| Message ID | 20210927204830.4018624-1-dmitry.baryshkov@linaro.org |
|---|---|
| State | New |
| Headers | show |
| Series | Revert "of: property: fw_devlink: Add support for remote-endpoint" | expand |
On Mon, Sep 27, 2021 at 1:48 PM Dmitry Baryshkov <dmitry.baryshkov@linaro.org> wrote: > > Since the commit f7514a663016 ("of: property: fw_devlink: Add support > for remote-endpoint") Linux kernel started parsing and adding devlinks > for the remote-endpoint properties. However this brings more harm than > good. > > For all the remote-endpoints in the graph two links are created. Thus > each and every remote-endpoint ends up in the cyclic graph (instead of > the original intent of catching a cycle of graph + non-graph link): Yes, I'm well aware of this. I even called this out in the commit text. This creating of cycles and then catching and relaxing it is intentional. https://lore.kernel.org/all/20210330185056.1022008-1-saravanak@google.com/ > > [ 0.381057] OF: remote-endpoint linking /soc@0/geniqup@9c0000/i2c@994000/hdmi-bridge@2b to /soc@0/mdss@ae00000/dsi@ae94000/ports/port@1/endpoint > [ 0.394421] OF: remote-endpoint linking /soc@0/geniqup@9c0000/i2c@994000/hdmi-bridge@2b to /hdmi-out/port/endpoint > [ 0.407007] OF: remote-endpoint linking /soc@0/phy@88e9000 to /soc@0/spmi@c440000/pmic@2/pmic-tcpm/connector/ports/port@0/endpoint@0 > [ 0.419648] OF: remote-endpoint linking /soc@0/usb@a6f8800/usb@a600000 to /soc@0/spmi@c440000/pmic@2/pmic-tcpm/ports/port@2/endpoint@0 > [ 0.432578] OF: remote-endpoint linking /soc@0/cci@ac4f000/i2c-bus@1/cam1@c0 to /soc@0/camss@ac6a000/ports/port@1/endpoint > [ 0.444450] OF: remote-endpoint linking /soc@0/camss@ac6a000 to /soc@0/cci@ac4f000/i2c-bus@1/cam1@c0/port/endpoint > [ 0.455292] OF: remote-endpoint linking /soc@0/mdss@ae00000/mdp@ae01000 to /soc@0/mdss@ae00000/dsi@ae94000/ports/port@0/endpoint > [ 0.467210] OF: remote-endpoint linking /soc@0/mdss@ae00000/mdp@ae01000 to /soc@0/mdss@ae00000/dsi@ae96000/ports/port@0/endpoint > [ 0.479239] OF: remote-endpoint linking /soc@0/mdss@ae00000/dsi@ae94000 to /soc@0/mdss@ae00000/mdp@ae01000/ports/port@0/endpoint > [ 0.491147] OF: remote-endpoint linking /soc@0/mdss@ae00000/dsi@ae94000 to /soc@0/geniqup@9c0000/i2c@994000/hdmi-bridge@2b/ports/port@0/endpoint > [ 0.504979] OF: remote-endpoint linking /soc@0/spmi@c440000/pmic@2/typec@1500 to /soc@0/spmi@c440000/pmic@2/pmic-tcpm/ports/port@0/endpoint > [ 0.517958] OF: remote-endpoint linking /soc@0/spmi@c440000/pmic@2/pdphy@1700 to /soc@0/spmi@c440000/pmic@2/pmic-tcpm/ports/port@1/endpoint > [ 0.565326] OF: remote-endpoint linking /hdmi-out to /soc@0/geniqup@9c0000/i2c@994000/hdmi-bridge@2b/ports/port@2/endpoint > > Under some conditions the device can become it's own supplier, > preventing this device to be probed at all: I'm not sure this analysis is correct -- this shouldn't be happening. If you go to the device link folder and cat "sync_state_only", I expect it to be "1" in this case. Can you confirm that? Which means it won't block probing. Yes, the link itself is useless and it'll get auto deleted once mdss probes and it's easy to not create it in the first place. But this is definitely not your issue. > $ ls -l /sys/bus/platform/devices/ae00000.mdss/ > lrwxrwxrwx 1 root root 0 Aug 4 15:13 consumer:platform:ae00000.mdss -> ../../../virtual/devlink/platform:ae00000.mdss--platform:ae00000.mdss > > I think that until of_link can be tought to handle bi-directional links > on its own, we should not parse remote-endpoint properties. Thus the > aforementioned commit should be reverted. Nak. remote-endpoint parsing is working as intended. I don't think the analysis is correct. Can you please enable the logs in all these functions and attach the log so we can see why it's not probing mdss? device_link_add device_links_check_suppliers func fw_devlink_relax_link fw_devlink_create_devlink -Saravana
