Re: [RFC PATCH 0/1] arm64: Fix /proc/cpuinfo

From: Catalin Marinas <catalin.marinas@arm.com>
To: Will Deacon <will.deacon@arm.com>
Cc: Mark Rutland <Mark.Rutland@arm.com>,
	"linux-arm-kernel@lists.infradead.org" 
	<linux-arm-kernel@lists.infradead.org>,
	"ghackmann@google.com" <ghackmann@google.com>,
	"ijc@hellion.org.uk" <ijc@hellion.org.uk>,
	Serban Constantinescu <Serban.Constantinescu@arm.com>,
	"cross-distro@lists.linaro.org" <cross-distro@lists.linaro.org>,
	"linux-api@vger.kernel.org" <linux-api@vger.kernel.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [RFC PATCH 0/1] arm64: Fix /proc/cpuinfo
Date: Thu, 6 Nov 2014 17:05:48 +0000	[thread overview]
Message-ID: <20141106170548.GF19702@e104818-lin.cambridge.arm.com> (raw)
In-Reply-To: <20141106165430.GH31605@arm.com>

On Thu, Nov 06, 2014 at 04:54:31PM +0000, Will Deacon wrote:
> On Thu, Nov 06, 2014 at 04:43:12PM +0000, Catalin Marinas wrote:
> > On Fri, Oct 24, 2014 at 02:56:39PM +0100, Mark Rutland wrote:
> > > [d] Print different hwcaps dependent on the personality.
> > > 
> > >     This would allow for 32-bit and 64-bit applications to function
> > >     correctly, but for some 32-bit applications the personality would
> > >     need to be set explicitly by the user.
> > 
> > Which makes this option actually in line with the uname -m behaviour. My
> > vote goes for [d] with option [b] as a close alternative.
> > 
> > > [1] arm, v3.17, Versatile Express A15x2 A7x3 coretile
> > > Features        : half thumb fastmult vfp edsp neon vfpv3 tls vfpv4 idiva idivt vfpd32 lpae evtstrm 
> > [...]
> > > [2] arm64, v3.17, Juno platform
> > > Features        : fp asimd evtstrm aes pmull sha1 sha2 crc32 
> > 
> > As an exercise, I'm trying to see what option [b] would look like when
> > CONFIG_COMPAT is enabled:
> > 
> > Features        : fp asimd evtstrm aes pmull sha1 sha2 crc32 half thumb fastmult vfp edsp neon vfpv3 tls vfpv4 idiva idivt vfpd32 lpae
> > 
> > The duplicate strings would only be listed once (evtstrm, aes, pmull,
> > sha1, sha2, crc32). New AArch64 features that we may expect to be
> > optional on AArch32 could be prefixed with "a64". If they are missing
> > entirely from AArch32, (like asimd), no need for the prefix.
> > 
> > The advantage is that we don't need to check the personality but we have
> > to assume that scripts would not search for substrings (sane people
> > shouldn't do this anyway as the Features string can always be extended).
> 
> And a big disadvantage is that I can imagine AArch64 applications checking
> for "neon" instead of "asimd", which will break if they're run under kernels
> without COMPAT support enabled.

That's because they do stupid things and I they probably deserve to
break ;). But I agree, there is a risk.

> So I'm inclined to stick with Mark's patch as it is.

If we don't hear otherwise, I propose sometime next week we queue Mark's
patch for -next.

-- 
Catalin