summaryrefslogtreecommitdiffstats
path: root/drivers/gpu/nvgpu/gk20a
Commit message (Collapse)AuthorAge
* gpu: nvgpu: IOCTL to set TSG timesliceDeepak Nibade2015-11-03
| | | | | | | | | | | | | | | | | | | | | | | | | Add new IOCTL NVGPU_IOCTL_TSG_SET_PRIORITY to allow setting timeslice for entire TSG Return error from channel specific IOCTL_CHANNEL_SET_PRIORITY if the channel is part of TSG Separate out API gk20a_channel_get_timescale_from_timeslice() to get timeslice_timeout and scale from timeslice period Use this API to get timeslice_timeout and scale for TSG and store it in tsg_gk20a structure Then trigger runlist update so that new timeslice values will be re-written to runlist for TSG Bug 200146615 Change-Id: I555467d034f81b372b31372f0835d72b1c159508 Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: http://git-master/r/824206 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: IOCTL to disable timeoutsDeepak Nibade2015-11-03
| | | | | | | | | | | | | | | | | | | | Add IOCTL NVGPU_DBG_GPU_IOCTL_TIMEOUT to support disabling/re-enabling scheduler timeout from user space If user space application is closed without re-enabling the timeouts, kernel will restore the timeouts' state while releasing the debug session This is needed for debugging purpose Bug 1514061 Change-Id: I32efb47ad09d793f3e7fd8f0aaa9720c8bc91272 Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: http://git-master/r/788176 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: Implement sparse PDEsTerje Bergstrom2015-10-30
| | | | | | | Change-Id: Idfeb3bf95751902d52a895d77045a529f69abc0b Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/758651 GVS: Gerrit_Virtual_Submit
* gpu: nvgpu: update thermal programmingSeshendra Gadagottu2015-10-27
| | | | | | | | | | | | | | | Add required fileds and values for thermal slow-down settings in thermal header file and implemented chip specific thermal register programming Reviewed-on: http://git-master/r/822199 (cherry picked from commit 9e8a745b8295af002b9780c83caa8dc7b22cc737) Change-Id: I016b18ed230fa6c104eada2e166ccd1a5f2ace36 Signed-off-by: Seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-on: http://git-master/r/823012 GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: Get rid of legacy gpfifo typeJanne Hellsten2015-10-27
| | | | | | | | | | | | | | | | | | | | | | | Get rid of the duplicate gpfifo struct to emphasize the fact that nvgpu_gpfifo is the only memory layout for gpfifo entries that works. This is the same layout that HW uses. Also, add a local pointer to the gpfifo memory in gk20a_submit_channel_gpfifo to get rid of repeated typecasts. Bug 1592391 Bug 1550886 Change-Id: I5432859ef8e7c1aab5907e44098994d7bb807f50 Signed-off-by: Janne Hellsten <jhellsten@nvidia.com> Signed-off-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-on: http://git-master/r/677341 (cherry picked from commit 724c8c6228af81dd440e825bddf545dd6b2b8bd7) Reviewed-on: http://git-master/r/822548 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Arto Merilainen <amerilainen@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: Add sysfs node to enable rail gatingSeshendra Gadagottu2015-10-26
| | | | | | | | | | | | | | | | Add sysfs node "railgate_enable" to enable gpu railgating dynamically. Bug 1552469 Reviewed-on: http://git-master/r/804746 (cherry picked from commit 53d76d1d1576a96c70f66b744411d2909ec8414f) Change-Id: I69ac63958b00ae4ea5d9ccbaed03316d35ddc5eb Signed-off-by: Seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-on: http://git-master/r/822208 GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: use railgate instead of reset_assertDeepak Nibade2015-10-26
