As documented here: http://www.mail-archive.com/debian-kernel@lists.debian.org/msg43737.html http://bugs.archlinux.org/task/12823 ... the padlock_sha module which accelerates SHA operations on VIA processors hangs if loaded on VIA processors(!) - or, at least, the C7. This is problematic for headless servers which have padlock_sha in /etc/modules.autoload.d/kernel-2.6. This should probably be checked for and entries removed by the ebuild, or at least a message should be output.
which is the last known working kernel?
2.6.27-gentoo-r8 definitely worked: I believe that this is a regression in 2.6.28.
According to this commit http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.29.y.git;a=commit;h=a760a6656e6f00bb0144a42a048cf0266646e22c this should be fixed on 2.6.29 branch. Could you please try the latest 2.6.29 sources ? That is 2.6.29-r3 Thanks
Please feel free to reopen after testings as described in comment #3
Apologies for the delay - I've today upgraded to 2.6.29-gentoo-r4, and that works perfectly.