On 12/1/2016 1:02 AM, Marek Szyprowski wrote:
> dwc2_hsotg_ep_enable can be called from interrupt context, so all
> allocations should be done with GFP_ATOMIC flags. This fixes following
> issue on ARM architecture:
>
> [<c010d830>] (unwind_backtrace) from [<c010a51c>] (show_stack+0x10/0x14)
> [<c010a51c>] (show_stack) from [<c032930c>] (dump_stack+0x74/0x94)
> [<c032930c>] (dump_stack) from [<c011cd30>] (__warn+0xd4/0x100)
> [<c011cd30>] (__warn) from [<c011cd7c>] (warn_slowpath_null+0x20/0x28)
> [<c011cd7c>] (warn_slowpath_null) from [<c0187e04>]
> (smp_call_function_many+0xcc/0x2a4)
> [<c0187e04>] (smp_call_function_many) from [<c0188014>]
> (on_each_cpu_mask+0x38/0xa8)
> [<c0188014>] (on_each_cpu_mask) from [<c01ddfe0>]
> (start_isolate_page_range+0x134/0x1b8)
> [<c01ddfe0>] (start_isolate_page_range) from [<c01a3c14>]
> (alloc_contig_range+0xac/0x2f8)
> [<c01a3c14>] (alloc_contig_range) from [<c01de3e4>] (cma_alloc+0xe0/0x1a8)
> [<c01de3e4>] (cma_alloc) from [<c0110acc>] (__alloc_from_contiguous+0x38/0xe0)
> [<c0110acc>] (__alloc_from_contiguous) from [<c0110ba4>]
> (cma_allocator_alloc+0x30/0x38)
> [<c0110ba4>] (cma_allocator_alloc) from [<c0111034>] (__dma_alloc+0x1c0/0x2c8)
> [<c0111034>] (__dma_alloc) from [<c01111b4>] (arm_dma_alloc+0x3c/0x48)
> [<c01111b4>] (arm_dma_alloc) from [<c04ad800>]
> (dwc2_hsotg_ep_enable+0xec/0x46c)
> [<c04ad800>] (dwc2_hsotg_ep_enable) from [<c04da610>]
> (usb_ep_enable+0x2c/0x3c)
> [<c04da610>] (usb_ep_enable) from [<c04dc0c0>] (ecm_set_alt+0xa8/0x154)
> [<c04dc0c0>] (ecm_set_alt) from [<c04d678c>] (composite_setup+0xd74/0x1540)
> [<c04d678c>] (composite_setup) from [<c04ae048>]
> (dwc2_hsotg_complete_setup+0xb8/0x370)
> [<c04ae048>] (dwc2_hsotg_complete_setup) from [<c04d987c>]
> (usb_gadget_giveback_request+0xc/0x10)
> [<c04d987c>] (usb_gadget_giveback_request) from [<c04acafc>]
> (dwc2_hsotg_complete_request+0x78/0x128)
> [<c04acafc>] (dwc2_hsotg_complete_request) from [<c04aed28>]
> (dwc2_hsotg_epint+0x69c/0x81c)
> [<c04aed28>] (dwc2_hsotg_epint) from [<c04af6c4>] (dwc2_hsotg_irq+0xfc/0x748)
> [<c04af6c4>] (dwc2_hsotg_irq) from [<c0163264>]
> (__handle_irq_event_percpu+0x58/0x140)
> [<c0163264>] (__handle_irq_event_percpu) from [<c0163368>]
> (handle_irq_event_percpu+0x1c/0x58)
> [<c0163368>] (handle_irq_event_percpu) from [<c01633dc>]
> (handle_irq_event+0x38/0x5c)
> [<c01633dc>] (handle_irq_event) from [<c01666e4>]
> (handle_fasteoi_irq+0xc4/0x19c)
> [<c01666e4>] (handle_fasteoi_irq) from [<c0162a2c>]
> (generic_handle_irq+0x18/0x28)
> [<c0162a2c>] (generic_handle_irq) from [<c0162b40>]
> (__handle_domain_irq+0x6c/0xe4)
> [<c0162b40>] (__handle_domain_irq) from [<c0101470>]
> (gic_handle_irq+0x50/0x9c)
> [<c0101470>] (gic_handle_irq) from [<c010b00c>] (__irq_svc+0x6c/0xa8)
>
> Fixes: 5f54c54b0ba83 ("usb: dwc2: gadget: Add DDMA chain pointers to
> dwc2_hsotg_ep structure")
> Signed-off-by: Marek Szyprowski <m.szyprow...@samsung.com>
> ---
> drivers/usb/dwc2/gadget.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/usb/dwc2/gadget.c b/drivers/usb/dwc2/gadget.c
> index b95930f20d90..c55db4aa54d6 100644
> --- a/drivers/usb/dwc2/gadget.c
> +++ b/drivers/usb/dwc2/gadget.c
> @@ -3753,7 +3753,7 @@ static int dwc2_hsotg_ep_enable(struct usb_ep *ep,
> hs_ep->desc_list = dma_alloc_coherent(hsotg->dev,
> MAX_DMA_DESC_NUM_GENERIC *
> sizeof(struct dwc2_dma_desc),
> - &hs_ep->desc_list_dma, GFP_KERNEL);
> + &hs_ep->desc_list_dma, GFP_ATOMIC);
> if (!hs_ep->desc_list) {
> ret = -ENOMEM;
> goto error2;
>
Acked-by: John Youn <johny...@synopsys.com>
Regards,
John
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
