]> git.ipfire.org Git - thirdparty/kernel/stable.git/commit
ftrace: Do not blindly read the ip address in ftrace_bug()
authorSteven Rostedt (VMware) <rostedt@goodmis.org>
Tue, 8 Jun 2021 01:39:08 +0000 (21:39 -0400)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Wed, 16 Jun 2021 09:36:36 +0000 (11:36 +0200)
commit4aedc2bc2b32c93555f47c95610efb89cc1ec09b
tree39725a1e04c0db71013816b4fa15fc40a574f5b9
parentb66107bf77730f9e02d53e52a7f2e6e4d2b9107f
ftrace: Do not blindly read the ip address in ftrace_bug()

commit 6c14133d2d3f768e0a35128faac8aa6ed4815051 upstream.

It was reported that a bug on arm64 caused a bad ip address to be used for
updating into a nop in ftrace_init(), but the error path (rightfully)
returned -EINVAL and not -EFAULT, as the bug caused more than one error to
occur. But because -EINVAL was returned, the ftrace_bug() tried to report
what was at the location of the ip address, and read it directly. This
caused the machine to panic, as the ip was not pointing to a valid memory
address.

Instead, read the ip address with copy_from_kernel_nofault() to safely
access the memory, and if it faults, report that the address faulted,
otherwise report what was in that location.

Link: https://lore.kernel.org/lkml/20210607032329.28671-1-mark-pk.tsai@mediatek.com/
Cc: stable@vger.kernel.org
Fixes: 05736a427f7e1 ("ftrace: warn on failure to disable mcount callers")
Reported-by: Mark-PK Tsai <mark-pk.tsai@mediatek.com>
Tested-by: Mark-PK Tsai <mark-pk.tsai@mediatek.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
kernel/trace/ftrace.c