| Message ID | 20200607133832.1730288-1-paul@crapouillou.net (mailing list archive) |
|---|---|
| Headers | show |
| Series | DSI/DBI and TinyDRM driver | expand |
Den 07.06.2020 15.38, skrev Paul Cercueil: > Hi, > > Here's a follow-up on the previous discussion about the current state of > DSI/DBI panel drivers, TinyDRM, and the need of a cleanup. > > This patchset introduces the following: > * It slightly tweaks the MIPI DSI code so that it supports MIPI DBI over > various buses. This patch has been tested with a non-upstream DRM > panel driver for a ILI9331 DBI/8080 panel, written with the DSI > framework (and doesn't include <drm/drm_mipi_dbi.h>), and non-upstream > DSI/DBI host driver for the Ingenic SoCs. > > * A SPI DBI host driver, using the current MIPI DSI framework. It allows > MIPI DSI/DBI drivers to be written with the DSI framework, even if > they are connected over SPI, instead of registering as SPI device > drivers. Since most of these panels can be connected over various > buses, it permits to reuse the same driver independently of the bus > used. > > * A TinyDRM driver for DSI/DBI panels, once again independent of the bus > used; the only dependency (currently) being that the panel must > understand DCS commands. > > * A DRM panel driver to test the stack. This driver controls Ilitek > ILI9341 based DBI panels, like the Adafruit YX240QV29-T 320x240 2.4" > TFT LCD panel. This panel was converted from > drivers/gpu/drm/tiny/ili9341.c. > > I would like to emphasize that while it has been compile-tested, I did > not test it with real hardware since I do not have any DBI panel > connected over SPI. I did runtime-test the code, just without any panel > connected. > > Another thing to note, is that it does not break Device Tree ABI. The > display node stays the same: > > display@0 { > compatible = "adafruit,yx240qv29", "ilitek,ili9341"; > reg = <0>; > spi-max-frequency = <32000000>; > dc-gpios = <&gpio0 9 GPIO_ACTIVE_HIGH>; > reset-gpios = <&gpio0 8 GPIO_ACTIVE_HIGH>; > rotation = <270>; > backlight = <&backlight>; > }; > > The reason it works, is that the "adafruit,yx240qv29" device is probed > on the SPI bus, so it will match with the SPI/DBI host driver. This will > in turn register the very same node with the DSI bus, and the ILI9341 > DRM panel driver will probe. The driver will detect that no controller > is linked to the panel, and eventually register the DBI/DSI TinyDRM > driver. > > I can't stress it enough that this is a RFC, so it still has very rough > edges. > I don't know bridge and dsi drivers so I can't comment on that, but one thing I didn't like is that the DT compatible string has to be added to 2 different modules. As an example, a MI0283QT panel (ILI9341) supports these interface options: 1. SPI Panel setup/control and framebuffer upload over SPI 2. SPI + DPI Panel setup/control over SPI, framebuffer scanout over DPI 3. Parallel bus Panel setup/control and framebuffer upload over parallel bus My suggestion is to have one panel driver module that can support all of these like this: For 1. and 2. a SPI driver is registered and if I understand your example correctly of_graph_get_port_by_id() can be used during probe to distinguish between the 2 options and register a full DRM driver for 1. and add a DRM panel for 2. For 3. a DSI driver is registered (adapted for DBI use like you're suggesting). Note that the interface part of the controller initialization will differ between these, the panel side init will be the same I assume. Noralf.
