Re: [PATCH v4 06/11] drm/ttm: factor out ttm_bo_mmap_vma_setup

[Koenig, Christian]

Am 16.10.19 um 14:18 schrieb Christian KÃnig:
> Am 16.10.19 um 13:51 schrieb Gerd Hoffmann:
>> Factor out ttm vma setup to a new function.
>> Reduces code duplication a bit.
>>
>> v2: don't change vm_flags (moved to separate patch).
>> v4: make ttm_bo_mmap_vma_setup static.
>>
>> Signed-off-by: Gerd Hoffmann <kraxel@xxxxxxxxxx>
>
> Reviewed-by: Christian KÃnig <christian.koenig@xxxxxxx> for this one 
> and #7 in the series.

Any objections that I add these two to my drm-ttm-next pull request or 
did you wanted to merge that through some other tree?

Thanks,
Christian.

>
>> ---
>> Â drivers/gpu/drm/ttm/ttm_bo_vm.c | 46 +++++++++++++++++----------------
>> Â 1 file changed, 24 insertions(+), 22 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/ttm/ttm_bo_vm.c 
>> b/drivers/gpu/drm/ttm/ttm_bo_vm.c
>> index 4aa007edffb0..53345c0854d5 100644
>> --- a/drivers/gpu/drm/ttm/ttm_bo_vm.c
>> +++ b/drivers/gpu/drm/ttm/ttm_bo_vm.c
>> @@ -426,6 +426,28 @@ static struct ttm_buffer_object 
>> *ttm_bo_vm_lookup(struct ttm_bo_device *bdev,
>> ÂÂÂÂÂ return bo;
>> Â }
>> Â +static void ttm_bo_mmap_vma_setup(struct ttm_buffer_object *bo, 
>> struct vm_area_struct *vma)
>> +{
>> +ÂÂÂ vma->vm_ops = &ttm_bo_vm_ops;
>> +
>> +ÂÂÂ /*
>> +ÂÂÂÂ * Note: We're transferring the bo reference to
>> +ÂÂÂÂ * vma->vm_private_data here.
>> +ÂÂÂÂ */
>> +
>> +ÂÂÂ vma->vm_private_data = bo;
>> +
>> +ÂÂÂ /*
>> +ÂÂÂÂ * We'd like to use VM_PFNMAP on shared mappings, where
>> +ÂÂÂÂ * (vma->vm_flags & VM_SHARED) != 0, for performance reasons,
>> +ÂÂÂÂ * but for some reason VM_PFNMAP + x86 PAT + write-combine is very
>> +ÂÂÂÂ * bad for performance. Until that has been sorted out, use
>> +ÂÂÂÂ * VM_MIXEDMAP on all mappings. See freedesktop.org bug #75719
>> +ÂÂÂÂ */
>> +ÂÂÂ vma->vm_flags |= VM_MIXEDMAP;
>> +ÂÂÂ vma->vm_flags |= VM_IO | VM_DONTEXPAND | VM_DONTDUMP;
>> +}
>> +
>> Â int ttm_bo_mmap(struct file *filp, struct vm_area_struct *vma,
>> ÂÂÂÂÂÂÂÂÂ struct ttm_bo_device *bdev)
>> Â {
>> @@ -449,24 +471,7 @@ int ttm_bo_mmap(struct file *filp, struct 
>> vm_area_struct *vma,
>> ÂÂÂÂÂ if (unlikely(ret != 0))
>> ÂÂÂÂÂÂÂÂÂ goto out_unref;
>> Â -ÂÂÂ vma->vm_ops = &ttm_bo_vm_ops;
>> -
>> -ÂÂÂ /*
>> -ÂÂÂÂ * Note: We're transferring the bo reference to
>> -ÂÂÂÂ * vma->vm_private_data here.
>> -ÂÂÂÂ */
>> -
>> -ÂÂÂ vma->vm_private_data = bo;
>> -
>> -ÂÂÂ /*
>> -ÂÂÂÂ * We'd like to use VM_PFNMAP on shared mappings, where
>> -ÂÂÂÂ * (vma->vm_flags & VM_SHARED) != 0, for performance reasons,
>> -ÂÂÂÂ * but for some reason VM_PFNMAP + x86 PAT + write-combine is very
>> -ÂÂÂÂ * bad for performance. Until that has been sorted out, use
>> -ÂÂÂÂ * VM_MIXEDMAP on all mappings. See freedesktop.org bug #75719
>> -ÂÂÂÂ */
>> -ÂÂÂ vma->vm_flags |= VM_MIXEDMAP;
>> -ÂÂÂ vma->vm_flags |= VM_IO | VM_DONTEXPAND | VM_DONTDUMP;
>> +ÂÂÂ ttm_bo_mmap_vma_setup(bo, vma);
>> ÂÂÂÂÂ return 0;
>> Â out_unref:
>> ÂÂÂÂÂ ttm_bo_put(bo);
>> @@ -481,10 +486,7 @@ int ttm_fbdev_mmap(struct vm_area_struct *vma, 
>> struct ttm_buffer_object *bo)
>> Â ÂÂÂÂÂ ttm_bo_get(bo);
>> Â -ÂÂÂ vma->vm_ops = &ttm_bo_vm_ops;
>> -ÂÂÂ vma->vm_private_data = bo;
>> -ÂÂÂ vma->vm_flags |= VM_MIXEDMAP;
>> -ÂÂÂ vma->vm_flags |= VM_IO | VM_DONTEXPAND;
>> +ÂÂÂ ttm_bo_mmap_vma_setup(bo, vma);
>> ÂÂÂÂÂ return 0;
>> Â }
>> Â EXPORT_SYMBOL(ttm_fbdev_mmap);
>