[PATCH v19 4/8] drm: bridge: Cadence: Add MHDP8501 DP/HDMI driver

Dmitry Baryshkov dmitry.baryshkov at linaro.org
Mon Dec 9 02:34:13 PST 2024


On Mon, Dec 09, 2024 at 08:38:01AM +0000, Sandor Yu wrote:
> 
> 
> > -----Original Message-----
> > From: Dmitry Baryshkov <dmitry.baryshkov at linaro.org>
> > Sent: 2024年11月27日 3:21
> > To: Sandor Yu <sandor.yu at nxp.com>
> > Cc: andrzej.hajda at intel.com; neil.armstrong at linaro.org; Laurent Pinchart
> > <laurent.pinchart at ideasonboard.com>; jonas at kwiboo.se;
> > jernej.skrabec at gmail.com; airlied at gmail.com; daniel at ffwll.ch;
> > robh+dt at kernel.org; krzysztof.kozlowski+dt at linaro.org;
> > shawnguo at kernel.org; s.hauer at pengutronix.de; festevam at gmail.com;
> > vkoul at kernel.org; dri-devel at lists.freedesktop.org;
> > devicetree at vger.kernel.org; linux-arm-kernel at lists.infradead.org;
> > linux-kernel at vger.kernel.org; linux-phy at lists.infradead.org;
> > mripard at kernel.org; kernel at pengutronix.de; dl-linux-imx
> > <linux-imx at nxp.com>; Oliver Brown <oliver.brown at nxp.com>;
> > alexander.stein at ew.tq-group.com; sam at ravnborg.org
> > Subject: [EXT] Re: [PATCH v19 4/8] drm: bridge: Cadence: Add MHDP8501
> > DP/HDMI driver

Please strip such headers from your response emails. They take a lot of
space and are useless. See how it's usually handled by other email
clients.

> > 
> > Caution: This is an external email. Please take care when clicking links or
> > opening attachments. When in doubt, report the message using the 'Report
> > this email' button
> > 
> > 
> > On Tue, Nov 26, 2024 at 10:11:49PM +0800, Sandor Yu wrote:
> > > Add a new DRM DisplayPort and HDMI bridge driver for Candence
> > MHDP8501
> > > used in i.MX8MQ SOC. MHDP8501 could support HDMI or DisplayPort
> > > standards according embedded Firmware running in the uCPU.
> > >
> > > For iMX8MQ SOC, the DisplayPort/HDMI FW was loaded and activated by
> > > SOC's ROM code. Bootload binary included respective specific firmware
> > > is required.
> > >
> > > Driver will check display connector type and
> > > then load the corresponding driver.
> > >
> > > Signed-off-by: Sandor Yu <Sandor.yu at nxp.com>
> > > ---

[...]

> > > +
> > > +static int reset_pipe(struct drm_crtc *crtc)
> > > +{
> > > +     struct drm_atomic_state *state;
> > > +     struct drm_crtc_state *crtc_state;
> > > +     struct drm_modeset_acquire_ctx ctx;
> > > +     int ret;
> > > +
> > > +     state = drm_atomic_state_alloc(crtc->dev);
> > > +     if (!state)
> > > +             return -ENOMEM;
> > > +
> > > +     drm_modeset_acquire_init(&ctx, 0);
> > > +
> > > +     state->acquire_ctx = &ctx;
> > > +
> > > +     crtc_state = drm_atomic_get_crtc_state(state, crtc);
> > > +     if (IS_ERR(crtc_state)) {
> > > +             ret = PTR_ERR(crtc_state);
> > > +             goto out;
> > > +     }
> > > +
> > > +     crtc_state->connectors_changed = true;
> > > +
> > > +     ret = drm_atomic_commit(state);
> >  
> > I'd say, this looks like a hack to me.
> 
> Basically, those code followed the sequence as vc4 HDMI and i915 according review comments in v15.
> HDMI driver should " Reenabling the scrambler when a display is disconnected and reconnected".
> 
> To be honest, I'm not 100% convinced by this implementation, 
> as the code doesn't seem to align perfectly with the current DRM framework. 
> However, adding this feature would indeed address the issue of applications (like modetest) that don't handle HPD events 
> but still require cable plug/unplug support when operating in scrambler-enabled video modes.

Ack. This matches vc4's reset_pipe().

> 
> > 
> > > +out:
> > > +     drm_atomic_state_put(state);
> > > +     drm_modeset_drop_locks(&ctx);
> > > +     drm_modeset_acquire_fini(&ctx);
> > > +
> > > +     return ret;
> > > +}
> > > +
> > > +void cdns_hdmi_reset_link(struct cdns_mhdp8501_device *mhdp)
> > > +{
> > > +     struct drm_connector *connector = mhdp->curr_conn;
> > > +     const struct drm_edid *drm_edid;
> > > +     struct drm_connector_state *conn_state;
> > > +     struct drm_crtc_state *crtc_state;
> > > +     struct drm_crtc *crtc;
> > > +
> > > +     if (!connector)
> > > +             return;
> > > +
> > > +     drm_edid = drm_edid_read_custom(connector,
> > cdns_hdmi_get_edid_block, mhdp);
> > > +     drm_edid_connector_update(connector, drm_edid);
> > 
> > Why?
> 
> MHDP8501 HDMI support scrambling.low_rates.
> The scrambler status can change when switching HDMI display monitors, even if the video mode stays the same. 
> The MHDP8501's HDMI code updates the EDID data to accommodate displays with different scrambler capabilities.

Ack. I'd suggest renaming cdns_hdmi_reset_link() to
cdns_hdmi_handle_hotplug() or any other similar name.

> 
> > 
> > > +
> > > +     if (!drm_edid)
> > > +             return;
> > > +
> > > +     drm_edid_free(drm_edid);
> > > +
> > > +     conn_state = connector->state;
> > > +     crtc = conn_state->crtc;
> > > +     if (!crtc)
> > > +             return;
> > > +
> > > +     crtc_state = crtc->state;
> > > +     if (!crtc_state->active)
> > > +             return;
> > > +
> > > +     /*
> > > +      * HDMI 2.0 says that one should not send scrambled data
> > > +      * prior to configuring the sink scrambling, and that
> > > +      * TMDS clock/data transmission should be suspended when
> > > +      * changing the TMDS clock rate in the sink. So let's
> > > +      * just do a full modeset here, even though some sinks
> > > +      * would be perfectly happy if were to just reconfigure
> > > +      * the SCDC settings on the fly.
> > > +      */
> > > +     reset_pipe(crtc);
> > 
> > We are not supposed to cause full-modeset flicker if we can avoid it. Is
> > it the case here?
> 
> My implementation differs slightly from the VC4 HDMI in that it doesn't check if the current video mode requires enabling the scrambler. 
> This is because we considered a scenario where the previous display was operating at 1080p60 with scrambling enabled, 
> and then the cable was plugged into a display that doesn't support scrambling at low rates. 
> This case also necessitates a full modeset.

Well, if this is a part of the hotplug, then of course there are no
flickering issues as it happens during hotplug.

> 
> I haven't found any video flickering issues in my actual tests.
> 
> > 
> > > +}
> > > +
> > > +static int cdns_hdmi_i2c_write(struct cdns_mhdp8501_device *mhdp,
> > > +                            struct i2c_msg *msgs)
> > > +{
> > > +     u8 msg[5], reg[5];
> > > +     int ret;
> > > +
> > > +     msg[0] = msgs->addr;
> > > +     msg[1] = msgs->buf[0];
> > > +     msg[2] = 0;
> > > +     msg[3] = 1;
> > > +     msg[4] = msgs->buf[1];
> > > +
> > > +     ret = cdns_mhdp_mailbox_send_recv(&mhdp->base,
> > > +
> > MB_MODULE_ID_HDMI_TX, HDMI_TX_WRITE,
> > > +                                       sizeof(msg), msg,
> > sizeof(reg), reg);
> > > +     if (ret) {
> > > +             dev_err(mhdp->dev, "I2C write failed: %d\n", ret);
> > > +             return ret;
> > > +     }
> > > +
> > > +     if (reg[0] != 0)
> > > +             return -EINVAL;
> > > +
> > > +     return 0;
> > > +}
> > > +
> > > +static int cdns_hdmi_i2c_read(struct cdns_mhdp8501_device *mhdp,
> > > +                           struct i2c_msg *msgs, int num)
> > > +{
> > > +     u8 msg[4], reg[5];
> > > +     u8 addr, offset, *buf, len;
> > > +     int ret, i;
> > > +
> > > +     for (i = 0; i < num; i++) {
> > > +             if (msgs[i].flags & I2C_M_RD) {
> > > +                     addr = msgs[i].addr;
> > > +                     buf = msgs[i].buf;
> > > +                     len = msgs[i].len;
> > > +             } else {
> > > +                     offset = msgs[i].buf[0];
> > > +             }
> > > +     }
> > > +
> > > +     msg[0] = addr;
> > > +     msg[1] = offset;
> > > +     put_unaligned_be16(len, msg + 2);
> > > +
> > > +     ret = cdns_mhdp_mailbox_send_recv_multi(&mhdp->base,
> > > +
> > MB_MODULE_ID_HDMI_TX, HDMI_TX_READ,
> > > +                                             sizeof(msg), msg,
> > > +                                             HDMI_TX_READ,
> > > +                                             sizeof(reg), reg,
> > > +                                             len, buf);
> > > +     if (ret) {
> > > +             dev_err(mhdp->dev, "I2c Read failed: %d\n", ret);
> > > +             return ret;
> > > +     }
> > > +
> > > +     return 0;
> > > +}
> > > +
> > > +#define  SCDC_I2C_SLAVE_ADDRESS      0x54
> > > +static int cdns_hdmi_i2c_xfer(struct i2c_adapter *adap,
> > > +                           struct i2c_msg *msgs, int num)
> > > +{
> > > +     struct cdns_mhdp8501_device *mhdp = i2c_get_adapdata(adap);
> > > +     struct cdns_hdmi_i2c *i2c = mhdp->hdmi.i2c;
> > > +     int i, ret = 0;
> > > +
> > > +     /* Only support SCDC I2C Read/Write */
> > > +     for (i = 0; i < num; i++) {
> > > +             if (msgs[i].addr != SCDC_I2C_SLAVE_ADDRESS) {
> > > +                     dev_err(mhdp->dev, "ADDR=%0x is not
> > supported\n", msgs[i].addr);
> > > +                     return -EINVAL;
> > 
> > Why?
> 
> MHDP FW offers mailbox APIs for SCDC register access but no direct I2C APIs. 
> Individual I2C registers can be read/written using HDMI general register APIs, 
> but block reads (e.g., EDID) are not supported, so it is not a full function I2C.
> i2c_adapter was implemented only for reuse drm_scdc_XXX functions.

Please put this info in the comment. From your 'Only support foo' it's
not obvious if the hw/fw doesn't support it or if you are artifically
limiting it on the driver's side.

> 

-- 
With best wishes
Dmitry



More information about the linux-phy mailing list