Re: [PATCH 1/1] elevator: indirect function calls reducing

[Antonio Vargas]

On 11/30/05, Ananiev, Leonid I <leonid.i.ananiev@xxxxxxxxx> wrote:
> Christoph,
> During that function calls 3 memory ridings are performed under
> spin_lock and having cache miss/conflict problem;
> and 2 only main memory ridings after patching.
>
> In a source a[b][c][d](arg);
> after patching number of memory ridings less by 1:
> a[c][d](arg);
> Do you agree with it?
> Have you other explanation of performance degradation 2.6.9 -> 2.6.10?
>
> Leonid
>

Leonid, you are "just" removing a memory fetch by embeding the struct
instead of pointing to it, not removing a whole indirect jump...
granted it's good to remove innecesary mem-fetchs, but then please
call the patch that, a removal of not-necessary mem-fetches.

>
> -----Original Message-----
> From: Christoph Hellwig [mailto:hch@xxxxxxxxxxxxx]
> Sent: Wednesday, November 30, 2005 7:26 PM
> To: Ananiev, Leonid I
> Cc: linux-kernel@xxxxxxxxxxxxxxx; axboe@xxxxxxx
> Subject: Re: [PATCH 1/1] elevator: indirect function calls reducing
>
>
> this _still_ isn't an indirect function call reduction and people
> have told you N times.  Please get your basics right first, to start
> with the patch description.
> -
> 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/
>



--
Greetz, Antonio Vargas aka winden of network

http://wind.codepixel.com/
windNOenSPAMntw@xxxxxxxxx
thesameasabove@xxxxxxxxxxxxx

Every day, every year
you have to work
you have to study
you have to scene.
-
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/