Re: Kmemcheck issue

[Vegard Nossum]

On 19 April 2010 15:23, naveen yadav <yad.naveen@xxxxxxxxx> wrote:
> Hi,
>
> à à à I Ãam Ãtrying to à run ÃKmemcheck on à a ÃCortex ( ARM-v7)
> board Ãrunning ÃLinux 2.6.30.9
>
> At boot-up time, Ãas Ãsoon Ãas ÃTimer-device is Ãenabled, Ã I Ãfind
> that à timer Ãgets Ãdisabled Ãbecause Ãsystem
> gets spurious Ãinterrupts Ãon Ãthat Ãirq Ãline.
>
>
> PID hash table entries: 1024 (order: 10, 4096 bytes)
> Linux System timer initialize
> irq 23: nobody cared (try booting with the "irqpoll" option)
> [<c002e148>] (unwind_backtrace+0x0/0xe8) from [<c002c4b8>]
> (dump_stack+0x18/0x1c)
> [<c002c4b8>] (dump_stack+0x18/0x1c) from [<c006ad74>]
> (__report_bad_irq+0x80/0xb0)
> [<c006ad74>] (__report_bad_irq+0x80/0xb0) from [<c006af28>]
> (note_interrupt+0x184/0x1f0)
> [<c006af28>] (note_interrupt+0x184/0x1f0) from [<c006ba38>]
> (handle_edge_irq+0xe8/0x160)
> [<c006ba38>] (handle_edge_irq+0xe8/0x160) from [<c0027060>]
> (asm_do_IRQ+0x60/0x74)
> [<c0027060>] (asm_do_IRQ+0x60/0x74) from [<c0027a30>] (__irq_svc+0x30/0x80)
> Exception stack(0xc0373ec0 to 0xc0373f08)
> 3ec0: 00000017 c0372000 00000002 00000000 00000002 c038e8f4 00000000 0aa00000
> 3ee0: c0372000 c03bfc40 0000001f c0373f44 00000100 c0373f08 c0047ffc c0047c98
> 3f00: 20000153 ffffffff
> [<c0027a30>] (__irq_svc+0x30/0x80) from [<c0047ffc>] (irq_exit+0x44/0x4c)
> [<c0047ffc>] (irq_exit+0x44/0x4c) from [<c0047ffc>] (irq_exit+0x44/0x4c)
> [<c0047ffc>] (irq_exit+0x44/0x4c) from [<c0027064>] (asm_do_IRQ+0x64/0x74)
> [<c0027064>] (asm_do_IRQ+0x64/0x74) from [<c0027a30>] (__irq_svc+0x30/0x80)
> Exception stack(0xc0373f78 to 0xc0373fc0)
> 3f60: Ã Ã Ã Ã Ã Ã Ã Ã Ã Ã Ã Ã Ã Ã Ã Ã Ã Ã Ã Ã Ã Ã Ã Ã Ã Ã Ã 00000000 c037c15c
> 3f80: 00000001 600001d3 955fffff c00242ec c0399f80 0aa00000 c0023000 412fc082
> 3fa0: 0000001f c0373ff4 c0373fc0 c0373fc0 c0008974 c0019798 20000153 ffffffff
> [<c0027a30>] (__irq_svc+0x30/0x80) from [<c0008974>] (start_kernel+0x168/0x3e4)
> [<c0008974>] (start_kernel+0x168/0x3e4) from [<60008034>] (0x60008034)
> handlers:
> [<c003a604>] (sdp_timer_interrupt+0x0/0x38)
> Disabling IRQ #23
> Console: colour dummy device 80x30
> console [ttyS1] enabled
> Dentry cache hash table entries: 32768 (order: 5, 131072 bytes)
> Inode-cache hash table entries: 16384 (order: 4, 65536 bytes)
>
>
> Later, Ãwhen Ãcalibrating delay Ãloop, Ãthe system is hung Ãas timer
> is not working.
>
> The problem happens Ãeven if ÃI Ãset kmemcheck_enabled = 0 at Ãboot.
> ( I also commented out à à pagetable_init(&meminfo) Ã).
>
> ÃIf Ãyou have seen this before, Ãplease let me know.
>
> ( I know that Ãyour patch was written for à ARM v6. à à My cpu is
> Cortex . Ã ÃI am aware that ÃI may have to make some change in the
> Ãinstruction simulation handling. )
>
> Thanks

Hi,

First up, I see a lot of funny à characters in your text.

I think that this is not about kmemcheck, but kmemleak. kmemcheck is
for catching references to uninitialized memory while kmemleak is for
detecting memory leaks. kmemcheck only works on x86 so far (with an
ARM patch by Ye Janboe, but it is not in mainline). kmemleak, however,
was written on ARM.

I put Catalin Marinas (kmemleak author) on Cc.

Please correct me if I am wrong. Good luck.


Vegard
--
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/