]> git.ipfire.org Git - thirdparty/openembedded/openembedded-core-contrib.git/commit
linux-yocto/6.4: fix uninitialized read in nohz_full/isolcpus setup
authorBruce Ashfield <bruce.ashfield@gmail.com>
Mon, 21 Aug 2023 15:44:41 +0000 (11:44 -0400)
committerRichard Purdie <richard.purdie@linuxfoundation.org>
Tue, 22 Aug 2023 14:13:51 +0000 (15:13 +0100)
commitbb5865b10316754213711551fa68a53a4fb741d9
tree4036daf3c60c61564e2910298dbef2581cc0912b
parent92642acd1166b7990fb520337b78a4d37ce7cbe4
linux-yocto/6.4: fix uninitialized read in nohz_full/isolcpus setup

Integrating the following commit(s) to linux-yocto/6.4:

    4675ec8d0b9a kernel/sched: Fix uninitialized read in nohz_full/isolcpus setup

The carry forward of commit from the v5.15 linux-yocto kernel:

https://git.yoctoproject.org/linux-yocto/commit/?id=97c96388922

...in which case the sanity checks are properly *after* the allocation
and processing of the bootargs into the cpumask.

However, it seems patch (or wiggle?) apparently decided to put the
sanity checks *before* the population of the cpumask during the
carry-forward and generation of the new v6.1 kernel.  Meaning they are
validating uninitialized memory and hence nohz_full= and isolcpus= are
subject to random failures even for valid input ranges.

(From OE-Core rev: b3a71c7d36e3a4833e90728499e08543257042a1)

Signed-off-by: Adrian Cinal <adriancinal1@gmail.com>
Acked-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: Bruce Ashfield <bruce.ashfield@gmail.com>
Signed-off-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
meta/recipes-kernel/linux/linux-yocto-rt_6.4.bb
meta/recipes-kernel/linux/linux-yocto-tiny_6.4.bb
meta/recipes-kernel/linux/linux-yocto_6.4.bb