Heap randomisation causes PCH testsuite failures

[Michael Hope]

See the upstream ticket. Logged in Linaro to track the backport.

[Pinskia (pinskia)]

Most likely when the linux arm kernel decided to randomize the heap.

[Mikpe (mikpe)]

I assume other Linux archs already randomize the heap. If so, why would that cause PCH failures on ARM? Could it be because there's no TRY_EMPTY_VM_SPACE
definition in gcc/config/host-linux.c for ARM?

[Pinskia (pinskia)]

(In reply to comment #2)
> I assume other Linux archs already randomize the heap.
 Could it be because there's no TRY_EMPTY_VM_SPACE definition in gcc/config/host-linux.c for ARM?

yes, a definition for TRY_EMPTY_VM_SPACE is needed for ARM. I just went through this for MIPS but I have not pushed it because I don't currently have a correct copyright assignment on file.

[Mikpe (mikpe)]

Ok. I'll cook something up for TRY_EMPTY_VM_SPACE on ARM.

[Mikpe (mikpe)]

(In reply to comment #1)
> Most likely when the linux arm kernel decided to randomize the heap.

Confirmed. Bisection of the kernel identified:

From: Nicolas Pitre <email address hidden>
Date: Tue, 15 Jun 2010 01:16:19 +0000 (-0400)
Subject: [ARM] add address randomization to mmap()
X-Git-Tag: v2.6.36-rc1~591^2^2~6^2~2
X-Git-Url: http://git.kernel.org/?p=linux%2Fkernel%2Fgit%2Ftorvalds%2Flinux-2.6.git;a=commitdiff_plain;h=cc92c28b2db5b406657ecc05235d4ca4e222ae34

[ARM] add address randomization to mmap()

Signed-off-by: Nicolas Pitre <email address hidden>
---

diff --git a/arch/arm/mm/mmap.c b/arch/arm/mm/mmap.c
index f5abc51..4f5b396 100644
--- a/arch/arm/mm/mmap.c
+++ b/arch/arm/mm/mmap.c
@@ -7,6 +7,7 @@
 #include <linux/shm.h>
 #include <linux/sched.h>
 #include <linux/io.h>
+#include <linux/random.h>
 #include <asm/cputype.h>
 #include <asm/system.h>

@@ -80,6 +81,9 @@ arch_get_unmapped_area(struct file *filp, unsigned long addr,
                start_addr = addr = TASK_UNMAPPED_BASE;
                mm->cached_hole_size = 0;
        }
+ /* 8 bits of randomness in 20 address space bits */
+ if (current->flags & PF_RANDOMIZE)
+ addr += (get_random_int() % (1 << 8)) << PAGE_SHIFT;

 full_search:
        if (do_align)

as the cause of this regression. Still, I think the kernel is Ok and it's really gcc's pch implementation which is broken.

[Mikpe (mikpe)]

Created attachment 22030
define TRY_EMPTY_VM_SPACE for ARM

Preliminary patch which defines TRY_EMPTY_VM_SPACE for ARM and fixes the PCH regressions in a PCH-only test run. I'll submit it after some more testing.

[Michael Hope (michaelh1)]

See the upstream ticket. Logged in Linaro to track the backport.

[Mikpe (mikpe)]

The patch tested ok and has now been submitted:
http://gcc.gnu.org/ml/gcc-patches/2010-10/msg02252.html

[Michael Hope (michaelh1)]

Patch sent upstream at:
 http://gcc.gnu.org/ml/gcc-patches/2010-10/msg02252.html

No ping, no acceptance, not in 4.6 as at r166926.

[Michael Hope (michaelh1)]

Asked the original author if he could ping it, else if we could take it over.

[Michael Hope (michaelh1)]

Ping is at:
 http://gcc.gnu.org/ml/gcc-patches/2010-11/msg02003.html

[Michael Hope (michaelh1)]

Currently stuck due to copyright assignment issues. Ubuntu could pull in this patch if they wanted.

Will continue to monitor.

[Ibolton (ibolton)]

Please can you confirm that this is now fixed in trunk, Mikael.

And has it been backported?

[Mikpe (mikpe)]

(In reply to comment #8)
> Please can you confirm that this is now fixed in trunk, Mikael.

No, it has not been fixed on trunk.

> And has it been backported?

Not in the official gcc repository. I do use the fix in my private 4.4 and 4.5 based compilers.

[Ira Rosen (irar)]

I am out of the office until 17/04/2011.

Note: This is an automated response to your message "[Bug 660156] Re: Heap
randomisation causes PCH testsuite failures" sent on 15/4/2011 5:41:01.

This is the only notification you will receive while this person is away.

[Ramana-gcc (ramana-gcc)]

Author: ramana
Date: Thu May 5 09:52:12 2011
New Revision: 173421

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=173421
Log:

For Michael Hope.

2011-05-05 Michael Hope <email address hidden>

       PR pch/45979
       * config/host-linux.c (TRY_EMPTY_VM_SPACE): Define for
       __ARM_EABI__ hosts.

Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/config/host-linux.c

[Pinskia (pinskia)]

Fixed.

[Michaelh-m (michaelh-m)]

Author: michaelh
Date: Fri Mar 23 02:19:46 2012
New Revision: 185724

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=185724
Log:
2012-03-16 Michael Hope <email address hidden>

 Backport from mainline
 2011-05-05 Michael Hope <email address hidden>

 PR pch/45979
 * config/host-linux.c (TRY_EMPTY_VM_SPACE): Define for
 __ARM_EABI__ hosts.

Modified:
    branches/gcc-4_6-branch/gcc/ChangeLog
    branches/gcc-4_6-branch/gcc/config/host-linux.c