Linux 2.3.21 Gripes.
Dara Hazeghi (sassan@cup.hp.com)
Tue, 12 Oct 1999 17:25:56 -0700
I've generally been pretty happy with the kernel especially in the nice
days of 2.2.0pre4 etc. However things have devolved since then. Stable
kernels have become capable of hard lockupsm, unstable kernels have
repeatedly broken things I am rather fond of and so forth. Here is a
list of things I would like to see remedied (in no particular order).
* AFFS support: Broken in 2.3.7 for good reason or so I was told. 14
updates later it has still not been repaired.
* HFS support: Same deal. If something is going to be broken, it
would be nice if people would be willing to fix it.
* NTFS support: The only one of these three filesystems I really
need. Same story. To make matters worse, my only working 2.2.X kernel
(RedHat 6.1 2.2.13+patches) doesn't come with a module for it and any
kernel I build from its source seems to bomb.
* 1MB+ Kernels: I don't know the precise breaking point, but it is
not nice after a 25minute compilation to discover your kernel is too
big. I know there are technical reasons, but this is an issue that'll
have to be addressed sooner or later, and getting started sooner does
have advantages.
* Slow recompilation: When I build a kernel and it doesn't complete,
usually I change the offending option to off, and recompile. However
every single thing gets rebuilt. Again, some of us have slow machines
and this is not a good thing for us.
* Slow fsck: When I have to do an fsck (which seems more and more
often) it often take more than 10 minutes on a 5GB root partition. This
is a Ultra2 SCSI drive which isn't that slow either. I know reiserfs and
Stephen Tweedie's journaled ext2 deal with this, but they have not yet
been ported to 2.3.X and so I'm stuck waiting 10minutes every time
another 2.2.13pre breaks on me.
Please don't take this the wrong way. I like using Linux for what it is,
and I will continue to appreciate it whether or not these issues are
addressed. It's just that from the user standpoint, some of these things
really do need to be addressed.
Dara Hazeghi
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/