Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
MIME-Version: 1.0
Message-ID: <trinity-4ed3af00-b67c-468b-be4c-1e4b2379eca6-1625752868027@3c-app-gmx-bap19>
From:   Frank Wunderlich <frank-w@public-files.de>
To:     Dafna Hirschfeld <dafna.hirschfeld@collabora.com>
Cc:     Chun-Kuang Hu <chunkuang.hu@kernel.org>,
        Thomas Zimmermann <tzimmermann@suse.de>,
        David Airlie <airlied@linux.ie>, linux-kernel@vger.kernel.org,
        dri-devel@lists.freedesktop.org,
        Enric Balletbo i Serra <enric.balletbo@collabora.com>,
        linux-mediatek@lists.infradead.org,
        Matthias Brugger <matthias.bgg@gmail.com>,
        Collabora Kernel ML <kernel@collabora.com>
Subject: Aw: Re:  Re: BUG: MTK DRM/HDMI broken on 5.13 (mt7623/bpi-r2)
Content-Type: text/plain; charset=UTF-8
Date:   Thu, 8 Jul 2021 16:01:08 +0200
Importance: normal
Sensitivity: Normal
In-Reply-To: <df431199-0c65-fb9a-02b4-9f84899f37ee@collabora.com>
References: <trinity-cc8f5927-9aaf-43ae-a107-6a6229f1b481-1625565279264@3c-app-gmx-bs60>
 <25d61873-38ae-5648-faab-03431b74f777@collabora.com>
 <trinity-e6443313-a436-4e9d-a93c-1bef1cce135d-1625736911475@3c-app-gmx-bap19>
 <df431199-0c65-fb9a-02b4-9f84899f37ee@collabora.com>
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

> Gesendet: Donnerstag, 08. Juli 2021 um 14:30 Uhr
> Von: "Dafna Hirschfeld" <dafna.hirschfeld@collabora.com>
> > i see both messages, but mtk_crtc_ddp_irq is never called and so the o=
ther 2 not.
>
> Yes, In my case the irq isr is also not called after resume which cause =
the warning
> even though "enable_vblank" do get called. Don't know why is that.


> > comp->funcs->enable_vblank should be mtk_drm_crtc_enable_vblank, right=
?
>
> No, this is a bit confusing , there are also the funcs of the components=
, see in file mtk_drm_ddp_comp.c
> so for mt7623  it is mtk_ovl_enable_vblank.

thanks for pointing to this. in this function another struct is filled wit=
h the callback+data, and this callback seems to be called mtk_disp_ovl_irq=
_handler which name suggests also a irq as trigger

412     ret =3D devm_request_irq(dev, irq, mtk_disp_ovl_irq_handler,
413                    IRQF_TRIGGER_NONE, dev_name(dev), priv);
414     if (ret < 0) {
415         dev_err(dev, "Failed to request irq %d: %d\n", irq, ret);
416         return ret;
417     }

as i don't see this error in dmesg, i guess the registration was successfu=
l. added again some debug and it looks like the interrupt callback (mtk_di=
sp_ovl_irq_handler) is not called

[    5.125002] DEBUG: Passed mtk_disp_ovl_probe 416 int reg:0
[    6.344029] DEBUG: Passed mtk_drm_crtc_enable_vblank 510
[    6.344051] DEBUG: Passed mtk_ddp_comp_enable_vblank 117
[    6.344057] DEBUG: Passed mtk_ovl_enable_vblank 107
[    6.344062] DEBUG: Passed mtk_ovl_enable_vblank 112
[    6.344066] DEBUG: Passed mtk_ddp_comp_enable_vblank 121

=2D-- a/drivers/gpu/drm/mediatek/mtk_disp_ovl.c
+++ b/drivers/gpu/drm/mediatek/mtk_disp_ovl.c
@@ -86,6 +86,7 @@ static irqreturn_t mtk_disp_ovl_irq_handler(int irq, voi=
d *dev_id)
 {
        struct mtk_disp_ovl *priv =3D dev_id;

+printk(KERN_ALERT "DEBUG: Passed %s %d \n",__FUNCTION__,__LINE__);
        /* Clear frame completion interrupt */
        writel(0x0, priv->regs + DISP_REG_OVL_INTSTA);

@@ -93,6 +94,7 @@ static irqreturn_t mtk_disp_ovl_irq_handler(int irq, voi=
d *dev_id)
                return IRQ_NONE;

        priv->vblank_cb(priv->vblank_cb_data);
+printk(KERN_ALERT "DEBUG: Passed %s %d \n",__FUNCTION__,__LINE__);

        return IRQ_HANDLED;
 }
@@ -102,11 +104,12 @@ void mtk_ovl_enable_vblank(struct device *dev,
                           void *vblank_cb_data)
 {
        struct mtk_disp_ovl *ovl =3D dev_get_drvdata(dev);
-
+printk(KERN_ALERT "DEBUG: Passed %s %d \n",__FUNCTION__,__LINE__);
        ovl->vblank_cb =3D vblank_cb;
        ovl->vblank_cb_data =3D vblank_cb_data;
        writel(0x0, ovl->regs + DISP_REG_OVL_INTSTA);
        writel_relaxed(OVL_FME_CPL_INT, ovl->regs + DISP_REG_OVL_INTEN);
+printk(KERN_ALERT "DEBUG: Passed %s %d \n",__FUNCTION__,__LINE__);
 }

 void mtk_ovl_disable_vblank(struct device *dev)
@@ -410,6 +413,7 @@ static int mtk_disp_ovl_probe(struct platform_device *=
pdev)

        ret =3D devm_request_irq(dev, irq, mtk_disp_ovl_irq_handler,
                               IRQF_TRIGGER_NONE, dev_name(dev), priv);
+printk(KERN_ALERT "DEBUG: Passed %s %d int reg:%d\n",__FUNCTION__,__LINE_=
_,ret);
        if (ret < 0) {
                dev_err(dev, "Failed to request irq %d: %d\n", irq, ret);
                return ret;


how can we trace this further? maybe watchdog related?

> >
> > "watchdog: watchdog0: watchdog did not stop!"
> >
> > i see this with my 5.13, 5.12-drm (5.12.0+mtk/core drm-patches) and 5.=
12.14 too (hdmi is working there), but not 5.12.0!
> > that means something in drm-patches (mtk/core) breaks watchdog. maybe =
the recursion mentioned above?