Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [High] drm/msm: hardware corruption due to zero-initialized LM3 base address
--
commit f01ed8464ded4b4b827835f9551a26ed60e5dded
Author: Muzaffer Kadir [off-list ref]
drm: msm: mdp5: Add MDP5 configuration for MSM8952
This commit adds the mdp5_cfg_hw entry for MDP5 version v1.12 found on
the MSM8952.
quoted hunk ↗ jump to hunk
diff --git a/drivers/gpu/drm/msm/disp/mdp5/mdp5_cfg.c b/drivers/gpu/drm/msm/disp/mdp5/mdp5_cfg.c
index 7c91fc1915f39..d34c94d6f7e2e 100644
--- a/drivers/gpu/drm/msm/disp/mdp5/mdp5_cfg.c
+++ b/drivers/gpu/drm/msm/disp/mdp5/mdp5_cfg.c
@@ -744,6 +744,86 @@ static const struct mdp5_cfg_hw msm8x76_config = {
.max_clk = 360000000,
};
+static const struct mdp5_cfg_hw msm8952_config = {
+ .name = "msm8952",
[ ... ]
quoted hunk ↗ jump to hunk
+ .lm = {
+ .count = 2,
+ .base = { 0x44000, 0x47000 },
+ .instances = {
+ { .id = 0, .pp = 0, .dspp = 0,
+ .caps = MDP_LM_CAP_DISPLAY, },
+ { .id = 3, .pp = -1, .dspp = -1,
+ .caps = MDP_LM_CAP_WB },
+ },
[Severity: High]
Will this lead to hardware register corruption when LM3 is brought up?
In msm8952_config, the .lm.base array is initialized sequentially as
{ 0x44000, 0x47000 }, which assigns base[0] = 0x44000 and base[1] = 0x47000.
This leaves base[3] implicitly zero-initialized.
However, the second LM instance is defined with .id = 3 (for LM3). When the
driver configures this mixer, it uses lm->id to access the LM registers.
Because base[3] is 0, the driver will mistakenly perform writes to offset 0x0
of the MDP block (which corresponds to top-level registers).
Should designated initializers be used for the base array here, such as
{ [0] = 0x44000, [3] = 0x47000 }?
quoted hunk ↗ jump to hunk
+ .nb_stages = 8,
+ .max_width = 2048,
+ .max_height = 0xFFFF,
+ },
--
Sashiko AI review · https://sashiko.dev/#/patchset/20260916-msm8952-initial-support-v2-0-798c2602dde1@mainlining.org?part=13