| | | | | | | | | | | | | | | | | | | | | | | | In gk20a_do_idle(), if can_railgate = false, we do reset_assert() on the GPU But asserting reset might have dependencies on specific h/w seqeunce for ensuring proper reset Hence avoid calling reset_assert() and directly call platform specific unrailgate() and railgate() APIs which take care of the correct reset sequence Bug 200142989 Bug 200137963 Bug 1678611 Change-Id: Ide886dd88b8422ad36de52d54378b1edd9c7bbd6 Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: http://git-master/r/820322 (cherry picked from commit d621ddd49da976a75c14aa7aaa37f700fb4e83f2) Reviewed-on: http://git-master/r/822515 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Sachin Nikam <snikam@nvidia.com>
* gpu: nvgpu: unrailgate only if pm_domains not enabledDeepak Nibade2015-10-26
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently we unrailgate the GPU if railgating is not enabled or pm_domains are not enabled But in case if railgating is not enabled and pm_domains are enabled, we explicitly unrailgate GPU in gk20a_pm_init() and then runtime PM unrailgates it again when first user space request arrives - setting unrailgate refcount to 2 Now for gk20a_do_idle(), we need to railgate the GPU in fist call but that does not happen since unrailgate refcount != 1 hence, in case railgating is not enabled, we should unrailgate the GPU from only one place i.e. when first user space request arrives Bug 200142989 Bug 200137963 Bug 1678611 Reviewed-on: http://git-master/r/820321 (cherry picked from commit 452a1ff8da8e3f47caed2371440f9ad150bf8699) Change-Id: Ic9fe2267c9df5629315c30c1404c2b3044c1265a Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: http://git-master/r/822296 GVS: Gerrit_Virtual_Submit Reviewed-by: Sachin Nikam <snikam@nvidia.com>
* gpu: nvgpu: Disable only channel at zcull bindTerje Bergstrom2015-10-23
| | | | | | | | | | | | | | At zcull bind we disable whole GR engine. This is unnecessary, so instead disable only the channel and make sure it's unloaded. Introduces also an API in fifo_gk20a.c to do the channel disable. gr_gk20a_ctx_zcull_setup() was always passed true as last parameter, so remove parameter. Change-Id: I7ae6e101ec7d1ab3f6ee4e9bcc442d23dbd21247 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/787570
* gpu: nvgpu: Protect sync by an own lockTerje Bergstrom2015-10-22
| | | | | | | | | | | Protect creation and deletion of sync by an own mutex. This prevents deadlock in channel abort when abort is called from submit path. Bug 200147887 Change-Id: I5d6308b773c1d1a6a89d4590e2e74c74d691f79d Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/821127
* gpu: nvgpu: Use gradual slowdownTerje Bergstrom2015-10-22
| | | | | | | | | Program clock slowdown to happen using gradual slowdown. It is significantly faster than the default slowdown. Change-Id: I9e5259889637fce2c0b083a424b54af12bb45c25 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/819698
* gpu: nvgpu: vgpu: re-factor gr ctx managementAingara Paramakuru2015-10-22
| | | | | | | | | | | | | | | | Move the gr ctx management to the GPU HAL. Also, add support for a new interface to allocate gr ctxsw buffers. Bug 1677153 Change-Id: I5a7980acf4de0de7dbd94b7dd20f91a6196dc989 Signed-off-by: Aingara Paramakuru <aparamakuru@nvidia.com> Reviewed-on: http://git-master/r/806961 Signed-off-by: Seema Khowala <seemaj@nvidia.com> Reviewed-on: http://git-master/r/817009 GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: Hard code FE_GO_IDLE_TIMEOUTTerje Bergstrom2015-10-20
| | | | | | | | Always use the PROD value for FE_GO_IDLE_TIMEOUT. Change-Id: I455c03ae07b35a8999cd0995e458c421a10e7ca2 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/813958
* gpu: nvgpu: Trigger recovery on HWW errorsTerje Bergstrom2015-10-20
| | | | | | | | | | | | | | | Trigger recovery on DS and MEMFMT HWW errors, and write an error line to UART for each HWW error. Also capture the channel id before clearing the exception. Bug 1683059 Change-Id: Ia00d88a76371a4bd7e047915dde0bf0d4b84bc10 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/816983 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit
