summaryrefslogtreecommitdiffstats
path: root/drivers
Commit message (Collapse)AuthorAge
...
* gpu: nvgpu: vgpu: call nvgpu_init_ltc_supportRichard Zhao2018-05-20
| | | | | | | | | | | | | | | vgpu needs to call nvgpu_init_ltc_support to floor sweep and set ltc_count. And set gops.ltc.set_enabled to null as guest is not allowed to change ltc settings. Jira VQRM-2345 Change-Id: I83517d631aa947db4a0a4c312f0cecda9ba03973 Signed-off-by: Richard Zhao <rizhao@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1703626 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: gv100: do not init idle filtersSeema Khowala2018-05-18
| | | | | | | | | | | | | | | | | | | | | gv100 cannot use idle filter values defined for gp106 since values are different. Also gv100 cannot use gv11b idle filter values since prod values are different between gv100 and gv11b. Finally since PROD values match with POR INIT values, there is no need to init idle filters for gv100. Bug 2115080 Change-Id: I9e7cfbde364d993ae04d80af14650739f32345cc Signed-off-by: Seema Khowala <seemaj@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1724060 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: release runlist_lock before issuing recoverySeema Khowala2018-05-18
| | | | | | | | | | | | | Release runlist_lock before issuing runlist update timeout recovery. Bug 2115080 Change-Id: I22cd0dd8ab6828412fcc98f587e4a5cdce907651 Signed-off-by: Seema Khowala <seemaj@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1722308 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Add timeouts_disabled_refcount for enabling timeoutSeema Khowala2018-05-18
| | | | | | | | | | | | | | | | -timeouts will be enabled only when timeouts_disabled_refcount will reach 0 -timeouts_enabled debugfs will change from u32 type to file type to avoid race enabling/disabling timeout from debugfs and ioctl -unify setting timeouts_enabled from debugfs and ioctl Bug 1982434 Change-Id: I54bab778f1ae533872146dfb8d80deafd2a685c7 Signed-off-by: Seema Khowala <seemaj@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1588690 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Code updates for MISRA violationsVinod G2018-05-18
| | | | | | | | | | | | | | | | | | | | | | | | | Code related to MC module is updated for handling MISRA violations Rule 10.1: Operands shalln't be an inappropriate essential type. Rule 10.3: Value of expression shalln't be assigned to an object with a narrow essential type. Rule 10.4: Both operands in an operator shall have the same essential type. Rule 14.4: Controlling if statement shall have essentially Boolean type. Rule 15.6: Enclose if() sequences with braces. JIRA NVGPU-646 JIRA NVGPU-659 JIRA NVGPU-671 Change-Id: Ia7ada40068eab5c164b8bad99bf8103b37a2fbc9 Signed-off-by: Vinod G <vinodg@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1720926 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: drop force_need_sync_fence in submit pathKonsta Holtta2018-05-18
| | | | | | | | | | | | | | | | For CDE work a sync fence is always requested, but kernel does not need it and submit flags from userspace will be passed to the submit function in cde path so a sync fence will get created if necessary. To reduce some complexity, remove the explicit boolean in favor of just NVGPU_SUBMIT_FLAGS_SYNC_FENCE. Jira NVGPU-705 Change-Id: I8aac85288513ed7cc640acd021d892cee86f41d8 Signed-off-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1721785 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: add HALs for devinit and preos bios operationsDeepak Nibade2018-05-18
| | | | | | | | | | | | | | | | | | | | | | | Add below new HALs for bios operations gops.bios.devinit() gops.bios.preos() gops.bios.verify_devinit() Export existing APIs gp106_bios_devinit() and gp106_bios_preos() and set them to above HALs on gp106 and gv100 And call new HALs from gp106_bios_init() if supported instead of directly calling APIs Jira NVGPUT-48 Change-Id: Ic89f1c86cf6e3e0785b3663fe733b201d6f2f773 Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1708382 GVS: Gerrit_Virtual_Submit Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: handle clk arb event posting in OS specific codeSourab Gupta2018-05-18
