Re: [tip:x86/cpu] x86/cpu: Strip any /proc/ cpuinfo model name field whitespace
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
- Subject: Re: [tip:x86/cpu] x86/cpu: Strip any /proc/ cpuinfo model name field whitespace
- From: "H. Peter Anvin" <hpa@xxxxxxxxx>
- Date: Thu, 28 May 2015 13:39:11 -0700
- Cc: Prarit Bhargava <prarit@xxxxxxxxxx>, Joe Perches <joe@xxxxxxxxxxx>, luto@xxxxxxxxxxxxxx, peterz@xxxxxxxxxxxxx, dvlasenk@xxxxxxxxxx, torvalds@xxxxxxxxxxxxxxxxxxxx, imammedo@xxxxxxxxxx, brgerst@xxxxxxxxx, mingo@xxxxxxxxxx, dave.hansen@xxxxxxxxxxxxxxx, fenghua.yu@xxxxxxxxx, linux-kernel@xxxxxxxxxxxxxxx, tglx@xxxxxxxxxxxxx, bp@xxxxxxx, linux-tip-commits@xxxxxxxxxxxxxxx
- In-reply-to: <20150528183321.GD31800@pd.tnic>
- User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.7.0
On 05/28/2015 11:33 AM, Borislav Petkov wrote:
> On Thu, May 28, 2015 at 09:57:15AM -0700, H. Peter Anvin wrote:
>> Why?!
>>
>> We are taking about 48 bytes run once per cpu. It isn't worth it to
>> optimize, in fact the extra code size hurts more.
>
> I wanted to save us the redundant copying of the exact same bytes.
> Because when there's no preceding whitespace, p and q point at the same
> thing so we end up doing *p = *p.
>
> OTOH, without the optimization, the code is even simpler.
>
> I can remove it if you wanna - I don't care all that much.
>
Yes, please. Actually, with a test inside the loop the way you have it,
the resulting code will almost certainly be slower -- a redundant write
to an already dirty cache line is way cheaper than a branch.
-hpa
--
To unsubscribe from this list: send the line "unsubscribe linux-tip-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
[Index of Archives]
[Linux Stable Commits]
[Linux Stable Kernel]
[Linux Kernel]
[Linux USB Devel]
[Linux Video &Media]
[Linux Audio Users]
[Yosemite News]
[Linux SCSI]