mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: David Laight <david.laight.linux@gmail.com>
To: Arnd Bergmann <arnd@kernel.org>
Cc: Vikash Garodia <vikash.garodia@oss.qualcomm.com>,
	Dikshita Agarwal <dikshita.agarwal@oss.qualcomm.com>,
	"Bryan O'Donoghue" <bod@kernel.org>,
	Mauro Carvalho Chehab <mchehab@kernel.org>,
	Sachin Kumar Garg <sachin.garg@oss.qualcomm.com>,
	Arnd Bergmann <arnd@arndb.de>,
	Abhinav Kumar <abhinav.kumar@linux.dev>,
	Wangao Wang <wangao.wang@oss.qualcomm.com>,
	Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>,
	Deepa Guthyappa Madivalara <deepa.madivalara@oss.qualcomm.com>,
	linux-media@vger.kernel.org, linux-arm-msm@vger.kernel.org,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH] media: iris: avoid plain 64-bit division
Date: Sat, 3 Oct 2026 18:37:49 +0100	[thread overview]
Message-ID: <20261003183749.4f113fcd@pumpkin> (raw)
In-Reply-To: <20261002124435.3435628-1-arnd@kernel.org>

On Fri,  2 Oct 2026 14:44:22 +0200
Arnd Bergmann <arnd@kernel.org> wrote:

> From: Arnd Bergmann <arnd@arndb.de>
> 
> inst->fw_caps[BITRATE].value is a 64-bit integer, which on 32-bit
> architectures cannot be divided:
> 
> ERROR: modpost: drivers/media/platform/qcom/iris/qcom-iris.ko: symbol '__divdi3' undefined!
> 
> In particular, doing three divisions in a row is particularly slow,
> even on 64-bit CPUs.
> 
> Rework the calculation to replace it with a single division, thus
> avoiding the build failure.
> 
> Fixes: 5f21cd9a4ae4 ("media: iris: add support for multi_slice in iris encoder")
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> ---
> Please check the math, as I'm not sure if this is the intended
> amount of rounding. Previously it was rounding down twice and up
> once, now it just rounds up during the combined division.

Can definitely change the result:
	DIV_ROUND_UP(7 / 3, 2) is 1
but
	DIV_ROUND_UP(7, 2 * 3) is 2
who knows what the value is supposed to be.

> 
> Ideally this would be replaced with a cast of the incoming value
> to u32, but doing this requires to prove that this does not overflow.

You only need to prove that in incoming value is less than 32G (Hz).
The '/ 8' is cheap(ish) and can be done first.

David

> 
> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
> ---
>  drivers/media/platform/qcom/iris/iris_ctrls.c | 5 +++--
>  1 file changed, 3 insertions(+), 2 deletions(-)
> 
> diff --git a/drivers/media/platform/qcom/iris/iris_ctrls.c b/drivers/media/platform/qcom/iris/iris_ctrls.c
> index 8c4c47944f22..5294999fdf7b 100644
> --- a/drivers/media/platform/qcom/iris/iris_ctrls.c
> +++ b/drivers/media/platform/qcom/iris/iris_ctrls.c
> @@ -1588,8 +1588,9 @@ int iris_set_slice_count(struct iris_inst *inst, enum platform_inst_fw_cap_type
>  	} else if (slice_mode == V4L2_MPEG_VIDEO_MULTI_SLICE_MODE_MAX_BYTES) {
>  		hfi_value = inst->fw_caps[SLICE_MAX_BYTES].value;
>  		if (rc_type != HFI_RC_OFF) {
> -			max_avg_slicesize = DIV_ROUND_UP((inst->fw_caps[BITRATE].value / fps) / 8,
> -							 slice_caps->max_slices_per_frame);
> +			max_avg_slicesize = DIV_U64_ROUND_UP(inst->fw_caps[BITRATE].value,
> +							     fps * 8 *
> +							     slice_caps->max_slices_per_frame);
>  		} else {
>  			/*
>  			 * No bitrate target exists under RC_OFF, so approximate a


      reply	other threads:[~2026-10-03 17:37 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-10-02 12:44 Arnd Bergmann
2026-10-03 17:37 ` David Laight [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20261003183749.4f113fcd@pumpkin \
    --to=david.laight.linux@gmail.com \
    --cc=abhinav.kumar@linux.dev \
    --cc=arnd@arndb.de \
    --cc=arnd@kernel.org \
    --cc=bod@kernel.org \
    --cc=deepa.madivalara@oss.qualcomm.com \
    --cc=dikshita.agarwal@oss.qualcomm.com \
    --cc=dmitry.baryshkov@oss.qualcomm.com \
    --cc=linux-arm-msm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-media@vger.kernel.org \
    --cc=mchehab@kernel.org \
    --cc=sachin.garg@oss.qualcomm.com \
    --cc=vikash.garodia@oss.qualcomm.com \
    --cc=wangao.wang@oss.qualcomm.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®