Arch Linux User Repository

I actually do not like DKMS... can you verify that the patch you supplied produces the desired results? I am glad to add it if so (code is dependent on that pkg being installed).

Can you kindly include the below patch to linux-ck.install so that DKMS autoinstall service runs after each kernel install? I prefer this method rather than invoking DKMS service on each system boot. --- linux-ck.install.old 2012-05-26 18:23:07.240130815 +0530 +++ linux-ck.install.new 2012-05-26 18:26:36.502330478 +0530 @@ -8,6 +8,13 @@ echo ">>> Generating initial ramdisk, using mkinitcpio. Please wait..." mkinitcpio -p linux${KERNEL_NAME} + pacman -Q dkms &>/dev/null + hasdkms=$? + if [ $hasdkms -eq 0 ]; then + echo ">>> Running the DKMS autoinstall service..." + dkms autoinstall -k ${KERNEL_VERSION} + fi + # compat symlinks for the official kernels only if [ -z "${KERNEL_NAME}" -o "${KERNEL_NAME}" = "-lts" ]; then loaders="$(find /boot -name syslinux.cfg -or -name extlinux.conf -or -name grub.cfg -or -name menu.lst)" @@ -53,6 +60,13 @@ depmod ${KERNEL_VERSION} echo ">>> Generating initial ramdisk, using mkinitcpio. Please wait..." mkinitcpio -p linux${KERNEL_NAME} + + pacman -Q dkms &>/dev/null + hasdkms=$? + if [ $hasdkms -eq 0 ]; then + echo ">>> Running the DKMS autoinstall service..." + dkms autoinstall -k ${KERNEL_VERSION} + fi } post_remove() {

@felixonmars @ValdikSS I think uksm is now official supported here: https://aur.archlinux.org/packages.php?ID=58802 And I've seen this interface on the user manual the first day I installed uksm, I am actually thinking this idea(remote debugging) is cool. Every other APPs have this feature nowadays, I just hate to dump my all messy stuff(i.e. /proc info, dmesg, core dumps....etc) if I see a bug and told to do so ---- that sounds more like they are debugging me! :P

@ValdikSS Hi, I e-mailed author of uksm and got his reply. This is documented but not that easy to find :P “usr_spt_enabled: Testing interface for future feature of “remote bug reporting”. Currently this interface is enabled by default for testing —- this will make uksmd send a meaningless udp message from your machine every 2 mintues. You can disable it by echo 0 > usr_spt_enabled. (via:http://kerneldedup.org/en/projects/uksm/uksmdoc/usage/).” And he stated they will split these codes so it won't be too confused.

Oh-oh-oh, what's this in uksm? + uksm_usr_spt_addr.sin_family = AF_INET; + uksm_usr_spt_addr.sin_port = htons(UKSM_USR_SPT_PORT); + uksm_usr_spt_addr.sin_addr.s_addr = in_aton("114.212.190.16"); + + uksm_usr_spt_iov.iov_base = &uksm_usr_spt; + uksm_usr_spt_iov.iov_len = sizeof(uksm_usr_spt); + uksm_usr_spt_msg.msg_name = (struct sockaddr *)&uksm_usr_spt_addr; + uksm_usr_spt_msg.msg_iov = &uksm_usr_spt_iov; + uksm_usr_spt_msg.msg_iovlen = 1; + uksm_usr_spt_msg.msg_control = NULL; + uksm_usr_spt_msg.msg_controllen = 0; + uksm_usr_spt_msg.msg_namelen = sizeof(uksm_usr_spt_addr); + uksm_usr_spt_msg.msg_flags = MSG_DONTWAIT; + + sock_sendmsg(uksm_usr_spt_sock, &uksm_usr_spt_msg, sizeof(uksm_usr_spt)); + uksm_usr_spt_last = jiffies;

ghen - why out of date?

Sorry guys, a single character typo caused mkinitcpio NOT to run. I have corrected it and re-uploaded 3.3.7-1 without bumping the pkgver. Sorry for the inconvenience.

Bump to v3.3.7-1 Changelog: https://lkml.org/lkml/2012/5/21/395 PKG Commit: http://pkgbuild.com/git/aur-mirror.git/commit/linux-ck?id=eb049a0cbcb676039312ec4ffbba61840711c341 As to the release of v3.4.0-1 ... CK emailed me to say that he is swamped with his personal life at the moment so it is unclear when we can expect a ck1 patchset for the 3.4 tree. As usual, please use the out-of-date button when CK pushes something.

Of course not, Latjay. Updating and testing. Should post here soon.