summaryrefslogtreecommitdiffstats
path: root/drivers/gpu/nvgpu/gv11b
Commit message (Collapse)AuthorAge
* gpu: nvgpu: Remove separation of t18x codeTerje Bergstrom2017-11-17
| | | | | | | | | | | Remove separation of t18x specific code and fields and the associated ifdefs. We can build T18x code in always. Change-Id: I4e8eae9c30335632a2da48b418c6138193831b4f Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1595431 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Add translation for NVGPU MM flagsAlex Waterman2017-11-17
| | | | | | | | | | | | | | | | | | | | | | | Add a translation layer to convert from the NVGPU_AS_* flags to to new set of NVGPU_VM_MAP_* and NVGPU_VM_AREA_ALLOC_* flags. This allows the common MM code to not depend on the UAPI header defined for Linux. In addition to this change a couple of other small changes were made: 1. Deprecate, print a warning, and ignore usage of the NVGPU_AS_MAP_BUFFER_FLAGS_MAPPABLE_COMPBITS flag. 2. Move the t19x IO coherence flag from the t19x UAPI header to the regular UAPI header. JIRA NVGPU-293 Change-Id: I146402b0e8617294374e63e78f8826c57cd3b291 Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1599802 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Fix some barrier usageAlex Waterman2017-11-16
| | | | | | | | | | | | | | | | | | | | | | | | | Commit 81868a187fa3b217368206f17b19309846e8e7fb updated barrier usage to use the nvgpu wrappers and in doing so downgraded many plain barriers {mb(), wmb(), rmb()} to the SMP versions of these barriers. The SMP version of the barriers in question are only issued when running on an SMP machine. In most of the cases mentioned above this is fine since the barriers are present to faciliate proper ordering across CPUs. A single CPU is always coherent with itself, so on a non-SMP case we don't need those barriers. However, there are a few places where the barriers in use (GMMU page table programming, IO accessors, userd) where the barrier usage is for communicating and establishing ordering for the GPU. We need these barriers for both SMP machines and non-SMP machines. Therefor we must use the plain barrier versions. Change-Id: I376129840b7dc64af8f3f23f88057e4e81360f89 Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1599744 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: gv11b: modify dma alloc flagsseshendra Gadagottu2017-11-15
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Call nvgpu_dma_alloc_flags_sys without NVGPU_DMA_NO_KERNEL_MAPPING flags, since it makes CMA memory handling simple in t194 and fixes error during subcontext header free: [ 340.378910] trying to free invalid coherent area: ffffffc0135ba000^M [ 340.378921] ------------[ cut here ]------------^M [ 340.378933] WARNING: CPU: 0 PID: 1618 at /code/volta/kernel/kernel-4.9/arch/arm64/mm/dma-mapping.c:1442 __arm_dma_free.isra.4+0x160/0x168^M [ 340.378950] Modules linked in: nvgpu^M [ 340.378958] ^M [ 340.378966] CPU: 0 PID: 1618 Comm: nvogtest Tainted: G W 4.9.52-tegra-g170e0c4 #20^M [ 340.378979] Hardware name: t194pre_si (DT)^M [ 340.378988] task: ffffffc018930d80 task.stack: ffffffc017e08000^M [ 340.378999] PC is at __arm_dma_free.isra.4+0x160/0x168^M [ 340.379009] LR is at __arm_dma_free.isra.4+0x160/0x168^M [ 340.379020] pc : [<ffffff800809cfd0>] lr : [<ffffff800809cfd0>] pstate: 60400045^M [ 340.379032] sp : ffffffc017e0bbe0^M [ 340.379039] x29: ffffffc017e0bbe0 x28: 00000000935ba000 ^M [ 340.379051] x27: 0000000000001000 x26: ffffffc0135b9580 ^M [ 340.379063] x25: ffffff8009ced1b0 x24: 0000000000000010 ^M [ 340.379075] x23: ffffffc070746010 x22: 0000000080000000 ^M [ 340.379088] x21: ffffffbf004d6e80 x20: ffffffc0135ba000 ^M [ 340.379100] x19: 0000000000001000 x18: ffffffffffffffff ^M [ 340.379112] x17: 0000007fa4d8fc60 x16: ffffff800823e370 ^M [ 340.379124] x15: ffffff8009cd8690 x14: ffffff8089fb34bf ^M [ 340.379135] x13: ffffff8009fb34cd x12: 0000000000000007 ^M [ 340.379147] x11: 0000000000000325 x10: 0000000005f5e0ff ^M [ 340.379159] x9 : 0000000000000326 x8 : 3331306366666666 ^M [ 340.379172] x7 : 