[AMD Official Use Only - General]
That requires extra work in MES and API level change to let driver send this info to MES . I think that's kind of unnecessary complicated.
The original problem is MES fw doesn't encapsulate their API defines good enough . Windows driver directly use MES internal structure to calculate the buffer size.
I already pushed MES team to have all the necessary info including this log buffer size defined in mes_api_def.h . We also agreed that maximum log buffer size won't exceed 0x4000 in the near future . This will happens on the new MES release and may take some time for driver side to pick it up , but before this I'd like to have a solution that can fix the issue ASAP .
Regards
Shaoyun.liu
-----Original Message-----
From: Kuehling, Felix <Felix.Kuehling@xxxxxxx>
Sent: Tuesday, March 26, 2024 2:07 PM
To: Liu, Shaoyun <Shaoyun.Liu@xxxxxxx>; amd-gfx@xxxxxxxxxxxxxxxxxxxxx
Subject: Re: [PATCH] drm/amdgpu : Increase the mes log buffer size as per new MES FW version
On 2024-03-25 19:33, Liu, Shaoyun wrote:
> [AMD Official Use Only - General]
>
> It can cause page fault when the log size exceed the page size .
I'd consider that a breaking change in the firmware that should be avoided. Is there a way the updated driver can tell the FW the log size that it allocated, so that old drivers continue to work with new firmware?
Regards,
Felix
>
> -----Original Message-----
> From: Kuehling, Felix <Felix.Kuehling@xxxxxxx>
> Sent: Monday, March 25, 2024 2:58 PM
> To: Liu, Shaoyun <Shaoyun.Liu@xxxxxxx>; amd-gfx@xxxxxxxxxxxxxxxxxxxxx
> Subject: Re: [PATCH] drm/amdgpu : Increase the mes log buffer size as
> per new MES FW version
>
>
> On 2024-03-22 12:49, shaoyunl wrote:
>> From MES version 0x54, the log entry increased and require the log
>> buffer size to be increased. The 16k is maximum size agreed
> What happens when you run the new firmware on an old kernel that only allocates 4KB?
>
> Regards,
> Felix
>
>
>> Signed-off-by: shaoyunl <shaoyun.liu@xxxxxxx>
>> ---
>> drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c | 5 ++---
>> drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h | 1 +
>> 2 files changed, 3 insertions(+), 3 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
>> b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
>> index 9ace848e174c..78e4f88f5134 100644
>> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
>> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.c
>> @@ -103,7 +103,7 @@ static int amdgpu_mes_event_log_init(struct amdgpu_device *adev)
>> if (!amdgpu_mes_log_enable)
>> return 0;
>>
>> - r = amdgpu_bo_create_kernel(adev, PAGE_SIZE, PAGE_SIZE,
>> + r = amdgpu_bo_create_kernel(adev, AMDGPU_MES_LOG_BUFFER_SIZE,
>> +PAGE_SIZE,
>> AMDGPU_GEM_DOMAIN_GTT,
>> &adev->mes.event_log_gpu_obj,
>> &adev->mes.event_log_gpu_addr, @@
>> -1548,12 +1548,11 @@ static int amdgpu_debugfs_mes_event_log_show(struct seq_file *m, void *unused)
>> uint32_t *mem = (uint32_t *)(adev->mes.event_log_cpu_addr);
>>
>> seq_hex_dump(m, "", DUMP_PREFIX_OFFSET, 32, 4,
>> - mem, PAGE_SIZE, false);
>> + mem, AMDGPU_MES_LOG_BUFFER_SIZE, false);
>>
>> return 0;
>> }
>>
>> -
>> DEFINE_SHOW_ATTRIBUTE(amdgpu_debugfs_mes_event_log);
>>
>> #endif
>> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h
>> b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h
>> index 7d4f93fea937..4c8fc3117ef8 100644
>> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h
>> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_mes.h
>> @@ -52,6 +52,7 @@ enum amdgpu_mes_priority_level {
>>
>> #define AMDGPU_MES_PROC_CTX_SIZE 0x1000 /* one page area */
>> #define AMDGPU_MES_GANG_CTX_SIZE 0x1000 /* one page area */
>> +#define AMDGPU_MES_LOG_BUFFER_SIZE 0x4000 /* Maximu log buffer size
>> +for MES */
>>
>> struct amdgpu_mes_funcs;
>>
