On 15/10/2024 18:50, Alex Deucher wrote:
On Tue, Oct 15, 2024 at 12:33 PM Shashank Sharma <shashank.sharma@xxxxxxx> wrote:This patch sets MES HQD mask to setup GFX queues for MES and KIQ operations. We are using one queue each for KIQ operations, and setting rest of the queues for MES scheduling. This also fixes a regression for missing Navi 4x MES mask from usermode queue series. V2: Rebase on staging, accommodate existing changes on Navi 3X Fixes: 1d316a52498f ("drm/amdgpu: fix MES GFX mask") Cc: Alex Deucher <alexander.deucher@xxxxxxx> Cc: Christian König <Christian.Koenig@xxxxxxx> Cc: Arvind Yadav <arvind.yadav@xxxxxxx> Signed-off-by: Shashank Sharma <shashank.sharma@xxxxxxx> --- drivers/gpu/drm/amd/amdgpu/mes_v11_0.c | 20 +++++++++++++------- drivers/gpu/drm/amd/amdgpu/mes_v12_0.c | 15 ++++++++++++--- 2 files changed, 25 insertions(+), 10 deletions(-) diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c index 03bf865fbdd4..1f8ce64a5bff 100644 --- a/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mes_v11_0.c @@ -654,6 +654,18 @@ static int mes_v11_0_misc_op(struct amdgpu_mes *mes, offsetof(union MESAPI__MISC, api_status)); } +static void mes_v11_0_set_gfx_hqd_mask(union MESAPI_SET_HW_RESOURCES *pkt) +{ + /* + * GFX pipe 0 queue 0 is being used by KIQMinor nit, pipe 0 queue 0 is the kernel gfx queue. KIQ is in the MEC and managed by MES on gfx11+.
Ah, my bad, noted.
+ * Set GFX pipe 0 queue 1 for MES scheduling + * mask = 10b + * GFX pipe 1 can't be used for MES due to HW limitation. + */ + pkt->gfx_hqd_mask[0] = 0x2; + pkt->gfx_hqd_mask[1] = 0; +} + static int mes_v11_0_set_hw_resources(struct amdgpu_mes *mes) { int i; @@ -678,13 +690,7 @@ static int mes_v11_0_set_hw_resources(struct amdgpu_mes *mes) mes_set_hw_res_pkt.compute_hqd_mask[i] = mes->compute_hqd_mask[i]; - /* - * GFX pipe 0 queue 0 is being used by kernel - * Set GFX pipe 0 queue 1 for MES scheduling - * GFX pipe 1 can't be used for MES due to HW limitation. - */ - mes_set_hw_res_pkt.gfx_hqd_mask[0] = 0x2; - mes_set_hw_res_pkt.gfx_hqd_mask[1] = 0; + mes_v11_0_set_gfx_hqd_mask(&mes_set_hw_res_pkt); for (i = 0; i < MAX_SDMA_PIPES; i++) mes_set_hw_res_pkt.sdma_hqd_mask[i] = mes->sdma_hqd_mask[i]; diff --git a/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c b/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c index 3daa8862e622..78aad9c4ddbc 100644 --- a/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c +++ b/drivers/gpu/drm/amd/amdgpu/mes_v12_0.c @@ -557,6 +557,17 @@ static int mes_v12_0_set_hw_resources_1(struct amdgpu_mes *mes, int pipe) offsetof(union MESAPI_SET_HW_RESOURCES_1, api_status)); } +static void mes_v12_0_set_gfx_hqd_mask(union MESAPI_SET_HW_RESOURCES *pkt) +{ + /* + * GFX V12 has only one GFX pipe, but 8 queues in it. + * GFX pipe 0 queue 0 is being used by KIQ.Same comment here.
Noted - Shashank
With those addressed, the patch is: Reviewed-by: Alex Deucher <alexander.deucher@xxxxxxx>+ * Set GFX pipe 0 queue 1-7 for MES scheduling + * mask = 1111 1110b + */ + pkt->gfx_hqd_mask[0] = 0xFE; +} + static int mes_v12_0_set_hw_resources(struct amdgpu_mes *mes, int pipe) { int i; @@ -579,9 +590,7 @@ static int mes_v12_0_set_hw_resources(struct amdgpu_mes *mes, int pipe) mes_set_hw_res_pkt.compute_hqd_mask[i] = mes->compute_hqd_mask[i]; - for (i = 0; i < MAX_GFX_PIPES; i++) - mes_set_hw_res_pkt.gfx_hqd_mask[i] = - mes->gfx_hqd_mask[i]; + mes_v12_0_set_gfx_hqd_mask(&mes_set_hw_res_pkt); for (i = 0; i < MAX_SDMA_PIPES; i++) mes_set_hw_res_pkt.sdma_hqd_mask[i] = -- 2.46.2