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | The mechanism of posting events to userspace is OS specific. In linux this works through poll fd, wherein we can make use of nvgpu_cond variables to poll and trigger the corresponding wait_queue. The post event functionality on QNX doesn't work on poll though. It uses iofunc_notify_trigger to post the events to the calling process. As such QNX can't work with nvgpu_cond's. To overcome this issue, it is proposed to create OS specific interface function for posting clk arb events. Linux can call nvgpu_cond based implementation, which makes sense since these are already initialized and poll'ed in Linux specific code only. QNX can implement this interface to call iofunc_notify_* functions, as per its need. Jira VQRM-3741 Change-Id: I7d9f71dae2ae7f6a09cd56662003fd1b7e50324c Signed-off-by: Sourab Gupta <sourabg@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1709656 GVS: Gerrit_Virtual_Submit Reviewed-by: Alex Waterman <alexw@nvidia.com> Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: remove usage of xchg in clk arb codeSourab Gupta2018-05-18
| | | | | | | | | | | | | | | | | | With the removal of rcu locks and using spinlocks in place, the usage of xchg for atomic pointer swap is unneccesary. A few places already have barriers in place before changing the pointer values, so a simple pointer assignment would do. Jira VQRM-3741 Change-Id: I03296202b273b5175f166ab3e094c0e4de910eb8 Signed-off-by: Sourab Gupta <sourabg@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1709655 GVS: Gerrit_Virtual_Submit Reviewed-by: Alex Waterman <alexw@nvidia.com> Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: use os agnostic api for retrieving timestampSourab Gupta2018-05-18
| | | | | | | | | | | | | | | | | currently clk arbiter is using the Linux specific sched_clock() api for retrieving current timestamp. Instead use the OS agnostic nvgpu_hr_timestamp(). Jira VQRM-3741 Change-Id: I315ca16327b30db06c39046af1eb05249d1a97ca Signed-off-by: Sourab Gupta <sourabg@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1709654 GVS: Gerrit_Virtual_Submit Reviewed-by: Alex Waterman <alexw@nvidia.com> Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: add conversion function for clk domainSourab Gupta2018-05-18
| | | | | | | | | | | | | | | Add a conversion function for NVGPU_GPU_CLK_DOMAIN_* defines present in uapi header. This enables movement of related code to the OS agnostic clk_arb.c Jira VQRM-3741 Change-Id: I922d1cfb91d6a5dda644cf418f2f3815d975fcfd Signed-off-by: Sourab Gupta <sourabg@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1709653 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: add macro for clk arb debug logsSourab Gupta2018-05-18
| | | | | | | | | | | | | | | Introduce a macro for clk arbiter debug logs. Jira VQRM-3741 Change-Id: I9f4ebf5f979e84b6383dc8755eb34c0ffa3d0f43 Signed-off-by: Sourab Gupta <sourabg@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1709652 GVS: Gerrit_Virtual_Submit Reviewed-by: Alex Waterman <alexw@nvidia.com> Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: move clk_arb.c to common codeSourab Gupta2018-05-18
| | | | | | | | | | | | | | | | Now that clk_arb.c is free of Linux'isms, move it to the clk/ directory. Jira VQRM-741 Change-Id: I53298c76f834322aa586781cdfd2e6031f4826a1 Signed-off-by: Sourab Gupta <sourabg@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1709651 GVS: Gerrit_Virtual_Submit Reviewed-by: Alex Waterman <alexw@nvidia.com> Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Update gv100 nvlink TLC buffer configTejal Kudav2018-05-18
| | | | | | | | | | | | | | | | | | TLC buffer sizes and credit init values do not match with the values recommended by IAS for dGPU-Xavier configuration. These buffer configuration values affect the latency over link. JIRA NVLINK-158 Change-Id: I7822747cb0ae5a5efdd2d57e2104d0cb30bf9352 Signed-off-by: Tejal Kudav <tkudav@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1686601 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Alex Waterman <alexw@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: add NVGPU_IOCTL_CHANNEL_RESCHEDULE_RUNLISTDavid Li2018-05-18
| | | | | | | | | | | | | | | | | | | | | | | Add NVGPU_IOCTL_CHANNEL_RESCHEDULE_RUNLIST ioctl to reschedule runlist, and optionally check host and FECS status to preempt pending load of context not belonging to the calling channel on GR engine during context switch. This should be called immediately after a submit to decrease worst case submit to start latency for high interleave channel. There is less than 0.002% chance that the ioctl blocks up to couple miliseconds due to race condition of FECS status changing while being read. For GV11B it will always preempt pending load of unwanted context since there is no chance that ioctl blocks due to race condition. Also fix bug with host reschedule for multiple runlists which needs to write both runlist registers. Bug 1987640 Bug 1924808 Change-Id: I0b7e2f91bd18b0b20928e5a3311b9426b1bf1848 Signed-off-by: David Li <davli@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1549050 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Fix gp106/clk_gp106.c compile errorsAlex Waterman2018-05-17
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fix the following compiler errors: /dvs/git/dirty/git-master_modular/kernel/nvgpu/drivers/gpu/nvgpu/gp106/clk_gp106.c: In function 'gp106_init_clk_support': /dvs/git/dirty/git-master_modular/kernel/nvgpu/drivers/gpu/nvgpu/gp106/clk_gp106.c:109:3: error: unknown field 'cntr' specified in initializer cc1: warnings being treated as errors /dvs/git/dirty/git-master_modular/kernel/nvgpu/drivers/gpu/nvgpu/gp106/clk_gp106.c:109:3: error: missing braces around initializer /dvs/git/dirty/git-master_modular/kernel/nvgpu/drivers/gpu/nvgpu/gp106/clk_gp106.c:109:3: error: (near initialization for '(anonymous).<anonymous>') /dvs/git/dirty/git-master_modular/kernel/nvgpu/drivers/gpu/nvgpu/gp106/clk_gp106.c:110:3: error: unknown field 'cntr' specified in initializer /dvs/git/dirty/git-master_modular/kernel/nvgpu/drivers/gpu/nvgpu/gp106/clk_gp106.c:112:3: error: unknown field 'cntr' specified in initializer /dvs/git/dirty/git-master_modular/kernel/nvgpu/drivers/gpu/nvgpu/gp106/clk_gp106.c:115:2: error: initialized field with side-effects overwritten /dvs/git/dirty/git-master_modular/kernel/nvgpu/drivers/gpu/nvgpu/gp106/clk_gp106.c:115:2: error: (near initialization for '(anonymous).scale') /dvs/git/dirty/git-master_modular/kernel/nvgpu/drivers/gpu/nvgpu/gp106/clk_gp106.c:122:3: error: unknown field 'cntr' specified in initializer /dvs/git/dirty/git-master_modular/kernel/nvgpu/drivers/gpu/nvgpu/gp106/clk_gp106.c:122:3: error: missing braces around initializer Not sure why this shows up for the L4T userspace compiler but not in any other compilers. JIRA NVGPU-525 Change-Id: I0609c1d9ad20232053768258103debadd6d9206f Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1720924 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Translate as_alloc flagsAlex Waterman2018-05-17
| | | | | | | | | | | | | | | | | | | Translate the as_alloc flags so that the common/as.c code no longer needs to include <uapi/linux/nvgpu.h>. This was an oversight from prior MM unification efforts which was caught by the userspace POSIX build on QNX. JIRA NVGPU-525 Change-Id: I6af6cb9904c2ae9edeb8dbb970846c31b56822bf Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1720918 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: Sourab Gupta <sourabg@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Update PMU firmware versionDeepak Goyal2018-05-17
| | | | | | | | | | | | | | | | | Removed PMU breakpoints if there is failure during GR save/restore during ELPG entry/exit. Bug 2108544 Signed-off-by: Deepak Goyal <dgoyal@nvidia.com> Change-Id: I08c342f5f79b7484d31e2437ede1881c4dceb6d0 Reviewed-on: https://git-master.nvidia.com/r/1719659 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Tested-by: Seema Khowala <seemaj@nvidia.com> Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Make sure there an nvlink.h header for POSIXAlex Waterman2018-05-17
