[PATCH] drm/amdkfd: enable heavy-weight TLB flush on Vega20

[PATCH] drm/amdkfd: enable heavy-weight TLB flush on Vega20

[Eric Huang]

It is to meet the requirement for memory allocation
optimization on MI50.

Signed-off-by: Eric Huang <jinhuieric.huang@amd.com>
---
 drivers/gpu/drm/amd/amdkfd/kfd_chardev.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
index 5b8ae0795c0a..d708f1a502cf 100644
--- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
+++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
@@ -1582,7 +1582,8 @@ static int kfd_ioctl_free_memory_of_gpu(struct file *filep,
 static bool kfd_flush_tlb_after_unmap(struct kfd_dev *dev) {
 	return KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 2) ||
 	       (KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 1) &&
-	        dev->adev->sdma.instance[0].fw_version >= 18);
+	        dev->adev->sdma.instance[0].fw_version >= 18) ||
+	        KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 0);
 }
 
 static int kfd_ioctl_map_memory_to_gpu(struct file *filep,
-- 
2.25.1

Re: [PATCH] drm/amdkfd: enable heavy-weight TLB flush on Vega20

[Alex Deucher]

On Fri, Jan 21, 2022 at 11:17 AM Eric Huang <jinhuieric.huang@amd.com> wrote:
>
> It is to meet the requirement for memory allocation
> optimization on MI50.
>
> Signed-off-by: Eric Huang <jinhuieric.huang@amd.com>

Assuming there is no firmware version requirement, the patch is:
Acked-by: Alex Deucher <alexander.deucher@amd.com>

> ---
>  drivers/gpu/drm/amd/amdkfd/kfd_chardev.c | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
> index 5b8ae0795c0a..d708f1a502cf 100644
> --- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
> +++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
> @@ -1582,7 +1582,8 @@ static int kfd_ioctl_free_memory_of_gpu(struct file *filep,
>  static bool kfd_flush_tlb_after_unmap(struct kfd_dev *dev) {
>         return KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 2) ||
>                (KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 1) &&
> -               dev->adev->sdma.instance[0].fw_version >= 18);
> +               dev->adev->sdma.instance[0].fw_version >= 18) ||
> +               KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 0);
>  }
>
>  static int kfd_ioctl_map_memory_to_gpu(struct file *filep,
> --
> 2.25.1
>

RE: [PATCH] drm/amdkfd: enable heavy-weight TLB flush on Vega20

[Chen, Guchun]

[Public]

Hi Eric,

Are you sure that there is no FW requirement for this patch on Vega20? KFDMemory test failed by this commit.

Regards,
Guchun

-----Original Message-----
From: amd-gfx <amd-gfx-bounces@lists.freedesktop.org> On Behalf Of Alex Deucher
Sent: Tuesday, January 25, 2022 4:08 AM
To: Huang, JinHuiEric <JinHuiEric.Huang@amd.com>
Cc: amd-gfx list <amd-gfx@lists.freedesktop.org>
Subject: Re: [PATCH] drm/amdkfd: enable heavy-weight TLB flush on Vega20

On Fri, Jan 21, 2022 at 11:17 AM Eric Huang <jinhuieric.huang@amd.com> wrote:
>
> It is to meet the requirement for memory allocation optimization on 
> MI50.
>
> Signed-off-by: Eric Huang <jinhuieric.huang@amd.com>

Assuming there is no firmware version requirement, the patch is:
Acked-by: Alex Deucher <alexander.deucher@amd.com>

> ---
>  drivers/gpu/drm/amd/amdkfd/kfd_chardev.c | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c 
> b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
> index 5b8ae0795c0a..d708f1a502cf 100644
> --- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
> +++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
> @@ -1582,7 +1582,8 @@ static int kfd_ioctl_free_memory_of_gpu(struct 
> file *filep,  static bool kfd_flush_tlb_after_unmap(struct kfd_dev *dev) {
>         return KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 2) ||
>                (KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 1) &&
> -               dev->adev->sdma.instance[0].fw_version >= 18);
> +               dev->adev->sdma.instance[0].fw_version >= 18) ||
> +               KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 0);
>  }
>
>  static int kfd_ioctl_map_memory_to_gpu(struct file *filep,
> --
> 2.25.1
>

Re: [PATCH] drm/amdkfd: enable heavy-weight TLB flush on Vega20

[Eric Huang]

Hi Guchun,

SDMA FW team confirms MI50/VG20 doesn't have the same bug as MI100, 
which cases asic hang issue when running RVS test. If this change makes 
KFDMemoryTest failed, please fill a Jira and assign to me.

Thanks,
Eric

On 2022-02-07 08:01, Chen, Guchun wrote:
> [Public]
>
> Hi Eric,
>
> Are you sure that there is no FW requirement for this patch on Vega20? KFDMemory test failed by this commit.
>
> Regards,
> Guchun
>
> -----Original Message-----
> From: amd-gfx <amd-gfx-bounces@lists.freedesktop.org> On Behalf Of Alex Deucher
> Sent: Tuesday, January 25, 2022 4:08 AM
> To: Huang, JinHuiEric <JinHuiEric.Huang@amd.com>
> Cc: amd-gfx list <amd-gfx@lists.freedesktop.org>
> Subject: Re: [PATCH] drm/amdkfd: enable heavy-weight TLB flush on Vega20
>
> On Fri, Jan 21, 2022 at 11:17 AM Eric Huang <jinhuieric.huang@amd.com> wrote:
>> It is to meet the requirement for memory allocation optimization on
>> MI50.
>>
>> Signed-off-by: Eric Huang <jinhuieric.huang@amd.com>
> Assuming there is no firmware version requirement, the patch is:
> Acked-by: Alex Deucher <alexander.deucher@amd.com>
>
>> ---
>>   drivers/gpu/drm/amd/amdkfd/kfd_chardev.c | 3 ++-
>>   1 file changed, 2 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
>> b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
>> index 5b8ae0795c0a..d708f1a502cf 100644
>> --- a/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
>> +++ b/drivers/gpu/drm/amd/amdkfd/kfd_chardev.c
>> @@ -1582,7 +1582,8 @@ static int kfd_ioctl_free_memory_of_gpu(struct
>> file *filep,  static bool kfd_flush_tlb_after_unmap(struct kfd_dev *dev) {
>>          return KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 2) ||
>>                 (KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 1) &&
>> -               dev->adev->sdma.instance[0].fw_version >= 18);
>> +               dev->adev->sdma.instance[0].fw_version >= 18) ||
>> +               KFD_GC_VERSION(dev) == IP_VERSION(9, 4, 0);
>>   }
>>
>>   static int kfd_ioctl_map_memory_to_gpu(struct file *filep,
>> --
>> 2.25.1
>>