Christoph,
Thanks for the review comments.
On 10/14/24 1:53 PM, Christoph Hellwig wrote:
> On Mon, Oct 14, 2024 at 11:47:28AM +0800, bingbu.cao@xxxxxxxxx wrote:
>> From: Bingbu Cao <bingbu.cao@xxxxxxxxx>
>>
>> DMA ops are a helper for architectures and not for drivers to override
>> the DMA implementation. Driver should not override the DMA
>> implementation. This patch remove the dma_ops override and expose the
>> IPU6 DMA mapping APIs.
>
> That's a good rationale, but it might make sense to mention what you're
> actually changing here as well.
Ack. I will reword the message.
>
>> + ret = dma_map_sgtable(&pdev->dev, sgt, DMA_TO_DEVICE, 0);
>> + if (ret) {
>> + sg_free_table(sgt);
>> + goto out;
>> + }
>> +
>> + ret = ipu6_dma_map_sgtable(sys, sgt, DMA_TO_DEVICE, 0);
>
> This looks like the only user of ipu6_dma_map_sgtable, any reason
> to not open code it here?
The Intel IPU6 driver contains 2 modules, this symbol is defined in
intel_ipu6 module. The map/unmap_sgtable are also used by another
driver - intel_ipu6_isys.
>
>> kfree(pages);
>> @@ -607,7 +615,10 @@ EXPORT_SYMBOL_NS_GPL(ipu6_buttress_map_fw_image, INTEL_IPU6);
>> void ipu6_buttress_unmap_fw_image(struct ipu6_bus_device *sys,
>> struct sg_table *sgt)
>> {
>> - dma_unmap_sgtable(&sys->auxdev.dev, sgt, DMA_TO_DEVICE, 0);
>> + struct pci_dev *pdev = sys->isp->pdev;
>> +
>> + ipu6_dma_unmap_sgtable(sys, sgt, DMA_TO_DEVICE, 0);
>
> Same for ipu6_dma_unmap_sgtable
>
--
Best regards,
Bingbu Cao
