summaryrefslogtreecommitdiffstats
path: root/drivers/gpu/nvgpu/gp106/clk_arb_gp106.c
Commit message (Collapse)AuthorAge
* gpu: nvgpu: fix clk_arb completion file private data access raceSagar Kamble2021-08-03
| | | | | | | | | | | | | | | | | | | | | | | | | | clk_arb completion file descriptor can get closed immediately after poll finishes in the work item gp10b_clk_arb_run_arbiter_cb. In that case, the refcount for nvgpu_clk_dev can become zero in the work item and can lead to invalid access while removing nvgpu_clk_dev from the lists. Remove nvgpu_clk_dev from the list before dropping the reference to it. Also, delete the nvgpu_clk_dev in completion file release handler within the session and requests spinlocks to avoid race with gp10b_clk_arb_run_arbiter_cb using it. bug 200757277 Change-Id: I054eee547f2a6fa633d7ef55df216ec36647a826 Signed-off-by: Sagar Kamble <skamble@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/c/linux-nvgpu/+/2569522 Tested-by: mobile promotions <svcmobile_promotions@nvidia.com> Reviewed-by: Debarshi Dutta <ddutta@nvidia.com> Reviewed-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-by: Bibek Basu <bbasu@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: move header location of gk20a.hDebarshi Dutta2018-09-25
| | | | | | | | | | | | | | | | change the path of gk20a.h to <nvgpu/gk20a.h> for files in the following directories. gp106/ gv100/ Jira NVGPU-597 Change-Id: I098c12c34cdce764e6fca4ea1d16c5f8dee81026 Signed-off-by: Debarshi Dutta <ddutta@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1836707 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: add igpu support for clk_arbiter.Debarshi Dutta2018-09-20
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch constructs clk_arbiter specific code for gp10b as well as gv11b and does the necessary plumbing in the clk_arbiter code. The changes made are as follows. 1) Constructed clk_arb_gp10b.* files which add support for clk_arb related HALS including the nvgpu_clk_arb_init and nvgpu_clk_arb_cb. This doesn't have support for debugfs nor the VFUpdateEvent yet and consequently no support for arb->notifications. 2) Added gpcclk specific variables corresponding to every gpc2clk in a given clk_arb related struct. 3) Linux specific support_clk_freq_controller is assigned true in platform_gp10b.c and platform_gv11b.c files. 4) Incremented the clk_arb_worker.put atomic variable during worker_deinit so as to allow the worker thread to be stopped. 5) Added the flag clk_arb_events_supported as part of struct nvgpu_clk_arb. This flag is used to selectively account for the extra refcounting present in OS specific code i.e. nvgpu_clk_arb_commit_request_fd. For igpus, the extra refcount is reduced during nvgpu_clk_arb_release_completion_dev. Bug 2061372 Change-Id: Id00acb106db2b46e55aa0324034a16a73723c078 Signed-off-by: Debarshi Dutta <ddutta@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1774281 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: protect clk_arb init with mutex.Debarshi Dutta2018-09-04
| | | | | | | | | | | | | | | | | | | g->clk_arb is currently initialized as a part of gk20a_finalize_poweron(). Any subsequent call to gk20a_finalize_poweron reinitializes the clk_arb and leading to memory leaks. This is resolved by protecting the g->clk_arb initialization with a mutex clk_arb_enable_lock in struct gk20a. We skip initializing the g->clk_arb if its not NULL. Bug 2061372 Change-Id: I59158e0a5e4c827fdbd6d9ea2d04c78d0986347a Signed-off-by: Debarshi Dutta <ddutta@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1811650 Reviewed-by: Deepak Nibade <dnibade@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: move gp106 specific clk_arbiter code into HALDebarshi Dutta2018-09-04
| | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently, clock arbiter code is extensively using dgpu specific implementation. This patch restructures the clk_arbiter code and moves gp106 specific code into HAL. Following changes are made in this patch 1) clk_domain_get_f_points is now invoked via HAL for gp106 i.e. g->ops.clk.clk_domain_get_f_points. 2) moved nvgpu_clk_arb_change_vf_point and other related static functions to clk_arb_gp106.c. 3) Instead of only checking if get_arbiter_clk_domain is empty, a check for support_clk_freq_controller is also added. This is to enable the clk_arbiter based on support from both the OS and the chips. Bug 2061372 Change-Id: I65b0a4e02145a86fbbfb420ed591b1fa3c86f6dc Signed-off-by: Debarshi Dutta <ddutta@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1774279 Reviewed-by: svc-misra-checker <svc-misra-checker@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: gp106: Fix MISRA 15.6 violationsSrirangan2018-08-29
| | | | | | | | | | | | | | | | | | MISRA Rule-15.6 requires that all if-else blocks be enclosed in braces, including single statement blocks. Fix errors due to single statement if blocks without braces, introducing the braces. JIRA NVGPU-671 Change-Id: I8493274995ed8de526902dd0ca0808b2972e28aa Signed-off-by: Srirangan <smadhavan@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1796806 Reviewed-by: svc-misra-checker <svc-misra-checker@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: make sure gpc2clk is greater than DVCO minThomas Fleury2018-07-01
| | | | | | | | | | | | | | | | | For V/f point with gpc2clk exactly equal to DVCO min, we observe lower than expected effective clock. Make sure min frequency for gpc2clk is greater than DVCO min. Bug 200412996 Change-Id: I85c33852c56c7a642aa5c85987d2da4147e73c22 Signed-off-by: Thomas Fleury <tfleury@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1764923 Reviewed-by: svc-mobile-coverity <svc-mobile-coverity@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com> Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: move clk_arb to linux specificMahantesh Kumbar2017-10-25
| | | | | | | | | | | | | | | | | - Clock arbiter has lot of linux dependent code so moved clk_arb.c to common/linux folder & clk_arb.h to include/nvgpu/clk_arb.h, this move helps to unblock QNX. - QNX must implement functions present under clk_arb.h as needed. JIRA NVGPU-33 Change-Id: I38369fafda9c2cb9ba2175b3e530e40d0c746601 Signed-off-by: Mahantesh Kumbar <mkumbar@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1582473 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Change license for common files to MITTerje Bergstrom2017-09-26
| | | | | | | | | | | | Change license of OS independent source code files to MIT. JIRA NVGPU-218 Change-Id: I1474065f4b552112786974a16cdf076c5179540e Signed-off-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1565880 Reviewed-by: mobile promotions <svcmobile_promotions@nvidia.com> Tested-by: mobile promotions <svcmobile_promotions@nvidia.com>
* gpu: nvgpu: Reorg clk HAL initializationSunny He2017-08-02
| | | | | | | | | | | | | | | | | | | Reorganize HAL initialization to remove inheritance and construct the gpu_ops struct at compile time. This patch only covers the clk and clk_arb 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: I553353df836b187b8eac61e16b63080b570c96b8 Signed-off-by: Sunny He <suhe@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1511076 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: HAL to get current pstateMahantesh Kumbar2017-01-17
| | | | | | | | | | | | | | | | | | - Added HAL support to get current pstate from clk_arb Note - This function is inherently unsafe to call while arbiter is running arbiter must be blocked before calling this function JIRA DNVGPU-165 Change-Id: I4e9f5eba7739280bddd9ee661fd314288c129516 Signed-off-by: Mahantesh Kumbar <mkumbar@nvidia.com> Reviewed-on: http://git-master/r/1286378 Reviewed-by: svccoveritychecker <svccoveritychecker@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* gpu: nvgpu: cap minimum gpc clocks to HW limitsDavid Nieto2016-12-27
| | | | | | | | | | | | | | | JIRA: DNVGPU-180 Change-Id: I1928e77cea4ac87bf2ba2b6b7b2f2942dfb97de9 Signed-off-by: David Nieto <dmartineznie@nvidia.com> Reviewed-on: http://git-master/r/1251493 (cherry picked from commit 7b8a105652a3169d9ec0cb7ce52c3b92e42ca310) Reviewed-on: http://git-master/r/1274545 Reviewed-by: Automatic_Commit_Validation_User Tested-by: Thomas Fleury <tfleury@nvidia.com> Reviewed-by: Vijayakumar Subbu <vsubbu@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
* nvgpu: gpu: arbiter for vf switch managementDavid Nieto2016-12-27
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | JIRA DNVGPU-143 The arbiter is charged with selecting the proper frequencies when multiple applications submit simultaneously clock change requests On the current implementation, the arbiter guarantees that the selected frequency will be always higher or equal to the request, as long as the request is in range. The current code is not yet realtime friendly, as requests are not pre-allocated. Summary of changes: (1) pstate/vf switch no longer selects boot frequency (2) changed mclk code change to accept input freq (3) added arbiter (4) now a single session can submit concurrent requests the last request is the one that applies for that session (5) modified locking mechanism to reduce lock contention (6) Added callback to notify the arbiter that the VF table has changed and is no longer valid (PMU/Thermals must call this when VF table is invalid) (7) changed internal API to work with MHz (8) added debugfs for stats Change-Id: I6a7b05c9447761e8536f84ef86b5ab0793164d63 Signed-off-by: David Nieto <dmartineznie@nvidia.com> Reviewed-on: http://git-master/r/1239461 Reviewed-by: Thomas Fleury <tfleury@nvidia.com> GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> Reviewed-on: http://git-master/r/1267120 Reviewed-by: Automatic_Commit_Validation_User
* gpu: nvgpu: clk requests completion and event fdsThomas Fleury2016-12-27
| | | | | | | | | | | | | | | | | | | Install one completion fd per SET request. Notifications on dedicated event fd. Changed frequencies unit to Hz from MHz. Remove sequence numbers from dummy arbiter. Added effective clock type (query frequency from counters). Jira DNVGPU-125 Change-Id: Id5445c6ae1d6bf06f7f59c82ff6c5d3b34e26483 Signed-off-by: Thomas Fleury <tfleury@nvidia.com> Reviewed-on: http://git-master/r/1230239 Reviewed-by: Automatic_Commit_Validation_User GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com> (cherry picked from commit d17083f4ceb69725c661678607a3e43148d38560) Reviewed-on: http://git-master/r/1243106
* gpu: nvgpu: clk arbiter skeletonThomas Fleury2016-12-27
Add clock arbiter skeleton with support of clock sessions, notifications on clock changes, request numbering, and asynchronous handling of clock requests. Provides minimum behaviour to allow unit tests implementation. Actual arbitration and clock settings will be done separately. For now, dummy arbiter keeps last requested target mhz. Actual arbiter may move to a lockless implementation. Jira DNVGPU-125 Change-Id: I6a8e443fb0d15dc5f1993e7260256d71acddd106 Signed-off-by: Thomas Fleury <tfleury@nvidia.com> Reviewed-on: http://git-master/r/1223476 (cherry picked from commit cb130825d84e4124d273bd443e2b62d493377461) Reviewed-on: http://git-master/r/1243105 GVS: Gerrit_Virtual_Submit Reviewed-by: Terje Bergstrom <tbergstrom@nvidia.com>
generated by cgit v1.2.2 (git 2.25.0) at 2024-07-04 22:32:11 -0400