Re: [PATCH] dmaengine: qcom_hidma: Simplify error handling path in hidma_probe

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



I apologize, I wrote my code hurriedly and did no explain the bug well.
I understood what the code is doing, but my fix was missing an if
condition.

On Tue, Apr 28, 2020 at 12:01:15PM -0400, Sinan Kaya wrote:
> On 4/28/2020 8:54 AM, Dan Carpenter wrote:
> >> @@ -897,7 +897,6 @@ static int hidma_probe(struct platform_device *pdev)
> >>  	if (msi)
> >             ^^^
> > This test doesn't work.  It will call free hidma_free_msis() if the
> > hidma_request_msi() call fails.  We should do:
> > 
> > 	if (msi) {
> > 		rc = hidma_request_msi(dmadev, pdev);
> > 		msi = false;

What I meant to say here was:

	if (msi) {
		rc = hidma_request_msi(dmadev, pdev);
		if (rc)
			msi = false;

Otherwise we end up checking freeing the msi in the error handling
code when we did not take it.

Hopefully, that clears things up?

regards,
dan carpenter




[Index of Archives]     [Linux ARM Kernel]     [Linux ARM]     [Linux Omap]     [Fedora ARM]     [Linux for Sparc]     [IETF Annouce]     [Security]     [Bugtraq]     [Linux MIPS]     [ECOS]     [Asterisk Internet PBX]     [Linux API]

  Powered by Linux