From: David Woodhouse <dwmw2-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org> To: Robin Murphy <robin.murphy-5wv7dgnIgG8@public.gmane.org>, joro-zLv9SwRftAIdnm+yROfE0A@public.gmane.org Cc: leedom-ut6Up61K2wZBDgjK7y7TUQ@public.gmane.org, herbert-lOAM2aK0SrRLBo1qDEOMRrpzq4S04n8Q@public.gmane.org, linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org, linux-crypto-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, Harsh-ut6Up61K2wZBDgjK7y7TUQ@public.gmane.org Subject: Re: [PATCH] iommu/vt-d: Fix scatterlist offset handling Date: Tue, 03 Oct 2017 13:55:34 +0100 [thread overview] Message-ID: <1507035334.29211.105.camel@infradead.org> (raw) In-Reply-To: <644c3e01654f8bd48d669c36e424959d6ef0e27e.1506607370.git.robin.murphy-5wv7dgnIgG8@public.gmane.org> [-- Attachment #1.1: Type: text/plain, Size: 2284 bytes --] On Thu, 2017-09-28 at 15:14 +0100, Robin Murphy wrote: > The intel-iommu DMA ops fail to correctly handle scatterlists where > sg->offset is greater than PAGE_SIZE - the IOVA allocation is computed > appropriately based on the page-aligned portion of the offset, but the > mapping is set up relative to sg->page, which means it fails to actually > cover the whole buffer (and in the worst case doesn't cover it at all): > > (sg->dma_address + sg->dma_len) ----+ > sg->dma_address ---------+ | > iov_pfn------+ | | > | | | > v v v > iova: a b c d e f > |--------|--------|--------|--------|--------| > <...calculated....> > [_____mapped______] > pfn: 0 1 2 3 4 5 > |--------|--------|--------|--------|--------| > ^ ^ ^ > | | | > sg->page ----+ | | > sg->offset --------------+ | > (sg->offset + sg->length) ----------+ I'd still dearly love to see some clear documentation of what it means for sg->offset to be outside the page referenced by sg->page. Or is it really not "outside", and it's *only* valid for the offset to be > PAGE_OFFSET when it's a huge page, so we can check that with a BUG_ON() ? In particular, I'd like to know what is intended in the Xen PV case, where there isn't a straight correspondence between pfn and mfn. Is the out-of-range sg->offset intended to refer to the next *pfn* after sg- >page, or to the next *mfn* after sg->page? I confess I've only followed this thread vaguely, but I haven't seen a *coherent* explanation except in the huge page case (in which case I want to see that BUG_ON in the patch) of why this isn't just totally bogus. [-- Attachment #1.2: smime.p7s --] [-- Type: application/x-pkcs7-signature, Size: 4938 bytes --] [-- Attachment #2: Type: text/plain, Size: 0 bytes --]
WARNING: multiple messages have this Message-ID (diff)
From: David Woodhouse <dwmw2@infradead.org> To: Robin Murphy <robin.murphy@arm.com>, joro@8bytes.org Cc: ashok.raj@intel.com, leedom@chelsio.com, Harsh@chelsio.com, herbert@gondor.apana.org.au, iommu@lists.linux-foundation.org, linux-crypto@vger.kernel.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH] iommu/vt-d: Fix scatterlist offset handling Date: Tue, 03 Oct 2017 13:55:34 +0100 [thread overview] Message-ID: <1507035334.29211.105.camel@infradead.org> (raw) In-Reply-To: <644c3e01654f8bd48d669c36e424959d6ef0e27e.1506607370.git.robin.murphy@arm.com> [-- Attachment #1: Type: text/plain, Size: 2284 bytes --] On Thu, 2017-09-28 at 15:14 +0100, Robin Murphy wrote: > The intel-iommu DMA ops fail to correctly handle scatterlists where > sg->offset is greater than PAGE_SIZE - the IOVA allocation is computed > appropriately based on the page-aligned portion of the offset, but the > mapping is set up relative to sg->page, which means it fails to actually > cover the whole buffer (and in the worst case doesn't cover it at all): > > (sg->dma_address + sg->dma_len) ----+ > sg->dma_address ---------+ | > iov_pfn------+ | | > | | | > v v v > iova: a b c d e f > |--------|--------|--------|--------|--------| > <...calculated....> > [_____mapped______] > pfn: 0 1 2 3 4 5 > |--------|--------|--------|--------|--------| > ^ ^ ^ > | | | > sg->page ----+ | | > sg->offset --------------+ | > (sg->offset + sg->length) ----------+ I'd still dearly love to see some clear documentation of what it means for sg->offset to be outside the page referenced by sg->page. Or is it really not "outside", and it's *only* valid for the offset to be > PAGE_OFFSET when it's a huge page, so we can check that with a BUG_ON() ? In particular, I'd like to know what is intended in the Xen PV case, where there isn't a straight correspondence between pfn and mfn. Is the out-of-range sg->offset intended to refer to the next *pfn* after sg- >page, or to the next *mfn* after sg->page? I confess I've only followed this thread vaguely, but I haven't seen a *coherent* explanation except in the huge page case (in which case I want to see that BUG_ON in the patch) of why this isn't just totally bogus. [-- Attachment #2: smime.p7s --] [-- Type: application/x-pkcs7-signature, Size: 4938 bytes --]
next prev parent reply other threads:[~2017-10-03 12:55 UTC|newest] Thread overview: 33+ messages / expand[flat|nested] mbox.gz Atom feed top 2017-09-28 14:14 [PATCH] iommu/vt-d: Fix scatterlist offset handling Robin Murphy 2017-09-28 16:17 ` Casey Leedom 2017-09-28 13:29 ` Raj, Ashok 2017-09-28 16:59 ` Robin Murphy 2017-09-28 15:43 ` Raj, Ashok 2017-10-03 19:36 ` Raj, Ashok 2017-09-29 8:14 ` Harsh Jain [not found] ` <fe25071a-18bf-e468-01e7-36515f2110e2-ut6Up61K2wZBDgjK7y7TUQ@public.gmane.org> 2017-09-29 16:18 ` Casey Leedom 2017-09-29 16:18 ` Casey Leedom [not found] ` <MWHPR12MB160034E91A834504FE85C07BC87E0-Gy0DoCVfaSVsWITs4OkDoAdYzm3356FpvxpqHgZTriW3zl9H0oFU5g@public.gmane.org> 2017-10-03 12:22 ` Harsh Jain 2017-10-03 22:22 ` Casey Leedom [not found] ` <644c3e01654f8bd48d669c36e424959d6ef0e27e.1506607370.git.robin.murphy-5wv7dgnIgG8@public.gmane.org> 2017-10-03 12:55 ` David Woodhouse [this message] 2017-10-03 12:55 ` David Woodhouse 2017-10-03 18:05 ` Robin Murphy 2017-10-03 22:16 ` David Woodhouse 2017-10-04 11:18 ` Robin Murphy 2017-10-06 14:43 ` Joerg Roedel 2017-10-06 12:54 ` Raj, Ashok [not found] ` <20171006144309.GA30803-zLv9SwRftAIdnm+yROfE0A@public.gmane.org> 2017-11-06 18:47 ` Jacob Pan 2017-11-06 18:47 ` Jacob Pan 2017-11-15 23:54 ` Jacob Pan 2017-11-15 23:54 ` Jacob Pan 2017-11-16 21:32 ` Alex Williamson [not found] ` <20171116143244.2583d044-1yVPhWWZRC1BDLzU/O5InQ@public.gmane.org> 2017-11-16 21:09 ` Raj, Ashok 2017-11-16 21:09 ` Raj, Ashok 2017-11-17 16:18 ` Alex Williamson 2017-11-17 16:18 ` Alex Williamson 2017-11-17 15:48 ` Raj, Ashok 2017-11-17 17:44 ` Casey Leedom 2017-11-17 17:44 ` Casey Leedom 2017-11-17 17:44 ` Casey Leedom [not found] ` <SN1PR12MB035214EF471935B6F4220E36C82F0-z7L1TMIYDg4e2a8M8f4RFAdYzm3356FpvxpqHgZTriW3zl9H0oFU5g@public.gmane.org> 2017-11-17 18:09 ` Jacob Pan 2017-11-17 18:09 ` Jacob Pan
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=1507035334.29211.105.camel@infradead.org \ --to=dwmw2-wegcikhe2lqwvfeawa7xhq@public.gmane.org \ --cc=Harsh-ut6Up61K2wZBDgjK7y7TUQ@public.gmane.org \ --cc=herbert-lOAM2aK0SrRLBo1qDEOMRrpzq4S04n8Q@public.gmane.org \ --cc=iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org \ --cc=joro-zLv9SwRftAIdnm+yROfE0A@public.gmane.org \ --cc=leedom-ut6Up61K2wZBDgjK7y7TUQ@public.gmane.org \ --cc=linux-crypto-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \ --cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \ --cc=robin.murphy-5wv7dgnIgG8@public.gmane.org \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.