Re: [Fwd: ext2fs patch for large stores, RC1+20040304]
Quoting Ognyan Kulev <ogi@fmi.uni-sofia.bg>:
> I try to post this announcement to bug-hurd since some days, but my
> mails are ignored. So I post it to debian-hurd.
There are some problems with the gnu.org mailserver.
> Changes in RC1+20040304:
>
> * ihash* -> hurd_ihash* (use revised libihash API)
> * block << log2_block_size -> (off_t) block << log2_block_size (`block'
> is usually 32-bit and stores larger than 4G didn't work with old code)
Great, you found it. :)
> * sleep(1) is called only when all cached blocks are referenced
> (improves performance).
Oh, nice. But is it possible to use conditions instead of sleep? Or
at least sleep shorter?
> This fixes what Andrew Mitchell reported to me on 2003-10-28. I'm not
> 100% sure if what Marco Gerards reported is fixed by this change.
> Marco, would you do some tests again?
Of course. This change is that 4GB problem you mentioned, right?
> And don't forget that if you test the patch, you'll probably want
> patched e2fsprogs too. You can get the patch from
> http://debian.fmi.uni-sofia.bg/~ogi/hurd/ext3fs/ . It applies cleanly
> to e2fsprogs_1.35-2 (with some offsets, but without *.rej files).
Right. I have not used this yet. It would be nice if someone could
make a package for this as well.
Thanks,
Marco
Reply to: