diff options
author | venkatesh.pallipadi@intel.com <venkatesh.pallipadi@intel.com> | 2009-07-02 20:08:31 -0400 |
---|---|---|
committer | Dave Jones <davej@redhat.com> | 2009-07-06 21:38:28 -0400 |
commit | 37c90e8887efd218dc4af949b7f498ca2da4af9f (patch) | |
tree | 437ba6c24556b22e5094f206e62711cabfb6586e | |
parent | 7d26e2d5e2da37e92c6c7644b26b294dedd8c982 (diff) |
[CPUFREQ] Mark policy_rwsem as going static in cpufreq.c wont be exported
lock_policy_rwsem_* and unlock_policy_rwsem_* routines in cpufreq.c are
currently exported to drivers. Improper use of those locks can
result in deadlocks and it is better to keep the locks localized.
Two previous in-kernel users of these interfaces (ondemand and conservative),
do not use this interfaces any more. Schedule them for removal.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Dave Jones <davej@redhat.com>
-rw-r--r-- | Documentation/feature-removal-schedule.txt | 10 |
1 files changed, 10 insertions, 0 deletions
diff --git a/Documentation/feature-removal-schedule.txt b/Documentation/feature-removal-schedule.txt index f8cd450be9aa..09e031c55887 100644 --- a/Documentation/feature-removal-schedule.txt +++ b/Documentation/feature-removal-schedule.txt | |||
@@ -458,3 +458,13 @@ Why: Remove the old legacy 32bit machine check code. This has been | |||
458 | but the old version has been kept around for easier testing. Note this | 458 | but the old version has been kept around for easier testing. Note this |
459 | doesn't impact the old P5 and WinChip machine check handlers. | 459 | doesn't impact the old P5 and WinChip machine check handlers. |
460 | Who: Andi Kleen <andi@firstfloor.org> | 460 | Who: Andi Kleen <andi@firstfloor.org> |
461 | |||
462 | ---------------------------- | ||
463 | |||
464 | What: lock_policy_rwsem_* and unlock_policy_rwsem_* will not be | ||
465 | exported interface anymore. | ||
466 | When: 2.6.33 | ||
467 | Why: cpu_policy_rwsem has a new cleaner definition making it local to | ||
468 | cpufreq core and contained inside cpufreq.c. Other dependent | ||
469 | drivers should not use it in order to safely avoid lockdep issues. | ||
470 | Who: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com> | ||