Linux 2.6.30 Kernel Benchmarks

[Apopas]

That's because kids are playing with not finished file systems. And there are data loses with xfs after an crash. It's the same as ext4.

I mean data loses for no obvious reason, not because of crash.
Recently I reformatted my old reiserfs storage disks to xfs and I'm very satisfied. Very fast!

[ebird]

I mean data loses for no obvious reason, not because of crash.
Recently I reformatted my old reiserfs storage disks to xfs and I'm very satisfied. Very fast!

I have heard of some data losses with ext4, while it was developed. It will take another year, till this file system is really stable.

The data loses with at splashdot and co, where cause by delayed allocation after a crash. There are test cases at launchpad like this.

1. edit a.new
2. mv a.new a
3. turn off the computer while it is running

=> file a is a 0 byte file

[val-gaav]

Actually I remember Sidux dev saying that from his experience ext4 will be fully stable and proven 3 kernels away from it's introduction as stable into the kernel...

Ext4 was introduced as stable in 2.6.28 so in 2.6.31 it should be all good ...

That being said I'm using ext4 in my jaunty box (2.6.28) and no data loss for me, though when I copy files from some pendrives it's ridiculous slow... Didn't have that problem with ext3 or raiserfs.

[nkef]

These test should be done with 64bit too. On gentoo, we found 2.6.30 over 600% faster in most intensive I/O load (like decompression). A 3 years old (introduced in 2.6.16) regression have been fixed.

This nasty bug makes my Ubuntu 9.04 x64 unusable during intensive I/O load.
Is there any patch for ATI catalyst 9.5 to make it compatible with 2.6.30?

[Kano]

If you use Kanotix you can use 2.6.30 + fglrx. As a kernel patch is required it is very unlikely that other distro kernels would work.