summaryrefslogtreecommitdiffstats
path: root/drivers/gpu/nvgpu/gv100
diff options
context:
space:
mode:
authorDeepak Nibade <dnibade@nvidia.com>2018-03-16 08:25:18 -0400
committermobile promotions <svcmobile_promotions@nvidia.com>2018-03-21 09:04:38 -0400
commit77b806fe7e68e853676f7c4bad14349aba1affa5 (patch)
tree2b0933a730d8b7f5144e0d51dfc89e3cb19a28f0 /drivers/gpu/nvgpu/gv100
parent66751bc05d7a1efca3668d59a2820e3e92985f91 (diff)
gpu: nvgpu: gv100: fix PMA list alignment in ctxsw buffer
GV100 ucode is changed so that it expects LIST_nv_perf_pma_ctx_reg list in ctxsw buffer to be 256 byte aligned but same change is not applied to other chip ucodes ADD new HAL (*add_ctxsw_reg_perf_pma) to configure PMA register list and define a common HAL gr_gk20a_add_ctxsw_reg_perf_pma() for all other chips except GV100 Define a separate HAL for GV100 gr_gv100_add_ctxsw_reg_perf_pma() and fix the required alignment in this function Bug 1998067 Change-Id: Ie172fe90e2cdbac2509f2ece953cd8552e66fc56 Signed-off-by: Deepak Nibade <dnibade@nvidia.com> Reviewed-on: https://git-master.nvidia.com/r/1676655 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>
Diffstat (limited to 'drivers/gpu/nvgpu/gv100')
-rw-r--r--drivers/gpu/nvgpu/gv100/gr_gv100.c10
-rw-r--r--drivers/gpu/nvgpu/gv100/gr_gv100.h4
-rw-r--r--drivers/gpu/nvgpu/gv100/hal_gv100.c1
3 files changed, 15 insertions, 0 deletions
diff --git a/drivers/gpu/nvgpu/gv100/gr_gv100.c b/drivers/gpu/nvgpu/gv100/gr_gv100.c
index 3aed7a19..c6273733 100644
--- a/drivers/gpu/nvgpu/gv100/gr_gv100.c
+++ b/drivers/gpu/nvgpu/gv100/gr_gv100.c
@@ -373,3 +373,13 @@ int gr_gv100_add_ctxsw_reg_pm_fbpa(struct gk20a *g,
373 *offset = off; 373 *offset = off;
374 return 0; 374 return 0;
375} 375}
376
377int gr_gv100_add_ctxsw_reg_perf_pma(struct ctxsw_buf_offset_map_entry *map,
378 struct aiv_list_gk20a *regs,
379 u32 *count, u32 *offset,
380 u32 max_cnt, u32 base, u32 mask)
381{
382 *offset = ALIGN(*offset, 256);
383 return gr_gk20a_add_ctxsw_reg_perf_pma(map, regs,
384 count, offset, max_cnt, base, mask);
385}
diff --git a/drivers/gpu/nvgpu/gv100/gr_gv100.h b/drivers/gpu/nvgpu/gv100/gr_gv100.h
index e1174686..7b107db2 100644
--- a/drivers/gpu/nvgpu/gv100/gr_gv100.h
+++ b/drivers/gpu/nvgpu/gv100/gr_gv100.h
@@ -39,4 +39,8 @@ int gr_gv100_add_ctxsw_reg_pm_fbpa(struct gk20a *g,
39 u32 *count, u32 *offset, 39 u32 *count, u32 *offset,
40 u32 max_cnt, u32 base, 40 u32 max_cnt, u32 base,
41 u32 num_fbpas, u32 stride, u32 mask); 41 u32 num_fbpas, u32 stride, u32 mask);
42int gr_gv100_add_ctxsw_reg_perf_pma(struct ctxsw_buf_offset_map_entry *map,
43 struct aiv_list_gk20a *regs,
44 u32 *count, u32 *offset,
45 u32 max_cnt, u32 base, u32 mask);
42#endif 46#endif
diff --git a/drivers/gpu/nvgpu/gv100/hal_gv100.c b/drivers/gpu/nvgpu/gv100/hal_gv100.c
index 08fc7c34..cfac8e0e 100644
--- a/drivers/gpu/nvgpu/gv100/hal_gv100.c
+++ b/drivers/gpu/nvgpu/gv100/hal_gv100.c
@@ -430,6 +430,7 @@ static const struct gpu_ops gv100_ops = {
430 .handle_notify_pending = gk20a_gr_handle_notify_pending, 430 .handle_notify_pending = gk20a_gr_handle_notify_pending,
431 .handle_semaphore_pending = gk20a_gr_handle_semaphore_pending, 431 .handle_semaphore_pending = gk20a_gr_handle_semaphore_pending,
432 .add_ctxsw_reg_pm_fbpa = gr_gv100_add_ctxsw_reg_pm_fbpa, 432 .add_ctxsw_reg_pm_fbpa = gr_gv100_add_ctxsw_reg_pm_fbpa,
433 .add_ctxsw_reg_perf_pma = gr_gv100_add_ctxsw_reg_perf_pma,
433 }, 434 },
434 .fb = { 435 .fb = {
435 .reset = gv100_fb_reset, 436 .reset = gv100_fb_reset,