> you might want to be careful with pagemap:
> http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-... .
> incidentally, that commit fixes security bugs without a corresponding
> stable commit, not to mention CVE.
You've missed similar commit re /proc/*/clear_refs .
Regardless, RTFS before making such statements.
2.6.25 is not affected not to NULL ->mm bug (security-irrelevant),
not to static walker bug (contains only pointers to hooks, data passed
explicitly to pagetable walkers), read(, , 0) is also fine, wraparound
in multiplication in kmalloc doesn't work too.
The only bug 2.6.25 has, which 2.6.26 doesn't have is read(, , 0)
returning 8. Actually, this was fixed accidently.
proc/*/clear_refs fixlet is unneeded in 2.6.25 also, again, walker
structure contains only hooks.
> business as usual, i guess.
How old are you?