Re: x86 memory barrier: why does Linux prefer MFENCE to Locked ADD?

From	Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Newsgroups	linux.kernel
Subject	Re: x86 memory barrier: why does Linux prefer MFENCE to Locked ADD?
Date	2016-08-04 01:30 +0200
Message-ID	<s2dLP-Nq-1@gated-at.bofh.it> (permalink)
References	(3 earlier) <r8Gki-2Ii-27@gated-at.bofh.it> <r8GNk-37C-23@gated-at.bofh.it> <s1W8h-5XG-27@gated-at.bofh.it> <s245P-2Wu-7@gated-at.bofh.it> <s24fw-31q-23@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

On Wed, 03 Aug 2016, Michael S. Tsirkin wrote:
> Are any of these used in kernel though?

These specific errata were not the point of my post, rather, it was the
fact that errata related to *FENCE and LOCKed instructions exists.

I didn't verify whether something attempts to use non-temporal loads or
stores from WC memory in the kernel.

-- 
  "One disk to rule them all, One disk to find them. One disk to bring
  them all and in the darkness grind them. In the Land of Redmond
  where the shadows lie." -- The Silicon Valley Tarot
  Henrique Holschuh

Back to linux.kernel | Previous | Next — Previous in thread | Find similar | Unroll thread

Thread

Re: x86 memory barrier: why does Linux prefer MFENCE to Locked ADD? "Michael S. Tsirkin" <mst@redhat.com> - 2016-08-03 06:40 +0200
  Re: x86 memory barrier: why does Linux prefer MFENCE to Locked ADD? Henrique de Moraes Holschuh <hmh@hmh.eng.br> - 2016-08-03 15:10 +0200
    Re: x86 memory barrier: why does Linux prefer MFENCE to Locked ADD? "Michael S. Tsirkin" <mst@redhat.com> - 2016-08-03 15:20 +0200
      Re: x86 memory barrier: why does Linux prefer MFENCE to Locked ADD? Henrique de Moraes Holschuh <hmh@hmh.eng.br> - 2016-08-04 01:30 +0200

csiph-web