From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Date: Tue, 18 Oct 2016 10:53:28 +0200 (CEST) From: Sebastian Ott Subject: Re: [PATCH 25/28] s390: pci: don't print uninitialized data for debugging In-Reply-To: <20161018084855.1a100670@mschwide> References: <20161017220342.1627073-1-arnd@arndb.de> <20161017221650.1902729-6-arnd@arndb.de> <20161018084855.1a100670@mschwide> MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Message-Id: Sender: linux-kernel-owner@vger.kernel.org List-Archive: List-Post: To: Martin Schwidefsky Cc: Arnd Bergmann , Heiko Carstens , Linus Torvalds , linux-kernel@vger.kernel.org, Gerald Schaefer , Joerg Roedel , Christian Borntraeger , linux-s390@vger.kernel.org List-ID: On Tue, 18 Oct 2016, Martin Schwidefsky wrote: > On Tue, 18 Oct 2016 00:16:13 +0200 > Arnd Bergmann wrote: > > > gcc correctly warns about an incorrect use of the 'pa' variable > > in case we pass an empty scatterlist to __s390_dma_map_sg: > > > > arch/s390/pci/pci_dma.c: In function '__s390_dma_map_sg': > > arch/s390/pci/pci_dma.c:309:13: warning: 'pa' may be used uninitialized in this function [-Wmaybe-uninitialized] > > > > This adds a bogus initialization to the function to sanitize > > the debug output. I would have preferred a solution without > > the initialization, but I only got the report from the > > kbuild bot after turning on the warning again, and didn't > > manage to reproduce it myself. > > > > Signed-off-by: Arnd Bergmann > > --- > > arch/s390/pci/pci_dma.c | 2 +- > > 1 file changed, 1 insertion(+), 1 deletion(-) > > > > diff --git a/arch/s390/pci/pci_dma.c b/arch/s390/pci/pci_dma.c > > index 7350c8b..6b2f72f 100644 > > --- a/arch/s390/pci/pci_dma.c > > +++ b/arch/s390/pci/pci_dma.c > > @@ -423,7 +423,7 @@ static int __s390_dma_map_sg(struct device *dev, struct scatterlist *sg, > > dma_addr_t dma_addr_base, dma_addr; > > int flags = ZPCI_PTE_VALID; > > struct scatterlist *s; > > - unsigned long pa; > > + unsigned long pa = 0; > > int ret; > > > > size = PAGE_ALIGN(size); > > The compiler is right. pa is set in the for-loop. If "dma_addr < dma_addr_base + size" > is never true and __dma_purge_tlb() returns with an error, pa *is* uninitialized. > The fix seems sensible to me. Although that could only happen if map_sg is called with zero length sg list entries I'm in favor getting rid of that warning. Acked-by: Sebastian Ott Regards, Sebastian