On Thursday 13 June 2013 06:46 PM, Russell King - ARM Linux wrote: > On Thu, Jun 13, 2013 at 06:39:20PM +0530, Rajendra Nayak wrote: >> diff --git a/arch/arm/plat-omap/dma.c b/arch/arm/plat-omap/dma.c >> index 8a71f75..8e16503 100644 >> --- a/arch/arm/plat-omap/dma.c >> +++ b/arch/arm/plat-omap/dma.c >> @@ -2111,8 +2111,8 @@ exit_dma_irq_fail: >> } >> >> exit_dma_lch_fail: >> - kfree(p); >> kfree(d); >> + kfree(p); > > Err. > > p = pdev->dev.platform_data; > d = p->dma_attr; > > Why is it kfree'ing platform data in the first place? This means that > a failed bind can't be reattempted later. It also means that an unbind > plus rebind in userspace will free the platform data leaving stale > pointers behind. Right, I just seemed to have overlooked the fact that p was pointing to platform data. Will remove all the kfree(p) and kfree(d) across the driver (I just realized this is also the case in .remove) > > This is totally nonsense. Don't kfree() data in your driver which you > haven't allocated yourself! > -- To unsubscribe from this list: send the line "unsubscribe linux-omap" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
