From: sashiko-bot@kernel.org
To: "Francesco Dolcini" <francesco@dolcini.it>
Cc: linux-phy@lists.infradead.org, vkoul@kernel.org,
neil.armstrong@linaro.org, olteanv@gmail.com
Subject: Re: [PATCH v1] Revert "phy: freescale: fsl-samsung-hdmi: initialize default rate"
Date: Mon, 16 Nov 2026 16:40:11 +0000 [thread overview]
Message-ID: <sashiko-outbox-217371@kernel.org> (raw)
In-Reply-To: <20261116162409.218059-1-francesco@dolcini.it>
Thank you for your contribution! Sashiko AI review found 2 potential
issues to consider:
- [Medium] phy: freescale: fsl-samsung-hdmi: apbclk leak in
fsl_samsung_hdmi_phy_resume()
- [Medium] phy: freescale: fsl-samsung-hdmi: NULL deref in
fsl_samsung_hdmi_phy_find_settings()
--
commit 9481293ffcfc6d865a5df21db581bbc72e6903fc
Author: Francesco Dolcini <francesco.dolcini@toradex.com>
Revert "phy: freescale: fsl-samsung-hdmi: initialize default rate"
This commit reverts commit c01ccf91b257 ("phy: freescale: fsl-samsung-hdmi:
initialize default rate") so that the driver no longer configures the
74.25 MHz default rate during probe or toggles refclk on resume.
> diff --git a/drivers/phy/freescale/phy-fsl-samsung-hdmi.c b/drivers/phy/freescale/phy-fsl-samsung-hdmi.c
> index 02539ad0b0fed..ded142020e725 100644
> --- a/drivers/phy/freescale/phy-fsl-samsung-hdmi.c
> +++ b/drivers/phy/freescale/phy-fsl-samsung-hdmi.c
[ ... ]
> @@ -664,23 +664,6 @@ static int fsl_samsung_hdmi_phy_probe(struct platform_device *pdev)
> return dev_err_probe(phy->dev, PTR_ERR(phy->refclk),
> "failed to get ref clk\n");
>
> - /*
> - * Establish the rate reported by recalc_rate() before registering the
> - * clock. Otherwise CCF may skip the first set_rate() when it requests
> - * the assumed default rate, leaving the PHY unconfigured.
> - */
> - ret = clk_prepare_enable(phy->refclk);
> - if (ret)
> - return dev_err_probe(phy->dev, ret,
> - "failed to enable ref clk\n");
> -
> - default_cfg = fsl_samsung_hdmi_phy_lookup_rate(PHY_DEFAULT_RATE);
> - ret = fsl_samsung_hdmi_phy_configure(phy, default_cfg);
> - clk_disable_unprepare(phy->refclk);
> - if (ret)
> - return dev_err_probe(phy->dev, ret,
> - "failed to configure default rate\n");
> -
> pm_runtime_get_noresume(phy->dev);
> pm_runtime_set_active(phy->dev);
> pm_runtime_enable(phy->dev);
[Severity: Medium]
With phy->cur_cfg left NULL until the first fsl_samsung_hdmi_phy_configure()
call, can fsl_samsung_hdmi_phy_find_settings() dereference a NULL
phy->cur_cfg when debug logging is enabled?
Both fsl_samsung_hdmi_phy_clk_determine_rate() and
fsl_samsung_hdmi_phy_clk_set_rate() call fsl_samsung_hdmi_phy_find_settings()
before fsl_samsung_hdmi_phy_configure() sets phy->cur_cfg. When the
requested rate is not an exact match and is closer to fract_div_phy->pixclk
than int_div_clk, fsl_samsung_hdmi_phy_find_settings() logs
phy->cur_cfg->pixclk instead of fract_div_phy->pixclk:
drivers/phy/freescale/phy-fsl-samsung-hdmi.c:fsl_samsung_hdmi_phy_find_settings() {
...
/* Calculate the absolute value of the differences and return whichever is closest */
if (abs((long)rate - (long)int_div_clk) <
abs((long)rate - (long)fract_div_phy->pixclk)) {
dev_dbg(phy->dev, "integer divider = %u\n", calculated_phy_pll_cfg.pixclk);
return &calculated_phy_pll_cfg;
}
dev_dbg(phy->dev, "fractional divider = %u\n", phy->cur_cfg->pixclk);
return fract_div_phy;
}
> @@ -719,7 +702,7 @@ static int __maybe_unused fsl_samsung_hdmi_phy_suspend(struct device *dev)
> static int __maybe_unused fsl_samsung_hdmi_phy_resume(struct device *dev)
> {
> struct fsl_samsung_hdmi_phy *phy = dev_get_drvdata(dev);
> - int ret;
> + int ret = 0;
>
> ret = clk_prepare_enable(phy->apbclk);
> if (ret) {
> @@ -727,23 +710,11 @@ static int __maybe_unused fsl_samsung_hdmi_phy_resume(struct device *dev)
> return ret;
> }
>
> - if (!phy->cur_cfg)
> - return 0;
> -
> - ret = clk_prepare_enable(phy->refclk);
> - if (ret) {
> - dev_err(phy->dev, "failed to enable ref clk\n");
> - goto disable_apbclk;
> - }
> -
> - ret = fsl_samsung_hdmi_phy_configure(phy, phy->cur_cfg);
> - clk_disable_unprepare(phy->refclk);
> - if (!ret)
> - return 0;
> + if (phy->cur_cfg)
> + ret = fsl_samsung_hdmi_phy_configure(phy, phy->cur_cfg);
>
> -disable_apbclk:
> - clk_disable_unprepare(phy->apbclk);
> return ret;
> +
> }
[Severity: Medium]
Does fsl_samsung_hdmi_phy_resume() leak the enable count on phy->apbclk when
fsl_samsung_hdmi_phy_configure() fails? If the PLL fails to lock on runtime
resume, the device stays suspended in runtime PM error state without
disabling and unpreparing phy->apbclk.
--
Sashiko AI review · https://sashiko.dev/#/patchset/20261116162409.218059-1-francesco@dolcini.it?part=1
--
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy
prev parent reply other threads:[~2026-11-16 16:40 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-11-16 16:24 [PATCH v1] Revert "phy: freescale: fsl-samsung-hdmi: initialize default rate" Francesco Dolcini
2026-11-16 16:40 ` sashiko-bot [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=sashiko-outbox-217371@kernel.org \
--to=sashiko-bot@kernel.org \
--cc=francesco@dolcini.it \
--cc=linux-phy@lists.infradead.org \
--cc=neil.armstrong@linaro.org \
--cc=olteanv@gmail.com \
--cc=sashiko-reviews@lists.linux.dev \
--cc=vkoul@kernel.org \
/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