| | | | | | | | | | | | | | | | | | | This header doesn't have anything in it yet but the header is now present. A change recently went in that only checked for __KERNEL__ before falling back to including the QNX header. This caused the POSIX build in GVS to attempt to include the QNX header. The QNX src is not synced in userspace dev-kernel tests builds resulting in a missing header. JIRA NVGPU-525 Change-Id: I60f29ad69cbed38b6ea47f95ca504dab51fa01e7 Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1714083 GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Update POSIX BIT() macro to ULLAlex Waterman2018-05-17
| | | | | | | | | | | | | | | | | | | | | | | | For most of the builds we have in GVS userspace is 64 bits. But it seems like at least some L4T userspace builds are either not 32 bits or have an UL that only covers 32 bits. This is seen in GVS: /dvs/git/dirty/git-master_modular/kernel/nvgpu/drivers/gpu/nvgpu/gv11b/mm_gv11b.c: In function 'gv11b_gpu_phys_addr': /dvs/git/dirty/git-master_modular/kernel/nvgpu/drivers/gpu/nvgpu/gv11b/mm_gv11b.c:273:3: error: left shift count >= width of type make[2]: *** [/dvs/git/dirty/git-master_modular/tmake/artifacts/CommonRules.tmk:318: mm_gv11b.o] Error 1 This patch simply bumps the UL to ULL in BIT() to make sure that we always have at least 64 bits available for the BIT() macro. JIRA NVGPU-525 Change-Id: I67de4338afc5bee4f1fa16faee6116e0e7dbf108 Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1718564 GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: runlist_lock released before preempt timeout recoverySeema Khowala2018-05-17
| | | | | | | | | | | | | | | | | Release runlist_lock and then initiate recovery if preempt timed out. Also do not issue preempt if ch, tsg or runlist id is invalid. tsgid could be invalid for below call trace gk20a_prepare_poweroff->gk20a_channel_suspend-> *_fifo_preempt_channel->*_fifo_preempt_tsg Bug 2065990 Bug 2043838 Change-Id: Ia1e3c134f06743e1258254a4a6f7256831706185 Signed-off-by: Seema Khowala <seemaj@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1662656 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: gp10b: fix priv error for blcg regSeema Khowala2018-05-17
| | | | | | | | | | | | | Bug 2112073 Change-Id: I4cf97d3e10e2e8d56ab97479d094c47ef842c662 Signed-off-by: Seema Khowala <seemaj@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1710636 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: gv100: use new MINION ucode formatAdeel Raza2018-05-16
| | | | | | | | | | | | | | | | Migrate to the new NVLINK MINION ucode format. The new format strips out an unnecessary ACR header from the ucode image. Moving to the new format will allow MINION ucode generation scripts to be unified. Bug 2113404 Change-Id: I9a72d6c3fa5edd50a4ec5eb835d157672931f994 Signed-off-by: Adeel Raza <araza@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1709986 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: implement new host semaphore methods for VoltaDeepak Nibade2018-05-16
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | Semaphore methods currently being used in Volta are deprecated for future chips And on Volta we support both old and new methods So replace old methods by new methods on Volta itself so that new methods get tested on silicon Implement below HALs for Volta with new semaphore methods gops.fifo.add_sema_cmd() to insert HOST semaphore acquire/release methods gops.fifo.get_sema_wait_cmd_size() to get size of acquire command buffer gops.fifo.get_sema_incr_cmd_size() to get size of release command buffer Also use new methods in these APIs gv11b_fifo_add_syncpt_wait_cmd() gv11b_fifo_add_syncpt_incr_cmd() And change corresponding APIs to reflect correct size of command buffer gv11b_fifo_get_syncpt_wait_cmd_size() gv11b_fifo_get_syncpt_incr_cmd_size() Jira NVGPUT-16 Change-Id: Ia3a37cd0560ddb54761dfea9bd28c4384cd8a11c Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1704518 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: add HAL to insert semaphore commandsDeepak Nibade2018-05-16
| | | | | | | | | | | | | | | | | | | | | Add below new HALs gops.fifo.add_sema_cmd() to insert HOST semaphore acquire/release methods gops.fifo.get_sema_wait_cmd_size() to get size of acquire command buffer gops.fifo.get_sema_incr_cmd_size() to get size of release command buffer Separate out new API gk20a_fifo_add_sema_cmd() to implement semaphore acquire/ release sequence and set it to gops.fifo.add_sema_cmd() Add gk20a_fifo_get_sema_wait_cmd_size() and gk20a_fifo_get_sema_incr_cmd_size() to return respective command buffer sizes Jira NVGPUT-16 Change-Id: Ia81a50921a6a56ebc237f2f90b137268aaa2d749 Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1704490 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: gv100: load mem_unlockSeema Khowala2018-05-16
| | | | | | | | | | | | | | | | | mem unlock bin should be written to install traps even if VPR isn’t actually supported Bug 2093809 Change-Id: I4024c66ff72a079c3f20f3b8ab356fba7ce05d4e Signed-off-by: Seema Khowala <seemaj@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1709765 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: fix build errors on qnx for thermRichard Zhao2018-05-15
| | | | | | | | | | | | | | | | | | | - include therm header file in source file - remove unreferenced function JIRA VQRM-2343 Change-Id: Ibfeaf1e431631fa2b9a8e8f1bf044c17edc018c5 Signed-off-by: Richard Zhao <rizhao@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1706201 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: Sourab Gupta <sourabg@nvidia.com> Tested-by: Sourab Gupta <sourabg@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Nirav Patel <nipatel@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: gp106 pmu f/w version updateMahantesh Kumbar2018-05-14
| | | | | | | | | | | | | | | | | | | - gp106 f/w version update for ucode https://git-master.nvidia.com/r/#/c/1708195/ - APP_VERSION_GP10X 24008084 to 24069912 - nvgpu driver cl's for current changes https://git-master.nvidia.com/r/#/c/1694546/ https://git-master.nvidia.com/r/#/c/1700746/ p4 CL# 24076634 Change-Id: If15663983a8753a256e47451938be1cf0102fadb Signed-off-by: Mahantesh Kumbar <mkumbar@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1708199 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* nvgpu: Add dummy variables to accomodate PS3.5 structureVaikundanathan S2018-05-14
| | | | | | | | | | | | Change-Id: I437f2aba6a63de87033721fa9a29c565cf8f4256 Signed-off-by: Vaikundanathan S <vaikuns@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1694546 Reviewed-by: Mahantesh Kumbar <mkumbar@nvidia.com> Tested-by: Mahantesh Kumbar <mkumbar@nvidia.com> Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: vf inject changesVaikundanathan S2018-05-14
| | | | | | | | | | | | | | | | | | | | - Added vf change inject support for gv10x - Updated clk_pmu_vf_inject() to fill required data for pascal or volta vf change inject support - Added new ctrl clk interface for gv10x clk domain list - Added pmu interface for gv10x clk domain list & vf change inject request - Modified clk cmd, msg & RPC id's to match with chips_a_23609936 branch Bug 200399373 Change-Id: Ib9dc10073386f63bdfd92110c7ec3e09b1c484ce Signed-off-by: Vaikundanathan S <vaikuns@nvidia.com> Signed-off-by: Mahantesh Kumbar <mkumbar@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1700746 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: move sync_gk20a under common/linux directoryDebarshi Dutta2018-05-14
| | | | | | | | | | | | | | | | sync_gk20a.* files are no longer used by core code and only invoked from linux specific implementations of the OS_FENCE framework which are under the common/linux directory. Hence, sync_gk20a.* files are also moved under common/linux. JIRA NVGPU-66 Change-Id: If623524611373d2da39b63cfb3c1e40089bf8d22 Signed-off-by: Debarshi Dutta <ddutta@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1712900 Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: nvhost: PC_SAMPLING ioctl failure.Prateek Sethi2018-05-14
