linux-kernel - [PATCH] arm64: Only HAVE_HARDLOCKUP_DETECTOR

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Date:   Tue, 23 May 2023 07:39:53 -0700
From:   Douglas Anderson <dianders@...omium.org>
To:     Andrew Morton <akpm@...ux-foundation.org>,
        Arnd Bergmann <arnd@...db.de>
Cc:     Douglas Anderson <dianders@...omium.org>,
        Catalin Marinas <catalin.marinas@....com>,
        Lecopzer Chen <lecopzer.chen@...iatek.com>,
        Pingfan Liu <kernelfans@...il.com>,
        Sumit Garg <sumit.garg@...aro.org>,
        Will Deacon <will@...nel.org>,
        linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: [PATCH] arm64: Only HAVE_HARDLOCKUP_DETECTOR_PERF if the PMU config is enabled

If you try to enable the arm64 perf-based hardlockup detector but you
don't enable CONFIG_ARM_PMU then you'll get an error:

arch/arm64/kernel/watchdog_hld.c: In function 'arch_perf_nmi_is_available':
arch/arm64/kernel/watchdog_hld.c:35:16: error: implicit declaration of function 'arm_pmu_irq_is_nmi' [-Werror=implicit-function-declaration]
   35 |         return arm_pmu_irq_is_nmi();

It doesn't make sense to enable HAVE_HARDLOCKUP_DETECTOR_PERF if the
PMU isn't enabled. Let's add a dependency. HW_PERF_EVENTS is a synonum
for ARM_PMU and makes the most logical sense here, so add the
dependency on that.

Reported-by: Arnd Bergmann <arnd@...db.de>
Closes: https://lore.kernel.org/r/20230522114922.1052421-1-arnd@kernel.org
Fixes: 02ea35ee19d9 ("arm64: enable perf events based hard lockup detector")
Signed-off-by: Douglas Anderson <dianders@...omium.org>
---
This is an alternative fix to the one Arnd proposed. I think it's a
little cleaner / more correct.

I'm not sure the exact way to mark "Fixes" for things in Andrew's tree
these days. I don't think the git hash is stable? I wouldn't object to
this being squashed into the patch it's fixing.

 arch/arm64/Kconfig | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index 3eacf40da850..0b055e6cda00 100644
--- a/arch/arm64/Kconfig
+++ b/arch/arm64/Kconfig
@@ -203,7 +203,8 @@ config ARM64
 	select HAVE_FUNCTION_ERROR_INJECTION
 	select HAVE_FUNCTION_GRAPH_TRACER
 	select HAVE_GCC_PLUGINS
-	select HAVE_HARDLOCKUP_DETECTOR_PERF if PERF_EVENTS && HAVE_PERF_EVENTS_NMI
+	select HAVE_HARDLOCKUP_DETECTOR_PERF if PERF_EVENTS && \
+		HW_PERF_EVENTS && HAVE_PERF_EVENTS_NMI
 	select HAVE_HW_BREAKPOINT if PERF_EVENTS
 	select HAVE_IOREMAP_PROT
 	select HAVE_IRQ_TIME_ACCOUNTING
-- 
2.40.1.698.g37aff9b760-goog