From: Andy Lutomirski <luto@amacapital.net>
To: Borislav Petkov <bp@alien8.de>
Cc: Michael Matz <matz@suse.de>, Ingo Molnar <mingo@kernel.org>,
Thomas Gleixner <tglx@linutronix.de>,
Oleg Nesterov <oleg@redhat.com>, Brian Gerst <brgerst@gmail.com>,
Luis Rodriguez <mcgrof@suse.com>,
Dave Hansen <dave.hansen@linux.intel.com>,
Andrew Morton <akpm@linux-foundation.org>,
Denys Vlasenko <dvlasenk@redhat.com>,
Peter Zijlstra <peterz@infradead.org>,
Linus Torvalds <torvalds@linux-foundation.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
Toshi Kani <toshi.kani@hp.com>,
Andrew Lutomirski <luto@kernel.org>,
Andrey Ryabinin <aryabinin@virtuozzo.com>,
"H. Peter Anvin" <hpa@zytor.com>,
"linux-tip-commits@vger.kernel.org"
<linux-tip-commits@vger.kernel.org>
Subject: Re: [PATCH -v1.1] x86/mm: Fix INVPCID asm constraint
Date: Wed, 10 Feb 2016 18:48:50 -0800 [thread overview]
Message-ID: <CALCETrVuq1Kckf5nYoCXV+cT=S2GTRqnZAvF6AEPW-ppBpE_HQ@mail.gmail.com> (raw)
In-Reply-To: <20160210145116.GF23914@pd.tnic>
On Wed, Feb 10, 2016 at 6:51 AM, Borislav Petkov <bp@alien8.de> wrote:
> On Wed, Feb 10, 2016 at 02:48:02PM +0100, Michael Matz wrote:
>> Hi,
>>
>> On Wed, 10 Feb 2016, Borislav Petkov wrote:
>>
>> > --- a/arch/x86/include/asm/tlbflush.h
>> > +++ b/arch/x86/include/asm/tlbflush.h
>> > @@ -23,7 +23,7 @@ static inline void __invpcid(unsigned long pcid, unsigned long addr,
>> > * invpcid (%rcx), %rax in long mode.
>> > */
>> > asm volatile (".byte 0x66, 0x0f, 0x38, 0x82, 0x01"
>> > - : : "m" (desc), "a" (type), "c" (desc) : "memory");
>> > + : : "m" (*desc), "a" (type), "c" (desc) : "memory");
>>
>> That still doesn't do what you want. Arrays in C are funny. *desc is
>> exactly equivalent to desc[0], _not_ to the whole array,
>
> Doh!
>
>> indeed there's no C syntax to name an lvalue of array type in normal
>> expressions. You need to jump through hoops for this:
>>
>> "m" (*(struct {unsigned long x[2];} *)desc)
>
> Aha! That's why we wrapped the array in clwb() in a struct too, btw:
>
> static inline void clwb(volatile void *__p)
> {
> volatile struct { char x[64]; } *p = __p;
> ...
>
>> It'd probably be easier to simply declare the descriptor as a struct,
>> rather than an array, then the original syntax would have been mostly
>> correct:
>>
>> struct {u64 d[2];} desc = { pcid, addr };
>> asm ... "m" (desc), "c" (&desc)
>
> Sounds better. Done. How does that below look like?
>
> Thanks Micha!
>
> ---
> From: Borislav Petkov <bp@suse.de>
> Date: Wed, 10 Feb 2016 12:53:48 +0100
> Subject: [PATCH -v1.1] x86/mm: Fix INVPCID asm constraint
> MIME-Version: 1.0
> Content-Type: text/plain; charset=UTF-8
> Content-Transfer-Encoding: 8bit
>
> So we want to specify the dependency on both @pcid and @addr so that the
> compiler doesn't reorder accesses to them *before* the TLB flush. But
> for that to work, we need to express this properly in the inline asm and
> deref the whole desc array, not the pointer to it. See clwb() for an
> example.
>
> This fixes the build error on 32-bit:
>
> arch/x86/include/asm/tlbflush.h: In function ‘__invpcid’:
> arch/x86/include/asm/tlbflush.h:26:18: error: memory input 0 is not directly addressable
>
Acked-by: Andy Lutomirski <luto@kernel.org>
next prev parent reply other threads:[~2016-02-11 2:49 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-01-29 19:42 [PATCH v3 0/3] x86/mm: INVPCID support Andy Lutomirski
2016-01-29 19:42 ` [PATCH v3 1/3] x86/mm: Add INVPCID helpers Andy Lutomirski
2016-02-09 16:07 ` [tip:x86/mm] " tip-bot for Andy Lutomirski
2016-02-10 12:08 ` Borislav Petkov
2016-02-10 13:48 ` Michael Matz
2016-02-10 14:51 ` [PATCH -v1.1] x86/mm: Fix INVPCID asm constraint Borislav Petkov
2016-02-11 2:48 ` Andy Lutomirski [this message]
2016-01-29 19:42 ` [PATCH v3 2/3] x86/mm: Add a noinvpcid option to turn off INVPCID Andy Lutomirski
2016-02-09 16:07 ` [tip:x86/mm] x86/mm: Add a 'noinvpcid' boot " tip-bot for Andy Lutomirski
2016-01-29 19:42 ` [PATCH v3 3/3] x86/mm: If INVPCID is available, use it to flush global mappings Andy Lutomirski
2016-02-09 16:08 ` [tip:x86/mm] " tip-bot for Andy Lutomirski
2016-02-01 10:51 ` [PATCH v3 0/3] x86/mm: INVPCID support Borislav Petkov
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='CALCETrVuq1Kckf5nYoCXV+cT=S2GTRqnZAvF6AEPW-ppBpE_HQ@mail.gmail.com' \
--to=luto@amacapital.net \
--cc=akpm@linux-foundation.org \
--cc=aryabinin@virtuozzo.com \
--cc=bp@alien8.de \
--cc=brgerst@gmail.com \
--cc=dave.hansen@linux.intel.com \
--cc=dvlasenk@redhat.com \
--cc=hpa@zytor.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-tip-commits@vger.kernel.org \
--cc=luto@kernel.org \
--cc=matz@suse.de \
--cc=mcgrof@suse.com \
--cc=mingo@kernel.org \
--cc=oleg@redhat.com \
--cc=peterz@infradead.org \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.org \
--cc=toshi.kani@hp.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).