6666203a61657261 x6 : ffffff8009fb3505 ^M [ 340.379184] x5 : 0000000000000012 x4 : 0000000000000000 ^M [ 340.379196] x3 : 0000000000010000 x2 : 0000000000040934 ^M [ 340.379207] x1 : 0000000000000000 x0 : 0000000000000036 ^M [ 340.379219] ^M [ 340.379224] ---[ end trace 9e7ab41f55eb32d2 ]---^M [ 340.379232] Call trace:^M [ 340.379241] [<ffffff800809cfd0>] __arm_dma_free.isra.4+0x160/0x168^M [ 340.379254] [<ffffff800809e3a0>] arm_dma_free+0x48/0x60^M [ 340.379827] [<ffffff8000f01250>] nvgpu_dma_free+0x260/0x410 [nvgpu]^M [ 340.380403] [<ffffff8000fb2fac>] gv11b_free_subctx_header+0x5c/0x80 [nvgpu]^M [ 340.380980] [<ffffff8000f3ed2c>] gk20a_free_channel_ctx+0x3c/0x150 [nvgpu]^M After changing dma alloc flags to none, this issue got fixed. Bug 1930032 Change-Id: I002236373c6a3ae5d7ec80a35f166429821662b7 Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1598193 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: deprecate TSG/CHANNEL_SET_PRIORITY IOCTLsDeepak Nibade2017-11-15
| | | | | | | | | | | | | | | | | | TSG/CHANNEL_SET_PRIORITY IOCTLs are deprecated and user space should be using combination of timeslice and interleave levels to decide the priority Hence remove the IOCTLs and all corresponding APIs Jira NVGPU-393 Change-Id: Idce925631653784e39864223dc418a99a7e7ca3c Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1598582 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: gv11b: detect stall intr during preemptionSeema Khowala2017-11-15
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Check for interrupts or hangs while waiting for the preempt to complete. During pbdma/eng preempt done polling, any stalling interrupts relating to the runlist must be detected and handled in order for the preemption to complete. When PBDMA fault or CE fault occurs, the PBDMA will save out automatically. TSG related to the context in which the fault occurred will not be scheduled again until the fault is handled. In the case of some other issue requiring the engine to be reset, TSG will need to be manually preempted. In all cases, a PBDMA interrupt may occur prior to the PBDMA being able to switch out. SW must handle these interrupts according to the relevant handling procedure before the PBDMA preempt can complete. Opt for eng reset instead of waiting for preemption to be finished when there is any stall interrupt pending during engine context preempt completion. Bug 200277163 Bug 1945121 Change-Id: Icaef79e3046d82987b8486d15cbfc8365aa26f2e Signed-off-by: Seema Khowala <seemaj@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1522914 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: David Martinez Nieto <dmartineznie@nvidia.com> Tested-by: David Martinez Nieto <dmartineznie@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: mc_elpg_enable & soc credit init moved to bpmpSeema Khowala2017-11-14
| | | | | | | | | | | | | | | | | | | | | | | | | -Program mc_elpg_enable and mss nvlink soc credits only when bpmp is not running or bpmp is running but underlying platorm is simulation. For simulation, bpmp does not execute hot reset sequence. As part of gpu unpowergate, bpmp will program mc_elpg_enable and also set mss nvlink soc credits after bringing mss nvlink out of reset -Remove updating mc_enable as writes to this register has no effect -Remove fifo_fb_iface_r read/write. This hack was added during initial bring up of emulation platforms Bug 2018223 Bug 200269361 Change-Id: Ie09c259e48295a93c6d15376308186152db973fa Signed-off-by: Seema Khowala <seemaj@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1594495 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: gv11b: define final netlistSeema Khowala2017-11-14
| | | | | | | | | | | | | | | Use NETD firmware on gv11b. GV11B_NETLIST_IMAGE_FW_NAME set to GK20A_NETLIST_IMAGE_D Change-Id: I0301999851ffb14713beaf61b5b2cc97efac74eb Signed-off-by: Seema Khowala <seemaj@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1597290 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: Seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: define preemption modes in common codeDeepak Nibade2017-11-14
