On 1/10/2022 1:17 PM, Tao Zhou wrote:
PMFW only returns 0 on master die and sends NACK back on other dies for
the message.
Signed-off-by: Tao Zhou <[email protected]>
---
.../gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c | 16 ++++++++++++----
1 file changed, 12 insertions(+), 4 deletions(-)
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c
b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c
index 261892977654..121b0ab5823f 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/aldebaran_ppt.c
@@ -1625,10 +1625,18 @@ static int aldebaran_set_df_cstate(struct smu_context
*smu,
static int aldebaran_allow_xgmi_power_down(struct smu_context *smu, bool en)
{
- return smu_cmn_send_smc_msg_with_param(smu,
- SMU_MSG_GmiPwrDnControl,
- en ? 0 : 1,
- NULL);
+ int ret;
+
+ ret = smu_cmn_send_smc_msg_with_param(smu,
+ SMU_MSG_GmiPwrDnControl,
+ en ? 0 : 1,
+ NULL);
+ /* The message only works on master die and NACK will be sent
+ back for other dies, ingore CMD failure for the message */
+ if (ret == -EIO)
+ ret = 0;
+
Could you confirm if this is hive master or primary die of each device?
If it's the latter, use aldebaran_is_primary(). For the former we can
add support similar to is_primary().
Thanks,
Lijo
+ return ret;
}
static const struct throttling_logging_label {
