| Message ID | 4D42DCD8.2010900@canonical.com |
|---|---|
| State | Accepted |
| Delegated to: | Stefan Bader |
| Headers | show |
On 01/28/2011 08:12 AM, Stefan Bader wrote: > On 01/28/2011 04:01 PM, Tim Gardner wrote: >> On 01/28/2011 07:36 AM, Stefan Bader wrote: >>> Code has changed reasonable since Maverick but same patch applies to >>> all older affected releases. >>> >>> From 2a8309b4f615072025743fae22147be5aa8e86cd Mon Sep 17 00:00:00 2001 >>> From: Linus Torvalds<torvalds@linux-foundation.org> >>> Date: Thu, 28 Oct 2010 15:40:55 +0000 >>> Subject: [PATCH] net: fix rds_iovec page count overflow >>> >>> BugLink: http://bugs.launchpad.net/bugs/709153 >>> CVE-2010-3865 >>> >>> As reported by Thomas Pollet, the rdma page counting can overflow. We >>> get the rdma sizes in 64-bit unsigned entities, but then limit it to >>> UINT_MAX bytes and shift them down to pages (so with a possible "+1" for >>> an unaligned address). >>> >>> So each individual page count fits comfortably in an 'unsigned int' (not >>> even close to overflowing into signed), but as they are added up, they >>> might end up resulting in a signed return value. Which would be wrong. >>> >>> Catch the case of tot_pages turning negative, and return the appropriate >>> error code. >>> >>> Reported-by: Thomas Pollet<thomas.pollet@gmail.com> >>> Signed-off-by: Linus Torvalds<torvalds@linux-foundation.org> >>> Signed-off-by: Andy Grover<andy.grover@oracle.com> >>> Signed-off-by: David S. Miller<davem@davemloft.net> >>> (backported from commit 1b1f693d7ad6d193862dcb1118540a030c5e761f upstream) >>> Signed-off-by: Stefan Bader<stefan.bader@canonical.com> >>> --- >>> net/rds/rdma.c | 10 ++++++++++ >>> 1 files changed, 10 insertions(+), 0 deletions(-) >>> >>> diff --git a/net/rds/rdma.c b/net/rds/rdma.c >>> index 3998967..0a403a7 100644 >>> --- a/net/rds/rdma.c >>> +++ b/net/rds/rdma.c >>> @@ -500,6 +500,16 @@ static struct rds_rdma_op *rds_rdma_prepare(struct >>> rds_sock *rs, >>> >>> max_pages = max(nr, max_pages); >>> nr_pages += nr; >>> + >>> + /* >>> + * nr for one entry in limited to (UINT_MAX>>PAGE_SHIFT)+1 >>> + * so nr_pages cannot overflow without first going negative. >>> + * If nr cannot overflow then max_pages should be ok. >>> + */ >>> + if (nr_pages< 0) { >>> + ret = -EINVAL; >>> + goto out; >>> + } >>> } >>> >>> pages = kcalloc(max_pages, sizeof(struct page *), GFP_KERNEL); >> >> I'm kind of uncomfortable comparing an 'unsigned int' against 0. IIRC the >> results are somewhat compiler dependent. Wouldn't it be clearer if it was 'if >> (nr_pages>= INT_MAX)' ? >> >> rtg > > Likely better. Just made it>INT_MAX (though that should not make that much > difference). > This change behaves as expected. See attached sample program. Acked-by: Tim Gardner <tim.gardner@canonical.com> rtg
On 01/28/2011 07:12 AM, Stefan Bader wrote: > On 01/28/2011 04:01 PM, Tim Gardner wrote: >> On 01/28/2011 07:36 AM, Stefan Bader wrote: >>> Code has changed reasonable since Maverick but same patch applies to >>> all older affected releases. >>> >>> From 2a8309b4f615072025743fae22147be5aa8e86cd Mon Sep 17 00:00:00 2001 >>> From: Linus Torvalds<torvalds@linux-foundation.org> >>> Date: Thu, 28 Oct 2010 15:40:55 +0000 >>> Subject: [PATCH] net: fix rds_iovec page count overflow >>> >>> BugLink: http://bugs.launchpad.net/bugs/709153 >>> CVE-2010-3865 >>> >>> As reported by Thomas Pollet, the rdma page counting can overflow. We >>> get the rdma sizes in 64-bit unsigned entities, but then limit it to >>> UINT_MAX bytes and shift them down to pages (so with a possible "+1" for >>> an unaligned address). >>> >>> So each individual page count fits comfortably in an 'unsigned int' (not >>> even close to overflowing into signed), but as they are added up, they >>> might end up resulting in a signed return value. Which would be wrong. >>> >>> Catch the case of tot_pages turning negative, and return the appropriate >>> error code. >>> >>> Reported-by: Thomas Pollet<thomas.pollet@gmail.com> >>> Signed-off-by: Linus Torvalds<torvalds@linux-foundation.org> >>> Signed-off-by: Andy Grover<andy.grover@oracle.com> >>> Signed-off-by: David S. Miller<davem@davemloft.net> >>> (backported from commit 1b1f693d7ad6d193862dcb1118540a030c5e761f upstream) >>> Signed-off-by: Stefan Bader<stefan.bader@canonical.com> >>> --- >>> net/rds/rdma.c | 10 ++++++++++ >>> 1 files changed, 10 insertions(+), 0 deletions(-) >>> >>> diff --git a/net/rds/rdma.c b/net/rds/rdma.c >>> index 3998967..0a403a7 100644 >>> --- a/net/rds/rdma.c >>> +++ b/net/rds/rdma.c >>> @@ -500,6 +500,16 @@ static struct rds_rdma_op *rds_rdma_prepare(struct >>> rds_sock *rs, >>> >>> max_pages = max(nr, max_pages); >>> nr_pages += nr; >>> + >>> + /* >>> + * nr for one entry in limited to (UINT_MAX>>PAGE_SHIFT)+1 >>> + * so nr_pages cannot overflow without first going negative. >>> + * If nr cannot overflow then max_pages should be ok. >>> + */ >>> + if (nr_pages< 0) { >>> + ret = -EINVAL; >>> + goto out; >>> + } >>> } >>> >>> pages = kcalloc(max_pages, sizeof(struct page *), GFP_KERNEL); >> >> I'm kind of uncomfortable comparing an 'unsigned int' against 0. IIRC the >> results are somewhat compiler dependent. Wouldn't it be clearer if it was 'if >> (nr_pages>= INT_MAX)' ? >> >> rtg > > Likely better. Just made it>INT_MAX (though that should not make that much > difference). > > Acked-by: Brad Figg <brad.figg@canonical.com>
From 45e8e04819666c5ad4c4464f52d2bf59d3728bb5 Mon Sep 17 00:00:00 2001 From: Linus Torvalds <torvalds@linux-foundation.org> Date: Thu, 28 Oct 2010 15:40:55 +0000 Subject: [PATCH] net: fix rds_iovec page count overflow BugLink: http://bugs.launchpad.net/bugs/709153 CVE-2010-3865 As reported by Thomas Pollet, the rdma page counting can overflow. We get the rdma sizes in 64-bit unsigned entities, but then limit it to UINT_MAX bytes and shift them down to pages (so with a possible "+1" for an unaligned address). So each individual page count fits comfortably in an 'unsigned int' (not even close to overflowing into signed), but as they are added up, they might end up resulting in a signed return value. Which would be wrong. Catch the case of tot_pages turning negative, and return the appropriate error code. Reported-by: Thomas Pollet <thomas.pollet@gmail.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org> Signed-off-by: Andy Grover <andy.grover@oracle.com> Signed-off-by: David S. Miller <davem@davemloft.net> (backported from commit 1b1f693d7ad6d193862dcb1118540a030c5e761f upstream) [v2: nr is unsigned in the old code] Signed-off-by: Stefan Bader <stefan.bader@canonical.com> --- net/rds/rdma.c | 11 +++++++++++ 1 files changed, 11 insertions(+), 0 deletions(-) diff --git a/net/rds/rdma.c b/net/rds/rdma.c index 6b09b94..ff5e3c9 100644 --- a/net/rds/rdma.c +++ b/net/rds/rdma.c @@ -473,6 +473,17 @@ static struct rds_rdma_op *rds_rdma_prepare(struct rds_sock *rs, max_pages = max(nr, max_pages); nr_pages += nr; + + /* + * nr for one entry in limited to (UINT_MAX>>PAGE_SHIFT)+1 + * so nr_pages cannot overflow without becoming bigger than + * INT_MAX first. If nr cannot overflow then max_pages should + * be ok. + */ + if (nr_pages > INT_MAX) { + ret = -EINVAL; + goto out; + } } pages = kcalloc(max_pages, sizeof(struct page *), GFP_KERNEL); -- 1.7.0.4