Re: Preempt Threshold Measurements

[William Lee Irwin III]

On Mon, Jul 12, 2004 at 07:43:25PM -0400, Gabriel Devenyi wrote:
> Keeping in mind that I'm using the nvidia-kernel drivers, here are my preempt 
> threshold violations.
> 6ms non-preemptible critical section violated 4 ms preempt threshold starting 
> at kernel_fpu_begin+0xd/0x50 and ending at _mmx_memcpy+0x127/0x170
>  [<c0241987>] _mmx_memcpy+0x127/0x170
>  [<c01163b0>] dec_preempt_count+0x110/0x120
>  [<c0241987>] _mmx_memcpy+0x127/0x170
>  [<c012d3b5>] load_module+0x835/0x900
>  [<c013c84e>] unmap_vmas+0x10e/0x1f0
>  [<c012d4fb>] sys_init_module+0x7b/0x230
>  [<c0103ee1>] sysenter_past_esp+0x52/0x71

Things tend to be slow and stupid in the interest of robustness during
system initialization.

I'd suggest ignoring those unless you're specifically interested in boot
time (in which case you should be doing things for yourself) and focusing
on ones reported during normal usage after the system is up.


-- wli
-
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/