From mboxrd@z Thu Jan 1 00:00:00 1970 From: linux@prisktech.co.nz (Tony Prisk) Date: Wed, 08 May 2013 17:11:25 +1200 Subject: [PATCH] USB: set device dma_mask without reference to global data In-Reply-To: <1367967232-10128-1-git-send-email-swarren@wwwdotorg.org> References: <1367967232-10128-1-git-send-email-swarren@wwwdotorg.org> Message-ID: <5189DE7D.7060205@prisktech.co.nz> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On 08/05/13 10:53, Stephen Warren wrote: > From: Stephen Warren > > Many USB host drivers contain code such as: > > if (!pdev->dev.dma_mask) > pdev->dev.dma_mask = &tegra_ehci_dma_mask; > > ... where tegra_ehci_dma_mask is a global. I suspect this code originated > in commit 4a53f4e "USB: ehci-tegra: add probing through device tree" and > was simply copied everywhere else. > > This works fine when the code is built-in, but can cause a crash when the > code is in a module. The first module load sets up the dma_mask pointer, > but if the module is removed and re-inserted, the value is now non-NULL, > and hence is not updated to point at the new location, and hence points > at a stale location within the previous module load address, which in > turn causes a crash if the pointer is de-referenced. > > The simplest way of solving this seems to be to copy the code from > ehci-platform.c, which uses the coherent_dma_mask as the target for the > dma_mask pointer. > > Suggested-by: Arnd Bergmann > Signed-off-by: Stephen Warren > --- In the case of uhci-platform you would be absolutely correct. I copied the example from tegra when we first had the problem on arch-vt8500.Because we have no NAND support yet, I have always booted myrootfs from USB so it's always been builtin and the problem wasnever a problem. The same problem would have existed on ehci-vt8500 but Arnd replaced it with ehci-platform due to the multiplatform issues. for uhci-platform.c Acked-by: Tony Prisk Regards Tony P