Hi Noralf, Le dim. 14 juin 2020 à 18:36, Noralf Trønnes <noralf@tronnes.org> a écrit : > > > Den 07.06.2020 15.38, skrev Paul Cercueil: >> Hi, >> >> Here's a follow-up on the previous discussion about the current >> state of >> DSI/DBI panel drivers, TinyDRM, and the need of a cleanup. >> >> This patchset introduces the following: >> * It slightly tweaks the MIPI DSI code so that it supports MIPI DBI >> over >> various buses. This patch has been tested with a non-upstream DRM >> panel driver for a ILI9331 DBI/8080 panel, written with the DSI >> framework (and doesn't include <drm/drm_mipi_dbi.h>), and >> non-upstream >> DSI/DBI host driver for the Ingenic SoCs. >> >> * A SPI DBI host driver, using the current MIPI DSI framework. It >> allows >> MIPI DSI/DBI drivers to be written with the DSI framework, even if >> they are connected over SPI, instead of registering as SPI device >> drivers. Since most of these panels can be connected over various >> buses, it permits to reuse the same driver independently of the >> bus >> used. >> >> * A TinyDRM driver for DSI/DBI panels, once again independent of >> the bus >> used; the only dependency (currently) being that the panel must >> understand DCS commands. >> >> * A DRM panel driver to test the stack. This driver controls Ilitek >> ILI9341 based DBI panels, like the Adafruit YX240QV29-T 320x240 >> 2.4" >> TFT LCD panel. This panel was converted from >> drivers/gpu/drm/tiny/ili9341.c. >> >> I would like to emphasize that while it has been compile-tested, I >> did >> not test it with real hardware since I do not have any DBI panel >> connected over SPI. I did runtime-test the code, just without any >> panel >> connected. >> >> Another thing to note, is that it does not break Device Tree ABI. >> The >> display node stays the same: >> >> display@0 { >> compatible = "adafruit,yx240qv29", "ilitek,ili9341"; >> reg = <0>; >> spi-max-frequency = <32000000>; >> dc-gpios = <&gpio0 9 GPIO_ACTIVE_HIGH>; >> reset-gpios = <&gpio0 8 GPIO_ACTIVE_HIGH>; >> rotation = <270>; >> backlight = <&backlight>; >> }; >> >> The reason it works, is that the "adafruit,yx240qv29" device is >> probed >> on the SPI bus, so it will match with the SPI/DBI host driver. This >> will >> in turn register the very same node with the DSI bus, and the >> ILI9341 >> DRM panel driver will probe. The driver will detect that no >> controller >> is linked to the panel, and eventually register the DBI/DSI TinyDRM >> driver. >> >> I can't stress it enough that this is a RFC, so it still has very >> rough >> edges. >> > > I don't know bridge and dsi drivers so I can't comment on that, but > one > thing I didn't like is that the DT compatible string has to be added > to > 2 different modules. That's a minimal drawback for a perfectly architectured design ;) > As an example, a MI0283QT panel (ILI9341) supports these interface > options: > > 1. SPI > Panel setup/control and framebuffer upload over SPI > > 2. SPI + DPI > Panel setup/control over SPI, framebuffer scanout over DPI > > 3. Parallel bus > Panel setup/control and framebuffer upload over parallel bus > > My suggestion is to have one panel driver module that can support all > of > these like this: > > For 1. and 2. a SPI driver is registered and if I understand your > example correctly of_graph_get_port_by_id() can be used during probe > to > distinguish between the 2 options and register a full DRM driver for > 1. > and add a DRM panel for 2. > > For 3. a DSI driver is registered (adapted for DBI use like you're > suggesting). That means basically having two entry points per DBI driver, one as DSI device and one as SPI device, the latter doing the job of the current DBI-SPI bridge. I think i would be cleaner to just have duplicated strings with the DBI-SPI bridge. Cheers, -Paul > Note that the interface part of the controller initialization will > differ between these, the panel side init will be the same I assume. > Noralf.
