[Public]
>> The message is really confusing during debug, and I remember getting stuck
>> with pmfw team , to explain them its not error just a message. As Lijo
>> pointed out we have not updated interface version for 13.0.6 and 13.0.12
>> from long time because we are not using it. I only use this message to check
>> pmfw version which is anyways available in sysfs interface.
This information was originally intended for developer use and problem
identification; this change was made specifically for 13.0.6/13.0.12 but did
not provide any benefit.
Best Regards,
Kevin
-----Original Message-----
From: Kamal, Asad <asad.ka...@amd.com>
Sent: Friday, September 19, 2025 3:09 PM
To: Lazar, Lijo <lijo.la...@amd.com>; Wang, Yang(Kevin)
<kevinyang.w...@amd.com>; amd-gfx@lists.freedesktop.org
Cc: Zhang, Hawking <hawking.zh...@amd.com>; Deucher, Alexander
<alexander.deuc...@amd.com>
Subject: RE: [PATCH] drm/amd/pm: Avoid interface mismatch messaging
[Public]
The message is really confusing during debug, and I remember getting stuck with
pmfw team , to explain them its not error just a message. As Lijo pointed out
we have not updated interface version for 13.0.6 and 13.0.12 from long time
because we are not using it. I only use this message to check pmfw version
which is anyways available in sysfs interface.
So, from my side
Reviewed-by: Asad Kamal <asad.ka...@amd.com>
Thanks & Regards
Asad
-----Original Message-----
From: Lazar, Lijo <lijo.la...@amd.com>
Sent: Friday, September 19, 2025 12:24 PM
To: Wang, Yang(Kevin) <kevinyang.w...@amd.com>; amd-gfx@lists.freedesktop.org
Cc: Zhang, Hawking <hawking.zh...@amd.com>; Deucher, Alexander
<alexander.deuc...@amd.com>; Kamal, Asad <asad.ka...@amd.com>
Subject: RE: [PATCH] drm/amd/pm: Avoid interface mismatch messaging
[Public]
First, it's not the developer who is using the system. Any mismatch information
in a dmesg log is always an alarm to the user who is not aware of the
implementation details.
This mismatch is bound to happen when PMFW is not loaded along with the driver.
In such cases, the mechanism to ensure compatibility is only PMFW version
checks and hence some IP implementations follow the same. If there is an issue,
the identification mechanism is the stack information which is the driver
version along with the all the firmware versions. The interface version is not
the first thing that is checked; in fact, I don't remember checking that at
all. That is why this code is keeping the same interface version for SMU
v13.0.12 and SMU v13.0.6 and not getting noticed so far. So instead of sticking
with it, it is rather good to clean up such messages or downgrade to debug
(which is again useless because as mentioned earlier it's not looked at all).
Thanks,
Lijo
-----Original Message-----
From: Wang, Yang(Kevin) <kevinyang.w...@amd.com>
Sent: Friday, September 19, 2025 12:14 PM
To: Lazar, Lijo <lijo.la...@amd.com>; amd-gfx@lists.freedesktop.org
Cc: Zhang, Hawking <hawking.zh...@amd.com>; Deucher, Alexander
<alexander.deuc...@amd.com>; Kamal, Asad <asad.ka...@amd.com>
Subject: RE: [PATCH] drm/amd/pm: Avoid interface mismatch messaging
[Public]
>> " is not used": Actually, this is used to display information to users and
>> developers, not the driver itself.
This "driver_if_version" is used to show what driver if source file version is
using on current driver, this information is used to *developers* and *user*
from dmesg.log, not driver *itself*.
>> This interface version is not used anywhere. Driver ensures backward
>> compatibility by PMFW version checks and shows the exact PMFW version in
>> dmesg log. PMFW version is a more authentic way to identify any
>> compatibility. So having a mismatch statement in the log is only leading to
>> confusion.
Pmfw/driver_if_version is the remote(pmfw) version, while driver_if_version is
the local (kmd) version, Because the driver allows for their existence without
matching, more information is needed.
They should not be confused as they do not display the same type of information.
Best Regards,
Kevin
-----Original Message-----
From: Lazar, Lijo <lijo.la...@amd.com>
Sent: Friday, September 19, 2025 2:34 PM
To: Wang, Yang(Kevin) <kevinyang.w...@amd.com>; amd-gfx@lists.freedesktop.org
Cc: Zhang, Hawking <hawking.zh...@amd.com>; Deucher, Alexander
<alexander.deuc...@amd.com>; Kamal, Asad <asad.ka...@amd.com>
Subject: RE: [PATCH] drm/amd/pm: Avoid interface mismatch messaging
[Public]
> Hiding this information seems unreasonable unless there is another
> strong reason
This interface version is not used anywhere. Driver ensures backward
compatibility by PMFW version checks and shows the exact PMFW version in dmesg
log. PMFW version is a more authentic way to identify any compatibility. So
having a mismatch statement in the log is only leading to confusion.
