netdev - Re: [PATCH net-next] can: rockchip_canfd: Use div_s64() in rkcanfd_timestamp

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]

Message-ID: <CAMZ6RqKA77BWAMxL4_iAcye78JXDGxGQcEd9BJg7uCL9EO11XA@mail.gmail.com>
Date: Tue, 10 Sep 2024 15:32:15 +0900
From: Vincent MAILHOL <mailhol.vincent@...adoo.fr>
To: Nathan Chancellor <nathan@...nel.org>
Cc: Marc Kleine-Budde <mkl@...gutronix.de>, kernel@...gutronix.de, 
	"David S. Miller" <davem@...emloft.net>, Eric Dumazet <edumazet@...gle.com>, 
	Jakub Kicinski <kuba@...nel.org>, Paolo Abeni <pabeni@...hat.com>, Heiko Stuebner <heiko@...ech.de>, 
	linux-can@...r.kernel.org, netdev@...r.kernel.org, 
	linux-arm-kernel@...ts.infradead.org, linux-rockchip@...ts.infradead.org, 
	llvm@...ts.linux.dev, patches@...ts.linux.dev
Subject: Re: [PATCH net-next] can: rockchip_canfd: Use div_s64() in rkcanfd_timestamp_init()

On Tue. 10 Sep. 2024 at 01:24, Nathan Chancellor <nathan@...nel.org> wrote:
> When building with clang for a 32-bit target, such as arm, a libcall is
> generated when dividing the result of clocksource_cyc2ns(), which
> returns a signed 64-bit integer:
>
>   ERROR: modpost: "__aeabi_ldivmod" [drivers/net/can/rockchip/rockchip_canfd.ko] undefined!
>
> Use div_s64() to avoid generating the libcall.
>
> Fixes: 4e1a18bab124 ("can: rockchip_canfd: add hardware timestamping support")
> Signed-off-by: Nathan Chancellor <nathan@...nel.org>
> ---
> This does not happen with GCC, likely because it implements
> optimizations for division by a constant that clang does not implement.
> ---
>  drivers/net/can/rockchip/rockchip_canfd-timestamp.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/net/can/rockchip/rockchip_canfd-timestamp.c b/drivers/net/can/rockchip/rockchip_canfd-timestamp.c
> index 81cccc5fd8384ee1fec919077db48632f0fe7cc2..4ca01d385ffffdeec964b7fd954d8ba3ba3a1381 100644
> --- a/drivers/net/can/rockchip/rockchip_canfd-timestamp.c
> +++ b/drivers/net/can/rockchip/rockchip_canfd-timestamp.c
> @@ -71,7 +71,7 @@ void rkcanfd_timestamp_init(struct rkcanfd_priv *priv)
>
>         max_cycles = div_u64(ULLONG_MAX, cc->mult);
>         max_cycles = min(max_cycles, cc->mask);
> -       work_delay_ns = clocksource_cyc2ns(max_cycles, cc->mult, cc->shift) / 3;
> +       work_delay_ns = div_s64(clocksource_cyc2ns(max_cycles, cc->mult, cc->shift), 3);

I was surprised at first to see div_s64() instead of div_u64(), but
clocksource_cyc2ns() indeed returns a s64. So this looks correct.

>         priv->work_delay_jiffies = nsecs_to_jiffies(work_delay_ns);
>         INIT_DELAYED_WORK(&priv->timestamp, rkcanfd_timestamp_work);

Reviewed-by: Vincent Mailhol <mailhol.vincent@...adoo.fr>