* gpu: nvgpu: Fix fake MMU fault for TSGsTerje Bergstrom2015-10-20
| | | | | | | | | | When we induce a fake MMU fault, we do not have pointer to a channel. Use the tsg pointer instead. Also remove the error print in case we do not have ch pointer. Change-Id: I14fd75d2b743244915bf32fe39de76097ef5c42f Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/819034
* gpu: nvgpu: restart timer instead of cancelDeepak Nibade2015-10-20
| | | | | | | | | | | | | | | | | In gk20a_fifo_handle_sched_error(), we currently cancel the timeout on all the channels But this could cause us to miss one of stuck channel hence, instead of cancelling, restart the timeout of channel on which it is already active Bug 200133289 Change-Id: I40e7e0e5394911fc110ab6fde39592b885dfaf7d Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: http://git-master/r/816133 Reviewed-by: Ishan Mittal <imittal@nvidia.com> Tested-by: Ishan Mittal <imittal@nvidia.com>
* gpu: nvgpu: fix deadlock on timeout lockDeepak Nibade2015-10-20
| | | | | | | | | | | | | | | | | | | | | | In gk20a_channel_timeout_stop(), we take the channel's timeout lock and then cancel the timeout worker thread Timeout worker thread also tries to acquire same timeout lock. Hence, while cancelling the timeout in gk20a_channel_timeout_stop() if the timeout_handler is already scheduled, we will have a deadlock Fix this by moving cancel_delayed_work_sync() out of the locks Bug 200133289 Bug 1695481 Change-Id: Iea78770180b483a63e5e176efba27831174e9dde Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: http://git-master/r/815922 Reviewed-by: Ishan Mittal <imittal@nvidia.com> Tested-by: Ishan Mittal <imittal@nvidia.com>
* gpu: nvgpu: fix pbdma intr handlingDeepak Nibade2015-10-16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | To handle any of the pbdma interrupt, we currently write zero to pbdma_method0 and then clear the interrupt But this is insufficient since we cannot use same intr clear method for all the interrupts Hence, add intr specific routines to handle those interrupts NV_PPBDMA_INTR_0_PBENTRY: - fix the pb_header to have a null opcode - fix the pbdma_method to have a valid nop NV_PPBDMA_INTR_0_METHOD: - fix the pbdma_method to have a valid nop NV_PPBDMA_INTR_0_DEVICE: - fix the pb_header to have a null opcode - go through all pbdma_method0/1/2/3 -- if they contain host s/w methods, replace those methods with a valid NOP Bug 200134238 Change-Id: I10c284a6cdc1441f9d437cea65aae00d3c33a8c8 Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: http://git-master/r/814561 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: add support to remove bar2 mmSeshendra Gadagottu2015-10-12
| | | | | | | | | | | | | Adding support to remove bar2 mm on gpu module remove. Change-Id: Id5f680b1abf7056da9871d5460d9fbc40422673e Signed-off-by: Seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-on: http://git-master/r/814571 (cherry picked from commit e7c6c87dd6b0893d26a9a3b4568121a691e1eb3c) Reviewed-on: http://git-master/r/815429 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: rework secure_page allocation pathDeepak Nibade2015-10-12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently, if can_railgate = false, then we have below sequence to allocate secure_page - unrailgate GPU (forever) - reset_assert() - allocate secure_page - reset_deassert() But if we allocate secure page even before unrailgating GPU for first time, then we can avoid reset_assert()/deassert() calls since GPU should already be in reset/railgated at boot time hence, rework this sequence as below - init required mutex, set platform->reset_control - allocate secure page (GPU should already be in reset at this point) - gk20a_pm_init() which unrailgates GPU in case of can_railgate = false Bug 200137963 Bug 1678611 Change-Id: I79d0543bb5cf1eaf1009e1e6ac142532d84514a5 Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: http://git-master/r/797153 (cherry picked from commit 368004501943d38c003747f6bec0384fed57ee65) Reviewed-on: http://git-master/r/816005 GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* nvgpu: gk20a: Optimize vm_put_buffers for zero buffersSami Kiminki2015-10-12