Hi all, Allow me to compare this approach with some work Linus W [1] did a while back, which I've just noticed. Pauls' approach: - Perhaps the shortest one possible Porting an existing DSI panel to DBI is 3 lines of code - compat line in the SPI/DSI bridge, a bus_type and mipi_dsi_maybe_register_tiny_driver() call The clever use of the DSI type (equal to zero) means that things will work w/o updating existing dsi hosts and devices in panel/. Yet in the very unlikely case that the panel does not support DSI, we will still allow it. Although thinking about the type I wonder if it can accommodate all use-cases: Since we can have a device (panel) capable of DSI+SPI it makes sense for it to expose the type bitmask, not the host. Although, what if the host itself supports DSI+SPI.? Now we can extrapolate that with a host (say fimd/exynos I think) which supports a SPI panel (s6e63m0) while having of_graph_get_port_by_id(0)? - Strange (ab)use of the DSI bus for DBI (SPI, 6800, 8080 etc) We care about existing users (DT) and it sounds unlikely (based on previous discussion) that DBI + SPI/6800... will make it into DT. So the current approach seems quite reasonable IMHO. Linus' approach: - Clear separation of DSI/SPI Compat strings are still duplicated, although in separate files - Minor code motion and slightly more invasive porting overall Much of the boilerplate can be reduced via helper lib and macros. Even then it's unlikely we'll reach the 3 line delta as with Paul's approach. - Does not handle tiny-dsi (dummy) drm driver It seems doable, with minor changes Personally I'm on the fence and a deciding factor for me is if Paul's approach can handle all the combinations of host/device type support. That said, the input from people likely to do the work would be highly appreciated. Once a decision is made, an illustrative series + todo entry would be great to have. -Emil [1] https://lists.freedesktop.org/archives/dri-devel/2020-May/266079.html
Hi Emil, Le mar. 16 juin 2020 à 18:47, Emil Velikov <emil.l.velikov@gmail.com> a écrit : > Hi all, > > Allow me to compare this approach with some work Linus W [1] did a > while back, which I've just noticed. > > Pauls' approach: > > - Perhaps the shortest one possible > Porting an existing DSI panel to DBI is 3 lines of code - compat line > in the SPI/DSI bridge, a bus_type and > mipi_dsi_maybe_register_tiny_driver() call > The clever use of the DSI type (equal to zero) means that things will > work w/o updating existing dsi hosts and devices in panel/. Yet in the > very unlikely case that the panel does not support DSI, we will still > allow it. Is there such a case? I assumed that all current DSI device and host drivers are for DSI panels. > Although thinking about the type I wonder if it can accommodate all > use-cases: > Since we can have a device (panel) capable of DSI+SPI it makes sense > for it to expose the type bitmask, not the host. Although, what if the > host itself supports DSI+SPI.? Yes, hosts can support more than one type (mine does), so it should expose a bitmask. As for the panel, even though some can do DSI, SPI and I8080 DBI, as far as I know the bus used is always set in hardware (with specific pins set to VCC/GND to select the mode), so this is not something the host can select. Therefore, the panel driver should register the mipi_dsi_device with one particular bus type. Note that the panel driver could very well infer the bus type from the compatible string. If the bus type can be changed at runtime (and there's a need for that), then we would need a bitmask on the panel driver side too, along with a 'set bus' callback, but I'm not sure it will be required. > Now we can extrapolate that with a host (say fimd/exynos I think) > which supports a SPI panel (s6e63m0) while having > of_graph_get_port_by_id(0)? I'm not sure I understand, if there is a port #0 in the panel node, then the tinyDRM driver is not created, and the SPI panel will be registered with the fimd/exynos host driver. So that should already work fine. > - Strange (ab)use of the DSI bus for DBI (SPI, 6800, 8080 etc) > We care about existing users (DT) and it sounds unlikely (based on > previous discussion) that DBI + SPI/6800... will make it into DT. So > the current approach seems quite reasonable IMHO. > > > Linus' approach: > - Clear separation of DSI/SPI > Compat strings are still duplicated, although in separate files > > - Minor code motion and slightly more invasive porting overall > Much of the boilerplate can be reduced via helper lib and macros. Even > then it's unlikely we'll reach the 3 line delta as with Paul's > approach. > > - Does not handle tiny-dsi (dummy) drm driver > It seems doable, with minor changes > > > Personally I'm on the fence and a deciding factor for me is if Paul's > approach can handle all the combinations of host/device type support. > That said, the input from people likely to do the work would be highly > appreciated. > > Once a decision is made, an illustrative series + todo entry would be > great to have. Sure. Thanks for the feedback! Cheers, -Paul > -Emil > > [1] > https://lists.freedesktop.org/archives/dri-devel/2020-May/266079.html
