linux-kernel - Re: [PATCH] powerpc: force inlining of csum_partial() to avoid multiple csum

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <1f745f94-c819-b708-815e-24946621a29d@csgroup.eu>
Date:   Thu, 15 Oct 2020 15:32:06 +0200
From:   Christophe Leroy <christophe.leroy@...roup.eu>
To:     Segher Boessenkool <segher@...nel.crashing.org>
Cc:     Benjamin Herrenschmidt <benh@...nel.crashing.org>,
        Paul Mackerras <paulus@...ba.org>,
        Michael Ellerman <mpe@...erman.id.au>,
        linuxppc-dev@...ts.ozlabs.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] powerpc: force inlining of csum_partial() to avoid
 multiple csum_partial() with GCC10



Le 15/10/2020 à 15:25, Segher Boessenkool a écrit :
> Hi!
> 
> On Thu, Oct 15, 2020 at 10:52:20AM +0000, Christophe Leroy wrote:
>> With gcc9 I get:
> 
> <snip>
> 
>> With gcc10 I get:
> 
> <snip>
> 
>> gcc10 defines multiple versions of csum_partial() which are just
>> an unconditionnal branch to __csum_partial().
> 
> It doesn't inline it, yes.
> 
> Could you open a GCC PR for this please?

Sure.

I also have get_order() 75 times in my vmlinux, all the same as the following:

c0016790 <get_order>:
c0016790:	38 63 ff ff 	addi    r3,r3,-1
c0016794:	54 63 a3 3e 	rlwinm  r3,r3,20,12,31
c0016798:	7c 63 00 34 	cntlzw  r3,r3
c001679c:	20 63 00 20 	subfic  r3,r3,32
c00167a0:	4e 80 00 20 	blr

Christophe