[PATCH v5 1/2] usb: make xhci platform driver use 64 bit or 32 bit DMA

On Sat, Aug 8, 2015 at 2:22 AM, Russell King - ARM Linux
[off-list ref] wrote:

On Fri, Aug 07, 2015 at 08:18:48PM -0700, Duc Dang wrote:

quoted

diff --git a/drivers/usb/host/xhci-plat.c b/drivers/usb/host/xhci-plat.c
index 890ad9d..5d03f8b 100644
--- a/drivers/usb/host/xhci-plat.c
+++ b/drivers/usb/host/xhci-plat.c

@@ -93,14 +93,14 @@ static int xhci_plat_probe(struct platform_device *pdev)
      if (irq < 0)
              return -ENODEV;

-     /* Initialize dma_mask and coherent_dma_mask to 32-bits */
-     ret = dma_set_coherent_mask(&pdev->dev, DMA_BIT_MASK(32));
-     if (ret)
-             return ret;
-     if (!pdev->dev.dma_mask)
-             pdev->dev.dma_mask = &pdev->dev.coherent_dma_mask;
-     else
-             dma_set_mask(&pdev->dev, DMA_BIT_MASK(32));
+     /* Try setting the coherent_dma_mask to 64 bits, then try 32 bits */
+     ret = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(64));
+     if (ret) {
+             ret = dma_set_mask_and_coherent(&pdev->dev, DMA_BIT_MASK(32));
+             if (ret)
+                     return ret;
+     }

Note that dma_set_mask_and_coherent() and the original code are not
equivalent because of this:

        if (!pdev->dev.dma_mask)
                pdev->dev.dma_mask = &pdev->dev.coherent_dma_mask;

If we know that pdev->dev.dma_mask will always be initialised at this
point, then the above change is fine.  If not, it's introducing a
regression - dma_set_mask_and_coherent() will fail if pdev->dev.dma_mask
is NULL (depending on the architectures implementation of dma_set_mask()).

Prefixing the above change with the two lines I mention above would
ensure equivalent behaviour.  Even if we do want to get rid of this,
I'd advise to do it as a separate patch after this change, which can
be independently reverted if there's problems with its removal.

Hi Russell,

I will add the 2 lines you mentioned back to next version of the
patch. It is safer to do it that way as I do not see
pdev->dev.dma_mask gets initialized before the call
dma_set_mask_and_coherent  inside this xhci_plat.c file.

--
FTTC broadband for 0.8mile line: currently at 10.5Mbps down 400kbps up
according to speedtest.net.

-- 
Regards,
Duc Dang.