Hi Emil, Le mar. 16 juin 2020 à 18:47, Emil Velikov <emil.l.velikov@gmail.com> a écrit : > Hi all, > > Allow me to compare this approach with some work Linus W [1] did a > while back, which I've just noticed. > > Pauls' approach: > > - Perhaps the shortest one possible > Porting an existing DSI panel to DBI is 3 lines of code - compat line > in the SPI/DSI bridge, a bus_type and > mipi_dsi_maybe_register_tiny_driver() call > The clever use of the DSI type (equal to zero) means that things will > work w/o updating existing dsi hosts and devices in panel/. Yet in the > very unlikely case that the panel does not support DSI, we will still > allow it. Actually the DSI type is not equal to zero, it's BIT(0) so 1. Right now in the patch, I added a WARN in mipi_dsi_attach() that checks that bus_type is non-zero. -Paul > Although thinking about the type I wonder if it can accommodate all > use-cases: > Since we can have a device (panel) capable of DSI+SPI it makes sense > for it to expose the type bitmask, not the host. Although, what if the > host itself supports DSI+SPI.? > Now we can extrapolate that with a host (say fimd/exynos I think) > which supports a SPI panel (s6e63m0) while having > of_graph_get_port_by_id(0)? > > - Strange (ab)use of the DSI bus for DBI (SPI, 6800, 8080 etc) > We care about existing users (DT) and it sounds unlikely (based on > previous discussion) that DBI + SPI/6800... will make it into DT. So > the current approach seems quite reasonable IMHO. > > > Linus' approach: > - Clear separation of DSI/SPI > Compat strings are still duplicated, although in separate files > > - Minor code motion and slightly more invasive porting overall > Much of the boilerplate can be reduced via helper lib and macros. Even > then it's unlikely we'll reach the 3 line delta as with Paul's > approach. > > - Does not handle tiny-dsi (dummy) drm driver > It seems doable, with minor changes > > > Personally I'm on the fence and a deciding factor for me is if Paul's > approach can handle all the combinations of host/device type support. > That said, the input from people likely to do the work would be highly > appreciated. > > Once a decision is made, an illustrative series + todo entry would be > great to have. > -Emil > > [1] > https://lists.freedesktop.org/archives/dri-devel/2020-May/266079.html
Hi Noralf/Paul. Trying to stir up this discussion again. On Sun, Jun 14, 2020 at 06:36:22PM +0200, Noralf Trønnes wrote: > > > Den 07.06.2020 15.38, skrev Paul Cercueil: > > Hi, > > > > Here's a follow-up on the previous discussion about the current state of > > DSI/DBI panel drivers, TinyDRM, and the need of a cleanup. > > > > This patchset introduces the following: > > * It slightly tweaks the MIPI DSI code so that it supports MIPI DBI over > > various buses. This patch has been tested with a non-upstream DRM > > panel driver for a ILI9331 DBI/8080 panel, written with the DSI > > framework (and doesn't include <drm/drm_mipi_dbi.h>), and non-upstream > > DSI/DBI host driver for the Ingenic SoCs. > > > > * A SPI DBI host driver, using the current MIPI DSI framework. It allows > > MIPI DSI/DBI drivers to be written with the DSI framework, even if > > they are connected over SPI, instead of registering as SPI device > > drivers. Since most of these panels can be connected over various > > buses, it permits to reuse the same driver independently of the bus > > used. > > > > * A TinyDRM driver for DSI/DBI panels, once again independent of the bus > > used; the only dependency (currently) being that the panel must > > understand DCS commands. > > > > * A DRM panel driver to test the stack. This driver controls Ilitek > > ILI9341 based DBI panels, like the Adafruit YX240QV29-T 320x240 2.4" > > TFT LCD panel. This panel was converted from > > drivers/gpu/drm/tiny/ili9341.c. > > > > I would like to emphasize that while it has been compile-tested, I did > > not test it with real hardware since I do not have any DBI panel > > connected over SPI. I did runtime-test the code, just without any panel > > connected. > > > > Another thing to note, is that it does not break Device Tree ABI. The > > display node stays the same: > > > > display@0 { > > compatible = "adafruit,yx240qv29", "ilitek,ili9341"; > > reg = <0>; > > spi-max-frequency = <32000000>; > > dc-gpios = <&gpio0 9 GPIO_ACTIVE_HIGH>; > > reset-gpios = <&gpio0 8 GPIO_ACTIVE_HIGH>; > > rotation = <270>; > > backlight = <&backlight>; > > }; > > > > The reason it works, is that the "adafruit,yx240qv29" device is probed > > on the SPI bus, so it will match with the SPI/DBI host driver. This will > > in turn register the very same node with the DSI