| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
| |
- Reenable debugging options.
|
|
|
|
| |
- Reenable debugging options.
|
|
|
|
|
|
|
|
|
|
|
| |
The benefit of THP has been somewhat questionable overall for a while,
and it's been known to cause performance issues with some workloads.
Upstream also considers it to be overly complicated and really not worth
it on machines with memory in the amounts found on typical desktops/SMB
servers.
Switch to using it via madvise, which most applications that care about
it should likely already be doing.
|
| |
|
| |
|
|
|
|
| |
This reverts commit ca43fd4b8dd1ce49ab0a9f239c6a056df104db72.
|
|
|
|
|
|
|
|
|
|
|
| |
Set CONFIG_BOOTPARAM_HOTPLUG_CPU0 to 'y'. Currently, to disable CPU0,
the "boot" cpu, one must specify cpu0_hotplug as a kernel parameter.
Setting this config variable to 1 enables it by default.
I've tested this on systems where it was expected to work and it seems to
function properly.
Signed-off-by: Prarit Bhargava <prarit@redhat.com>
|
|
|
|
|
|
|
|
|
|
|
| |
Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=1051284
Set CONFIG_X86_UV to 'y' and CONFIG_UV_MMTIMER to 'm' for large SGI boxes.
These settings will enable the booting of more than 16 processors.
Signed-off-by: Prarit Bhargava <prarit@redhat.com>
Cc: glaubitz@physik.fu-berlin.de
Cc: jwboyer@redhat.com
|
| |
|
| |
|
| |
|
|
|
|
| |
- Drop ACPI blacklist year options and patch (removed with upstream commit 4c47cb197e13 )
|
| |
|
|
|
|
|
|
|
|
| |
Replace x86-allow-1024-cpus.patch with the patch that contains the one hunk
not already applied upstream.
Don't install the 'trace' alias symlink for 'perf trace'. Seems much too
generic to just throw that into /usr/bin
|
|
|
|
| |
- Reenable debugging options.
|
|
|
|
| |
- Set NR_CPUS=1024 on non-debug x86_64 builds (MAXSMP is set on debug)
|
| |
|
| |
|
|
|
|
| |
- First DRM merge
|
|
|
|
| |
Enable 10gigE on 64-bit only.
|
| |
|
|
|
|
| |
- Enable the rtl8192e driver (rhbz 913753)
|
|
|
|
|
|
|
|
|
|
| |
So apparently we've been enabling CONFIG_NAMESPACES in Fedora kernels for
a very long time even though it was never explicitly set. The kconfig entry
for it looks like it should depend on CONFIG_EXPERT (which isn't set), but
either the fact that we've had CONFIG_NET_NS, etc enabled explicitly or the
fact that it defaults to !EXPERT seems to do something weird with menuconfig
and enables it in the final config. Whatever, just turn it back on. It
seems to break rtkit-daemon for some reason.
|
|
|
|
|
| |
- Enable NAMESPACES and CHECKPOINT_RESTORE on x86_64 for F19 CRIU feature
- Enable CONFIG_DEBUG_ATOMIC_SLEEP
|
| |
|
| |
|
|
|
|
| |
- Fix module blacklist patch to not leak a reference to the blacklist keyring
|
| |
|
| |
|
| |
|
|
|
|
| |
The utrace patch is left unapplied as it doesn't apply cleanly any longer
|
|
|
|
|
|
| |
Rawhide builds now use MAXSMP on x86.
For release builds, set x86-64 to support 64 CPUs.
If larger systems become widespread, we can increase in an update.
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
| |
Make CONFIG_EDAC_SBRIDGE only build on x86_64 since it has 64-bit divide errors
on 32-bit kernels.
|
| |
|
|
|
|
|
|
| |
First take on the upcomming Linux 3.2 kernel. The moving of the ethernet
drivers into vendor subdirectories caused a lot of churn in config-generic, but
hopefully it's a bit cleaner now.
|
| |
|
|
|
|
|
|
| |
Remove unnecessary duplicates, consolidate.
This should reduce having to set the same option twice on 32/64.
|
| |
|
|
|
|
| |
Adjust Makefile munging for 3.x numbering scheme
|
|
|
|
|
| |
Add epoll patch until it gets upstreamed
Add trial patch to fix the scheduling while atomic issues around pidmap_init
|
| |
|
|
|
|
|
|
| |
Testing that this part of the spec still works...
Hopefully the next release is indeed 3.1.0...
|
|
|
|
| |
revert CONFIG_X86_RESERVE_LOW=640, it breaks boot.
|