Re: [PATCH 4/6] usbip: vhci_hcd: clear URB_DMA_MAP_SG on all giveback paths

[sashiko-bot@kernel.org]

> In vhci_send_cmd_submit(), URB_DMA_MAP_SG is set on urb->transfer_flags
> when urb->num_sgs is non-zero so that setup_cmd_submit_pdu() packs
> USBIP_URB_DMA_MAP_SG into the wire PDU, even though vhci_hcd never maps
> the scatterlist with dma_map_sg().
> 
> Currently, only vhci_recv_ret_submit() clears URB_DMA_MAP_SG before
> calling usb_hcd_giveback_urb(). If an URB with scatter-gather entries
> is instead given back via vhci_recv_ret_unlink(),
> vhci_cleanup_unlink_list(), vhci_cleanup_priv_list(), or
> vhci_urb_dequeue(), __usb_hcd_giveback_urb() invokes
> usb_hcd_unmap_urb_for_dma(), which sees URB_DMA_MAP_SG and calls
> dma_unmap_sg() on an unmapped scatterlist.
> 
> Move the URB_DMA_MAP_SG clearing from vhci_recv_ret_submit() into
> pickup_urb_and_free_priv() (covering vhci_recv_ret_submit(),
> vhci_recv_ret_unlink(), and vhci_cleanup_unlink_list()), and also clear
> URB_DMA_MAP_SG in vhci_urb_dequeue() and vhci_cleanup_priv_list() so all
> URB giveback paths clear the flag.
> 
> Fixes: ea44d190764b ("usbip: Implement SG support to vhci-hcd and stub driver")
> Cc: stable@vger.kernel.org
> Assisted-by: Gemini
> Signed-off-by: Raul E Rangel <rrangel@chromium.org>

Sashiko has reviewed this patch and found no issues.

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20261021150704.3129446-1-rrangel@chromium.org?part=4