Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 8 Jul 2010 03:35:00 +0000 (UTC)
From:      Alan Cox <alc@FreeBSD.org>
To:        cvs-src-old@freebsd.org
Subject:   cvs commit: src/sys/amd64/amd64 cpu_switch.S pmap.c
Message-ID:  <201007080357.o683vSA4007382@repoman.freebsd.org>

next in thread | raw e-mail | index | archive | help
alc         2010-07-08 03:35:00 UTC

  FreeBSD src repository

  Modified files:
    sys/amd64/amd64      cpu_switch.S pmap.c 
  Log:
  SVN rev 209789 on 2010-07-08 03:35:00Z by alc
  
  Correctly maintain the per-cpu field "curpmap" on amd64 just like we
  do on i386.  The consequences of not doing so on amd64 became apparent
  with the introduction of the COUNT_IPIS and COUNT_XINVLTLB_HITS
  options.  Specifically, single-threaded applications were generating
  unnecessary IPIs to shoot-down the TLB on other processors.  However,
  this is clearly nonsensical because a single-threaded application is
  only running on the current processor.  The reason that this happens
  is that pmap_activate() is unable to properly update the old pmap's
  field "pm_active" without the correct "curpmap".  So, in effect, stale
  bits in "pm_active" were leading pmap_protect(), pmap_remove(),
  pmap_remove_pages(), etc. to flush the TLB contents on some arbitrary
  processor that wasn't even running the same application.
  
  Reviewed by:    kib
  MFC after:      3 weeks
  
  Revision  Changes    Path
  1.171     +10 -12    src/sys/amd64/amd64/cpu_switch.S
  1.706     +2 -0      src/sys/amd64/amd64/pmap.c



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201007080357.o683vSA4007382>