Gh0st
/
linux
mirror of https://github.com/torvalds/linux.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

drm/amdgpu: fix the indexing issue during rlcg access ctrl init 

In case that the GET_INST() is used for looping, only loops for the
times of actual num of xcc, otherwise GET_INST() will return the invalid
index, a.k.a -1

And also remove the redundant mask checking in case of GET_INST()

Signed-off-by: Shiwu Zhang <shiwu.zhang@amd.com>
Reviewed-by: Le Ma <le.ma@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This commit is contained in:
Shiwu Zhang 2023-07-20 15:25:15 +08:00 committed by Alex Deucher
parent 818c158fd4
commit 9bc12db4e2
1 changed files with 3 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c
View File

@ -1075,12 +1075,11 @@ static void gfx_v9_4_3_xcc_unset_safe_mode(struct amdgpu_device *adev,
static void gfx_v9_4_3_init_rlcg_reg_access_ctrl(struct amdgpu_device *adev)
{
int xcc_id;
int xcc_id, num_xcc;
struct amdgpu_rlcg_reg_access_ctrl *reg_access_ctrl;
for (xcc_id = 0; xcc_id < AMDGPU_MAX_RLC_INSTANCES; xcc_id++) {
if (((1 << xcc_id) & adev->gfx.xcc_mask) == 0)
continue;
num_xcc = NUM_XCC(adev->gfx.xcc_mask);
for (xcc_id = 0; xcc_id < num_xcc; xcc_id++) {
reg_access_ctrl = &adev->gfx.rlc.reg_access_ctrl[GET_INST(GC, xcc_id)];
reg_access_ctrl->scratch_reg0 = SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regSCRATCH_REG0);
reg_access_ctrl->scratch_reg1 = SOC15_REG_OFFSET(GC, GET_INST(GC, xcc_id), regSCRATCH_REG1);