From mboxrd@z Thu Jan 1 00:00:00 1970 From: Christoph Hellwig Date: Sun, 25 Jul 2021 06:07:47 +0000 Subject: Re: [PATCH v2 01/21] dma-mapping: Allow map_sg() ops to return negative error codes Message-Id: <20210725060747.GA10852@lst.de> List-Id: References: <20210723175008.22410-1-logang@deltatee.com> <20210723175008.22410-2-logang@deltatee.com> In-Reply-To: <20210723175008.22410-2-logang@deltatee.com> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Logan Gunthorpe Cc: linux-kernel@vger.kernel.org, linux-alpha@vger.kernel.org, linux-arm-kernel@lists.infradead.org, linux-ia64@vger.kernel.org, linux-mips@vger.kernel.org, linuxppc-dev@lists.ozlabs.org, linux-s390@vger.kernel.org, sparclinux@vger.kernel.org, iommu@lists.linux-foundation.org, linux-parisc@vger.kernel.org, xen-devel@lists.xenproject.org, Christoph Hellwig , Marek Szyprowski , Robin Murphy , Stephen Bates , Martin Oliveira > +int dma_map_sgtable(struct device *dev, struct sg_table *sgt, > + enum dma_data_direction dir, unsigned long attrs) > +{ > + int nents; > + > + nents = __dma_map_sg_attrs(dev, sgt->sgl, sgt->orig_nents, dir, attrs); > + if (nents = 0) > + return -EIO; > + else if (nents < 0) { > + if (WARN_ON_ONCE(nents != -EINVAL && nents != -ENOMEM && > + nents != -EIO)) > + return -EIO; I think this validation of the errnos needs to go into __dma_map_sg_attrs, so that we catch it for the classic dma_map_sg callers as well.