summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAge
* gpu: nvgpu: adding USERMODE enable bit.Prateek Sethi2018-06-14
| | | | | | | | | | | | | | | | | | This change is required to keep enable USERMODE on QNX as part of gpu characteristics unification. Jira VQRM-3996 Change-Id: I9881c8d3a2dfcadf618a56deee847cd9b575ed7b Signed-off-by: Prateek Sethi <prsethi@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1735573 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: Alex Waterman <alexw@nvidia.com> Reviewed-by: Richard Zhao <rizhao@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: fix crash due to accessing incorrect TSG pointerDeepak Nibade2018-06-14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | In gk20a_gr_isr(), we handle various errors including GPC/TPC errors. And then if BPT errors are pending we call gk20a_gr_post_bpt_events() at the end and pass channel pointer to it gk20a_gr_post_bpt_events() extracts TSG pointer based on ch->tsgid But in some race conditions it is possible that we clear the error and trigger recovery and as a result channel is unbounded from TSG and closed by user space before calling gk20a_gr_post_bpt_events() And in that case the code above results in getting incorrect TSG pointer and hence crashes as below Unable to handle kernel paging request at virtual address ffffff8012000c08 ... [<ffffff8008081f84>] el1_da+0x24/0xb4 [<ffffff80086e72e0>] gk20a_tsg_get_event_data_from_id+0x30/0xb0 [<ffffff80086e7560>] gk20a_tsg_event_id_post_event+0x50/0xc8 [<ffffff800872922c>] gk20a_gr_isr+0x27c/0x12e0 To fix this extract the TSG pointer before handling all the errors and pass this pointer to gk20a_gr_post_bpt_events() will post the events if they are enabled and if TSG is still open Bug 200404720 Change-Id: I4861c72e338a2cec96f31cb9488af665c5f2be39 Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1735415 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Vinod Gopalakrishnakurup <vinodg@nvidia.com> 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: vgpu: implement .free_syncpt_bufRichard Zhao2018-06-14
| | | | | | | | | | | | | | | vgpu needs to free the va range besides unmap the syncpt buffer. Jira VFND-4870 Change-Id: Ia9edb687b66db6da8529bf256d348999e24a0bc3 Signed-off-by: Richard Zhao <rizhao@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1693094 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: initialize HAL for NEXT_2 gpuDeepak Nibade2018-06-14
| | | | | | | | | | | | | | | | | In gpu_init_hal(), call NVGPU_NEXT_2_INIT_HAL() if we detect chip NVGPU_GPUID_NEXT_2 Jira NVGPUT-95 Change-Id: Ie1121591e53a1587766ea03bb62d0aae01d9ccbf Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1734099 Reviewed-by: Automatic_Commit_Validation_User 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: add g->fifo_eng_timeout_usThomas Fleury2018-06-14
| | | | | | | | | | | | | | Add g->fifo_eng_timeout_us to define engine timeout in microseconds. It is initialized with GRFIFO_TIMEOUT_CHECK_PERIOD_US. In RM server case, it can be overriden with value defined in device tree. Jira EVLR-2674 Change-Id: I69ac2ce779fe575566c8ba48e8cd2d0e6b2d93cf Signed-off-by: Thomas Fleury <tfleury@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1728391 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Remove check to disallow gv100 link0/1Tejal Kudav2018-06-14
| | | | | | | | | | | | | | | | | | On GV100, we could not enable reflck repeater at source of PLL which is shared by link 0/1. So we do not allow link 0 and 1 to be used on GV100. This refclk repeater is present only on GV100. Remove the check as we currently use link3 on GV100 and do not plan to use any other link. JIRA NVLINK-162 Change-Id: I9ffcc0b20d084a208271d2c594ec64b5bafaabfb Signed-off-by: Tejal Kudav <tkudav@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1734538 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: support CAU ctxsw listDeepak Nibade2018-05-30
| | | | | | | | | | | | | | | | | | | | | | | CAU (Counter Aggregation Unit) registers might be split out from SMPC registers and moved into their own list on some platforms In gr_gk20a_init_ctx_vars_fw() add support to check if pm_cau list is available If list is available, count will be set to non-zero here In add_ctxsw_buffer_map_entries_gpcs(), parse the pm_cau list if count is non-zero Bug 2139870 Change-Id: Ia630e7d03481a6f927c6739d28ebfe49f221326f Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1733208 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Matthew Braun (SW-GPU) <matthewb@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: gv11b/gv100: add SHADER_CUT_COLLECTOR compute classSeema Khowala2018-05-30
| | | | | | | | | | | | | | | | | | | | Support NVC3C0_SET_SHADER_CUT_COLLECTOR s/w method for compute class. This is needed to enable/disable SHADER_CUT_COLLECTOR_STATE. Bug 2108381 Bug 2099681 Change-Id: I228dcc55b6df605aeeee094f872157023e3c783c Signed-off-by: Seema Khowala <seemaj@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1730634 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: Wei Sun <wsun@nvidia.com> Tested-by: Wei Sun <wsun@nvidia.com> Reviewed-by: Seshendra Gadagottu <sgadagottu@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: Make the header compile for userspace buildsSami Kiminki2018-05-30
| | | | | | | | | | | | | | | | | | | | Do the quick fixes to make the header compile on userspace builds without changes. Namely: - define __packed only if not yet defined - Fix the NVGPU_NO_TIMEOUT definition ((u32)~0) --> ((__u32)~0U) - Remove the left-over definition of NVGPU_AS_IOCTL_MAP_BUFFER. It refers to a struct that no longer exists. Bug 1777616 Bug 1902982 Change-Id: I964aa5eff7bfade6af4fd0635680fd6af29d623b Signed-off-by: Sami Kiminki <skiminki@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1729654 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: posix: undef min, max macros is definedAlex Waterman2018-05-30
| | | | | | | | | | | | | | | | | | | These macros are sometimes defined by the std library headers. So when they are defined #undef them and use our own version. Also explicitly include types.h from gmmu.c since it uses the min and max macros. JIRA NVGPU-525 Change-Id: I905ff23ef7a4a96467be59c4a124cb09b63f1f96 Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1721015 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: posix: add impl for nvgpu_is_soc_t194_a01()Alex Waterman2018-05-29
| | | | | | | | | | | | | | | | | This function was added for Linux but not for posix. Thankfully the implementation for posix is trivial: just return false. JIRA NVGPU-525 Change-Id: Ib5af66dedeabb754ce36cbf6f0351e4af3926c52 Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1730742 Reviewed-by: Vinod Gopalakrishnakurup <vinodg@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Seshendra Gadagottu <sgadagottu@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: Fix include path for clockgating headersVinod G2018-05-29
| | | | | | | | | | | | Removed the reference for $NVGPU/drivers/gpu/nvgpu/common from Makefile and fixed the include path for the clockgating headers in HAL files. Change-Id: I34c179866d28c932501b38184ec18d20dce81480 Signed-off-by: Vinod G <vinodg@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1730759 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: gv11b: move to use gops.fifo.check_tsg_ctxsw_timeoutRichard Zhao2018-05-25
| | | | | | | | | | | | | Fixing RM server crash when handles ctxsw timeout. RM server needs to call its own HAL of .check_tsg_ctxsw_timeout. Jira EVLR-2691 Change-Id: Ibbc09cbcda39d44dc2a721c9b270d065e13fa9e2 Signed-off-by: Richard Zhao <rizhao@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1729900 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: abstract submit profilingKonsta Holtta2018-05-25
| | | | | | | | | | | | | | | | Add gk20a_fifo_profile_snapshot() to store the submit time in a profiling entry that was acquired from gk20a_fifo_profile_acquire(). Also get rid of ifdef CONFIG_DEBUG_FS by stubbing the acquire and free functions when debugfs is not enabled. This reduces some cyclomatic complexity in the submit path. Jira NVGPU-708 Change-Id: I39829a6475cfe3aa582620219e420bde62228e52 Signed-off-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1729545 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: remove empty gk20a_mm_debugfs_initKonsta Holtta2018-05-25
| | | | | | | | | | | | gk20a_mm_debugfs_init() used to initialize one mm-specific debugfs knob, but now that there aren't any, delete the function and the related files. Change-Id: Ic3177945bbab501c506b5199b44c87450df75452 Signed-off-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1729530 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Remove unused function declarationsAlex Waterman2018-05-25
| | | | | | | | | | | | Change-Id: I36f2d13ed3797719137c670afef9b644d48ea16e Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1727485 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Konsta Holtta <kholtta@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: posix: Use GCC builtins for ffs(), fls()Alex Waterman2018-05-25
| | | | | | | | | | | | | | | | These intrinsics will be fast for the given platform that the compiler is targeting. This also reduces complexity in the code. Change-Id: I6cfb761d6f881056446fa9a5de53dca50ed93c34 Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1727383 Reviewed-by: Konsta Holtta <kholtta@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: posix: Don't do the POSIX build on integrityAlex Waterman2018-05-25
| | | | | | | | | | | | | | | | | For now interity needs a lot of work. It may make sense to come back to this in the future though. JIRA NVGPU-525 Change-Id: Ib533d7a85f105d17debe8a17431f06dc348534a9 Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1727382 Reviewed-by: Richard Zhao <rizhao@nvidia.com> Reviewed-by: Konsta Holtta <kholtta@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: Move bus HAL to commonTerje Bergstrom2018-05-25
| | | | | | | | | | | Move implementation of bus HAL to common/bus. Change-Id: Ia89350f9d94f3ccfd5500a340e6a677cd7d4cfaa Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1726337 GVS: Gerrit_Virtual_Submit Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Add types.h to clock gating headersAlex Waterman2018-05-24
| | | | | | | | | | | | | | The headers use u32 and bool but do not include <nvgpu/types.h>. Moving around header includes exposed this issue in the cascade builds. This patch fixes the problem in all clock gating headers to avoid this being a concern in the future. Change-Id: Id56074df393d95bf65baf4062ac811d80d87e96b Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1729748 Reviewed-by: Bo Yan <byan@nvidia.com>
* gpu: nvgpu: Don't use -lpthread for QNXAlex Waterman2018-05-24
| | | | | | | | | | | | | | | | | The QNX compiler seems to automatically link against this library and as such the extra -lpthread is not necessary. Instead it causes a link failure since the pthread library is not present. JIRA NVGPU-525 Change-Id: Id5a6fcdffb067ed961665a3ee44a9d44301b725b Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1722157 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: Remove __uXX typedefsAlex Waterman2018-05-24
| | | | | | | | | | | | | | | | | Integrity already typedefs these and complains if you override them even with the same underlying type. Since we only use these in the regops_gk20a.h header file (outside of the Linux specific code, that is) this patch just changes the __uXX to uXX. With that we can delete the now unnecessary __uXX defs. JIRA NVGPU-525 Change-Id: I01dd2723b68db2170449342f73c711ee5a589adb Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1721186 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Add posix condition in sim.hAlex Waterman2018-05-24
| | | | | | | | | | | | | | | Without this the default fall back includes are the rmos headers which are obviously not present for the POSUX build. Change-Id: Iaf7d459e09c62dd57c5b33e21934e40f5780840a Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1727427 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: Vinod Gopalakrishnakurup <vinodg@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 proper include path for clockgating headerAlex Waterman2018-05-24
| | | | | | | | | | | | | | | | | | Instead of referencing the header from $NVGPU/drivers/gpu/nvgpu/common reference it from $NVGPU/drivers/gpu/nvgpu. This makes the POSIX compilation happy since we don't do a -Idrivers/gpu/nvgpu/common. Not sure exactly why the regular kernbel build does this but it probably should not. Change-Id: I00aee373b651e3b7710669fa04c5b75fc1c814d9 Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1727426 GVS: Gerrit_Virtual_Submit Reviewed-by: Vinod Gopalakrishnakurup <vinodg@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: fix oob access in submit profiling dumpKonsta Holtta2018-05-24
| | | | | | | | | | | | | | | | The number of samples has to be at least the number of percentile ranges (here 20) for the reporting to work as expected and also to not cause negative indices in reading the sorted profile data. If there are not enough samples, just report all zeroes. Change-Id: Ie893859d95074f5ceabf6abe873941873668861d Signed-off-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1721892 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: add HALs to enable/disable hub interruptsDeepak Nibade2018-05-24
| | | | | | | | | | | | | | | | | | Add below two new HALs gops.fb.enable_hub_intr() to enable hub interrupts gops.fb.disable_hub_intr() to disable hub interrupts Set existing APIs gv11b_fb_enable/disable_hub_intr() to these HALs Call the HALs everywhere instead of calling the APIs directly Jira NVGPUT-44 Change-Id: Id299c6d228733ed365a71be6b180186776cc1306 Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1725977 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: export hub intr handling APIsDeepak Nibade2018-05-24
| | | | | | | | | | | | | | | | | | | | | | | | Export below APIs in fb_gv11b.h gv11b_fb_handle_dropped_mmu_fault() gv11b_fb_handle_other_fault_notify() gv11b_fb_handle_mmu_nonreplay_replay_fault() gv11b_fb_handle_nonreplay_fault_overflow() gv11b_fb_handle_replay_fault_overflow() gv11b_handle_l2tlb_ecc_isr() gv11b_handle_hubtlb_ecc_isr() gv11b_handle_fillunit_ecc_isr() Jira NVGPUT-44 Change-Id: Ib50e3f3c2f698d486ffe718ebf4a651ccfe8cd93 Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1725976 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: Code updates for MISRA violationsVinod G2018-05-24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Regenerated the gating_reglist.c files for various chips after fixing the script for MISRA C-2012 violations Rule 15.5: Multiple points of exit detected Rule 15.6: "if" body without compound statement Rule 10.3: Implicit conversions of 64bit to 32bit int Rule 7.2: Const must be declared with "U" Rule 5.7: Tags with name xxx already declared Add preprocessor conditional gaurds in gating_reglist header files JIRA NVGPU-671 JIRA NVGPU-656 JIRA NVGPU-688 JIRA NVGPU-686 JIRA NVGPU-644 Change-Id: Ie5a688cb8c39f072d2a15d86fb0ee0f2039a2cf1 Signed-off-by: Vinod G <vinodg@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1724444 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: Delete unused static variableAlex Waterman2018-05-24
| | | | | | | | | | | | | | | | This variable is never used. So remove it. JIRA NVGPU-525 Change-Id: I7ace77ffe1c2da58d8d9cee3bbbcf8361886bddf Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1724094 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: Cast unsigned arith to signed for abs()Alex Waterman2018-05-24
| | | | | | | | | | | | | | | | | | | | | | | | | Before passing in an unsigned value to abs() cast the result of the subtraction to signed. In Linux this happens automatically but on non-Linux platforms abs() does not necessarily do this. clang flags this case as a pointless operation: abs(x) obviously must equal x for any unsigned x. This change should hopefully preserve the Linux behavior but avoid the compiler warning from clang. JIRA NVGPU-525 Change-Id: I71320964c0922f1e4890c8b25d801f17e54ed3c0 Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1724093 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: Aleksandr Frid <afrid@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: remove broken force_pramin featureKonsta Holtta2018-05-24
| | | | | | | | | | | | | | The forced PRAMIN reads and writes for sysmem buffers haven't worked in a while since the PRAMIN access code was refactored to work with vidmem-only sgt allocs. This feature was only ever meant for testing and debugging PRAMIN access and early dGPU support, but that is stable enough now so just delete the broken feature instead of fixing it. Change-Id: Ib31dae4550f3b6fea3c426a2e4ad126864bf85d2 Signed-off-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1723725 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: use nvgpu_current_time_ns in submit profilingKonsta Holtta2018-05-24
| | | | | | | | | | | | | Replace Linux-specific and dubious sched_clock() with common nvgpu_current_time_ns(). sched_clock() used also nanoseconds. Jira NVGPU-708 Change-Id: I70f992fe42cc9c3ffed374fdebd582867475e84f Signed-off-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1723202 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: add nvgpu_current_time_ns()Konsta Holtta2018-05-24
| | | | | | | | | | | | | | | | | | | | Add an abstraction over a monotonic system clock in nanosecond units. Use ktime_get() for the Linux implementation, similarly to nvgpu_current_time_ms(). Jira NVGPU-708 Change-Id: I3165c20abf2652f1a1fa04e66c04cd34a8fe6dcc Signed-off-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1723201 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-by: Sourab Gupta <sourabg@nvidia.com> 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: populate gpu rev based on soc checkseshendra Gadagottu2018-05-24
| | | | | | | | | | | | | | | | | | Populate gpu rev as 0xa2 for gv11b with t194 A02 soc. Bug 2053668 Change-Id: I22a2bc7026162e34e9a605dfda3d83fa989b5248 Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com> Signed-off-by: Seema Khowala <seemaj@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1713096 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Wei Sun <wsun@nvidia.com> Reviewed-by: Bharat Nihalani <bnihalani@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: gv11b: chip revision check for invalidatesseshendra Gadagottu2018-05-24
| | | | | | | | | | | | | | | Only for T194 A01 version following invalidates are disabled: -CBM alpha and beta invalidations for L2 -SCC pagepool invalidates -SWDX spill buffer invalidates Bug 2053668 Change-Id: I7122b223946a1bfa4b11ed8ee782572215313dc1 Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1680500 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: use nvgpu types.h in clock gating sourceSourab Gupta2018-05-22
| | | | | | | | | | | | | | | | Use the nvgpu/types.h instead of linux/types.h in the clock gating sources Jira VQRM-3700 Change-Id: Ib399cc4367c77f0d08454aa7639bb619367f673b Signed-off-by: Sourab Gupta <sourabg@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1726782 Reviewed-by: Shashank Singh <shashsingh@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: Add GPU Fmax@Vmin access through BPMPAlex Frid2018-05-22
| | | | | | | | | | | | | | On Tegra platforms that have clock management under BPMP, and do not support Tegra DVFS, GPU driver cannot access Fmax@Vmin (get interface always returns "0"). Added such access through BPMP DVFS shim driver. Bug 2045903 Change-Id: I0222f2e2917cda15d18ea3296dd1fe53b2ea6b45 Signed-off-by: Alex Frid <afrid@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1722431 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* nvgpu: Added QNX specific include in sim.hAntony Clince Alex2018-05-21
| | | | | | | | | | | | | | | | | | | | Added QNX specific include in the common sim.h Forward declared "platform_device". JIRA VQRM-3836 Change-Id: I6e965ccc41df2445b36111d88e9cebf9866dd877 Signed-off-by: Antony Clince Alex <aalex@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1725686 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Sourab Gupta <sourabg@nvidia.com> GVS: Gerrit_Virtual_Submit Tested-by: Sourab Gupta <sourabg@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: Code updates for MISRA violationsVinod G2018-05-21
| | | | | | | | | | | | | | | | As part of the MISRA fixes, moving all the gating_reglist files to common/clock_gating dir, the new directory structure suggested to follow. Removed unused gating_reglist files for gk20a JIRA NVGPU-646 Change-Id: I388855befcf991ee68eeffed10fe9ac456210649 Signed-off-by: Vinod G <vinodg@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1722330 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: vgpu: init gr->ch_tlb_lockRichard Zhao2018-05-21
| | | | | | | | | | | | | | | | | | The bug was exposed when enable CONFIG_DEBUG_SPINLOCK. Jira VFND-4943 Change-Id: I01720f93fe6de9b85987d490df852c8d1c8fb1c2 Signed-off-by: Richard Zhao <rizhao@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1703656 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Thomas Fleury <tfleury@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Sourab Gupta <sourabg@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: 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>