per device dirty throttling -v10
Get series with b4
b4 am [email protected]Versions newest first
- v1 Latest23 patches 4 repliesIn review
Patches in v1 No applies recorded yet
| # | Subject and files | Size | Replies | Applied |
|---|---|---|---|---|
| 1/23 | nfs: remove congestion_end()fs/ include/ mm/ | +1 | 0 replies | |
| 2/23 | lib: percpu_counter_addfs/ fs/ fs/ +5 more | +17 | 0 replies | |
| 3/23 | lib: percpu_counter_subfs/ fs/ fs/ +1 more | +8 | 0 replies | |
| 4/23 | lib: percpu_counter variable batchinclude/ lib/ | +12 | 0 replies | |
| 5/23 | lib: make percpu_counter_add take s64include/ lib/ | +5 | 0 replies | |
| 6/23 | lib: percpu_counter_setinclude/ lib/ | +20 | 0 replies | |
| 7/23 | lib: percpu_counter_sum_positivefs/ fs/ fs/ +4 more | +12 | 0 replies | |
| 8/23 | lib: percpu_count_sum()include/ lib/ | +20 | 0 replies | |
| 9/23 | lib: percpu_counter_init error handlingfs/ fs/ fs/ +2 more | +52 | 0 replies | |
| 10/23 | lib: percpu_counter_init_irqinclude/ lib/ | +15 | 0 replies | |
| 11/23 | mm: bdi init hooksblock/ drivers/ drivers/ +16 more | +131 | 0 replies | |
| 12/23 | containers: bdi init hookskernel/ | +11 | 0 replies | |
| 13/23 | mtd: bdi init hooksdrivers/ | +9 | 0 replies | |
| 14/23 | mtd: clean up the backing_dev_info usagedrivers/ include/ | +7 | 0 replies | |
| 15/23 | mtd: give mtdconcat devices their own backing_dev_infodrivers/ | +18 | 0 replies | |
| 16/23 | mm: scalable bdi statistics counters.include/ mm/ | +109 | 0 replies | |
| 17/23 | mm: count reclaimable pages per BDIfs/ fs/ include/ +2 more | +16 | 0 replies | |
| 18/23 | mm: count writeback pages per BDIinclude/ mm/ | +11 | 0 replies | |
| 19/23 | mm: expose BDI statistics in sysfs.block/ | +30 | 0 replies | |
| 20/23 | lib: floating proportionsinclude/ lib/ lib/ | +505 | 0 replies | |
| 21/23 | mm: per device dirty thresholdinclude/ include/ kernel/ +2 more | +196 | 2 replies | |
| 22/23 | mm: dirty balancing for tasksinclude/ include/ kernel/ +1 more | +62 | 0 replies | |
| 23/23 | debug: sysfs files for the current ratio/size/totalblock/ mm/ | +83 | 0 replies |
Reviews trailers per patch
No Reviewed-by, Acked-by or Tested-by trailers for v1 yet.
Cover letter [PATCH 00/23] per device dirty throttling -v10
Per device dirty throttling patches
These patches aim to improve balance_dirty_pages() and directly address three issues:
1) inter device starvation
2) stacked device deadlocks
3) inter process starvation
1 and 2 are a direct result from removing the global dirty limit and using per device dirty limits. By giving each device its own dirty limit is will no longer starve another device, and the cyclic dependancy on the dirty limit is broken.
In order to efficiently distribute the dirty limit across the independant devices a floating proportion is used, this will allocate a share of the total limit proportional to the device's recent activity.
3 is done by also scaling the dirty limit proportional to the current task's recent dirty rate.
Changes since -v9:
- cleaned up the perpcu_counter_init code
- little fixups
- fwd port to .23-rc4-mm1
Changes since -v8:
- cleanup of the proportion code
- fix percpu_counter_add(&counter, -(unsigned long))
- fix per task dirty rate code
- fwd port to .23-rc2-mm2
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>