[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <35205870-97E8-4E8E-B4EF-9961878FB591@zytor.com>
Date: Mon, 06 Mar 2017 10:53:57 -0800
From: hpa@...or.com
To: Logan Gunthorpe <logang@...tatee.com>,
Borislav Petkov <bp@...e.de>,
Linus Torvalds <torvalds@...ux-foundation.org>
CC: Thomas Gleixner <tglx@...utronix.de>,
Ingo Molnar <mingo@...hat.com>,
Tony Luck <tony.luck@...el.com>,
Al Viro <viro@...iv.linux.org.uk>,
the arch/x86 maintainers <x86@...nel.org>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: Question Regarding ERMS memcpy
On March 6, 2017 9:12:41 AM PST, Logan Gunthorpe <logang@...tatee.com> wrote:
>
>
>On 06/03/17 12:28 AM, H. Peter Anvin wrote:
>> On 03/05/17 23:01, Logan Gunthorpe wrote:
>>>
>>> On 05/03/17 12:54 PM, Borislav Petkov wrote:
>>>> Logan, wanna give that a try, see if it takes care of your issue?
>>>
>>> Well honestly my issue was solved by fixing my kernel config. I have
>no
>>> idea why I had optimize for size in there in the first place.
>>>
>>
>> Yes, to gcc "optimize for size" means exactly that... intended for
>cases
>> where saving storage (e.g. ROM) or code download time is paramount.
>
>I agree and understand, however placing a poorly performing _inline_
>memcpy instead of a single call instruction to a performant memcopy
>probably took more code space in the end. So like Linus, I just have to
>scratch my head at the -Os optimization option.
>
>Logan
No, it will be smaller: -Os counts bytes.
If you think about it, there is no way that replacing a five-byte subroutine call with a two-byte instruction opcode can make it bigger! The only other difference between the two from a size perspective is that the compiler doesn't have to worry about clobbered registers other than the argument registers.
--
Sent from my Android device with K-9 Mail. Please excuse my brevity.
Powered by blists - more mailing lists