Re: [PATCH] drm/v3d: set the V3D GPU sg segment limitation

[xiaolei wang <xiaolei.wang@windriver.com>]

On 12/1/25 21:35, Maíra Canal wrote:
> CAUTION: This email comes from a non Wind River email account!
> Do not click links or open attachments unless you recognize the sender 
> and know the content is safe.
>
> Hi Xiaolei,
>
> On 13/11/25 06:25, Xiaolei Wang wrote:
>> When using V3D rendering, the kernel occasionally reports a
>> segment size mismatch. This is because 'max_seg_size' is not
>
> Could you mention that this only happens when CONFIG_DMA_API_DEBUG is
> enabled?

Yes, I only found that when CONFIG_DMA_API_DEBUG was enabled

thanks

xiaolei

>
> Best Regards,
> - Maíra
>
>> set. The kernel defaults to 64K. Based on V3D's addressing
>> capabilities, setting 'max_seg_size' to the maximum will
>> prevent 'debug_dma_map_sg()' from complaining about the
>> over-mapping of the V3D segment length.
>>
>> DMA-API: v3d 1002000000.v3d: mapping sg segment longer than device
>>   claims to support [len=8290304] [max=65536]
>> WARNING: CPU: 0 PID: 493 at kernel/dma/debug.c:1179 
>> debug_dma_map_sg+0x330/0x388
>> CPU: 0 UID: 0 PID: 493 Comm: Xorg Not tainted 6.12.53-yocto-standard #1
>> Hardware name: Raspberry Pi 5 Model B Rev 1.0 (DT)
>> pstate: 60400009 (nZCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--)
>> pc : debug_dma_map_sg+0x330/0x388
>> lr : debug_dma_map_sg+0x330/0x388
>> sp : ffff8000829a3ac0
>> x29: ffff8000829a3ac0 x28: 0000000000000001 x27: ffff8000813fe000
>> x26: ffffc1ffc0000000 x25: ffff00010fdeb760 x24: 0000000000000000
>> x23: ffff8000816a9bf0 x22: 0000000000000001 x21: 0000000000000002
>> x20: 0000000000000002 x19: ffff00010185e810 x18: ffffffffffffffff
>> x17: 69766564206e6168 x16: 74207265676e6f6c x15: 20746e656d676573
>> x14: 20677320676e6970 x13: 5d34303334393134 x12: 0000000000000000
>> x11: 00000000000000c0 x10: 00000000000009c0 x9 : ffff8000800e0b7c
>> x8 : ffff00010a315ca0 x7 : ffff8000816a5110 x6 : 0000000000000001
>> x5 : 000000000000002b x4 : 0000000000000002 x3 : 0000000000000008
>> x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff00010a315280
>> Call trace:
>>   debug_dma_map_sg+0x330/0x388
>>   __dma_map_sg_attrs+0xc0/0x278
>>   dma_map_sgtable+0x30/0x58
>>   drm_gem_shmem_get_pages_sgt+0xb4/0x140
>>   v3d_bo_create_finish+0x28/0x130 [v3d]
>>   v3d_create_bo_ioctl+0x54/0x180 [v3d]
>>   drm_ioctl_kernel+0xc8/0x140
>>   drm_ioctl+0x2d4/0x4d8
>>
>> Signed-off-by: Xiaolei Wang <xiaolei.wang@windriver.com>
>> ---
>>   drivers/gpu/drm/v3d/v3d_drv.c | 2 ++
>>   1 file changed, 2 insertions(+)
>>
>> diff --git a/drivers/gpu/drm/v3d/v3d_drv.c 
>> b/drivers/gpu/drm/v3d/v3d_drv.c
>> index c5a3bbbc74c5..f4da7a94e401 100644
>> --- a/drivers/gpu/drm/v3d/v3d_drv.c
>> +++ b/drivers/gpu/drm/v3d/v3d_drv.c
>> @@ -377,6 +377,8 @@ static int v3d_platform_drm_probe(struct 
>> platform_device *pdev)
>>       if (ret)
>>               goto clk_disable;
>>
>> +     dma_set_max_seg_size(&pdev->dev, UINT_MAX);
>
> I believe I'd
>
>> +
>>       v3d->va_width = 30 + V3D_GET_FIELD(mmu_debug, V3D_MMU_VA_WIDTH);
>>
>>       ident1 = V3D_READ(V3D_HUB_IDENT1);
>