| | | | | | | | | | | | | Use common preemption modes in common code instead of using linux specific definitions Jira NVGPU-392 Change-Id: Iff65ab4278973f2e2d7db33f6fedb561b2164c42 Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1596931 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: ctx_patch_write fixesPeter Daifuku2017-11-13
| | | | | | | | | | | | | | - Update commit_global_timeslice to remove unused patch parameter - Update calls to ctx_patch_write_begin/end to add update_patch_count param JIRA ESRM-74 Bug 2012077 Change-Id: Ie2e640dfa0ab7193a062a58f588575f220e5efd3 Signed-off-by: Peter Daifuku <pdaifuku@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1594791 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Include UAPI explicitlyTerje Bergstrom2017-11-13
| | | | | | | | | | | | | | Add explicit #includes for <uapi/linux/nvgpu.h> for source code files that depend on it. JIRA NVGPU-388 Change-Id: I5d834e6f3b413cee9b1e4e055d710fc9f2c8f7c2 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1596246 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Remove PTE kind code for GV100/GV11BSami Kiminki2017-11-10
| | | | | | | | | | | | | | | | Remove gv11b_init_uncompressed_kind_map(), gv11b_init_kind_attr(), and the related kind setup code. They are not needed anymore. While we're doing these changes, remove a redundant assignment of g->bootstrap_owner in hal_gv100.c. Bug 1902982 Change-Id: Ib40d8f55cfbfa34143a3765c2b4913926ca021fd Signed-off-by: Sami Kiminki <skiminki@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1560931 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: gv11b: clear channel statusseshendra Gadagottu2017-11-09
| | | | | | | | | | | | | | | | | | | | | | | After unbinding channel, following fields in channel status needs to be cleared manually: ccsr_channel_enable_clr_true ccsr_channel_pbdma_faulted_reset ccsr_channel_eng_faulted_reset Unbinding channel expected to clear all other channel status fields. Bug 1972365 Change-Id: Ibfd84df2f41adc2eb437a026acde3f3d618d7758 Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1594671 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: gv11b: update prod settingsseshendra Gadagottu2017-11-09
| | | | | | | | | | | | | | | | | | Updated clock gating prod settings for HWCL # 39314184. This is corrected output after fixing issue in register generator tool. Bug 1994238 Change-Id: I646c4e1a134570016425367be636250205205005 Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1594605 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: gv11b: Move sm_arch to nvgpu_gpu_paramsTerje Bergstrom2017-11-09
| | | | | | | | | | | | | Move sm_arch_* fields to nvgpu_gpu_params to make them available from common code without accessing Linux specific GPU characteristics. JIRA NVGPU-259 Change-Id: I8e7b542642b620f161d62954400777079065f49d Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1593692 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Return GPU classes in get_litter_valueTerje Bergstrom2017-11-09
| | | | | | | | | | | | | Return GPU classes in HAL get_litter_value() instead of assigning them to GPU characteristics at HAL initialization time. JIRA NVGPU-259 Change-Id: I92cbadf3bd07292a8715d30843972def879795f5 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1593691 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: gv1xx: Move fuse override DT handlingTerje Bergstrom2017-11-09
| | | | | | | | | | | | | | | | | Move fuse override DT handling to Linux code. All the chip specific fuse override functions did the same thing, so delete the HAL and call the same function to read the DT overrides on all chips. Also remove the fuse override functionality from dGPU. There are no DT entries for PCIe devices, so it would've failed anyway. JIRA NVGPU-259 Change-Id: Ic672e25090cdfc207d9771ab61b6cf53185113a4 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1593693 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Switch to newer NVGPU_AS_MAP_BUFFER flagsSami Kiminki2017-11-08
