Bug 5021 - Safely change pm_idle
Summary: Safely change pm_idle
Status: REJECTED INVALID
Alias: None
Product: ACPI
Classification: Unclassified
Component: Power-Processor (show other bugs)
Hardware: i386 Linux
: P2 normal
Assignee: Len Brown
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2005-08-07 23:16 UTC by Shaohua
Modified: 2007-03-08 23:01 UTC (History)
1 user (show)

See Also:
Kernel Version: 2.6.13-rc5
Subsystem:
Regression: ---
Bisected commit-id:

Attachments
patch to use cpu_idle_wait (2.68 KB, patch)
2005-08-07 23:19 UTC, Shaohua 		Details | Diff
patch to use cpu_idle_wait (2.68 KB, patch)
2005-08-07 23:19 UTC, Shaohua 		Details | Diff
updated patch (7.83 KB, patch)
2005-09-07 00:10 UTC, Shaohua 		Details | Diff
updated patch (7.82 KB, patch)
2005-09-14 00:09 UTC, Shaohua 		Details | Diff
Show Obsolete (3) Add an attachment (proposed patch, testcase, etc.)

Description Shaohua 2005-08-07 23:16:23 UTC 
synchronize_sched() only guarantees all CPUs have exited non-preemptive block. 
From my understanding, we want the affected CPU exits idle (in 
acpi_processor_cst_has_changed), so synchronize_sched() doesn
Comment 1 Shaohua 2005-08-07 23:19:11 UTC 
Created attachment 5542 [details]
patch to use cpu_idle_wait

The patch does:
It does:
1. change pm idle fix. 
2. cleanup for acpi_processor_get_power_info_cst to make it use less stack
space (acpi_processor_cx is a big structure)
Comment 2 Shaohua 2005-08-07 23:19:20 UTC 
Created attachment 5543 [details]
patch to use cpu_idle_wait

The patch does:
It does:
1. change pm idle fix. 
2. cleanup for acpi_processor_get_power_info_cst to make it use less stack
space (acpi_processor_cx is a big structure)
Comment 3 Len Brown 2005-09-02 23:04:38 UTC 
patch needs to be re-based on Lindented processor_idle.c
Comment 4 Shaohua 2005-09-07 00:10:20 UTC 
Created attachment 5921 [details]
updated patch

A updated patch. It tries to fix all races.
Comment 5 Shaohua 2005-09-14 00:09:31 UTC 
Created attachment 6005 [details]
updated patch

My last patch still has a small race window when one CPU is reading per-cpu
idle and another CPU is changing it. This updated one fixes it.
Comment 6 Len Brown 2007-03-08 21:24:39 UTC 
patch does not apply to current tree -- re-opening
Comment 7 Shaohua 2007-03-08 23:01:54 UTC 
with cpuidle, this patch isn't required.
Note You need to log in before you can comment on or make changes to this bug.