Messages in this thread | | | Date | Wed, 20 Sep 2006 03:42:06 +1000 | From | Nick Piggin <> | Subject | Re: Linux 2.4.34-pre3 |
| |
Willy Tarreau wrote: > Hello ! > > I've been a little bit silent and quite busy too. As announced with -pre2, > here comes -pre3 with only GCC4 fixes. Other fixes I received are minor > and can wait for -pre4. I really want people to test -pre3 without adding > any noise to the test. There should be *no* regression at all with existing > compilers. > > Gcc 4.1 is known to build x86, x86_64, ppc, sparc64, and sparc. Only sparc > has received no testing yet, while sparc64 is OK. It is possible that some > sparc/sparc64 drivers have not been caught because of undetected options > combinations (Davem CC'd for any possible advice on this mater). Status > for other archs is unknown but at least must not be affected for existing > setups. > > You will notice that most of the changes below appear under my name while > I do not deserve any credit for the changes. It is just because I've cut > all the fixes to sort them, and committed them myself individually. But > the real work has been done by Mikael. > > We have worked *very* carefully on this merge, and we hope to get all > possible feedback. People who encounter build problems on archs listed > above are free to report them, possibly with the fix. When providing a > fix, *please* provide the whole error output in the commit message so > that we can track what has been fixed. People who want to include support > for other archs will have to provide patches, as (at least for me) we > are not equipped to build on other archs (except for alpha when my RAM > arrives). > > I plan to wait up to the end of this month before providing -pre4 if there > is no feedback. Important fixes will be subject to another -stable release > anyway, so it's safe to wait for feedback here.
I wonder if 2.4 doesn't need the memory ordering fix to prevent pagecache corruption in reclaim? (http://www.gatago.com/linux/kernel/14682626.html)
What would need to be done is to test page_count before testing PageDirty, and putting an smp_rmb between the two.
-- SUSE Labs, Novell Inc. Send instant messages to your online friends http://au.messenger.yahoo.com
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |