A kernel I’ve built from `sys-kernel/hardened-sources-3.15.5-r2` fails, for me, to boot properly — my computer’s display goes black after (I assume) the firmware passes control to the kernel, and stays black. blueness asked that I file a bug report after I mentioned my issue on IRC (in <ircs://chat.freenode.net/gentoo-hardened>). This issue also occurred for me with version 3.14.5-r2, resulting in me staying on version 3.13.6-r3, which did not exhibit this issue. The only patches I’ve applied to the kernel source tree are for filesystem (ZFS, AUFS) support. These patches were also applied to versions 3.14.5-r2 and 3.13.6-r3. I can attach my kernel configuration file, if that would be helpful.
worksforme 3.14.5-hardened-r2 do you have intel? do you have CONFIG_DRM_I915_FBDEV enabled?
(In reply to 8573dd from comment #0) > A kernel I’ve built from `sys-kernel/hardened-sources-3.15.5-r2` fails, > for me, to boot properly — my computer’s display goes black after (I > assume) the firmware passes control to the kernel, and stays black. > > blueness asked that I file a bug report after I mentioned my issue on > IRC (in <ircs://chat.freenode.net/gentoo-hardened>). > > This issue also occurred for me with version 3.14.5-r2, resulting in me > staying on version 3.13.6-r3, which did not exhibit this issue. > > The only patches I’ve applied to the kernel source tree are for > filesystem (ZFS, AUFS) support. These patches were also applied to > versions 3.14.5-r2 and 3.13.6-r3. > > I can attach my kernel configuration file, if that would be helpful. Can you try the equivalent vanilla kernel versions with the same config file and see if ti works. Also what video card do you have?
(In reply to Matthew Thode ( prometheanfire ) from comment #1) > do you have intel? do you have CONFIG_DRM_I915_FBDEV enabled? (In reply to Anthony Basile from comment #2) > Also what video card do you have? This computer has integrated Intel HD Graphics 4000, plus a discrete NVIDIA Quadro K1000M, the latter being disabled in firmware because I didn’t want to bother with it. My kernel configuration includes `CONFIG_DRM_I915_FBDEV=y`. (In reply to Anthony Basile from comment #2) > Can you try the equivalent vanilla kernel versions with the same config file > and see if ti works. The vanilla kernel being `sys-kernel/vanilla-sources`, and the equivalent versions being 3.14.17 and 3.15.10?
(In reply to 8573dd from comment #3) > > The vanilla kernel being `sys-kernel/vanilla-sources`, and the > equivalent versions being 3.14.17 and 3.15.10? Yes.
(In reply to Anthony Basile from comment #4) > (In reply to 8573dd from comment #3) > > > > The vanilla kernel being `sys-kernel/vanilla-sources`, and the > > equivalent versions being 3.14.17 and 3.15.10? > > Yes. Those version numbers were from `eix`’s output; when I tried to `emerge` vanilla-sources, `emerge`’s output seemed to show only 3.15.2, 3.14.9, 3.12.23, 3.10.45, 3.4.95, and 3.2.60 as available versions. Additionally, `emerge` refused to install vanilla-sources due to all versions of it being “masked by: ~amd64 keyword”. --- Also, after reading my logs of #gentoo-hardened, I tried hardened-sources 3.15.5-r2 again with: - The kernel configuration setting `CONFIG_GRKERNSEC_KSTACKOVERFLOW=n` — this did not seem to affect the issue. - The kernel command-line parameter `efi=old_map` — with this, the kernel did boot, but init failed (whether it’s `efi=old_map`’s fault, I wouldn’t know): INIT: version 2.88 booting OpenRC 0.12.4 is starting up Gentoo Linux (x86_64) Press I to enter interactive boot mode * /proc is already mounted * Mounting /run ... /lib64/rc/sh/init.sh: line 68: 3642 Killed […many spaces…] mount -t tmpfs -o mode=0755,nosuid,nodev,size=10% tmpfs /run - Both `CONFIG_GRKERNSEC_KSTACKOVERFLOW=n` and `efi=old_map` — this exhibited the same behavior as `efi=old_map` alone.
(In reply to 8573dd from comment #5) > (In reply to Anthony Basile from comment #4) > > (In reply to 8573dd from comment #3) > > > > > > The vanilla kernel being `sys-kernel/vanilla-sources`, and the > > > equivalent versions being 3.14.17 and 3.15.10? > > > > Yes. > > Those version numbers were from `eix`’s output; when I tried to `emerge` > vanilla-sources, `emerge`’s output seemed to show only 3.15.2, 3.14.9, > 3.12.23, 3.10.45, 3.4.95, and 3.2.60 as available versions. download the source from kernel.org > > Additionally, `emerge` refused to install vanilla-sources due to all > versions of it being “masked by: ~amd64 keyword”. > Add `sys-kernel/vanilla ~amd64` to /etc/portage/package.accept_keywords
This bug never did crystalize. I'm not sure what its about. Reopen it if we get more information.
