| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-Id: <20111028003935.a75d16b6.akpm@linux-foundation.org>
Date: Fri, 28 Oct 2011 00:39:35 -0700
From: Andrew Morton <akpm@...ux-foundation.org>
To: Ingo Molnar <mingo@...e.hu>
Cc: Konrad Rzeszutek Wilk <konrad.wilk@...cle.com>,
linux-kernel@...r.kernel.org, a.p.zijlstra@...llo.nl,
hpa@...or.com, jeremy.fitzhardinge@...rix.com, mingo@...hat.com,
stable@...nel.org, tglx@...utronix.de,
Stephen Rothwell <sfr@...b.auug.org.au>,
Rusty Russell <rusty@...tcorp.com.au>
Subject: Re: Not really merged? Re: [merged]
x86-paravirt-pte-updates-in-kunmap_atomic-need-to-be-synchronous-regardless-of-lazy_mmu-mode.patch
removed from -mm tree
On Fri, 28 Oct 2011 09:08:39 +0200 Ingo Molnar <mingo@...e.hu> wrote:
>
> * Andrew Morton <akpm@...ux-foundation.org> wrote:
>
> > On Tue, 25 Oct 2011 14:24:50 -0400
> > Konrad Rzeszutek Wilk <konrad.wilk@...cle.com> wrote:
> >
> > > On Fri, Oct 14, 2011 at 12:51:48PM -0700, akpm@...gle.com wrote:
> > > >
> > > > The patch titled
> > > > Subject: x86/paravirt: PTE updates in k(un)map_atomic need to be synchronous, regardless of lazy_mmu mode
> > > > has been removed from the -mm tree. Its filename was
> > > > x86-paravirt-pte-updates-in-kunmap_atomic-need-to-be-synchronous-regardless-of-lazy_mmu-mode.patch
> > > >
> > > > This patch was dropped because it was merged into mainline or a subsystem tree
> > >
> > > Hey Andrew,
> > >
> > > I am actually not seeing this in mainline? Was it accidently dropped out of your tree?
> >
> > hm, well spotted. I'm not sure what happened here - possibly the
> > patch was merged into an x86 tree (and hence linux-next) but later
> > got lost. Or possibly not, and I just screwed up.
>
> No, a patch with the -i 'paravirt.*lazy' pattern never touched -tip,
> even temporarily.
>
> Could it be that someone else (say the Xen guys) picked it up, it
> went into linux-next, you thought it's applied - but then they
> dropped it?
>
> Do we have a full log of all linux-next patches?
Don't know.
The patch was present in the linux-next which I pulled on 14 Oct. It
is no longer in linux-next.
Here's my copy:
From: Konrad Rzeszutek Wilk <konrad.wilk@...cle.com>
Subject: x86/paravirt: PTE updates in k(un)map_atomic need to be synchronous, regardless of lazy_mmu mode
Fix an outstanding issue that has been reported since 2.6.37. Under a
heavy loaded machine processing "fork()" calls could crash with:
BUG: unable to handle kernel paging request at f573fc8c
IP: [<c01abc54>] swap_count_continued+0x104/0x180
*pdpt = 000000002a3b9027 *pde = 0000000001bed067 *pte = 0000000000000000
Oops: 0000 [#1] SMP
Modules linked in:
Pid: 1638, comm: apache2 Not tainted 3.0.4-linode37 #1
EIP: 0061:[<c01abc54>] EFLAGS: 00210246 CPU: 3
EIP is at swap_count_continued+0x104/0x180
.. snip..
Call Trace:
[<c01ac222>] ? __swap_duplicate+0xc2/0x160
[<c01040f7>] ? pte_mfn_to_pfn+0x87/0xe0
[<c01ac2e4>] ? swap_duplicate+0x14/0x40
[<c01a0a6b>] ? copy_pte_range+0x45b/0x500
[<c01a0ca5>] ? copy_page_range+0x195/0x200
[<c01328c6>] ? dup_mmap+0x1c6/0x2c0
[<c0132cf8>] ? dup_mm+0xa8/0x130
[<c013376a>] ? copy_process+0x98a/0xb30
[<c013395f>] ? do_fork+0x4f/0x280
[<c01573b3>] ? getnstimeofday+0x43/0x100
[<c010f770>] ? sys_clone+0x30/0x40
[<c06c048d>] ? ptregs_clone+0x15/0x48
[<c06bfb71>] ? syscall_call+0x7/0xb
The problem is that in copy_page_range() we turn lazy mode on, and then in
swap_entry_free() we call swap_count_continued() which ends up in:
map = kmap_atomic(page, KM_USER0) + offset;
and then later we touch *map.
Since we are running in batched mode (lazy) we don't actually set up the
PTE mappings and the kmap_atomic is not done synchronously and ends up
trying to dereference a page that has not been set.
Looking at kmap_atomic_prot_pfn(), it uses 'arch_flush_lazy_mmu_mode' and
doing the same in kmap_atomic_prot() and __kunmap_atomic() makes the problem
go away.
Interestingly, commit b8bcfe997e4615 ("x86/paravirt: remove lazy mode in
interrupts") removed part of this to fix an interrupt issue - but it went
to far and did not consider this scenario.
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@...cle.com>
Cc: Thomas Gleixner <tglx@...utronix.de>
Cc: Ingo Molnar <mingo@...hat.com>
Cc: "H. Peter Anvin" <hpa@...or.com>
Cc: Peter Zijlstra <a.p.zijlstra@...llo.nl>
Cc: Jeremy Fitzhardinge <jeremy.fitzhardinge@...rix.com>
Cc: <stable@...nel.org>
Signed-off-by: Andrew Morton <akpm@...ux-foundation.org>
---
arch/x86/mm/highmem_32.c | 2 ++
1 file changed, 2 insertions(+)
diff -puN arch/x86/mm/highmem_32.c~x86-paravirt-pte-updates-in-kunmap_atomic-need-to-be-synchronous-regardless-of-lazy_mmu-mode arch/x86/mm/highmem_32.c
--- a/arch/x86/mm/highmem_32.c~x86-paravirt-pte-updates-in-kunmap_atomic-need-to-be-synchronous-regardless-of-lazy_mmu-mode
+++ a/arch/x86/mm/highmem_32.c
@@ -45,6 +45,7 @@ void *kmap_atomic_prot(struct page *page
vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx);
BUG_ON(!pte_none(*(kmap_pte-idx)));
set_pte(kmap_pte-idx, mk_pte(page, prot));
+ arch_flush_lazy_mmu_mode();
return (void *)vaddr;
}
@@ -88,6 +89,7 @@ void __kunmap_atomic(void *kvaddr)
*/
kpte_clear_flush(kmap_pte-idx, vaddr);
kmap_atomic_idx_pop();
+ arch_flush_lazy_mmu_mode();
}
#ifdef CONFIG_DEBUG_HIGHMEM
else {
_
> But IMO it's at least as important to figure out what went wrong. I
> somehow doubt it that you spuriously dropped it - that someone else
> messes up has a far higher likelihood.
My drop was legitimate.
Here's the commit from the Oct 14 linux-next:
commit ab67482036cee590753dd42b7f66aada97e6dcde
Author: Konrad Rzeszutek Wilk <konrad.wilk@...cle.com>
AuthorDate: Fri Sep 23 17:02:29 2011 -0400
Commit: Konrad Rzeszutek Wilk <konrad.wilk@...cle.com>
CommitDate: Mon Sep 26 09:12:37 2011 -0400
x86/paravirt: Partially revert "remove lazy mode in interrupts"
which has git commit b8bcfe997e46150fedcc3f5b26b846400122fdd9.
The unintended consequence of removing the flushing of MMU
updates when doing kmap_atomic (or kunmap_atomic) is that we can
hit a dereference bug when processing a "fork()" under a heavy loaded
machine. Specifically we can hit:
BUG: unable to handle kernel paging request at f573fc8c
IP: [<c01abc54>] swap_count_continued+0x104/0x180
*pdpt = 000000002a3b9027 *pde = 0000000001bed067 *pte = 0000000000000000
Oops: 0000 [#1] SMP
Modules linked in:
Pid: 1638, comm: apache2 Not tainted 3.0.4-linode37 #1
EIP: 0061:[<c01abc54>] EFLAGS: 00210246 CPU: 3
EIP is at swap_count_continued+0x104/0x180
.. snip..
Call Trace:
[<c01ac222>] ? __swap_duplicate+0xc2/0x160
[<c01040f7>] ? pte_mfn_to_pfn+0x87/0xe0
[<c01ac2e4>] ? swap_duplicate+0x14/0x40
[<c01a0a6b>] ? copy_pte_range+0x45b/0x500
[<c01a0ca5>] ? copy_page_range+0x195/0x200
[<c01328c6>] ? dup_mmap+0x1c6/0x2c0
[<c0132cf8>] ? dup_mm+0xa8/0x130
[<c013376a>] ? copy_process+0x98a/0xb30
[<c013395f>] ? do_fork+0x4f/0x280
[<c01573b3>] ? getnstimeofday+0x43/0x100
[<c010f770>] ? sys_clone+0x30/0x40
[<c06c048d>] ? ptregs_clone+0x15/0x48
[<c06bfb71>] ? syscall_call+0x7/0xb
The problem looks that in copy_page_range we turn lazy mode on, and then
in swap_entry_free we call swap_count_continued which ends up in:
map = kmap_atomic(page, KM_USER0) + offset;
and then later touches *map.
Since we are running in batched mode (lazy) we don't actually set up the
PTE mappings and the kmap_atomic is not done synchronously and ends up
trying to dereference a page that has not been set.
Looking at kmap_atomic_prot_pfn, it uses 'arch_flush_lazy_mmu_mode' and
sprinkling that in kmap_atomic_prot and __kunmap_atomic makes the problem
go away.
CC: Thomas Gleixner <tglx@...utronix.de>
CC: Ingo Molnar <mingo@...hat.com>
CC: "H. Peter Anvin" <hpa@...or.com>
CC: x86@...nel.org
CC: Peter Zijlstra <a.p.zijlstra@...llo.nl>
CC: Jeremy Fitzhardinge <jeremy.fitzhardinge@...rix.com>
CC: stable@...nel.org
Signed-off-by: Konrad Rzeszutek Wilk <konrad.wilk@...cle.com>
diff --git a/arch/x86/mm/highmem_32.c b/arch/x86/mm/highmem_32.c
index b499626..f4f29b1 100644
--- a/arch/x86/mm/highmem_32.c
+++ b/arch/x86/mm/highmem_32.c
@@ -45,6 +45,7 @@ void *kmap_atomic_prot(struct page *page, pgprot_t prot)
vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx);
BUG_ON(!pte_none(*(kmap_pte-idx)));
set_pte(kmap_pte-idx, mk_pte(page, prot));
+ arch_flush_lazy_mmu_mode();
return (void *)vaddr;
}
@@ -88,6 +89,7 @@ void __kunmap_atomic(void *kvaddr)
*/
kpte_clear_flush(kmap_pte-idx, vaddr);
kmap_atomic_idx_pop();
+ arch_flush_lazy_mmu_mode();
}
#ifdef CONFIG_DEBUG_HIGHMEM
else {
I'm not sure what to make of that. The signoffs imply that Konrad is
running his own git tree, but I don't think he is. Or someone (Jeremy
or Rusty I think) merged it but didn't add a signoff.
Note that the patch was merged using its old name "x86/paravirt:
Partially revert "remove lazy mode in interrupts"". The patch got
renamed to "x86/paravirt: PTE updates in k(un)map_atomic need to be
synchronous, regardless of lazy_mmu mode" and perhaps this prompted
someone to drop the old-named patch then lose the new-named one.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists