linux-kernel - [PATCH] clocksource/drivers/sp804: make CONFIG_ARM_TIMER

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

Date:   Fri, 16 Aug 2019 08:47:30 +0200
From:   Helmut Grohne <helmut.grohne@...enta.de>
To:     Daniel Lezcano <daniel.lezcano@...aro.org>,
        Thomas Gleixner <tglx@...utronix.de>
CC:     <linux-kernel@...r.kernel.org>
Subject: [PATCH] clocksource/drivers/sp804: make CONFIG_ARM_TIMER_SP804
 selectable again

Adding a dependency on CONFIG_COMPILE_TEST makes the relevant item
unselectable for practical purposes. The correct solution is to add a
dependency alternative on the relevant architecture.

Fixes: dfc82faad72520 ("clocksource/drivers/sp804: Add COMPILE_TEST to CONFIG_ARM_TIMER_SP804")
Link: https://lore.kernel.org/lkml/20190618120719.a4kgyiuljm5uivfq@laureti-dev/
Link: https://lore.kernel.org/lkml/alpine.DEB.2.21.1908152227590.1908@nanos.tec.linutronix.de/
Signed-off-by: Helmut Grohne <helmut.grohne@...enta.de>
---
 drivers/clocksource/Kconfig | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

Hi Thomas,

On Thu, Aug 15, 2019 at 10:30:39PM +0200, Thomas Gleixner wrote:
> The obvious fix is to add
> 
>       depends on ARM || ARM64 || COMPILE_TEST
> 
> instead of reverting the whole thing. Care to do that?

Incidentally, that's what I proposed earlier as RFC. Resending that
variant now.

I also note that there are likely more instances for this pattern.
Should they be fixed in a similar way? You can find a lot using the
following incantation:

    $ git describe --tags
    v5.3-rc4
    $ git ls-files -- "*/Kconfig" | xargs git grep --cached 'bool .* if COMPILE_TEST$' -- | wc -l
    185
    $

Seems like an anti-pattern to me. It is particularly common in the
clocksource subtree.

Helmut

diff --git a/drivers/clocksource/Kconfig b/drivers/clocksource/Kconfig
index 5e9317dc3d39..7081a250573b 100644
--- a/drivers/clocksource/Kconfig
+++ b/drivers/clocksource/Kconfig
@@ -393,7 +393,8 @@ config ARM_GLOBAL_TIMER
 	  This options enables support for the ARM global timer unit
 
 config ARM_TIMER_SP804
-	bool "Support for Dual Timer SP804 module" if COMPILE_TEST
+	bool "Support for Dual Timer SP804 module"
+	depends on ARM || ARM64 || COMPILE_TEST
 	depends on GENERIC_SCHED_CLOCK && CLKDEV_LOOKUP
 	select CLKSRC_MMIO
 	select TIMER_OF if OF
-- 
2.11.0