Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Thu, 07 May 2020 12:31:19 -0700
User-Agent: K-9 Mail for Android
In-Reply-To: <9c701ca55bc442c1899a70896f3ea73e@AcuMS.aculab.com>
References: <20200505174423.199985-1-ndesaulniers@google.com> <CAMzpN2idWF2_4wtPebM2B2HVyksknr9hAqK8HJi_vjQ06bgu2g@mail.gmail.com> <6A99766A-59FB-42DF-9350-80EA671A42B0@zytor.com> <CAMzpN2iCgr0rb=MCYPGMx8tcfLq2qdzv0h7YnX5hkzBB+O7JJQ@mail.gmail.com> <9c701ca55bc442c1899a70896f3ea73e@AcuMS.aculab.com>
MIME-Version: 1.0
Content-Type: text/plain;
 charset=utf-8
Content-Transfer-Encoding: quoted-printable
Subject: RE: [PATCH] x86: bitops: fix build regression
To:     David Laight <David.Laight@ACULAB.COM>,
        "'Brian Gerst'" <brgerst@gmail.com>
CC:     Nick Desaulniers <ndesaulniers@google.com>,
        Thomas Gleixner <tglx@linutronix.de>,
        Ingo Molnar <mingo@redhat.com>, Borislav Petkov <bp@alien8.de>,
        Sedat Dilek <sedat.dilek@gmail.com>,
        stable <stable@vger.kernel.org>,
        Jesse Brandeburg <jesse.brandeburg@intel.com>,
        "kernelci . org bot" <bot@kernelci.org>,
        Andy Shevchenko <andriy.shevchenko@intel.com>,
        Ilie Halip <ilie.halip@gmail.com>,
        the arch/x86 maintainers <x86@kernel.org>,
        Marco Elver <elver@google.com>,
        "Paul E. McKenney" <paulmck@kernel.org>,
        "Peter Zijlstra (Intel)" <peterz@infradead.org>,
        Daniel Axtens <dja@axtens.net>,
        Masahiro Yamada <yamada.masahiro@socionext.com>,
        Luc Van Oostenryck <luc.vanoostenryck@gmail.com>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        "clang-built-linux@googlegroups.com" 
        <clang-built-linux@googlegroups.com>
From:   hpa@zytor.com
Message-ID: <A6FD91D2-8D3A-4767-B6AD-D35B056C58C4@zytor.com>
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On May 7, 2020 8:09:35 AM PDT, David Laight <David=2ELaight@ACULAB=2ECOM> w=
rote:
>From: Brian Gerst
>> Sent: 07 May 2020 14:32
>=2E=2E=2E
>> I think the bug this worked around was that the compiler didn't
>detect
>> that CONST_MASK(nr) was also constant and doesn't need to be put into
>> a register=2E  The question is does that bug still exist on compiler
>> versions we care about?
>
>Hmmm=2E=2E=2E
>That ought to have been fixed instead of worrying about the fact
>that an invalid register was used=2E
>
>Alternatively is there any reason not to use the bts/btc instructions?
>Yes, I know they'll do wider accesses, but variable bit numbers do=2E
>It is also possible that the assembler will support constant bit
>numbers >=3D 32 by adding to the address offset=2E
>
>	David
>
>-
>Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes,
>MK1 1PT, UK
>Registration No: 1397386 (Wales)

They're slower, and for unaligned locked fields can be severely so=2E
--=20
Sent from my Android device with K-9 Mail=2E Please excuse my brevity=2E