* REF_PRISTINE node ...
@ 2003-11-19 16:38 Sherwin Liu
2003-11-19 17:09 ` David Woodhouse
0 siblings, 1 reply; 2+ messages in thread
From: Sherwin Liu @ 2003-11-19 16:38 UTC (permalink / raw)
To: linux-mtd
I'm running Linux v2.4.18 under Montavista Linux on a
PPC440GP. I have many JFFS2 partitions mounted and a
test program that opens files on these partitions and
constantly writes and reads these files. Once in a
while I get this isolated message:
"REF_PRISTINE node at XX had a previous non-hole frag
in the same page. Tell dwmw2."
Is this just a debugging message, or is it an
indication of something wrong?
Thanks in advance...
-Sherwin
swl@dolby.com
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: REF_PRISTINE node ...
2003-11-19 16:38 REF_PRISTINE node Sherwin Liu
@ 2003-11-19 17:09 ` David Woodhouse
0 siblings, 0 replies; 2+ messages in thread
From: David Woodhouse @ 2003-11-19 17:09 UTC (permalink / raw)
To: Sherwin Liu; +Cc: linux-mtd
On Wed, 2003-11-19 at 08:38 -0800, Sherwin Liu wrote:
> "REF_PRISTINE node at XX had a previous non-hole frag
> in the same page. Tell dwmw2."
>
> Is this just a debugging message, or is it an
> indication of something wrong?
It's basically harmless, and should be gone if you update to the current
CVS JFFS2 code.
A while ago, I implemented a garbage-collection optimisation; if a node
is expected to be identical when we write out a replacement, we can
avoid decompressing and recompressing it if we just copy the whole thing
intact. So I start keeping track of such 'pristine' nodes which were an
optimal representation of their data... and that message warns me of one
case where such a node marked REF_PRISTINE but shouldn't have been.
--
dwmw2
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2003-11-19 17:09 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2003-11-19 16:38 REF_PRISTINE node Sherwin Liu
2003-11-19 17:09 ` David Woodhouse
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox