summaryrefslogtreecommitdiffstats
path: root/drivers
Commit message (Collapse)AuthorAge
* gpu: nvgpu: debugfs code to dump HAL functionsSunny He2017-08-24
| | | | | | | | | | | | | | | | | | | | | Prints addresses of device-specific HAL functions to debugfs file hal/gops. The list of functions is produced by dumping the contents of the gpu_ops substruct of the gk20a struct. This interface makes the assumption that there are only function pointers in gpu_ops. Companion Python script nvgpu_debug_hal.py analyzes gk20a.h to determine operation counts and prettyify debugfs interface's output. Jira NVGPU-107 Change-Id: I0910e86638d144979e8630bbc5b330bccfd3ad94 Signed-off-by: Sunny He <suhe@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1542990 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Reorg gr HAL initializationSunny He2017-08-24
| | | | | | | | | | | | | | | | | | | | | Reorganize HAL initialization to remove inheritance and construct the gpu_ops struct at compile time. This patch only covers the gr sub-module of the gpu_ops struct. Perform HAL function assignments in hal_gxxxx.c through the population of a chip-specific copy of gpu_ops. Jira NVGPU-74 Change-Id: Ie37638f442fd68aca8a7ade5f297118447bdc91e Signed-off-by: Sunny He <suhe@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1542989 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com> Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
* gpu: nvgpu: Limit max CBC clear job size to 16384 linesSami Kiminki2017-08-24
| | | | | | | | | | | | | | | | | | | | | Limit the maximum job size of CBC ctrl clear to 16 klines. This avoids timeouts and excessive lock hold duration when clearing comptags for huge surface. 16 klines corresponds to a 1-GB surface for 64-kB compression page size. If the requested CBC ctrl job is larger than 16 klines, split it to at most 16-kline chunks. Bug 1860962 Bug 200334740 Change-Id: Ibc69adc8bf59527b1acec5b2097b5aefa2169960 Signed-off-by: Sami Kiminki <skiminki@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1540432 Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com> Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: Replace kref for refcounting in nvgpuDebarshi Dutta2017-08-24
| | | | | | | | | | | | | | | | - added wrapper struct nvgpu_ref over nvgpu_atomic_t - added nvgpu_ref_* APIs to access the above struct JIRA NVGPU-140 Change-Id: Id47f897995dd4721751f7610b6d4d4fbfe4d6b9a Signed-off-by: Debarshi Dutta <ddutta@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1540899 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
* gpu: nvgpu: Add mem usage to page allocator debugAlex Waterman2017-08-24
| | | | | | | | | | | | | | Add the amount of memory used to the page allocator debug dump. Change-Id: Icd4b4a0489068aaa3f60221b792de7f8dbf0092c Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1543695 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com> Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
* gpu: nvgpu: correct NUM_FBPASPeter Daifuku2017-08-22
| | | | | | | | | | | | | | | Although igpu does not have an FBPA unit, the hardware reports one, and the ucode leaves space for one in the HWPM context save buffer. So let NUM_FBPAS reflect this, so that registers that follow this section in the context buffer are offset properly JIRA EVLR-1716 Change-Id: I40105cf2ee7f92a44d1c7cc2d23a17276bbb9e84 Signed-off-by: Peter Daifuku <pdaifuku@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1535273 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Remove support for old kernel versionTerje Bergstrom2017-08-22
| | | | | | | | | | | Remove support for pre-4.4 kernels. This allows deleting the checks for kernel version, and usage of linux/version.h. Change-Id: I4d6cb30512ea164d27549f4f4d096e5931bb1379 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1543499 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Delete VBIOS overlay supportTerje Bergstrom2017-08-22
| | | | | | | | | | | | | | | VBIOS overlay support is not used anywhere, and it introduces a dependency to Linux PCIe. Instead of fixing it, just delete the code. Change-Id: I751e490f8bb8271a19b917686f5d2c8a85d92cb7 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1541512 Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com> 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
* gpu: nvgpu: Use os_linux.h only for debugfsTerje Bergstrom2017-08-22
| | | | | | | | | | | | | | | gm206_bios_init() uses nvgpu_os_linux only for accessing debugfs. That's the only reason os_linux.h is included. Make the #include conditional and enabled only when debugfs is enabled. Change-Id: I2939d51a25865ab634c92676daa9a7c10551fee5 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1541511 Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com> 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
* gpu: nvgpu: Copy vbios_min_version to struct gk20aTerje Bergstrom2017-08-22
| | | | | | | | | | | | Accessing vbios_min_version in gk20a_platform creates an extra dependency to Linux. Copy it to struct gk20a at driver initialization. Change-Id: I9ff5dbeb1fecc6dc44a62f7affc24fd52c2bab26 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1542837 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Use nvgpu flags for run_preosTerje Bergstrom2017-08-22
| | | | | | | | | | | | | | Accessing run_preos from gk20a_platform causes unnecessary Linux dependency, so copy the flag to abstract flags. Change-Id: I4818fb6735201f36e552c1ff45138a44a3d94db1 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1542836 Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com> 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
* gpu: nvgpu: Nvgpu abstraction for linux barriers.Debarshi Dutta2017-08-22
| | | | | | | | | | | | | | | | | construct wrapper nvgpu_* methods to replace mb,rmb,wmb,smp_mb,smp_rmb,smp_wmb,read_barrier_depends and smp_read_barrier_depends. NVGPU-122 Change-Id: I8d24dd70fef5cb0fadaacc15f3ab11531667a0df Signed-off-by: Debarshi <ddutta@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1541199 Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com> Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Sourab Gupta <sourabg@nvidia.com> Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
* gpu: nvgpu: Reorg pmu HAL initializationSunny He2017-08-21
| | | | | | | | | | | | | | | | | | | | Reorganize HAL initialization to remove inheritance and construct the gpu_ops struct at compile time. This patch only covers the pmu sub-module of the gpu_ops struct. Perform HAL function assignments in hal_gxxxx.c through the population of a chip-specific copy of gpu_ops. Jira NVGPU-74 Change-Id: I8839ac99e87153637005e23b3013237f57275c54 Signed-off-by: Sunny He <suhe@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1530982 Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com> 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>
* gpu: nvgpu: Move non-fp pmu members from gpu_opsSunny He2017-08-21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Move non-function pointer members out of the pmu and pmu_ver substructs of gpu_ops. Ideally gpu_ops will have only function ponters, better matching its intended purpose and improving readability. - g.ops.pmu_ver.cmd_id_zbc_table_update has been changed to g.pmu_ver_cmd_id_zbc_table_update - g.ops.pmu.lspmuwprinitdone has been changed to g.pmu_lsf_pmu_wpr_init_done - g.ops.pmu.lsfloadedfalconid has been changed to g.pmu_lsf_loaded_falcon_id Boolean flags have been implemented using the enabled.h API - g.ops.pmu_ver.is_pmu_zbc_save_supported moved to common flag NVGPU_PMU_ZBC_SAVE - g.ops.pmu.fecsbootstrapdone moved to common flag NVGPU_PMU_FECS_BOOTSTRAP_DONE Jira NVGPU-74 Change-Id: I08fb20f8f382277f2c579f06d561914c000ea6e0 Signed-off-by: Sunny He <suhe@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1530981 Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com> 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>
* gpu: nvgpu: protect header inclusion in __kernel__Debarshi Dutta2017-08-21
| | | | | | | | | | | | | | The header <linux/atomic.h> needs to be included inside __kernel__ JIRA-121 Change-Id: I978e7c12c6fff7d7ccea300d0be5784d45e559ed Signed-off-by: Debarshi <ddutta@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1541220 Reviewed-by: Sourab Gupta <sourabg@nvidia.com> Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
* gpu: nvgpu: fix include file recursion for clk_arbThomas Fleury2017-08-18
| | | | | | | | | | | | | | | Remove inclusion of gk20a.h from clk_arb.h to fix coverity warning. Bug 200291879 Change-Id: I81e6e193f252ca022155659bfb8413cab62c9f91 Signed-off-by: Thomas Fleury <tfleury@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1539690 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Richard Zhao <rizhao@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Alex Waterman <alexw@nvidia.com> Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: channel_worker_enqueue: warn to infoPeter Daifuku2017-08-17
| | | | | | | | | | | | | | | | | | gk20a_channel_worker_enqueue currently WARNS if the call to gk20a_channel_get fails. But in fact a failure here is legitimate, in particular if the channel is in the process of being freed, and the referenceable flag changed between the time the caller grabbed a reference to the channel and the time this function tries to grab its own reference. Instead, turn the warning into an info message. Bug 200329971 Change-Id: Iab4ca6b32bb98f42f05b12b7971538db962b9795 Signed-off-by: Peter Daifuku <pdaifuku@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1540458 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Fix potential map failing for vGPUAlex Waterman2017-08-17
