I habitually install HTTP, started by loading installation linux and initrd from HD after fetching from mirrors with wget, most often with 13.1's Grub 0.97-194 (probably 32 bit in most or all cases, but I'm not sure how to tell after the fact, and have no idea if there is any difference, or if it could matter) installed to a primary partition that never gets mounted as /boot, which is where I place the linux/initrd installation set. In recent minutes in order to follow up to an old bug, I initialized installation on 3 different Intel CPU machines, two using the 64 TW installation set, 1 using 32 bit TW set. The first 64 took a very long time for the the kernel and initrd load message from Grub to clear and kernel messages to start appearing. I didn't time it. On the second 64 I timed it: more than 7 minutes. On the 32, it took less than 15 seconds. I've seen similarly perplexing similarity on a bunch of 64 bit installations of Fedora, 13.2 and TW, delaying as much as 12+ minutes, but none that I can remember on 42.1, and none with 32 bit kernels, all on only two SATA HDs in systems, plus a third HD that is a backup/clone of one of those two. Moving that clone HD to a different but very similar machine preserved the delays on the other machine. On one of the two machines having the problem, I think I've eliminated the problem by "restoring backups" ("cloning" via rsync from a different machine's HD's similar installations) after reformatting the originally EXT4 / partitions on which the slow-loaders live, again EXT4. On the other, I did a similar restore, except only for one / partition, and to a new partition formatted EXT3, so that on that machine, all problem installations as yet remain problematic. I first reported strangely long load times 6 months ago https://www.mail-archive.com/initramfs@vger.kernel.org/msg04168.html and after nothing happening there, last month on kernel https://lists.opensuse.org/opensuse-kernel/2016-03/msg00024.html and more this month https://lists.opensuse.org/opensuse-kernel/2016-04/msg00002.html https://lists.opensuse.org/opensuse-kernel/2016-04/msg00004.html Any ideas what could be causing these exclusive to 64 bit delays, or how to go about trying to find out what their cause is? Do the portions of Grub responsible for kernel/initrd loading differ between their 32 bit and 64 bit source rpms? I expect not, as whether the primary partition's Grub loads the kernels directly or chainloads to the Grub installed on each respective / partition has produced no observable differences in the delays. http://fm.no-ip.com/Tmp/Linux/Delays/ has a bunch of files that track which installations and kernels do or don't delay, plus various hardware characteristics, e.g. partitioning. I've checked BIOS of the involved machines. On none can I remember seeing any explicit mention of AHCI, mainly only choices between enhanced or compatible, or similar language. All are set to best guess equivalent to AHCI enabled. -- "The wise are known for their understanding, and pleasant words are persuasive." Proverbs 16:21 (New Living Translation) Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata *** http://fm.no-ip.com/ -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org