gk104: kick channel upon removalAlexandre Courbot1-0/+1
A channel may still be processed by the PBDMA even after removal, unless it is properly kicked. Some chips are more sensible to this than others, with GM20B triggering the issue very easily (the PBDMA will try to fetch methods from the previously-removed channel after a new one is added). Make sure this cannot happen by kicking the channel right after it is disabled, and before the new runlist is submitted. Signed-off-by: Alexandre Courbot <acourbot@nvidia.com> Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
2016-03-14drm/nouveau/fifo/gk104: take runlist target into accountAlexandre Courbot1-1/+5
Bits 28:29 of RUNLIST_BASE specify the memory target of the runlist. Set it to 0x3 (SYS_MEM_NONCOHERENT) if the runlist object resides in system memory. Signed-off-by: Alexandre Courbot <acourbot@nvidia.com> Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
generated by cgit v1.2.3 (git 2.39.0) at 2025-09-27 06:11:45 +0300