linux-kernel - Re: [PATCH 1/7] lib: Add shared copies of some GCC library routines

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]

Date:   Fri, 02 Jun 2017 19:18:07 -0700 (PDT)
From:   Palmer Dabbelt <palmer@...belt.com>
To:     matt.redfearn@...tec.com
CC:     linux-arch@...r.kernel.org
Subject:     Re: [PATCH 1/7] lib: Add shared copies of some GCC library routines

On Wed, 24 May 2017 01:52:13 PDT (-0700), matt.redfearn@...tec.com wrote:
> Hi Palmer
>
>
> On 23/05/17 23:05, Palmer Dabbelt wrote:
>> Many ports (m32r, microblaze, mips, parisc, score, and sparc) use
>> functionally identical copies of various GCC library routine files,
>> which came up as we were submitting the RISC-V port (which also uses
>> some of these).
>>
>> This patch adds a new copy of these library routine files, which are
>> functionally identical to the various other copies.  These are
>> availiable via Kconfig as CONFIG_LIB_$ROUTINE, which currently isn't
>> used anywhere.
>
> Note that, on MIPS, we had to mark the compiler intrinsics as notrace
> (see commit aedcfbe06558a9f53002e82d5be64c6c94687726) because if the
> compiler requires the intrinsic in the ftrace path, it then tries to
> trace itself leading to infinite recursion and stack overflow. So I'd
> suggest you mark the generic versions notrace as well.

Sorry, I didn't notice that.  I got a bit swamped responding to the RISC-V
port's code reviews, but assuming nobody has merged any of this I'll submit a
v2 patch set that fixes this (and some other errors I believe I introduced)
once I get through all my email.

Thanks!