[linux-mm-cc] compcache with swap free notify support

Nitin Gupta nitingupta910 at gmail.com
Wed Jun 18 00:29:52 EDT 2008


On Tue, Jun 17, 2008 at 11:32 AM, John McCabe-Dansted <gmatht at gmail.com> wrote:
> On Mon, Jun 16, 2008 at 4:59 PM, Nitin Gupta <nitingupta910 at gmail.com> wrote:
>> Hi All,
>>
>> Attached is the patch for much refined version of compcache (for
>
> I have applied the patch to the Hardy kernel (2.6.24-18-generic). It
> applied cleanly except for:
> 1 out of 1 hunk FAILED -- saving rejects to file drivers/block/Kconfig.rej
> 1 out of 1 hunk FAILED -- saving rejects to file drivers/block/Makefile.rej
> 1 out of 1 hunk FAILED -- saving rejects to file init/Kconfig.rej
>
> This was easy to fix (see attached linux-2.6.24.cc.diff.gz). This then
> compiles fine, but I get crashes:
>
> Jun 17 13:21:32 xp-desktop kernel: [ 5327.447776] Adding 63820k swap
> on /dev/ramzswap0.  Priority:-2 extents:1 across:63820k
> Jun 17 13:22:19 xp-desktop kernel: [ 5374.949327] Trying to vfree()
> bad address (c5a57ff8)
> Jun 17 13:22:19 xp-desktop kernel: [ 5374.949355] WARNING: at
> mm/vmalloc.c:331 __vunmap()
> Jun 17 13:22:19 xp-desktop kernel: [ 5374.949381] Pid: 120, comm:
> pdflush Not tainted 2.6.24.3cc #1
> Jun 17 13:22:19 xp-desktop kernel: [ 5374.949472]
> [tlsf_free+0x246/0x350] tlsf_free+0x246/0x350
>

I will try to resolve this problem soon.


> See kern.log.gz for more info.
>
> Also, I compiled compcache with "y" instead of "m" (module). This
> means I cannot set the compcache size right?
>

Ah. I missed this point. I will see how we can make compcache size
configurable when its built-in with "y" option.


Thanks for feedback.

Nitin


More information about the linux-mm-cc mailing list