Summary: | Supermount Kernel patches | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Aniruddha Shankar <k> |
Component: | [OLD] Core system | Assignee: | Brandon Low (RETIRED) <lostlogic> |
Status: | RESOLVED FIXED | ||
Severity: | enhancement | CC: | christian-neumair, jensflorian, styx |
Priority: | High | ||
Version: | 1.2 | ||
Hardware: | x86 | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Attachments: | Kernel Configuration |
Description
Aniruddha Shankar
2002-06-28 08:22:07 UTC
As this is requested often (just look for 'supermount' at the gentoo forums) it would be really nice to see it we should raise priority. I'll try to make a few experiments. see you, - Manny - Anyhow my post got cropped and trashed (I assume a conspiracy ;) As this is requested often (just look for 'supermount' at the gentoo forums) it would be really nice to see this patch included in the next kernel revision. Additionally we should raise priority. I'll try to make a few experiments. see you, - Manny - I'll stick it in mjc-sources possibly. Doesn't help too much with floppies though.... I made an ebuild for the vanill sources. See here: http://bugs.gentoo.org/show_bug.cgi?id=5410 I'm looking ahead to get also the gentoo-sources patched, but at the moment it won't work and I'm no kernel hacker. under consideration for gentoo-sources-2.4.20 series. *** Bug 5410 has been marked as a duplicate of this bug. *** in latest lolo-sources, please test and let me know how it is going. Created attachment 4233 [details]
Kernel Configuration
Kernel .config - make bzImage fails
make bzImage fails (gcc 2.95.3) on the i810 DRM section - on a related note, can we please incorporate the i810 DRM from Mandrake's stable kernel ? The computers in my lab are using 2.4.18-6mdk and it's got by far the _best_ DRM performance for the Intel 810 of all the kernels I've tried (vanilla, gentoo, mjc) gcc -D__KERNEL__ -I/usr/src/linux-2.4.19-lolo-r10_pre4/include -Wall -Wstrict-prototypes -Wno-trigraphs -Os -fno-strict-aliasing -fno-common -fomit-frame-pointer -pipe -mpreferred-stack-boundary=2 -march=i686 -nostdinc -I /usr/lib/gcc-lib/i686-pc-linux-gnu/2.95.3/include -DKBUILD_BASENAME=i810_dma -c -o i810_dma.o i810_dma.c i810_dma.c: In function `i810_unmap_buffer': i810_dma.c:236: too few arguments to function `do_munmap' i810_dma.c: In function `i810_free_page': i810_dma.c:300: warning: implicit declaration of function `unlock_page' make[4]: *** [i810_dma.o] Error 1 make[4]: Leaving directory `/usr/src/linux-2.4.19-lolo-r10_pre4/drivers/char/drm' make[3]: *** [first_rule] Error 2 make[3]: Leaving directory `/usr/src/linux-2.4.19-lolo-r10_pre4/drivers/char/drm' make[2]: *** [_subdir_drm] Error 2 make[2]: Leaving directory `/usr/src/linux-2.4.19-lolo-r10_pre4/drivers/char' make[1]: *** [_subdir_char] Error 2 make[1]: Leaving directory `/usr/src/linux-2.4.19-lolo-r10_pre4/drivers' make: *** [_dir_drivers] Error 2 in lolo-sources, joining gentoo-sources soon. pls have a look at bug #60462 - there is a supermount-patch against vanilla 2.6.8.1 kernel sources. please try and provide some feedback to me - perhaps there are some people who want to help me maintaining the patch for the next few kernel versions? reagrds andre i volunteer to help with the supermount patches... I was quite upset that 2.6.8-gentoo was marked stable without them. cheers, Aniruddha Shankar New Delhi, India checked out supermount-ng205-1 from ck-sources and applied patch to linux-2.6.8-gentoo-r4. It worked out - only some line offsets - no- thing big... Will test it this week and give some feedback. for ck-sources, go to http://members.optusnet.com.au/ckolivas/kernel/ greets, andre I have tested this for a while now:
> checked out supermount-ng205-1 from ck-sources and applied patch to
> linux-2.6.8-gentoo-r4. It worked out - only some line offsets - no-
> thing big...
It works out very fine :-) no problems - think it can be used by all
people who want to.
There are some other solutions for doing things to make devices appear
to the user like supermount does. None of them worked for me:
1) dbus/hal/ivman
It simply did not work. No more, no less. If it worked, it would use
the system mount utility, which would not work out for me because we
have the old behavior except that we do not have to mount the device
by ourselves.
2) subfs
Did not work for me - seems to be a lightweight automounter and
keeps opening and closing the tray of some old hardware. And, in
addition to this - simply does not work for me because it has the
locking problem described below, too.
I NEED supermount because it does not lock up on device removal if the
device is usb-storage and mounted.
Consider this:
- plug in your usb-stick with whatsoever fs/data on it
- mount it
- read from it
- remove it
- read from it ==> locked up mount/dev/usb bus
- even if you plug the SAME device in again ==> locked up mount/dev/usb bus
Do the same with supermounted device:
- no problems - just does what it should do :-)
That is why i NEED supermount. So please put it back into at least
gentoo-dev-sources. a good point for starting may be the supermount
version i tested successfully with 2.6.8-gentoo-r4.
Hope this answers the question encoutered couple of weeks ago in some
related bug, which was what supermount can do and the other solutions
cannot do.
Sorry for reopening this bug - but i really, really tried to replace
supermount with one of the other solutions. It did not work.
greets
andre lammel
|