[Adding Stephen and linux-arm-msm to the CC list, missed on the patch Cc list] On 28/09/2021 00:58, Saravana Kannan wrote: > On Mon, Sep 27, 2021 at 1:48 PM Dmitry Baryshkov > <dmitry.baryshkov@linaro.org> wrote: >> >> Since the commit f7514a663016 ("of: property: fw_devlink: Add support >> for remote-endpoint") Linux kernel started parsing and adding devlinks >> for the remote-endpoint properties. However this brings more harm than >> good. >> >> For all the remote-endpoints in the graph two links are created. Thus >> each and every remote-endpoint ends up in the cyclic graph (instead of >> the original intent of catching a cycle of graph + non-graph link): > > Yes, I'm well aware of this. I even called this out in the commit > text. This creating of cycles and then catching and relaxing it is > intentional. > https://lore.kernel.org/all/20210330185056.1022008-1-saravanak@google.com/ What would be the reason two always create a cycle which gives no additional information? Maybe I'm just misunderstanding this piece of code. Regarding your commit message. Even if there is a non-remote-endpoint dependency, it will be hidden by the remote-endpoint cycle. And another consequence of remote-endpoint loops. Consider this part part of dmesg. One warning is correct (real cyclic dependency). Others are remote-endpoint spam. Can you spot, which ones? [ 7.032225] platform 1d87000.phy: Fixing up cyclic dependency with 1d84000.ufshc [ 21.760326] platform c440000.spmi:pmic@2:typec@1500: Fixing up cyclic dependency with c440000.spmi:pmic@2:pmic-tcpm [ 21.944849] platform c440000.spmi:pmic@2:pdphy@1700: Fixing up cyclic dependency with c440000.spmi:pmic@2:pmic-tcpm [ 23.541968] platform a600000.usb: Fixing up cyclic dependency with c440000.spmi:pmic@2:pmic-tcpm [ 30.354170] i2c 5-002b: Fixing up cyclic dependency with hdmi-out >> >> [ 0.381057] OF: remote-endpoint linking /soc@0/geniqup@9c0000/i2c@994000/hdmi-bridge@2b to /soc@0/mdss@ae00000/dsi@ae94000/ports/port@1/endpoint >> [ 0.394421] OF: remote-endpoint linking /soc@0/geniqup@9c0000/i2c@994000/hdmi-bridge@2b to /hdmi-out/port/endpoint >> [ 0.407007] OF: remote-endpoint linking /soc@0/phy@88e9000 to /soc@0/spmi@c440000/pmic@2/pmic-tcpm/connector/ports/port@0/endpoint@0 >> [ 0.419648] OF: remote-endpoint linking /soc@0/usb@a6f8800/usb@a600000 to /soc@0/spmi@c440000/pmic@2/pmic-tcpm/ports/port@2/endpoint@0 >> [ 0.432578] OF: remote-endpoint linking /soc@0/cci@ac4f000/i2c-bus@1/cam1@c0 to /soc@0/camss@ac6a000/ports/port@1/endpoint >> [ 0.444450] OF: remote-endpoint linking /soc@0/camss@ac6a000 to /soc@0/cci@ac4f000/i2c-bus@1/cam1@c0/port/endpoint >> [ 0.455292] OF: remote-endpoint linking /soc@0/mdss@ae00000/mdp@ae01000 to /soc@0/mdss@ae00000/dsi@ae94000/ports/port@0/endpoint >> [ 0.467210] OF: remote-endpoint linking /soc@0/mdss@ae00000/mdp@ae01000 to /soc@0/mdss@ae00000/dsi@ae96000/ports/port@0/endpoint >> [ 0.479239] OF: remote-endpoint linking /soc@0/mdss@ae00000/dsi@ae94000 to /soc@0/mdss@ae00000/mdp@ae01000/ports/port@0/endpoint >> [ 0.491147] OF: remote-endpoint linking /soc@0/mdss@ae00000/dsi@ae94000 to /soc@0/geniqup@9c0000/i2c@994000/hdmi-bridge@2b/ports/port@0/endpoint >> [ 0.504979] OF: remote-endpoint linking /soc@0/spmi@c440000/pmic@2/typec@1500 to /soc@0/spmi@c440000/pmic@2/pmic-tcpm/ports/port@0/endpoint >> [ 0.517958] OF: remote-endpoint linking /soc@0/spmi@c440000/pmic@2/pdphy@1700 to /soc@0/spmi@c440000/pmic@2/pmic-tcpm/ports/port@1/endpoint >> [ 0.565326] OF: remote-endpoint linking /hdmi-out to /soc@0/geniqup@9c0000/i2c@994000/hdmi-bridge@2b/ports/port@2/endpoint >> >> Under some conditions the device can become it's own supplier, >> preventing this device to be probed at all: > > I'm not sure this analysis is correct -- this shouldn't be happening. > If you go to the device link folder and cat "sync_state_only", I > expect it to be "1" in this case. Can you confirm that? It is "1". > Which means it won't block probing. Yes, the link itself is useless > and it'll get auto deleted once mdss probes and it's easy to not > create it in the first place. But this is definitely not your issue. > >> $ ls -l /sys/bus/platform/devices/ae00000.mdss/ >> lrwxrwxrwx 1 root root 0 Aug 4 15:13 consumer:platform:ae00000.mdss -> ../../../virtual/devlink/platform:ae00000.mdss--platform:ae00000.mdss >> >> I think that until of_link can be tought to handle bi-directional links >> on its own, we should not parse remote-endpoint properties. Thus the >> aforementioned commit should be reverted. > > Nak. remote-endpoint parsing is working as intended. I don't think the > analysis is correct. > > Can you please enable the logs in all these functions and attach the > log so we can see why it's not probing mdss? > device_link_add > device_links_check_suppliers > func fw_devlink_relax_link > fw_devlink_create_devlink After doing the analysis, I can confirm that I was too quick regarding the mdss links preventing it from being probed. Sorry about that. It all went up to the DP phy having a link with usb-c-connector. I was running the kernel 5.15-rc1, so your tcpm fix is already present. However my colleague has disabled the tcpm device (which I did not notice). So the driver did not call fw_devlink_purge_absent_suppliers(). The devlink still exists: [ 53.426446] platform 88e9000.phy: probe deferral - wait for supplier connector However it is not present in the sysfs: root@qcom-armv8a:~# ls -l /sys/bus/platform/devices/88e9000.phy/ lrwxrwxrwx 1 root root 0 Aug 4 15:13 consumer:platform:a600000.usb -> ../../../virtual/devlink/platform:88e9000.phy--platform:a600000.usb lrwxrwxrwx 1 root root 0 Aug 4 15:13 consumer:platform:af00000.clock-controller -> ../../../virtual/devlink/platform:88e9000.phy--platform:af00000.clock-controller -rw-r--r-- 1 root root 4096 Aug 4 15:13 driver_override -r--r--r-- 1 root root 4096 Aug 4 15:13 modalias lrwxrwxrwx 1 root root 0 Aug 4 15:13 of_node -> ../../../../firmware/devicetree/base/soc@0/phy@88e9000 drwxr-xr-x 2 root root 0 Aug 4 15:13 power lrwxrwxrwx 1 root root 0 Aug 4 15:10 subsystem -> ../../../../bus/platform lrwxrwxrwx 1 root root 0 Aug 4 15:13 supplier:platform:100000.clock-controller -> ../../../virtual/devlink/platform:100000.clock-controller--platform:88e9000.phy lrwxrwxrwx 1 root root 0 Aug 4 15:13 supplier:platform:18200000.rsc:clock-controller -> ../../../virtual/devlink/platform:18200000.rsc:clock-controller--platform:88e9000.phy lrwxrwxrwx 1 root root 0 Aug 4 15:13 supplier:platform:18200000.rsc:pm8150-rpmh-regulators -> ../../../virtual/devlink/platform:18200000.rsc:pm8150-rpmh-regulators--platform:88e9000.phy -rw-r--r-- 1 root root 4096 Aug 4 15:10 uevent -r--r--r-- 1 root root 4096 Aug 4 15:13 waiting_for_supplier Thus it is not possible to spot this device link without CONFIG_DEBUG_DRIVER=y (or any similar debugging technique). If I re-enabled tcpm device or if I reverted remote-endpoint parsing, DP PHY probing would go fine. The DP PHY does not really depend on the connector (or TCPM) being present in the system. The driver will continue working w/o it. However it does not have a change to declare that. Furthermore I went back to the original case that caused you to add remote-endpoint support. The DSI-eDP bridge and eDP panel using the GPIO provided by that bridge. I think the proper fix for the original problem was implemented by the commit bf73537f411b ("drm/bridge: ti-sn65dsi86: Break GPIO and MIPI-to-eDP bridge into sub-drivers"). It split the DSI-eDP bridge driver into functional parts (devices), so that GPIO part and eDP parts are independent, thus breaking this cyclic dependency in a functional way. The remote-endpoint parsing is no longer necessary in this case (Stephen, please correct me if I'm wrong). I still think that remote endpoint parsing does more harm and noise than good and thus should be reverted. -- With best wishes Dmitry
On Tue, Sep 28, 2021 at 5:30 AM Greg Kroah-Hartman <gregkh@linuxfoundation.org> wrote: > > On Mon, Sep 27, 2021 at 06:13:12PM -0700, Saravana Kannan wrote: > > On Mon, Sep 27, 2021 at 5:56 PM Dmitry Baryshkov > > > root@qcom-armv8a:~# ls -l /sys/bus/platform/devices/88e9000.phy/ > > > lrwxrwxrwx 1 root root 0 Aug 4 15:13 > > > consumer:platform:a600000.usb -> > > > ../../../virtual/devlink/platform:88e9000.phy--platform:a600000.usb > > > lrwxrwxrwx 1 root root 0 Aug 4 15:13 > > > consumer:platform:af00000.clock-controller -> > > > ../../../virtual/devlink/platform:88e9000.phy--platform:af00000.clock-controller > > > -rw-r--r-- 1 root root 4096 Aug 4 15:13 driver_override > > > -r--r--r-- 1 root root 4096 Aug 4 15:13 modalias > > > lrwxrwxrwx 1 root root 0 Aug 4 15:13 of_node -> > > > ../../../../firmware/devicetree/base/soc@0/phy@88e9000 > > > drwxr-xr-x 2 root root 0 Aug 4 15:13 power > > > lrwxrwxrwx 1 root root 0 Aug 4 15:10 subsystem -> > > > ../../../../bus/platform > > > lrwxrwxrwx 1 root root 0 Aug 4 15:13 > > > supplier:platform:100000.clock-controller -> > > > ../../../virtual/devlink/platform:100000.clock-controller--platform:88e9000.phy > > > lrwxrwxrwx 1 root root 0 Aug 4 15:13 > > > supplier:platform:18200000.rsc:clock-controller -> > > > ../../../virtual/devlink/platform:18200000.rsc:clock-controller--platform:88e9000.phy > > > lrwxrwxrwx 1 root root 0 Aug 4 15:13 > > > supplier:platform:18200000.rsc:pm8150-rpmh-regulators -> > > > ../../../virtual/devlink/platform:18200000.rsc:pm8150-rpmh-regulators--platform:88e9000.phy > > > -rw-r--r-- 1 root root 4096 Aug 4 15:10 uevent > > > -r--r--r-- 1 root root 4096 Aug 4 15:13 > > > waiting_for_supplier > > > > > > Thus it is not possible to spot this device link without > > > CONFIG_DEBUG_DRIVER=y (or any similar debugging technique). > > > > I sent out some patches to make this easier. But doesn't look like > > it'll land in 5.15. > > https://lore.kernel.org/lkml/20210915172808.620546-1-saravanak@google.com/ > > I have now queued these up to make it into 5.15-final, this thread has > convinced me :) Thanks :) -Saravana
On Fri, Oct 1, 2021 at 6:04 PM Dmitry Baryshkov <dmitry.baryshkov@linaro.org> wrote: > > On 28/09/2021 04:13, Saravana Kannan wrote: > > On Mon, Sep 27, 2021 at 5:56 PM Dmitry Baryshkov <snip> > >> After doing the analysis, I can confirm that I was too quick regarding > >> the mdss links preventing it from being probed. Sorry about that. > >> > >> It all went up to the DP phy having a link with usb-c-connector. I was > >> running the kernel 5.15-rc1, so your tcpm fix is already present. > >> However my colleague has disabled the tcpm device (which I did not > >> notice). So the driver did not call fw_devlink_purge_absent_suppliers(). > >> The devlink still exists: > > > > Let me take a closer look at this before the end of this week. Can you > > point me to the exact DT changes that were made that's causing this > > issue? It should help me debug the issue. I have a guess on what the > > issue might be. > > Here is the kernel source: > https://git.linaro.org/people/bryan.odonoghue/kernel.git/log/?h=5.15-rc1-camss-v2 > > The change that causes PHY driver to silently stop probing, causing an > avalanche of devices not being probed: > > https://git.linaro.org/people/bryan.odonoghue/kernel.git/commit/?h=5.15-rc1-camss-v2&id=d0bf3fc47c132968c302965154eeb5c88007fa73 Sorry, I haven't had a chance to look into this yet, but I still have a strong hunch that this is related to how of_device_is_available() doesn't recurse up till the root to check if a node is disabled (if a parent is disabled, the child should also be considered disabled). And I think patch series should help your case due to a side effect (it wasn't meant as a fix for your issue). Can you give it a shot? https://lore.kernel.org/lkml/20210929000735.585237-1-saravanak@google.com/ -Saravana
diff --git a/drivers/of/property.c b/drivers/of/property.c index a3483484a5a2..6aaaf01ef36e 100644 --- a/drivers/of/property.c +++ b/drivers/of/property.c @@ -1056,25 +1056,6 @@ static bool of_is_ancestor_of(struct device_node *test_ancestor, return false; } -static struct device_node *of_get_compat_node(struct device_node *np) -{ - of_node_get(np); - - while (np) { - if (!of_device_is_available(np)) { - of_node_put(np); - np = NULL; - } - - if (of_find_property(np, "compatible", NULL)) - break; - - np = of_get_next_parent(np); - } - - return np; -} - /** * of_link_to_phandle - Add fwnode link to supplier from supplier phandle * @con_np: consumer device tree node @@ -1098,11 +1079,25 @@ static int of_link_to_phandle(struct device_node *con_np, struct device *sup_dev; struct device_node *tmp_np = sup_np; + of_node_get(sup_np); /* * Find the device node that contains the supplier phandle. It may be * @sup_np or it may be an ancestor of @sup_np. */ - sup_np = of_get_compat_node(sup_np); + while (sup_np) { + + /* Don't allow linking to a disabled supplier */ + if (!of_device_is_available(sup_np)) { + of_node_put(sup_np); + sup_np = NULL; + } + + if (of_find_property(sup_np, "compatible", NULL)) + break; + + sup_np = of_get_next_parent(sup_np); + } + if (!sup_np) { pr_debug("Not linking %pOFP to %pOFP - No device\n", con_np, tmp_np); @@ -1249,7 +1244,6 @@ static struct device_node *parse_##fname(struct device_node *np, \ * @parse_prop.index: For properties holding a list of phandles, this is the * index into the list * @optional: Describes whether a supplier is mandatory or not - * @node_not_dev: The consumer node containing the property is never a device. * * Returns: * parse_prop() return values are @@ -1261,7 +1255,6 @@ struct supplier_bindings { struct device_node *(*parse_prop)(struct device_node *np, const char *prop_name, int index); bool optional; - bool node_not_dev; }; DEFINE_SIMPLE_PROP(clocks, "clocks", "#clock-cells") @@ -1286,7 +1279,6 @@ DEFINE_SIMPLE_PROP(pinctrl5, "pinctrl-5", NULL) DEFINE_SIMPLE_PROP(pinctrl6, "pinctrl-6", NULL) DEFINE_SIMPLE_PROP(pinctrl7, "pinctrl-7", NULL) DEFINE_SIMPLE_PROP(pinctrl8, "pinctrl-8", NULL) -DEFINE_SIMPLE_PROP(remote_endpoint, "remote-endpoint", NULL) DEFINE_SIMPLE_PROP(pwms, "pwms", "#pwm-cells") DEFINE_SIMPLE_PROP(resets, "resets", "#reset-cells") DEFINE_SIMPLE_PROP(leds, "leds", NULL) @@ -1374,7 +1366,6 @@ static const struct supplier_bindings of_supplier_bindings[] = { { .parse_prop = parse_pinctrl6, }, { .parse_prop = parse_pinctrl7, }, { .parse_prop = parse_pinctrl8, }, - { .parse_prop = parse_remote_endpoint, .node_not_dev = true, }, { .parse_prop = parse_pwms, }, { .parse_prop = parse_resets, }, { .parse_prop = parse_leds, }, @@ -1421,16 +1412,10 @@ static int of_link_property(struct device_node *con_np, const char *prop_name) } while ((phandle = s->parse_prop(con_np, prop_name, i))) { - struct device_node *con_dev_np; - - con_dev_np = s->node_not_dev - ? of_get_compat_node(con_np) - : of_node_get(con_np); matched = true; i++; - of_link_to_phandle(con_dev_np, phandle); + of_link_to_phandle(con_np, phandle); of_node_put(phandle); - of_node_put(con_dev_np); } s++; }
Since the commit f7514a663016 ("of: property: fw_devlink: Add support for remote-endpoint") Linux kernel started parsing and adding devlinks for the remote-endpoint properties. However this brings more harm than good. For all the remote-endpoints in the graph two links are created. Thus each and every remote-endpoint ends up in the cyclic graph (instead of the original intent of catching a cycle of graph + non-graph link): [ 0.381057] OF: remote-endpoint linking /soc@0/geniqup@9c0000/i2c@994000/hdmi-bridge@2b to /soc@0/mdss@ae00000/dsi@ae94000/ports/port@1/endpoint [ 0.394421] OF: remote-endpoint linking /soc@0/geniqup@9c0000/i2c@994000/hdmi-bridge@2b to /hdmi-out/port/endpoint [ 0.407007] OF: remote-endpoint linking /soc@0/phy@88e9000 to /soc@0/spmi@c440000/pmic@2/pmic-tcpm/connector/ports/port@0/endpoint@0 [ 0.419648] OF: remote-endpoint linking /soc@0/usb@a6f8800/usb@a600000 to /soc@0/spmi@c440000/pmic@2/pmic-tcpm/ports/port@2/endpoint@0 [ 0.432578] OF: remote-endpoint linking /soc@0/cci@ac4f000/i2c-bus@1/cam1@c0 to /soc@0/camss@ac6a000/ports/port@1/endpoint [ 0.444450] OF: remote-endpoint linking /soc@0/camss@ac6a000 to /soc@0/cci@ac4f000/i2c-bus@1/cam1@c0/port/endpoint [ 0.455292] OF: remote-endpoint linking /soc@0/mdss@ae00000/mdp@ae01000 to /soc@0/mdss@ae00000/dsi@ae94000/ports/port@0/endpoint [ 0.467210] OF: remote-endpoint linking /soc@0/mdss@ae00000/mdp@ae01000 to /soc@0/mdss@ae00000/dsi@ae96000/ports/port@0/endpoint [ 0.479239] OF: remote-endpoint linking /soc@0/mdss@ae00000/dsi@ae94000 to /soc@0/mdss@ae00000/mdp@ae01000/ports/port@0/endpoint [ 0.491147] OF: remote-endpoint linking /soc@0/mdss@ae00000/dsi@ae94000 to /soc@0/geniqup@9c0000/i2c@994000/hdmi-bridge@2b/ports/port@0/endpoint [ 0.504979] OF: remote-endpoint linking /soc@0/spmi@c440000/pmic@2/typec@1500 to /soc@0/spmi@c440000/pmic@2/pmic-tcpm/ports/port@0/endpoint [ 0.517958] OF: remote-endpoint linking /soc@0/spmi@c440000/pmic@2/pdphy@1700 to /soc@0/spmi@c440000/pmic@2/pmic-tcpm/ports/port@1/endpoint [ 0.565326] OF: remote-endpoint linking /hdmi-out to /soc@0/geniqup@9c0000/i2c@994000/hdmi-bridge@2b/ports/port@2/endpoint Under some conditions the device can become it's own supplier, preventing this device to be probed at all: $ ls -l /sys/bus/platform/devices/ae00000.mdss/ lrwxrwxrwx 1 root root 0 Aug 4 15:13 consumer:platform:ae00000.mdss -> ../../../virtual/devlink/platform:ae00000.mdss--platform:ae00000.mdss I think that until of_link can be tought to handle bi-directional links on its own, we should not parse remote-endpoint properties. Thus the aforementioned commit should be reverted. Cc: Saravana Kannan <saravanak@google.com> Fixes: f7514a663016 ("of: property: fw_devlink: Add support for remote-endpoint") Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org> --- drivers/of/property.c | 47 +++++++++++++++---------------------------- 1 file changed, 16 insertions(+), 31 deletions(-) -- 2.33.0