Re: [GIT PULL] x86/vdso changes for v3.1

[Andrew Lutomirski]

On Thu, Aug 25, 2011 at 1:58 AM, Arnaud Lacombe <lacombar@xxxxxxxxx> wrote:
> Hi,
>
> On Fri, Jul 22, 2011 at 11:42 AM, Ingo Molnar <mingo@xxxxxxx> wrote:
>> Linus,
>>
>> Please pull the latest x86-vdso-for-linus git tree from:
>>
>>   git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git x86-vdso-for-linus
>>
>>
>> out-of-topic modifications in x86-vdso-for-linus:
>> -------------------------------------------------
>> arch/ia64/Kconfig                  # ae7bd11: clocksource: Change __ARCH_HAS_CL
>> arch/ia64/include/asm/clocksource.h# ae7bd11: clocksource: Change __ARCH_HAS_CL
>>                                   # 574c44f: ia64: Replace clocksource.fsys_mm
>> arch/ia64/kernel/cyclone.c         # 574c44f: ia64: Replace clocksource.fsys_mm
>> arch/ia64/kernel/time.c            # 574c44f: ia64: Replace clocksource.fsys_mm
>> arch/ia64/sn/kernel/sn2/timer.c    # 574c44f: ia64: Replace clocksource.fsys_mm
>> drivers/char/hpet.c                # 574c44f: ia64: Replace clocksource.fsys_mm
>> include/linux/clocksource.h        # ae7bd11: clocksource: Change __ARCH_HAS_CL
>>                                   # 574c44f: ia64: Replace clocksource.fsys_mm
>>                                   # 433bd80: clocksource: Replace vread with g
>> include/linux/seccomp.h            # 5cec93c: x86-64: Emulate legacy vsyscalls
>>
>>  Thanks,
>>
>>        Ingo
>>
>> ------------------>
>> Andy Lutomirski (17):
>>      x86-64: Fix alignment of jiffies variable
>>      x86-64: Document some of entry_64.S
>>      x86-64: Give vvars their own page
>>      x86-64: Remove kernel.vsyscall64 sysctl
>>      x86-64: Map the HPET NX
>>      x86-64: Remove vsyscall number 3 (venosys)
>>      x86-64: Fill unused parts of the vsyscall page with 0xcc
>>      x86-64: Emulate legacy vsyscalls
>>      x86-64: Improve vsyscall emulation CS and RIP handling
>>      x86: Make alternative instruction pointers relative
>
> Bisecting this BUG:
>
> CPU: Intel QEMU Virtual CPU version 0.13.0 stepping 03
> BUG: unable to handle kernel paging request at 8277e54c
> IP: [<c13765fb>] apply_alternatives+0xc3/0x18a
> *pde = 00000000
> Oops: 0000 [#1] DEBUG_PAGEALLOC
>
> Pid: 0, comm: swapper Not tainted 3.1.0-rc3 #20 Bochs Bochs
> EIP: 0060:[<c13765fb>] EFLAGS: 00000286 CPU: 0
> EIP is at apply_alternatives+0xc3/0x18a
> EAX: ffffffff EBX: c13bee40 ECX: 00000005 EDX: 8277e54c
> ESI: 8277e54c EDI: c1349e9e EBP: c1349fa8 ESP: c1349e70
>  DS: 007b ES: 007b FS: 0000 GS: 0000 SS: 0068
> Process swapper (pid: 0, ti=c1348000 task=c134e340 task.ti=c1348000)
> Stack:
>  00000002 c135d7c0 c11418a0 00000db8 c1349e9e 82655b12 00000005 8277e54c
>  c13bf6ec c1349e05 05000086 f6890046 00f69000 c1349eb0 00000086 00000000
>  00000046 c1349ec0 00000046 00000000 c1349edc c101f29c 00000dc5 00000000
> Call Trace:
>  [<c11418a0>] ? serial8250_start_tx+0xe0/0xe0
>  [<c101f29c>] ? console_unlock+0x10c/0x150
>  [<c10039c3>] ? do_IRQ+0x43/0xa0
