BUG: isofs broken (2.2 and 2.4)
Quoted 1 line
both 2.2.x and 2.4.x kernels can't read `real sky' CDs
Yes. 2.0.38 is OK. I just made a patch that seems to work.
Harald, could you try
ftp.xx.kernel.org/.../people/aeb/linux-2.4.0test9-isofs-patch
and report?
Linus, Alan - I made patches for 2.2 and 2.4 but want to polish and check them a bit more before submitting. There also seem to be a lot of bug reports in newsgroups and mailing lists - must check whether people complain about the same thing or whether there are more problems.
Andries
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
Please read the FAQ at http://www.tux.org/lkml/
Quoted 7 lines · On Nov 17, [email protected] wrote:
Quoted 1 line
both 2.2.x and 2.4.x kernels can't read `real sky' CDs
Yes. 2.0.38 is OK. I just made a patch that seems to work.
Harald, could you try
ftp.xx.kernel.org/.../people/aeb/linux-2.4.0test9-isofs-patchand report?
works -- sort of:( I've tried both test9 and test10 with your patch
on a PPro200 with 128MB ram and I get the same effects both times:
directory listing seems to be possible (haven't checked data contents yet) BUT if I try "du /cdrom/" (either using real cdrom device or loopback mount of my sample isofs image) there seems to be a huge memory leak !
first observation is that "du" is awfuly slow -- it takes ~90 secs real time and ~60 system cpu secs to "du" through the first ~70 of 106 direcories, then the 128MB memory are almost used up and the systems starts to swap heavily. this meory doesn't get freed up even after umount/losetup -d or whatever -- only reboot "helps"...
I'll attach log files showing output of "free", /proc/meminfo and output of ALT-SYSREQ-M plus full "ps" output for both -test9 and -test10 with your patch in the situation when almost all memory is "gone" (du already killed). hope this helps...
Harald
Signature · 8 lines
All SCSI disks will from now on ___ _____
be required to send an email notice 0--,| /OOOOOOO\
24 hours prior to complete hardware failure! <_/ / /OOOOOOOOOOO\
\ \/OOOOOOOOOOOOOOO\
\ OOOOOOOOOOOOOOOOO|//
Harald Koenig, \/\/\/\/\/\/\/\/\/
Inst.f.Theoret.Astrophysik // / \\ \
[email protected] ^^^^^ ^^^^^