| | | | | | | | | | | | | | | | | | Switch two cases using the old NVGPU_MAP_BUFFER_FLAGS_CACHEABLE_* flags to the newer definitions, that is, NVGPU_AS_MAP_BUFFER_FLAGS_CACHEABLE. The legacy NVGPU_MAP_BUFFER_FLAGS_* definitions have been deleted. Bug 1902982 Change-Id: Ifbd2678b10005b4af2375600888469b01dd09f4e Signed-off-by: Sami Kiminki <skiminki@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1592655 Reviewed-by: Alex Waterman <alexw@nvidia.com> Reviewed-by: Konsta Holtta <kholtta@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>
* nvgpu: gv11b: implement railgate/unrailgateSeema Khowala2017-11-02
| | | | | | | | | | | | | | | | | | | | | | | | | | Implement gv11b platform specific rail gating functions by calling relevant powergate and unpowergate functions and linux clock frmework functions: gv11b_tegra_is_railgated gv11b_tegra_railgate gv11b_tegra_unrailgate These calls will take care of hot reset sequence required for gpu powergate and gpu unpowergate. Bug 200269361 Bug 200273571 Change-Id: Ib1825e4324d51fc508b3b5dc9e5e2fdb252eeff4 Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com> Signed-off-by: Seema Khowala <seemaj@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1589509 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: move platform_gk20a.h to linuxDeepak Nibade2017-11-02
| | | | | | | | | | | | | Fix #includes in all the files to include platform_gk20a.h file with correct path NVGPU-316 Change-Id: Icb26d3c75076b8fdc8da992f751e1cfea22996be Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1589939 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: use struct gk20a for create_gr_sysfsDeepak Nibade2017-11-02
| | | | | | | | | | | | | | | | | | | | | | API gr_gv11b_create_sysfs() and GR HAL create_gr_sysfs() right now receive linux specific struct device But since this function is called from/declared in common code, we need to remove linux dependency from it Hence update the API and GR HAL to receive struct gk20a pointer instead of device pointer Jira NVGPU-259 Change-Id: I65d717ad9f263f0397f8efa5761c64e55c7846eb Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1588465 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> GVS: Gerrit_Virtual_Submit 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: Split ctxsw_trace API into non-Linux componentAlex Waterman2017-11-01
| | | | | | | | | | | | | | T19x component for similar change in the main nvgpu code. JIRA NVGPU-287 Change-Id: Ib126b3d1fb562850fbb3ab89103f2a7fdaa13306 Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1589430 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: use nvgpu_* APIs to allocate/free memoryDeepak Nibade2017-11-01
| | | | | | | | | | | | | | | | Use nvgpu specific nvgpu_kcalloc()/nvgpu_kfree() calls instead of linux specific kcalloc()/kfree() Jira NVGPU-259 Change-Id: I73034ea23561d1269230b9ac10360f8b171b8d41 Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1588221 Reviewed-by: Konsta Holtta <kholtta@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: gv11b: replay invalid pte faults onlySeema Khowala2017-11-01
| | | | | | | | | | | | | | | | | | Try to fix invalid pte type repalayable faults only. All other replayable faults will be cancelled so that next mmu fault for same fault address will be triggered as non-replayable fault and ch/tsg teardown will take place. Bug 1958308 Change-Id: I63b90ce7c639ee183f87db3e771f253fd04c3567 Signed-off-by: Seema Khowala <seemaj@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1566576 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: gv11b: fix faulted channel's id/typeSeema Khowala2017-11-01
| | | | | | | | | | | | | | | | | | | Teardown function should be passed appropriate id and id_type. E.g. if a channel is marked as tsg, channel teardown/rc function should be passed it's tsgid as id and type_tsg as id_type Bug 200277163 Change-Id: I2e83561c03d515fac28cbb8ce75a9f2c7bf746ac Signed-off-by: Seema Khowala <seemaj@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1557296 Reviewed-by: Automatic_Commit_Validation_User 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: Move ctxsw_trace_gk20a.c to common/linuxAlex Waterman2017-10-29
| | | | | | | | | | | | Fixups for the change of name subject in nvgpu. JIRA NVGPU-287 Change-Id: I6c19733079061a42786b94fc48db374d715ccbef Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1586548 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: fix implicit declaration of nvgpu_inst_block_addrSeema Khowala2017-10-27
| | | | | | | | | | | | | | | | | t19x changes necessary for change in core MM code. JIRA NVGPU-30 Change-Id: Id0d66543582abcef522e3182da0b01d0042f4b14 Signed-off-by: Seema Khowala <seemaj@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1585476 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: Konsta Holtta <kholtta@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: gv1xx: resize patch bufferDavid Nieto2017-10-26
| | | | | | | | | | | | | Follow the sizing consideration in bug 1753763 to support dynamic TPC modes and subcontexts. bug 200350539 Change-Id: Ibbdbf02f9c2ea3f082c1b2810ae7176b0775d461 Signed-off-by: David Nieto <dmartineznie@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1584034 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Linux specific GPU characteristics flagsTerje Bergstrom2017-10-26
| | | | | | | | | | | | Make GPU characteristics flags specific to Linux code only. The rest of driver is moved to using nvgpu_is_enabled() API. JIRA NVGPU-259 Change-Id: I46a5a90bb34f170e9e755e7683be142ed6b18cce Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1583992 GVS: Gerrit_Virtual_Submit
* gpu: nvgpu: Linux specific sm_error_state_recordTerje Bergstrom2017-10-26
| | | | | | | | | | | | | | Create an nvgpu internal nvgpu_gr_sm_error_state to store and propagate SM error state within driver. Use nvgpu_dbg_gpu_sm_error_state_record only in Linux code. JIRA NVGPU-259 Change-Id: Ia2b347d0054365bdc790b4d6f2653a568935bdb0 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1585646 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: init_cyclestats fixesPeter Daifuku2017-10-25
| | | | | | | | | | | | | | | | | | - in the native case, replace calls for init_cyclestats with the gm20b version, as each chip had identical versions of the code. - in the virtual case, use the vgpu version of the function in order to get the new max_css_buffer_size characteristic set to the mempool size. JIRA ESRM-54 Bug 200296210 Change-Id: I475876cb392978fb1350ede58e37d0962ae095c3 Signed-off-by: Peter Daifuku <pdaifuku@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1578934 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: gv11b: update regops whitelistSeema Khowala2017-10-25
| | | | | | | | | | | | | Updated regops whitelist for HW CL 39314184 i.e. snap_0913 and VDK_R11 Change-Id: Ie22f0a000c4bb151023a92e0d7e877bbceb157f2 Signed-off-by: Seema Khowala <seemaj@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1565684 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Seshendra Gadagottu <sgadagottu@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: gv11b: update clock gating prod settingsSeema Khowala2017-10-25
| | | | | | | | | | | | Updated clock gating prod settings for HW CL 39314184 i.e. snap_0913 and VDK_R11 Change-Id: Iae6fd9e95ee5e1ec20bafbb24cd761bdce8fdc5f Signed-off-by: Seema Khowala <seemaj@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1565683 Reviewed-by: Seshendra Gadagottu <sgadagottu@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: gv11b: use scg perf for smid numberingseshendra Gadagottu2017-10-25
| | | | | | | | | | | | | | | | | | | | | | For SCG to work, smid numbering needs to be done based on scg performance of tpcs. For gv11b and gv11b vgpu, reuse gv100 function "gr_gv100_init_sm_id_table" to do this. Used local variable "index" to avoid multiple computations in the function: gr_gv100_init_sm_id_table index = sm_id + sm Add deug info for printing initialized gpc/tpc/sm/global_tpc indexs. Bug 1842197 Change-Id: Ibf10f47f10a8ca58b86c307a22e159b2cc0d0f43 Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1583916 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Cleanup generic MM codeAlex Waterman2017-10-24
| | | | | | | | | | | | | | t19x changes necessary for change in core MM code. JIRA NVGPU-30 Change-Id: I62f419450c1a33d0826390d7cbb5ad93569f8c89 Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1577265 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: David Martinez Nieto <dmartineznie@nvidia.com> Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: gv11b: use correct acr_dmem_desc ver.Deepak Goyal2017-10-23
| | | | | | | | | | | | | | | gv11b should use acr_dmem_desc_v1 instead of acr_dmem_desc. JIRA GPUT19X-5 Change-Id: I3ccae72541607aec12e25845ea4cb875ff11d67c Signed-off-by: Deepak Goyal <dgoyal@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1583642 Reviewed-by: Supriya Sharatkumar <ssharatkumar@nvidia.com> Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: gv10x: alloc fault buffer in sysmemDavid Nieto2017-10-23
| | | | | | | | | | | | | | | With coherency issues solved, it is no longer needed to allocate the fault buffer in vidmem as a workaround. JIRA: NVGPUGV100-36 Change-Id: I1c83e9bac61f27b75f38fce963899485afeed009 Signed-off-by: David Nieto <dmartineznie@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1582769 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com> GVS: Gerrit_Virtual_Submit
* gpu: nvgpu: gv100: fix timeout handlingDavid Nieto2017-10-23
| | | | | | | | | | | | | GV100 has a larger vidmem size and a slower sideband to sysmem so timeouts need to be adjusted to avoid false positives. JIRA: NVGPUGV100-36 Change-Id: I3cbc19aa1158c89bc48ae1fa6ec4bc755cd9389d Signed-off-by: David Nieto <dmartineznie@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1582092 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: gv100 memory unlock supportMahantesh Kumbar2017-10-21
| | | | | | | | | | | | | | | | | | - Added method to load mem unlock binary into nvdec falcon & execute to perform mem unlock if VPR enabled. - Updated .mem_unlock gv100 HAL to point method gv100_fb_memory_unlock(). - Updated .mem_unlock gv11b HAL to NULL. - Added vpr info hw registers - Added nvdec enable hw register Change-Id: Ia4bf820ae103baede679d300d1d390fd748c919a Signed-off-by: Mahantesh Kumbar <mkumbar@nvidia.com> (cherry picked from commit 2e176ad9d47316bf4d001692a2ae07e6c1fb1ccb) Reviewed-on: https://git-master.nvidia.com/r/1573101 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: gv11b: memory aperture for perfbufseshendra Gadagottu2017-10-20
| | | | | | | | | | | | | | | Updated perf_pmasys_mem_block_target in perf_pmasys_mem_block based on memory aperture used for perfbuf inst_block. Bug 200327596 Change-Id: Ic3df332a2248c5ea2d6d38ceab8ba04c618ffefc Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1582788 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: Alex Waterman <alexw@nvidia.com> Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: gv11b: update css opsseshendra Gadagottu2017-10-20
| | | | | | | | | | | | | | | | | | | | | | Updated following hal functions for css gv11b and reused them for gv100: enable_snapshot disable_snapshot check_data_available These changes are needed because of following reasons: 1. Register offsets for perf_pmasys_* are changed for gv11b/gv100 from gk20a. 2. Updated memory type for perf_pmasys_mem_block_target based on memory aperture used for hwpm inst_block. Bug 200327596 Change-Id: I500d17670e2f389d8d0e77884374bcc3504a41f8 Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1507546 GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: gv1xx: Remove HAL for restore_context_headerseshendra Gadagottu2017-10-20
| | | | | | | | | | | | gr restore_context_header is not required any more after enabling per context va mode for subcontext. Cleaning-up unused function pointers from gv100 and gv11b HAL. Change-Id: I65cc7d12d3c96726d323defd99726c3e259e7e63 Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1581432 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Refactoring nvgpu_vm functionsAlex Waterman2017-10-18
| | | | | | | | | | | | | | | | | Change required for equivalent change on nvgpu. This is required since a few HALs were added that must be populated for all chips. This patch adds those HAL definitions for gv11b, gv100, and the vgpu. JIRA NVGPU-30 JIRA NVGPU-138 Change-Id: I65374764350a5cacce8624b15d98947fada35a4a Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1579865 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: gv11b: update dbg opsseshendra Gadagottu2017-10-18
| | | | | | | | | | | | | | | | | | | | | Updated following hal functions for gv11b and reused them for gv100: perfbuffer_enable perfbuffer_disable These changes are needed because of following reasons: 1. Register offsets for perf_pmasys_* are changed for gv11b/gv100 from gk20a. 2. Updated memory type for perf_pmasys_mem_block_target to sys_ncoh_f(). Bug 200327596 Change-Id: Ia672ac561917c8ed36caea9cc7e74b7fc7ce8188 Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1571074 GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: gv11b: Use nvgpu_rwsem as TSG channel lockTerje Bergstrom2017-10-17
| | | | | | | | | | | | | Use abstract nvgpu_rwsem as TSG channel list lock instead of the Linux specific rw_semaphore. JIRA NVGPU-259 Change-Id: I5f6c918464315e3d140bea0c61a619c3712619c1 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1579934 GVS: Gerrit_Virtual_Submit Reviewed-by: Seshendra Gadagottu <sgadagottu@nvidia.com>
* gpu: nvgpu: gv11b: Use internal nvgpu_warpstateTerje Bergstrom2017-10-17
| | | | | | | | | | | | | Replace use of ioctl structure warpstate with internal nvgpu_warptate. JIRA NVGPU-259 Change-Id: I003c15152042e566124c04d6124e515e36157c88 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1578683 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: vgpu: flatten out t19x vgpu halPeter Daifuku2017-10-13
| | | | | | | | | | | | | | | | | Instead of calling the native HAL init function then adding multiple layers of modification for VGPU, flatten out the sequence so that all entry points are set statically and visible in a single file. JIRA ESRM-30 Change-Id: I8d277aaccb0e63b2d504e7aba32eb31ef82f4ec0 Signed-off-by: Peter Daifuku <pdaifuku@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1574619 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: Richard Zhao <rizhao@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: gv11b: Abstract IO aperture accessorsTerje Bergstrom2017-10-13
| | | | | | | | | | | | | | Implement T19x specific usermode aperture initialization functions. Move usermode_regs field to nvgpu_os_linux_t19x, because it is Linux specific. JIRA NVGPU-259 Change-Id: I9d6ce243a692ab48209d468288ed85f89fb26770 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1569699 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: gv11b: add syncpt shim ro mapseshendra Gadagottu2017-10-13
| | | | | | | | | | | | | | | | | For sync-point read map, create read only map per vm and share with all channels that are using same vm. Now restrict rw map to single syncpoint shim memory range. JIRA GPUT19X-2 Change-Id: Ibd0b82d1cdb8861e1dbb073b27da1f9c9ab1d2ab Signed-off-by: seshendra Gadagottu <sgadagottu@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1514339 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: Alex Waterman <alexw@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* Revert "gpu: nvgpu: gv11b: disable cycle stat"Timo Alho2017-10-12
| | | | | | | | | | | | | | | | This reverts commit 6647e5c9569258fbf3db096275a79f86f86ed3a6. Bug 200352825 Change-Id: Ia44d61eafce78f99be2271e0afaf69cd5c102080 Signed-off-by: Timo Alho <talho@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1577920 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-by: Seshendra Gadagottu <sgadagottu@nvidia.com> Tested-by: Seshendra Gadagottu <sgadagottu@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Srikar Srimath Tirumala <srikars@nvidia.com>
generated by cgit v1.2.2 (git 2.25.0) at 2024-08-26 07:48:50 -0400