Thanks,
Lijo
-----Original Message-----
From: Wang, Yang(Kevin) <kevinyang.w...@amd.com>
Sent: Friday, September 19, 2025 11:59 AM
To: Lazar, Lijo <lijo.la...@amd.com>; amd-gfx@lists.freedesktop.org
Cc: Zhang, Hawking <hawking.zh...@amd.com>; Deucher, Alexander
<alexander.deuc...@amd.com>; Kamal, Asad <asad.ka...@amd.com>
Subject: RE: [PATCH] drm/amd/pm: Avoid interface mismatch messaging
[AMD Official Use Only - AMD Internal Distribution Only]
>> PMFW interface version is not used by some IP implementations like SMU
>> v13.0.6/12, instead rely on PMFW version checks. Avoid the log if interface
>> version is not used.
" is not used": Actually, this is used to display information to users and
developers, not the driver itself.
Smu 'driver_if_version': it is used to show driver source file interface
version on kernel driver side.
Smu 'If_version': it is returned from pmfw side, that means what driver if
version is using on pmfw side.
The most perfect aesthetic situation is using same if version between driver
and pmfw side, but it can still work normally even if it is not matched. (with
limited functionality)
According to the SMU driver design, KMD will not reject interface version
mismatches to ensure smooth driver loading.
Preserving this information helps the driver understand the KMD/PMFW operating
environment and detect problems early.
Hiding this information seems unreasonable unless there is another strong
reason.
Best Regards,
Kevin
-----Original Message-----
From: Lazar, Lijo <lijo.la...@amd.com>
Sent: Friday, September 19, 2025 1:45 PM
To: amd-gfx@lists.freedesktop.org
Cc: Zhang, Hawking <hawking.zh...@amd.com>; Deucher, Alexander
<alexander.deuc...@amd.com>; Kamal, Asad <asad.ka...@amd.com>; Wang,
Yang(Kevin) <kevinyang.w...@amd.com>
Subject: [PATCH] drm/amd/pm: Avoid interface mismatch messaging
PMFW interface version is not used by some IP implementations like SMU
v13.0.6/12, instead rely on PMFW version checks. Avoid the log if interface
version is not used.
Signed-off-by: Lijo Lazar <lijo.la...@amd.com>
---
drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c | 3 ++-
drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c | 2 +-
drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h | 2 ++
3 files changed, 5 insertions(+), 2 deletions(-)
diff --git a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c
b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c
index 1a1f2a6b2e52..a89075e25717 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0.c
@@ -288,7 +288,8 @@ int smu_v13_0_check_fw_version(struct smu_context *smu)
* Considering above, we just leave user a verbal message instead
* of halt driver loading.
*/
- if (if_version != smu->smc_driver_if_version) {
+ if (smu->smc_driver_if_version != SMU_IGNORE_IF_VERSION &&
+ if_version != smu->smc_driver_if_version) {
dev_info(adev->dev, "smu driver if version = 0x%08x, smu fw if
version = 0x%08x, "
"smu fw program = %d, smu fw version = 0x%08x
(%d.%d.%d)\n",
smu->smc_driver_if_version, if_version, diff --git
a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
index 349b6b8be010..062f92635215 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu13/smu_v13_0_6_ppt.c
@@ -3930,7 +3930,7 @@ void smu_v13_0_6_set_ppt_funcs(struct smu_context *smu)
smu->feature_map = (amdgpu_ip_version(smu->adev, MP1_HWIP, 0) ==
IP_VERSION(13, 0, 12)) ?
smu_v13_0_12_feature_mask_map : smu_v13_0_6_feature_mask_map;
smu->table_map = smu_v13_0_6_table_map;
- smu->smc_driver_if_version = SMU13_0_6_DRIVER_IF_VERSION;
+ smu->smc_driver_if_version = SMU_IGNORE_IF_VERSION;
smu->smc_fw_caps |= SMU_FW_CAP_RAS_PRI;
smu_v13_0_set_smu_mailbox_registers(smu);
smu_v13_0_6_set_temp_funcs(smu); diff --git
a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h
b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h
index d588f74b98de..0ae91c8b6d72 100644
--- a/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h
+++ b/drivers/gpu/drm/amd/pm/swsmu/smu_cmn.h
@@ -40,6 +40,8 @@
#define SMU_IH_INTERRUPT_CONTEXT_ID_FAN_ABNORMAL 0x8
#define SMU_IH_INTERRUPT_CONTEXT_ID_FAN_RECOVERY 0x9
+#define SMU_IGNORE_IF_VERSION 0xFFFFFFFF
+
#define smu_cmn_init_soft_gpu_metrics(ptr, frev, crev) \
do { \
typecheck(struct gpu_metrics_v##frev##_##crev *, (ptr)); \
--
2.49.0
