The HSF kernel modules compile ok against the 2.6.15 kernel, but when you try to load them, it causes a kernel panic. There seems to be a patch available on the linuxant site (http://www.linuxant.com/drivers/hsf/downloads-patches.php) I have tried the suggested method of patching the driver (hsfconfig --patch), but it fails with the following error:- mr_flibble hsfmodem # hsfconfig --patch Conexant HSF softmodem driver, version 7.18.00.07full If you need license keys, assistance or more information, please go to: http://www.linuxant.com/ When reporting a problem for the first time, please send us the file generated by "hsfconfig --dumpdiag". /usr/sbin/hsfconfig: line 658: cd: /var/tmp/portage/hsfmodem-7.18.00.07/work/hsfmodem-7.18.00.07full/scripts: No such file or directory mr_flibble hsfmodem # Any chance of these patches somehow being incorporated into the ebuild? Or maybe have a resolution to the missing scripts file/folder problem?
Please test hsfmodem-7.18.00.07-r1. This version contains both upstream patches (class_device_parent and refrigerator) merged into one patch because th original ones couldn't be merged simultaneously. I believe the class_device_parent solves the kernel panic.
Ok, tests fine now. Thanks for the very quick response. You guys & gals do a fantastic job....
hsfmodem-7.18.00.07-r1 should be marked as stable on x86 after the probation period (1 month). I'll close this bug then.
I end up with: * Applying hsfmodem-7.18.00.07-upstream-20060107.patch ... * Failed Patch: hsfmodem-7.18.00.07-upstream-20060107.patch ! * ( /usr/portage/net-dialup/hsfmodem/files/hsfmodem-7.18.00.07-upstream-20060107.patch ) * However, patching manually with -p1 instead of -p0 works... (I have amd64, but I do not think it is relevant in this case...)
I keep forgetting about S dependance of the arch! my appologies... the fixed ebuild (that changes the current dir to MY_ARCH_S before applying the patch) is now in cvs.
Unfortunately, I still end up with "Unable to handle kernel paging request" on amd64. I have "only" 2.6.14-gentoo-r5 kernel...
are you speaking about something described in some other bug? I am unaware of such bug...
Oh, sorry, of course... well, I opened #118999.
I've marked hsfmodem-7.18.00.07-r1 stable on x86 before the scheduled time for allowing kernel 2.6.15 to be marked stable (bug 118671).