linux-kernel - Re: [PATCH] MIPS: VDSO: Always select -msoft-float

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, 4 Nov 2016 11:33:54 -0700
From:   Guenter Roeck <linux@...ck-us.net>
To:     Matthew Fortune <Matthew.Fortune@...tec.com>
Cc:     Maciej Rozycki <Maciej.Rozycki@...tec.com>,
        Ralf Baechle <ralf@...ux-mips.org>,
        "linux-mips@...ux-mips.org" <linux-mips@...ux-mips.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        James Hogan <James.Hogan@...tec.com>
Subject: Re: [PATCH] MIPS: VDSO: Always select -msoft-float

On Fri, Nov 04, 2016 at 04:55:05PM +0000, Matthew Fortune wrote:
> Maciej Rozycki <Maciej.Rozycki@...tec.com> writes:
> > On Fri, 4 Nov 2016, Guenter Roeck wrote:
> > 
> > > > As above, unless absolutely critical to have floating point code
> > > > then the vDSO should just avoid all FP related issues and build
> > soft-float.
> ...
> > > Anyway, isn't the kernel supposed to not use floating point operations
> > > in the first place ? Is this different for vDSO ?
> > 
> >  This code is executed in the user mode so while floating-point code may
> > not be needed there, not at least right now, there's actually nothing
> > which should stop us from from adding some should such a need arise.
> 
> Indeed. For now though the switch to -msoft-float is the simplest solution
> isn't it?
> 
One should think so, especially since 1) the code has to be built as part of the
kernel build anyway, and 2) there would be other implications besides just
adding floating point operations into the kernel (see the other e-mail).
Overall I am glad that I don't have to maintain this code.

Either case, it would be great if we can come up with a solution to avoid build
failures due to toolchain configurations. If that means that vDSO won't be
compilable with debian toolchains, so be it. After all, that would not be
different to the current situation, only it would be official instead of causing
build failures.

Thanks,
Guenter