| | | | | | | | | | | | | | | Return immediately in case there are no buffers to put. This skips acquiring mutexes and map batch start/finish overheads. Bug 1614735 Bug 1623949 Bug 1660392 Change-Id: Ief04e36d995e65c1510496c17cb3f5bb90486c69 Signed-off-by: Sami Kiminki <skiminki@nvidia.com> Reviewed-on: http://git-master/r/815376 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: API for gpmu bootstrapSeshendra Gadagottu2015-10-09
| | | | | | | | | | | | | | Expose API for gpmu bootstrap. Bug 1685722 Change-Id: I46ca6f8b36e14cd1c6a12eb0d5cd178da2e0be1c Signed-off-by: Seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-on: http://git-master/r/812270 (cherry picked from commit bd7ac9992923cc32f2739926400bbf9b5cadc0c1) Reviewed-on: http://git-master/r/813977 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* Revert "gpu: nvgpu: WAR for bad GPFIFO entries from userspace"Terje Bergstrom2015-10-08
| | | | | | | | | | | This reverts commit aeb74fc7952718ffab6281c687951499510c4333. User space was fixed not to send zero-length GPFIFO entries. Bug 1662670 Change-Id: Iec6bf1870a19db4e8daa2ed4512650b92a37ba92 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/811783
* gpu: nvgpu: cancel all wdt timeouts while handling SCHED errorsDeepak Nibade2015-10-07
| | | | | | | | | | | | | | | | | | | A SCHED error might cause multiple channels' watchdogs to trigger simultaneously Hence, to avoid this conflict cancel watchdog timeout on all channels before recovering from SCHED errors Also, define API gk20a_channel_timeout_stop_all_channels() to cancel wdt timeout on all channels Bug 200133289 Change-Id: I8324c397891f0a711327b77d0677cd6718af6d01 Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: http://git-master/r/810959 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: debugfs for ch_wdt timeoutDeepak Nibade2015-10-07
| | | | | | | | | | | | | | | | export debugfs /d/gpu.0/ch_wdt_timeout_ms to modify all channels' watchdog timeout this is needed for testing purpose only Bug 200133289 Change-Id: I8776b567d5d5a1c304334835b0bcab7b242cf0ab Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: http://git-master/r/810957 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: make wdt timeout per-platformDeepak Nibade2015-10-07
| | | | | | | | | | | | | | | | Channel watchdog timeout is set to a costant value of 5s as of now Make this timeout platform specific and set it to 5s for gm20b and 7s for gk20a Bug 200133289 Change-Id: I6e7f0fed93a8d5b197ae46807131311196c6636f Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: http://git-master/r/810956 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: Avoid resetting CDE flagsujeet baranwal2015-10-07
| | | | | | | | | | | | | | | While loading the context, erstwhile set CDE flag was being overwritten by copying code of golden context, thus losing the information. This was not letting the CDE info reach to the ucode, and T1 was not configured to 128B mem access. Bug 200096226 Change-Id: I5ceb234a62450ff7875aeba05ec616758cb319d9 Signed-off-by: sujeet baranwal <sbaranwal@nvidia.com> Reviewed-on: http://git-master/r/811767 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: set correct timeslice valueKirill Artamonov2015-10-06
| | | | | | | | | | | | | | | | | | | | Scale timeslice register value based on platform specific ptimer scale koefficient. Expose timeslice values through debugfs to simplify performance tuning. bug 1605552 bug 1603226 Change-Id: I49f86f22d58d26a366ee1b5f5a9ab9d7f896ad25 Signed-off-by: Kirill Artamonov <kartamonov@nvidia.com> Reviewed-on: http://git-master/r/800007 (cherry picked from commit 00c85ef24cf28ffaa81eb53fff7edef1c699220a) Reviewed-on: http://git-master/r/808251 GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: scale ptimer based timeoutsVijayakumar2015-10-06
| | | | | | | | | | | | | | | | | bug 1603226 host based timeouts use ptimer for detecting timeouts. on gk20a and gm20b ptimer runs 2.6x slower. scale the fifo_eng_timeout to account for this Change-Id: Ie44718382953e36436ea47d6e89b9a225d5c2070 Signed-off-by: Vijayakumar <vsubbu@nvidia.com> Reviewed-on: http://git-master/r/799983 (cherry picked from commit d1d837fd09ff0f035feff1757c67488404c23cc6) Reviewed-on: http://git-master/r/808250 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: fix ctag computation overflow with 8GBJussi Rasanen2015-10-06
| | | | | | | | | | | | | | Bug 1689976 Change-Id: I97ad14c9698030b630d3396199a2a5296c661392 Signed-off-by: Jussi Rasanen <jrasanen@nvidia.com> Reviewed-on: http://git-master/r/806590 (cherry picked from commit c90cd5ee674d6357db3be2243950ff0d81ef15ef) Reviewed-on: http://git-master/r/808249 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: gk20a: dont disable pmu in pmu_destroyVijayakumar2015-10-06
| | | | | | | | | | | | | | | | | | | bug 1688374 disabling pmu will break RAM suspend on chips implementing split rails. pmu will be powered down along with rest of the GPU anyway. pmu_destroy is not be used outside of rail gating or gpu suspend Change-Id: I9e89859b7c701f731276ae1d1063d9ccd88d4334 Signed-off-by: Vijayakumar <vsubbu@nvidia.com> Reviewed-on: http://git-master/r/805940 (cherry picked from commit 8ded353878ff7df73e55b702041008ddc3cbf069) Reviewed-on: http://git-master/r/808248 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: support reset_control APIDeepak Nibade2015-10-06
| | | | | | | | | | | | | | Bug 200137963 Change-Id: I3197af905c945540b97ba191e5695d970d77af8e Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: http://git-master/r/797154 (cherry picked from commit 8a50245ea636deb87a3d9435fb115b4eac88fac9) Reviewed-on: http://git-master/r/808247 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: fix 4k compressionJussi Rasanen2015-10-06
| | | | | | | | | | | | | | | | | Add CPU dcache flush after populating scatterBuffer so that the GPU will see the buffer contents. Bug 1679453 Change-Id: I564394ed1fcff4d08d753e753bd3243b460d76df Signed-off-by: Jussi Rasanen <jrasanen@nvidia.com> Reviewed-on: http://git-master/r/805197 (cherry picked from commit d6a5513745aa77c84ac5408a62f72f24839ef439) Reviewed-on: http://git-master/r/808246 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: update slcg xbar prod settingsSeshendra Gadagottu2015-10-06
| | | | | | | | | | | | | | Bug 1689806 Change-Id: I368ad8fb64e49b21ba61c519def1f86e1ca6e492 Signed-off-by: Seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-on: http://git-master/r/806116 (cherry picked from commit 1a3bbe989a795d379703e7f4b915f6e1bb38c2c3) Reviewed-on: http://git-master/r/805480 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: let shutdown callback call vgpu_pm_prepare_poweroff for vgpuRichard Zhao2015-10-06
| | | | | | | | | | | | | | | It fixed the issue that system hang when reboot. Bug 1638850 Change-Id: If53a31e86c10b2fce4a22fe4fcf92106d86c95ef Signed-off-by: Richard Zhao <rizhao@nvidia.com> Reviewed-on: http://git-master/r/803234 (cherry picked from commit 4dbea2c7037a5244ccb9d6e886023c29ba584892) Reviewed-on: http://git-master/r/808245 GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: Commit cb manager at context createTerje Bergstrom2015-10-01
| | | | | | | | | | | | Call commit_cb_manager() at context creation time instead of hardware initialization. This allows per-channel sizes for buffers. Bug 1686189 Change-Id: Ie4d08e87f237bc63bac0268128f59d4fe8536c95 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/801777 Reviewed-on: http://git-master/r/806181
* gpu: nvgpu: Write patch_count after updating ctxswTerje Bergstrom2015-10-01
| | | | | | | | | | | Bug 1686189 Change-Id: Idf92d3277a7e8932d11ece13e3b988609e49c74e Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/802550 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-on: http://git-master/r/806180
* gpu: nvgpu: prevent extra user unmapsDeepak Nibade2015-09-30
| | | | | | | | | | | | | | | | | | It is possible that user space requests more unmaps on a buffer than it requested maps In this case, we end up dropping one extra refcount which could lead to releasing buffer early Fix this by checking and returning if buffer's user_mapped refcount is already zero Bug 200130521 Change-Id: Ic8ef2dbfe0476b16d852ad899b1ed0404b5bb7de Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: http://git-master/r/788904 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: ELPG init & statistics updateMahantesh Kumbar2015-09-30
| | | | | | | | | | | | | | | | - Required init param to start elpg - change in statistics dump Bug 1684939 Change-Id: I26dca52079f08b8962e9cb758831910207610220 Signed-off-by: Mahantesh Kumbar <mkumbar@nvidia.com> Reviewed-on: http://git-master/r/802456 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/806179 Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: PMU ucode version updateMahantesh Kumbar2015-09-30
| | | | | | | | | | | | | | | | - PMU ucode version update to sync with LS production signature Bug 200140416 Change-Id: Ib77fa81f7b05ed3cf45c373f3d759a2cfb69b238 Signed-off-by: Mahantesh Kumbar <mkumbar@nvidia.com> Reviewed-on: http://git-master/r/801738 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/806177 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit
* gpu: nvgpu: interface update to sync CL #19870492Mahantesh Kumbar2015-09-30
| | | | | | | | | | | | | | | | - pg statistics update - perfmon update - ADD GR inti params interface to enable ELPG Bug n/a Change-Id: I39ae1d4518733480a42f06a0be7bd794fc93ff6f Signed-off-by: Mahantesh Kumbar <mkumbar@nvidia.com> Reviewed-on: http://git-master/r/799684 GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/806176 Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: unmapped ptes handlingSeshendra Gadagottu2015-09-30
| | | | | | | | | | | | | | | | Correct logic for supporting unmapped ptes during gmmu map. Bug 1587825 Change-Id: I1b0b603f7758a65d9666046d0d908663f8e460e3 Signed-off-by: Seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-on: http://git-master/r/796577 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/759345 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit
* gpu: nvgpu: create debugfs node earlyKirill Artamonov2015-09-30
| | | | | | | | | | | | | | | | | | | Create debugfs node before platform->probe() is called. Allow chip specific debugfs entries go to correct directory. bug 1525327 bug 1581799 Change-Id: I2d91bdc1e72dac6787938eff01218c9f871029cb Signed-off-by: Kirill Artamonov <kartamonov@nvidia.com> Reviewed-on: http://git-master/r/796092 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/778729 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit
* gpu: nvgpu: pmu version updateMahantesh Kumbar2015-09-30
| | | | | | | | | | | | | - pmu version update P4 CL #19870492 - pmu allocation update P4 CL #19870492 Signed-off-by: Mahantesh Kumbar <mkumbar@nvidia.com> Reviewed-on: http://git-master/r/788791 Reviewed-on: http://git-master/r/786342 Change-Id: If6607cfbb134f22e25148b74d6101a6b9709e155 Reviewed-on: http://git-master/r/807474 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: Add CDE bits in FECS headersujeet baranwal2015-09-29
| | | | | | | | | | | | | | | | In case of CDE channel, T1 (Tex) unit needs to be promoted to 128B aligned, otherwise causes a HW deadlock. Gpu driver makes changes in FECS header which FECS uses to configure the T1 promotions to aligned 128B accesses. Bug 200096226 Change-Id: I8a8deaf6fb91f4bbceacd491db7eb6f7bca5001b Signed-off-by: sujeet baranwal <sbaranwal@nvidia.com> Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Signed-off-by: sujeet baranwal <sbaranwal@nvidia.com> Reviewed-on: http://git-master/r/804625 Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: Dump GR register on ucode timeoutGagan Grover2015-09-29
| | | | | | | | | | | | | | | | | | | Dump GR registers on ucode timeout. GR dump is needed during ucode timeout to get more details. Bug 200124360 Change-Id: Id19f5bc0d092c060de2ec07a5e63a0a155f86b76 Signed-off-by: Gagan Grover <ggrover@nvidia.com> Reviewed-on: http://git-master/r/771969 (cherry picked from commit 3f0f13073a174a357623d76db47b2148cb24503c) Reviewed-on: http://git-master/r/777785 (cherry picked from commit d5b7247757cdccbc3ea98c4b9e018468d5554933) Reviewed-on: http://git-master/r/795355 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Bharat Nihalani <bnihalani@nvidia.com> Tested-by: Bharat Nihalani <bnihalani@nvidia.com>
* gpu: nvgpu: Handling null pointerGagan Grover2015-09-29
| | | | | | | | | | | | | | | Handling null pointer in gk20a_fence_is_expired. Bug 200117724 Change-Id: I0f9307a5f8b82bf990b6ddaea1a408d4f3f376fb Signed-off-by: Gagan Grover <ggrover@nvidia.com> Reviewed-on: http://git-master/r/777796 (cherry picked from commit dbf5bae53e0e7862754faba78eab84284786ecb3) Reviewed-on: http://git-master/r/795356 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: Bharat Nihalani <bnihalani@nvidia.com>
* gpu: nvgpu: Add support for CDE scatter buffersJussi Rasanen2015-09-28
| | | | | | | | | | | | | | | | Add support for CDE scatter buffers. When the bus addresses for surfaces are not contiguous as seen by the GPU (e.g., when SMMU is bypassed), CDE swizzling needs additional per-page information. This information is populated in a scatter buffer when required. Bug 1604102 Change-Id: I3384e2cfb5d5f628ed0f21375bdac8e36b77ae4f Signed-off-by: Jussi Rasanen <jrasanen@nvidia.com> Reviewed-on: http://git-master/r/789436 Reviewed-on: http://git-master/r/791243 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: implement per-channel watchdogDeepak Nibade2015-09-28
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Implement per-channel watchdog/timer as per below rules : - start the timer while submitting first job on channel or if no timer is already running - cancel the timer when job completes - re-start the timer if there is any incomplete job left in the channel's queue - trigger appropriate recovery method as part of timeout handling mechanism Handle the timeout as per below : - get timed out channel, and job data - disable activity on all engines - check if fence is really pending - get information on failing engine - if no engine is failing, just abort the channel - if engine is failing, trigger the recovery Also, add flag "ch_wdt_enabled" to enable/disable channel watchdog mechanism. Watchdog can also be disabled using global flag "timeouts_enabled" Set the watchdog time to be 5s using macro NVGPU_CHANNEL_WATCHDOG_DEFAULT_TIMEOUT_MS Bug 200133289 Change-Id: I401cf14dd34a210bc429f31bd5216a361edf1237 Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: http://git-master/r/797072 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: APIs to disable/enable all engines' activityDeepak Nibade2015-09-28
| | | | | | | | | | | | | | | Add below APIs to disable/re-enable activity on all engines gk20a_fifo_disable_all_engine_activity() gk20a_fifo_enable_all_engine_activity() Bug 200133289 Change-Id: Ie01a260d587807a3c1712ee32fe870fbcb08f9cd Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: http://git-master/r/798747 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
generated by cgit v1.2.2 (git 2.25.0) at 2024-07-07 19:18:43 -0400