>  [<c12975a9>] ? common_interrupt+0x29/0x30
>  [<c1294d59>] ? printk+0x18/0x1f
>  [<c139070b>] ? print_cpu_info+0x93/0x11f
>  [<c13766d9>] alternative_instructions+0x17/0x32
>  [<c1377552>] check_bugs+0x91/0x93
>  [<c13716b6>] start_kernel+0x289/0x294
>  [<c13711b7>] ? loglevel+0x1a/0x1a
>  [<c13710b3>] i386_start_kernel+0xb3/0xbb
> Code: ec fe ff ff 8d 0c 0b 89 8d dc fe ff ff 8d 54 13 04 8b bd d8 fe
> ff ff 89 95 e4 fe ff ff 89 b5 e0 fe ff ff 8b 8d e0 fe ff ff 89 d6 <f3>
> a4 80 bd f6 fe ff ff e8 75 17 80 bd ec fe ff ff 05 75 0e 89
> EIP: [<c13765fb>] apply_alternatives+0xc3/0x18a SS:ESP 0068:c1349e70
> CR2: 000000008277e54c
> ---[ end trace 4eaa2a86a8e2da22 ]---
> Kernel panic - not syncing: Attempted to kill the idle task!
> Pid: 0, comm: swapper Tainted: G      D     3.1.0-rc3 #20
> Call Trace:
>  [<c1294d59>] ? printk+0x18/0x1f
>  [<c1294c4a>] panic+0x57/0x14e
>  [<c10221a1>] do_exit+0x201/0x290
>  [<c1004ace>] oops_end+0x6e/0x90
>  [<c1294d59>] ? printk+0x18/0x1f
>  [<c1013efc>] no_context+0xbc/0x150
>  [<c1013fc0>] __bad_area_nosemaphore+0x30/0x170
>  [<c1014450>] ? vmalloc_sync_all+0xf0/0xf0
>  [<c1014112>] bad_area_nosemaphore+0x12/0x20
>  [<c101466f>] do_page_fault+0x21f/0x360
>  [<c105c070>] ? get_page_from_freelist+0x110/0x300
>  [<c1023b24>] ? irq_exit+0x54/0x90
>  [<c10039c3>] ? do_IRQ+0x43/0xa0
>  [<c1140e31>] ? wait_for_xmitr+0x31/0x90
>  [<c1014450>] ? vmalloc_sync_all+0xf0/0xf0
>  [<c1296df8>] error_code+0x58/0x60
>  [<c1014450>] ? vmalloc_sync_all+0xf0/0xf0
>  [<c13765fb>] ? apply_alternatives+0xc3/0x18a
>  [<c11418a0>] ? serial8250_start_tx+0xe0/0xe0
>  [<c101f29c>] ? console_unlock+0x10c/0x150
>  [<c10039c3>] ? do_IRQ+0x43/0xa0
>  [<c12975a9>] ? common_interrupt+0x29/0x30
>  [<c1294d59>] ? printk+0x18/0x1f
>  [<c139070b>] ? print_cpu_info+0x93/0x11f
>  [<c13766d9>] alternative_instructions+0x17/0x32
>  [<c1377552>] check_bugs+0x91/0x93
>  [<c13716b6>] start_kernel+0x289/0x294
>  [<c13711b7>] ? loglevel+0x1a/0x1a
>  [<c13710b3>] i386_start_kernel+0xb3/0xbb
> QEMU: Terminated
>
> lead me to that commit:
>
> commit 59e97e4d6fbcd5b74a94cb48bcbfc6f8478a5e93
> Author: Andy Lutomirski <luto@xxxxxxx>
> Date:   Wed Jul 13 09:24:10 2011 -0400
>
>    x86: Make alternative instruction pointers relative
>
>    This save a few bytes on x86-64 and means that future patches can
>    apply alternatives to unrelocated code.
>
>    Signed-off-by: Andy Lutomirski <luto@xxxxxxx>
>    Link: http://lkml.kernel.org/r/ff64a6b9a1a3860ca4a7b8b6dc7b4754f9491cd7.1310563276.git.luto@xxxxxxx
>    Signed-off-by: H. Peter Anvin <hpa@xxxxxxxxxxxxxxx>
>

I probably missed a hardcoded alternative entry that depends on
config.  Can you send me your System.map and the output of objdump -s
-j .altinstructions vmlinux?

I'm also having issues building from your .config.  Building
59e97e4d6fbcd5b74a94cb48bcbfc6f8478a5e93 asks me about Moorestown MID
platform, Ethernet, etc.

--Andy
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/