releases/2.6.32.19/acpi-fix-regression-where-_ppc-is-not-read-at-boot-even-when-ignore_ppc-0.patch

   1 From 455c0d71d46e86b0b7ff2c9dcfc19bc162302ee9 Mon Sep 17 00:00:00 2001
   2 From: Darrick J. Wong <djwong@us.ibm.com>
   3 Date: Thu, 18 Feb 2010 10:28:20 -0800
   4 Subject: ACPI: Fix regression where _PPC is not read at boot even when ignore_ppc=0
   5
   6 From: Darrick J. Wong <djwong@us.ibm.com>
   7
   8 commit 455c0d71d46e86b0b7ff2c9dcfc19bc162302ee9 upstream.
   9
  10 Earlier, Ingo Molnar posted a patch to make it so that the kernel would avoid
  11 reading _PPC on his broken T60.  Unfortunately, it seems that with Thomas
  12 Renninger's patch last July to eliminate _PPC evaluations when the processor
  13 driver loads, the kernel never actually reads _PPC at all!  This is problematic
  14 if you happen to boot your non-T60 computer in a state where the BIOS _wants_
  15 _PPC to be something other than zero.
  16
  17 So, put the _PPC evaluation back into acpi_processor_get_performance_info if
  18 ignore_ppc isn't 1.
  19
  20 Signed-off-by: Darrick J. Wong <djwong@us.ibm.com>
  21 Signed-off-by: Len Brown <len.brown@intel.com>
  22 Acked-by: Jeff Mahoney <jeffm@suse.com>
  23 Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
  24
  25 ---
  26  drivers/acpi/processor_perflib.c |    6 +++++-
  27  1 file changed, 5 insertions(+), 1 deletion(-)
  28
  29 --- a/drivers/acpi/processor_perflib.c
  30 +++ b/drivers/acpi/processor_perflib.c
  31 @@ -356,7 +356,11 @@ static int acpi_processor_get_performanc
  32         if (result)
  33                 goto update_bios;
  34
  35 -       return 0;
  36 +       /* We need to call _PPC once when cpufreq starts */
  37 +       if (ignore_ppc != 1)
  38 +               result = acpi_processor_get_platform_limit(pr);
  39 +
  40 +       return result;
  41
  42         /*
  43          * Having _PPC but missing frequencies (_PSS, _PCT) is a very good hint that