[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <939ec057-3851-d8fb-7b45-993fa07c4cb5@intel.com>
Date: Fri, 21 May 2021 09:14:12 -0700
From: Dave Hansen <dave.hansen@...el.com>
To: Florian Weimer <fweimer@...hat.com>,
Dave Hansen via Libc-alpha <libc-alpha@...rceware.org>
Cc: Len Brown <lenb@...nel.org>, Rich Felker <dalias@...c.org>,
Linux API <linux-api@...r.kernel.org>,
"Bae, Chang Seok" <chang.seok.bae@...el.com>,
X86 ML <x86@...nel.org>, LKML <linux-kernel@...r.kernel.org>,
Kyle Huey <me@...ehuey.com>, Borislav Petkov <bp@...en8.de>,
Andy Lutomirski <luto@...nel.org>,
Thomas Gleixner <tglx@...utronix.de>,
Keno Fischer <keno@...iacomputing.com>,
Arjan van de Ven <arjan@...ux.intel.com>,
Willy Tarreau <w@....eu>
Subject: Re: Candidate Linux ABI for Intel AMX and hypothetical new related
features
On 5/21/21 7:44 AM, Florian Weimer wrote:
> * Dave Hansen via Libc-alpha:
>> Our system calls are *REALLY* fast. We can even do a vsyscall for this
>> if we want to get the overhead down near zero. Userspace can also cache
>> the "I did the prctl()" state in thread-local storage if it wants to
>> avoid the syscall.
> Why can't userspace look at XCR0 to make the decision?
The thing we're trying to avoid is a #NM exception from XFD (the new
first-use detection feature) that occurs on the first use of AMX. XCR0
will have XCR0[AMX]=1, even if XFD is "armed" and ready to generate the #NM.
Powered by blists - more mailing lists