| | | | | | | | | | | | | | | | | | | | | | NVGPU_DBG_GPU_IOCTL_PC_SAMPLING ioctl is not handled properly for HV case for both Linux and QNX. Currently guest vm is trying to perform gpu memory read and write operations which supposed to be done by RM server, causing the crash. This patch is supposed to fix ioctl failure. Bug 2052040 Change-Id: Ia0773959b84739a1bced858331764751520a3561 Signed-off-by: Prateek Sethi <prsethi@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1708102 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Sourab Gupta <sourabg@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Sourab Gupta <sourabg@nvidia.com> Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Fix Gpu sysfs access to Fmax@VminAlex Frid2018-05-11
| | | | | | | | | | | | | | | | | | | | | | | | | Currently gpu sysfs retrieves Fmax@Vmin by direct call into Tegra DVFS driver that introduces compile time dependencies on CONFIG_TEGRA_DVFS. In addition incorrect clock is used for DVFS information access. Re-factored sysfs node to use generic GPU clock operation for Fmax@Vmin read. This would fix a bug in target clock selection, and allows to remove dependency of sysfs on CONFIG_TEGRA_DVFS. Updated nvgpu_linux_get_fmax_at_vmin_safe operation itself so it can be called on platforms that does not support Tegra DVFS, although 0 will still be returned as Fmax@Vmin on such platforms. Bug 2045903 Change-Id: I32cce25320df026288c82458c913b0cde9ad4f72 Signed-off-by: Alex Frid <afrid@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1710924 Reviewed-by: Alex Waterman <alexw@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Don't put any functions in the exports file yetAlex Waterman2018-05-11
| | | | | | | | | | | | | | | | | This will make it easier to stage the make build of userspace nvgpu through module branching. Otherwise we will have a nightmare of integration with almost certain dev-kernel lockdowns. As functions get unit tested they should be added to the exports file. Change-Id: I7833fa0895c44282ec6a690dc5de43e01336292a Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1711314 GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: poll watchdog status activelyKonsta Holtta2018-05-11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Read GP_GET and GET from hardware every time when the poll timer expires instead of when the watchdog timer expires. Restart the watchdog timer if the get pointers have increased since the previous read. This way stuck channels are detected quicker. Previously it could have taken at most twice the watchdog timeout limit for a stuck channel to get recovered; with this change, a coarse sliding window is used. The polling period is still 100 ms. The difference is illustrated in the following diagram: time 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 get a b b b b b b b b b b b b b b b b b b b b prev - n n n n n n n n n A n n n n n n n n n S next - A s s s s s s s s s S "time" represents wall time in polling units; 0 is submit time. For simplicity, watchdog timeout is ten units. "get" is the GP_GET that advances a little from a and then gets stuck at b. "prev" is the previous behaviour, "next" is after this patch. "A" is when the channel is detected as advanced, and "S" when it's found stuck and recovered; small "s" is when it's found stuck but when the time limit has not yet expired and "n" is when the hw state is not read. Bug 1700277 Bug 1982826 Change-Id: Ie2921920d5396cee652729c6a7162b740d7a1f06 Signed-off-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1710554 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: pmc access through pmc driverkalyani chidambaram2018-05-11
| | | | | | | | | | | | | The pmc access has to be through pmc driver. This prevents the direct access to pmc registers and supports onlt tz access to it in t210 platforms. Bug 1811649 Change-Id: I235768a07fa93ce4cced656aade160bbd853be62 Signed-off-by: kalyani chidambaram <kalyanic@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1704315 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: dump fecs method data and push adrSeema Khowala2018-05-10
| | | | | | | | | | | Dump fecs method data and adr if ucode times out or errors out. This is good to have for debugging. Change-Id: I79c4bc3bc30cfd09f273f4eb6b53863653227ecd Signed-off-by: Seema Khowala <seemaj@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1707761 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: remove access to mc_enable_pb_r()Deepak Nibade2018-05-10
