linux-kernel - Re: [PATCH v2] usb: dwc3: host: Set the dma

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:	Wed, 8 Jun 2016 16:18:55 +0800
From:	Baolin Wang <baolin.wang@...aro.org>
To:	Felipe Balbi <balbi@...nel.org>
Cc:	Greg KH <gregkh@...uxfoundation.org>,
	Mark Brown <broonie@...nel.org>,
	USB <linux-usb@...r.kernel.org>,
	LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v2] usb: dwc3: host: Set the dma_ops for xhci device

Hi,

On 6 June 2016 at 22:59, Felipe Balbi <balbi@...nel.org> wrote:
>
> Hi,
>
> Baolin Wang <baolin.wang@...aro.org> writes:
>> On ARM64 platform, it will set 'dummy_dma_ops' for device dma_ops if
>> it did not call 'arch_setup_dma_ops' at device creation time, that will
>> cause failure when setting the dma mask for device.
>>
>> Thus this patch set the xhci device dma_ops from the parent device if
>> the xhci device dma_ops is 'dummy_dma_ops'.
>>
>> Changes since v1:
>>  - Add CONFIG_ARM64 macro.
>>
>> Signed-off-by: Baolin Wang <baolin.wang@...aro.org>
>> ---
>>  drivers/usb/dwc3/host.c |    5 +++++
>>  1 file changed, 5 insertions(+)
>>
>> diff --git a/drivers/usb/dwc3/host.c b/drivers/usb/dwc3/host.c
>> index c679f63..edb666d 100644
>> --- a/drivers/usb/dwc3/host.c
>> +++ b/drivers/usb/dwc3/host.c
>> @@ -32,6 +32,11 @@ int dwc3_host_init(struct dwc3 *dwc)
>>               return -ENOMEM;
>>       }
>>
>> +#ifdef CONFIG_ARM64
>> +     if (get_dma_ops(&xhci->dev) == get_dma_ops(NULL))
>> +             xhci->dev.archdata.dma_ops = get_dma_ops(dwc->dev);
>> +#endif
>
> I don't like the ifdef and also don't like that this is done in dwc3
> itself. Seems like we need something like this done from the
> platform_bus core.

I've sent one patch to fix this issue in platform_bus core, and Robin
explained "DMA-capable devices are real hardware, therefore don't
spring out of thin air without being described in DT or ACPI."
But the problem is here, how can we handle this issue to make the host
work well? What about below modification? Thanks.

diff --git a/drivers/usb/dwc3/host.c b/drivers/usb/dwc3/host.c
index c679f63..f7c58f9 100644
--- a/drivers/usb/dwc3/host.c
+++ b/drivers/usb/dwc3/host.c
@@ -17,6 +17,7 @@

 #include <linux/platform_device.h>
 #include <linux/usb/xhci_pdriver.h>
+#include <linux/of_device.h>

 #include "core.h"

@@ -37,6 +38,7 @@ int dwc3_host_init(struct dwc3 *dwc)
        xhci->dev.parent        = dwc->dev;
        xhci->dev.dma_mask      = dwc->dev->dma_mask;
        xhci->dev.dma_parms     = dwc->dev->dma_parms;
+       of_dma_configure(&xhci->dev, dwc->dev->of_node);

        dwc->xhci = xhci;

-- 
Baolin.wang
Best Regards