4.19-stable review patch. If anyone has any objections, please let me know.
------------------ From: Mathieu Desnoyers <[email protected]> commit 0ac569bf6a7983c0c5747d6df8db9dc05bc92b6c upstream. commit e46daee53bb5 ("ARM: 8806/1: kprobes: Fix false positive with FORTIFY_SOURCE") introduced a regression in optimized kprobes. It triggers "invalid instruction" oopses when using kprobes instrumentation through lttng and perf. This commit was introduced in kernel v4.20, and has been backported to stable kernels 4.19 and 4.14. This crash was also reported by Hongzhi Song on the redhat bugzilla where the patch was originally introduced. Link: https://bugzilla.redhat.com/show_bug.cgi?id=1639397 Link: https://bugs.lttng.org/issues/1174 Link: https://lore.kernel.org/lkml/[email protected] Fixes: e46daee53bb5 ("ARM: 8806/1: kprobes: Fix false positive with FORTIFY_SOURCE") Signed-off-by: Mathieu Desnoyers <[email protected]> Reported-by: Robert Berger <[email protected]> Tested-by: Robert Berger <[email protected]> Acked-by: Kees Cook <[email protected]> Cc: Robert Berger <[email protected]> Cc: Masami Hiramatsu <[email protected]> Cc: William Cohen <[email protected]> Cc: Laura Abbott <[email protected]> Cc: Kees Cook <[email protected]> Cc: <[email protected]> # v4.14+ Cc: [email protected] Cc: [email protected] Signed-off-by: Russell King <[email protected]> Signed-off-by: Greg Kroah-Hartman <[email protected]> --- arch/arm/probes/kprobes/opt-arm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) --- a/arch/arm/probes/kprobes/opt-arm.c +++ b/arch/arm/probes/kprobes/opt-arm.c @@ -247,7 +247,7 @@ int arch_prepare_optimized_kprobe(struct } /* Copy arch-dep-instance from template. */ - memcpy(code, (unsigned char *)optprobe_template_entry, + memcpy(code, (unsigned long *)&optprobe_template_entry, TMPL_END_IDX * sizeof(kprobe_opcode_t)); /* Adjust buffer according to instruction. */