| | | | | | | | | | | | | We don't need to configure mc_enable_pb_r() register in any of the supported chips, so remove access to this register Jira NVGPUT-52 Change-Id: I8a7a524367ce7953f926143242c6d63bc8fd5ed1 Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1711245 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: remove sync_fence dependencies from fence_gk20aDebarshi Dutta2018-05-10
| | | | | | | | | | | | | | | | | Replaced all instances of sync_fence in gk20a_fence* code with nvgpu_os_fence. Added the API install_fence for the nvgpu_os_fence abstraction. sync_fence mechanism and its dependencies are completely removed from the fence_gk20a methods. Due to the recent os_fence changes and the changes to fence_gk20a, we can finally get rid of all the CONFIG_SYNCS present in the submit path. JIRA NVGPU-66 Change-Id: I3551dab04b93b1e94db83fc102a41872be89e9ed Signed-off-by: Debarshi Dutta <ddutta@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1701245 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: adapt gk20a_channel_syncpt to use os_fenceDebarshi Dutta2018-05-10
| | | | | | | | | | | | | This patch adapts gk20a_channel_syncpt to use os_fence for post fence as well as pre-fence(wait) use cases. Jira NVGPU-66 Change-Id: I49627d1f88d52a53511a02f5de60fed6df8350de Signed-off-by: Debarshi Dutta <ddutta@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1676631 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: create a wrapper over sync_fencesDebarshi Dutta2018-05-10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch constructs an abstraction to hide the sync_fence functionality from the common code. struct nvgpu_os_fence acts as an abstraction for struct sync_fence. struct nvgpu_os_fence consists of an ops structure named nvgpu_os_fence_ops which contains an API to do pushbuffer programming to generate wait commands for the fence. The current implementation of nvgpu only allows for wait method on a sync_fence which was generated using a similar backend(i.e. either Nvhost Syncpoints or Semaphores). In this patch, a generic API is introduced which will decide the type of the underlying implementation of the struct nvgpu_os_fence at runtime and run the corresponding wait implementation on it. This patch changes the channel_sync_gk20a's semaphore specific implementation to use the abstract API. A subsequent patch will make the changes for the nvhost_syncpoint based implementations as well. JIRA NVGPU-66 Change-Id: If6675bfde5885c3d15d2ca380bb6c7c0e240e734 Signed-off-by: Debarshi Dutta <ddutta@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1667218 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: support TSG IOCTL to unbind channelDeepak Nibade2018-05-10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We so far do not support unbinding a channel from TSG explicitly through IOCTL Channel is unbound from TSG when channel fd is closed But this could create issue in case process is forked and fd is duplicated In worst case it is possible that duplicate fd is closed sometime later when TSG is active and that could lead to corruption of TSG state Fix this by implementing NVGPU_TSG_IOCTL_UNBIND_CHANNEL API This API will simply call gk20a_tsg_unbind_channel() API to unbind the channel Note that we also mark the channel has_timedout since unbound channel does not have any context of its own and it cannot serve any job Remove call to gk20a_disable_channel() while closing the channel. We do not support bare channels without TSG, and if channel is not part of TSG then it means channel is already unbound from TSG with call to NVGPU_TSG_IOCTL_UNBIND_CHANNEL, and there is nothing to do while closing the channel Bug 200327095 Jira NVGPU-229 Change-Id: Ib7f6710f0dc8125caafabe7bee737622c3dd9fa3 Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1708902 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: vgpu: clean up redundant declarationsPeter Daifuku2018-05-09
| | | | | | | | | | | | | | | | | Clean up redundand function declarations in vgpu.h JIRA ESRM-299 Change-Id: I9b3a2aee6be68dffce47897baed2e7b9290026e9 Signed-off-by: Peter Daifuku <pdaifuku@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1710885 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Alex Waterman <alexw@nvidia.com> Reviewed-by: Richard Zhao <rizhao@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: added nvgpu_sim_init_lateAntony Clince Alex2018-05-09
| | | | | | | | | | | | | | | | | | | | Split sim initialization to two parts, first part gets invoked as part of probe and second part gets invoked in the finalize_poweron after the hal has been initialized. This is done because some of the sim init code uses mm api's which are assigned as part of hal init. replaced sim buffer allocation api's with nvgpu_dma_sys_alloc. Change-Id: Ib019fbb747bdf6dcc74e7deba732ab41f0869e96 Signed-off-by: Antony Clince Alex <aalex@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1705424 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Remove gk20a_dbg* functionsTerje Bergstrom2018-05-09
| | | | | | | | | | | | | | | Switch all logging to nvgpu_log*(). gk20a_dbg* macros are intentionally left there because of use from other repositories. Because the new functions do not work without a pointer to struct gk20a, and piping it just for logging is excessive, some log messages are deleted. Change-Id: I00e22e75fe4596a330bb0282ab4774b3639ee31e Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1704148 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: created os-agnostic sim header.Antony Clince Alex2018-05-09
| | | | | | | | | | | | | | added a os-agnostic sim.h header which could be included in by any platform, moved out os specific headers to nvgpu/linux. JIRA VQRM-2368 Change-Id: I3861bfa75a6b8d2d909bc7223467fd68c208275b Signed-off-by: Antony Clince Alex <aalex@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1702816 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: os-agnostic segregation of sim/sim_pciAntony Clince Alex2018-05-09
| | | | | | | | | | | | | | | | | | segregated os-agnostic function from linux/sim.c and linux/sim_pci.c to sim.c and sim_pci.c, while retaining os-specific functions. renamed all gk20a_* api's to nvgpu_*. renamed hw_sim_gk20a.h to nvgpu/hw_sim.h moved hw_sim_pci.h to nvgpu/hw_sim_pci.h JIRA VQRM-2368 Change-Id: I040a6b12b19111a0b99280245808ea2b0f344cdd Signed-off-by: Antony Clince Alex <aalex@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1702425 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: refactored struct sim_gk20aAntony Clince Alex2018-05-09
| | | | | | | | | | | | | | | | | | moved sim buffer(send, recv and msg) from os-specific structure to OS agnostic structure. JIRA VQRM-2368 Change-Id: I10ff23fe24d86f2bd372f1bae0369cc45aadfb80 Signed-off-by: Antony Clince Alex <aalex@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1702178 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: Sourab Gupta <sourabg@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: refectored sim_readl/sim_writelAntony Clince Alex2018-05-09
| | | | | | | | | | | | | | | | | | | | | refactored sim_readl and sim_writel to use os-agnostic structures. converted all sim buffers to the type nvgpu_mem and replaced all alloc_page and free_page calls with corresponding nvgpu_dma_alloc/nvgpu_dma_free calls. JIRA VQRM-2368 Change-Id: Ia9d29119d31f239ed16be932cfd16c334002c727 Signed-off-by: Antony Clince Alex <aalex@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1702050 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: Sourab Gupta <sourabg@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: nvlink endpoint ops to common codeThomas Fleury2018-05-09
| | | | | | | | | | | | | Move nvlink endpoint operations to common code. These operations are invoked when handling nvlink core driver requests. Jira VQRM-3523 Change-Id: I93024bf88a8caa3765b33c1264dde452c1a85ee3 Signed-off-by: Thomas Fleury <tfleury@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1698686 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
generated by cgit v1.2.2 (git 2.25.0) at 2024-07-07 18:18:03 -0400