Fix overflow issues with sysctl values in centiseconds/seconds
Get series with b4
b4 am [email protected]Versions newest first
- v1 Latest3 patches 6 repliesIn review
Patches in v1 No applies recorded yet
| # | Subject and files | Size | Replies | Applied |
|---|---|---|---|---|
| 1/3 | Represent dirty_*_centisecs as jiffies internallyinclude/ kernel/ mm/ | +19 | 0 replies | |
| 2/3 | Represent laptop_mode as jiffies internallykernel/ mm/ | +5 | 0 replies | |
| 3/3 | Range checking in do_proc_dointvec_(userhz_)jiffies_convkernel/ | +4 | 0 replies |
Reviews trailers per patch
No Reviewed-by, Acked-by or Tested-by trailers for v1 yet.
Cover letter [PATCH 0/3] Fix overflow issues with sysctl values in centiseconds/seconds
Hi Andrew,
Here's a threesome of patches to fix up some issues with the following sysctl values:
/proc/sys/vm/laptop_mode/proc/sys/vm/dirty_writeback_centisecs/proc/sys/vm/dirty_expire_centisecs
The issues:
1. The values are not range checked when they are set. They all have
a range smaller than the full integer range.
2. Conversion from these centisecond/second values is done on-the-fly wherever they are used. This wastes some resources.
3. The conversions are done badly. Conversion from USER_HZ to HZ is done by doing "value * USER_HZ / HZ". One day expressed in centiseconds already causes an overflow at HZ = 250. This should use clock_t_to_jiffies() instead.
The approach:
1. Represent everything in jiffies internally.
2. Do the conversion and range checking in the sysctl interface.
Cheers,
Bart