| | | | | | | | | | | | | | Ensure that the mapping size passed to the vGPU mapping code is page aligned. The vGPU mapping code returns -EINVAL otherwise. Change-Id: I87a90085882fa0ff538b181a55240468392c4135 Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1540423 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: Add wrapper over atomic_t and atomic64_tDebarshi Dutta2017-08-17
| | | | | | | | | | | | | | | | | - added wrapper structs nvgpu_atomic_t and nvgpu_atomic64_t over atomic_t and atomic64_t - added nvgpu_atomic_* and nvgpu_atomic64_* APIs to access the above wrappers. JIRA NVGPU-121 Change-Id: I61667bb0a84c2fc475365abb79bffb42b8b4786a Signed-off-by: Debarshi Dutta <ddutta@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1533044 Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com> Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> GVS: Gerrit_Virtual_Submit
* gpu: nvgpu: fix pmu state change for dgpu bootPeter Daifuku2017-08-16
| | | | | | | | | | | | | | | | | | | | In pmu_handle_pg_elpg_msg, when processing ELPG_MSG_DISALLOW_ACK, if current pmu state is ELPG_BOOTING and GR_POWER_GATING is not supported, make sure nvgpu_pmu_state_change updates the state_change flag, since PMU is now fully initialized. In particular, this fixes PMU boot for dgpu, which does not support GR_POWER_GATING. JIRA EVLR-1776 Change-Id: I2feb97b0fb8248e9cb7945ac3189877c21815a4a Signed-off-by: Peter Daifuku <pdaifuku@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1539102 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com> Reviewed-by: Konsta Holtta <kholtta@nvidia.com> Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Nirav Patel <nipatel@nvidia.com>
* gpu: nvgpu: Deterministic submit fixAlex Waterman2017-08-16
| | | | | | | | | | | | | | | | | | | | | | | | | | Fix some simple pointer arithmetic errors in the deterministic submit path. The lockless allocator was doing a subtraction to determine the correct offset of the element to free. However, this subtraction was using the base address and a numeric offset - not a pointer offset. Thus the difference computed was in bytes not in elements of the block size. The fix is simple: just divide by the block size. Also this modifies the debugging statement a bit so that a bit more information is printed at more useful times. Lastly, a pointer to numeric cast was fixed in the fence code. Change-Id: I514724205f1b73805b21e979481a13ac689f3482 Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1538905 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com> Reviewed-by: Aingara Paramakuru <aparamakuru@nvidia.com> Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: Konsta Holtta <kholtta@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: Disable rd_coalesce for all chipsAlex Waterman2017-08-16
| | | | | | | | | | | | Disable read coalescing for all chips. Bug 200314091 Change-Id: Iaa3f58f94369ae1edae0620083eca4594be730fd Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1518308 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Add su_rd_coalesce register fieldAlex Waterman2017-08-16
| | | | | | | | | | | | | | | Add the surface rd coalesce field in the register that controls read coalescing. Bug 200314091 Change-Id: I185ad7e6ef64ecae9369e26d22a7381611ddc693 Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1518305 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com>
* gpu: nvgpu: Update GM20B GPCPLL rev C1 parametersAlex Frid2017-08-14
| | | | | | | | | | | | | | | | - Set GM20B GPCPLL rev C1 DFS coefficients. - Updated VCO control setting - Decreased output frequency minimum to 76.8 MHz Bug 1971441 Change-Id: Ie1fa04db11d9cd76db0424acd9f24c02c6e6054a Signed-off-by: Alex Frid <afrid@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1533493 (cherry picked from commit cc495b86ad97a0a713fd46f74a4fd6d17336ff02) Reviewed-on: https://git-master.nvidia.com/r/1538242 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Reorg mm HAL initializationSunny He2017-08-14
| | | | | | | | | | | | | | | | | | | Reorganize HAL initialization to remove inheritance and construct the gpu_ops struct at compile time. This patch only covers the mm sub-module of the gpu_ops struct. Perform HAL function assignments in hal_gxxxx.c through the population of a chip-specific copy of gpu_ops. Jira NVGPU-74 Change-Id: Ieb87a62f047510e51c52e6563d8e3fd5a65b5f28 Signed-off-by: Sunny He <suhe@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1537753 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: Reorg fb HAL initializationSunny He2017-08-14
| | | | | | | | | | | | | | | | | | | Reorganize HAL initialization to remove inheritance and construct the gpu_ops struct at compile time. This patch only covers the fb sub-module of the gpu_ops struct. Perform HAL function assignments in hal_gxxxx.c through the population of a chip-specific copy of gpu_ops. Jira NVGPU-74 Change-Id: I593d00ffccc40a3721e792ef7ca259583852eb28 Signed-off-by: Sunny He <suhe@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1537746 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: remove kind map inheritanceSunny He2017-08-14
| | | | | | | | | | | | | | | | | | | | Currently, the kind map initialization inherits from earlier chips' kind map definitions and simply adds on newly supported features. This is dangerous as changes in older architectures may affect newer architectures in unpredictable ways. This patch removes inheritance between subsequent architectures' kind map initializations. Jira NVGPU-74 Change-Id: Ie5c33f613db7ba109564b4cb2312e47ee3332afc Signed-off-by: Sunny He <suhe@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1537745 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: Add struct gk20a ptr to FUSE APIsAlex Waterman2017-08-14
| | | | | | | | | | | | | | | | | | | Add a pointer to struct gk20a to the FUSE APIs. This helps QNX builds avoid any static data definitions. Also this change plumbs struct gk20a in some of the Linux clk code and fixes a few minor style nits. Change-Id: I27dfb2c4e9a352f784d6cead150460d8e9e808d3 Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1537611 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Richard Zhao <rizhao@nvidia.com> Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Sourab Gupta <sourabg@nvidia.com> Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: Fix offset units in PRAMIN codeAlex Waterman2017-08-14
| | | | | | | | | | | | | | | | | | The offset units in the nvgpu_pramin_access_batched() code changes midway through the function. In the first section it is treated as bytes but then in the while-loop iterating over the PRAMIN window and page_alloc_chunks it becomes an offset in words. This patch leaves the offset field in bytes and converts to words where needed. Change-Id: Iba964171679dfc27645238b297ed467a450b5cbc Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1537079 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: Fix length passed to VIDMEM mapAlex Waterman2017-08-14
| | | | | | | | | | | | | | | | | | The call to __set_pd_level() for vidmem allocs had the wrong length being passed in. This was a silent error since the subsequent __set_pd_level() calls overwrote the bad mappings. However this caused significantly more PDE/PTE writes than necessary since each chunk could be mapped N times where N is the number of chunks in an SGL. Change-Id: Ied7247b70825dc91b9eea1c3350f4ef370ab1a52 Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1537078 Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com> Reviewed-by: Konsta Holtta <kholtta@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: shorten PTE debug printAlex Waterman2017-08-14
| | | | | | | | | | | | | If the debug message printed to the Linux kernel is too long then the end of the message gets truncated. This causes loss of valuable PTE information. Change-Id: I05e890819ec176c6aa909b04352e8a40f1936ecf Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1537077 Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: gk20a: Fix up qos/sysfs limitsSai Gurrappadi2017-08-11
| | | | | | | | | | | | | | | | | | | | | | | | Userspace sysfs limits devfreq->{min|max}_freq and pmqos min/max limits are treated as the same priority. In case these two ranges don't intersect prefer the max_freq clamp over the min_freq clamp. This is done in order to ensure any thermal throttling isn't bypassed by a userspace sysfs write. MaxQ/MaxP configurations only care about the Fmax limit and not necessarily Fmin, so preferring max_freq over min_freq is consistent with what is required. Bug 1933216 Change-Id: Ie8685306f10dee0f8d340c4b8e0eae533dbfc6ac Signed-off-by: Sai Gurrappadi <sgurrappadi@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1535937 Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com> Reviewed-by: Aaron Huang <aaronh@nvidia.com> Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> Reviewed-by: svc-mobile-misra <svc-mobile-misra@nvidia.com> Reviewed-by: Deepak Nibade <dnibade@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* Revert "gpu: nvgpu: remove kind map inheritance"Sunny He2017-08-11
| | | | | | | | | | | | Conflicts with gv100 changes This reverts commit 83e49b39181a86d32f6350c65659b5892f8f7670. Change-Id: Ic5720fd3c214fa306645fb628a28ecf7d6d0c26d Signed-off-by: Sunny He <suhe@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1537673 Reviewed-by: Shu Zhong <shuz@nvidia.com> Tested-by: Shu Zhong <shuz@nvidia.com>
* Revert "gpu: nvgpu: Reorg fb HAL initialization"Sunny He2017-08-11
| | | | | | | | | | | | Conflicts with gv100 changes This reverts commit 63b74d4b768e0c96367d4983fdd8f1db1d317d01. Change-Id: I5e6a1c93ff613daaa100dee436f4941af74f0ac4 Signed-off-by: Sunny He <suhe@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1537671 Reviewed-by: Shu Zhong <shuz@nvidia.com> Tested-by: Shu Zhong <shuz@nvidia.com>
* Revert "gpu: nvgpu: Reorg mm HAL initialization"Sunny He2017-08-11
| | | | | | | | | | | | Conflicts with gv100 changes This reverts commit 8d63cd3995d4a650b478ad69d7e29ed2b1b2d927. Change-Id: Ie2f88d281b2b87a9a794d79164a61c4d883626b7 Signed-off-by: Sunny He <suhe@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1537668 Reviewed-by: Shu Zhong <shuz@nvidia.com> Tested-by: Shu Zhong <shuz@nvidia.com>
* gpu: nvgpu: Reorg mm HAL initializationSunny He2017-08-11
| | | | | | | | | | | | | | | | | Reorganize HAL initialization to remove inheritance and construct the gpu_ops struct at compile time. This patch only covers the mm sub-module of the gpu_ops struct. Perform HAL function assignments in hal_gxxxx.c through the population of a chip-specific copy of gpu_ops. Jira NVGPU-74 Change-Id: I289284e6e528fc7951c959c8765ccf9349eec33b Signed-off-by: Sunny He <suhe@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1533351 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Reorg fb HAL initializationSunny He2017-08-11
| | | | | | | | | | | | | | | | | | | Reorganize HAL initialization to remove inheritance and construct the gpu_ops struct at compile time. This patch only covers the fb sub-module of the gpu_ops struct. Perform HAL function assignments in hal_gxxxx.c through the population of a chip-specific copy of gpu_ops. Jira NVGPU-74 Change-Id: Ib746798b849810401c12abf5e9cce42d827c6fb1 Signed-off-by: Sunny He <suhe@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1533350 Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com> Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: remove kind map inheritanceSunny He2017-08-11
| | | | | | | | | | | | | | | | | | | | Currently, the kind map initialization inherits from earlier chips' kind map definitions and simply adds on newly supported features. This is dangerous as changes in older architectures may affect newer architectures in unpredictable ways. This patch removes inheritance between subsequent architectures' kind map initializations. Jira NVGPU-74 Change-Id: I21951c0920f69ced73088cb6670ca1894159c7b9 Signed-off-by: Sunny He <suhe@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1533349 Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com> Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: PG503 supportDavid Nieto2017-08-11
| | | | | | | | | | | | Adds basic PG503 support allowing devinit to complete. JIRA: EVLR-1693 Change-Id: Ice8a9ba18c8bba11f6bc174ba2c2d8802a738706 Signed-off-by: David Nieto <dmartineznie@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1532746 Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Tested-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: remove railgate lock from gm20b_tegra_postscaleDeepak Nibade2017-08-11
| | | | | | | | | | | | | | | | | | | | | | | | In gm20b_tegra_postscale(), we use platform->railgate_lock to check if GPU is railgated or not But platform->railgate_lock was introduced only to prevent unrailgating in midst of gk20a_do_idle() sequence This lock is not the right way to check railgate status since it is still possible to railgate GPU with this lock being held Hence remove acquire/release of platform->railgate_lock from gm20b_tegra_postscale() Bug 1962265 Change-Id: I6208063de3fa77ed71e8fb0c011367fb66151193 Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1536573 (cherry picked from commit 68bce66be338e48f4921f645b10b3fa5994fe1d4) Reviewed-on: https://git-master.nvidia.com/r/1537297 Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Bharat Nihalani <bnihalani@nvidia.com>
* gpu: nvgpu: fix pbdma_id searching when init engine infoRichard Zhao2017-08-11
| | | | | | | | | | | | | | The searching loop should break once find a match. Jira VFND-3797 Change-Id: I43ef553a3e90afb00ee9a4df7d269b7c6616b18e Signed-off-by: Richard Zhao <rizhao@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1535304 Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com> Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: fix debugfs to disable big pagesThomas Fleury2017-08-07
| | | | | | | | | | | | | | | | | | | | | After setting 'Y' in disable_bigpage, in native SMMU case, we could still see 64K GMMU pages beeing used. Fixed the following: - enforce disable_bigpage in nvgpu_vm_map - update GPU characteristics so that new clients know whether or not big pages are enabled. For instance this may affect how CUDA requests memory mapping. JIRA EVLR-1694 Change-Id: I62841096add3bd798c5c11090054f82c8a2be832 Signed-off-by: Thomas Fleury <tfleury@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1532429 Reviewed-by: Richard Zhao <rizhao@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com> Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: Increase small page apertureAlex Waterman2017-08-07
| | | | | | | | | | | | | Increase the small page aperture to 56GB to facilitate easier fixed address mapping for userspace (primarily CUDA). Bug 200320732 Change-Id: I1f0aaa4f28c8a294cc880b35f26942b562396b48 Signed-off-by: Sunny He <suhe@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1502432 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Remove mm.get_iova_addrAlex Waterman2017-08-04
| | | | | | | | | | | | | | | | | | | | | | Remove the mm.get_iova_addr() HAL and replace it with a new HAL called mm.gpu_phys_addr(). This new HAL provides the real phys address that should be passed to the GPU from a physical address obtained from a scatter list. It also provides a mechanism by which the HAL code can add extra bits to a GPU physical address based on the attributes passed in. This is necessary during GMMU page table programming. Also remove the flags argument from the various address functions. This flag was used for adding an IO coherence bit to the GPU physical address which is not supported. JIRA NVGPU-30 Change-Id: I69af5b1c6bd905c4077c26c098fac101c6b41a33 Signed-off-by: Alex Waterman <alexw@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1530864 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Pass struct gk20a to gk20a_wait_for_idle()Terje Bergstrom2017-08-04
| | | | | | | | | | | | Pass struct gk20a pointer instead of struct device to gk20a_wait_for_idle(). The code is not Linux specific and does not need pointer to struct device. Change-Id: I2cafd6c7db019c9de76b6e68a1ae73f0b4cea37d Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1533173 GVS: Gerrit_Virtual_Submit Reviewed-by: Alex Waterman <alexw@nvidia.com>
* gpu: nvgpu: Remove unnecessary use of ACCESS_ONCE()Terje Bergstrom2017-08-04
| | | | | | | | | | | | | | | ACCESS_ONCE() is used for making sure that in a given place of code access a variable exactly once. It prevents compiler rearranging the read from happening earlier. Remove its use from cases where rearranging of the read does not create problems. Change-Id: I340f375e8fecc31f3a3fab543256069cb4c682dc Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1531649 GVS: Gerrit_Virtual_Submit Reviewed-by: Alex Waterman <alexw@nvidia.com>
* gpu: nvgpu: Pass gk20a to XVE initTerje Bergstrom2017-08-04
| | | | | | | | | | | Pass gk20a pointer instead of struct device pointer to XVE init. Change-Id: I5f5efc1d09430a45a49321830b1d7f17c7a57e2f Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1528264 Reviewed-by: Automatic_Commit_Validation_User Reviewed-by: Alex Waterman <alexw@nvidia.com> GVS: Gerrit_Virtual_Submit
* gpu: nvgpu: Move debugfs fields to os_linuxTerje Bergstrom2017-08-04
| | | | | | | | | | | | | Move all Linux specific debugfs dentry fields to struct nvgpu_os_linux. JIRA NVGPU-62 Change-Id: I615620005f5d042943dd2e478c1629bcc912c550 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1528263 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Make LTC disabling common codeTerje Bergstrom2017-08-04
| | | | | | | | | | | | | | | | | | Refactor the sync_debugfs LTC HAL op so that the logic to enable or disable LTC goes to common code nvgpu_ltc_sync_enabled() and the LTC HAL set_enabled only performs the hardware register access. Create a new common function nvgpu_init_ltc_support() to initialize the LTC software variable, and move hardware initialization of LTC to be called from it. JIRA NVGPU-62 Change-Id: Ib1cf4f5b83ca3dac08407464ed56a732e0a33923 Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1528262 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: fix warnings for GPUs with real vidmemPeter Daifuku2017-08-03
| | | | | | | | | | | | | | | | | | | | | Fix kernel warnings for GPUs with real vidmem: - dma.c: in nvgpu_dma_alloc_flags, ignore incoming flags when using vidmem, since anything but NVGPU_DMA_NO_KERNEL_MAPPING will end up generating kernel warnings, and the vidmem mapping functions ignore the other flags anyway. - gmmu.c: in __nvgpu_gmmu_update_page_table, use appropriate function for memory type to retrieve physical address Bug 1967748 Change-Id: I6fc01fd5f2c5cd7b81cba70ab59cc3c8fe4cda19 Signed-off-by: Peter Daifuku <pdaifuku@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1530877 Reviewed-by: Alex Waterman <alexw@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
generated by cgit v1.2.2 (git 2.25.0) at 2024-07-16 22:18:21 -0400