bus, and the ILI9341 > > DRM panel driver will probe. The driver will detect that no controller > > is linked to the panel, and eventually register the DBI/DSI TinyDRM > > driver. > > > > I can't stress it enough that this is a RFC, so it still has very rough > > edges. > > > > I don't know bridge and dsi drivers so I can't comment on that, but one > thing I didn't like is that the DT compatible string has to be added to > 2 different modules. > > As an example, a MI0283QT panel (ILI9341) supports these interface options: > > 1. SPI > Panel setup/control and framebuffer upload over SPI > > 2. SPI + DPI > Panel setup/control over SPI, framebuffer scanout over DPI > > 3. Parallel bus > Panel setup/control and framebuffer upload over parallel bus To continue the configurations we should support: - Panels where the chip can be configured to SPI, SPI+DPI, Parallel bus (as detailed by Noralf above) - Panels that supports only 6800 or 8080 - connected via GPIO pins or memory mapped (maybe behind some special IP to support this) Command set is often special. We will see a number of chips with many different types of displays. So the drivers should be chip specific with configuration depending on the connected display. What I hope we can find a solution for is a single file/driver that can support all the relevant interface types for a chip. So we would end up with a single file that included the necessary support for ili9341 in all interface configurations with the necessary support for the relevant displays. I do not know how far we are from this as I have not dived into the details of any of the proposals. > > My suggestion is to have one panel driver module that can support all of > these like this: So I think we agree here. > > For 1. and 2. a SPI driver is registered and if I understand your > example correctly of_graph_get_port_by_id() can be used during probe to > distinguish between the 2 options and register a full DRM driver for 1. > and add a DRM panel for 2. > > For 3. a DSI driver is registered (adapted for DBI use like you're > suggesting). > > Note that the interface part of the controller initialization will > differ between these, the panel side init will be the same I assume. Sam
(cc Dillon) Den 03.07.2020 19.26, skrev Sam Ravnborg: > Hi Noralf/Paul. > > Trying to stir up this discussion again. > > On Sun, Jun 14, 2020 at 06:36:22PM +0200, Noralf Trønnes wrote: >> >> >> Den 07.06.2020 15.38, skrev Paul Cercueil: >>> Hi, >>> >>> Here's a follow-up on the previous discussion about the current state of >>> DSI/DBI panel drivers, TinyDRM, and the need of a cleanup. >>> >>> This patchset introduces the following: >>> * It slightly tweaks the MIPI DSI code so that it supports MIPI DBI over >>> various buses. This patch has been tested with a non-upstream DRM >>> panel driver for a ILI9331 DBI/8080 panel, written with the DSI >>> framework (and doesn't include <drm/drm_mipi_dbi.h>), and non-upstream >>> DSI/DBI host driver for the Ingenic SoCs. >>> >>> * A SPI DBI host driver, using the current MIPI DSI framework. It allows >>> MIPI DSI/DBI drivers to be written with the DSI framework, even if >>> they are connected over SPI, instead of registering as SPI device >>> drivers. Since most of these panels can be connected over various >>> buses, it permits to reuse the same driver independently of the bus >>> used. >>> >>> * A TinyDRM driver for DSI/DBI panels, once again independent of the bus >>> used; the only dependency (currently) being that the panel must >>> understand DCS commands. >>> >>> * A DRM panel driver to test the stack. This driver controls Ilitek >>> ILI9341 based DBI panels, like the Adafruit YX240QV29-T 320x240 2.4" >>> TFT LCD panel. This panel was converted from >>> drivers/gpu/drm/tiny/ili9341.c. >>> >>> I would like to emphasize that while it has been compile-tested, I did >>> not test it with real hardware since I do not have any DBI panel >>> connected over SPI. I did runtime-test the code, just without any panel >>> connected. >>> >>> Another thing to note, is that it does not break Device Tree ABI. The >>> display node stays the same: >>> >>> display@0 { >>> compatible = "adafruit,yx240qv29", "ilitek,ili9341"; >>> reg = <0>; >>> spi-max-frequency = <32000000>; >>> dc-gpios = <&gpio0 9 GPIO_ACTIVE_HIGH>; >>> reset-gpios = <&gpio0 8 GPIO_ACTIVE_HIGH>; >>> rotation = <270>; >>> backlight = <&backlight>; >>> }; >>> >>> The reason it works, is that the "adafruit,yx240qv29" device is probed >>> on the SPI bus, so it will match with the SPI/DBI host driver. This will >>> in turn register the very same node with the DSI bus, and the ILI9341 >>> DRM panel driver will probe. The driver will detect that no controller >>> is linked to the panel, and eventually register the DBI/DSI TinyDRM >>> driver. >>> >>> I can't stress it enough that this is a RFC, so it still has very rough >>> edges. >>> >> >> I don't know bridge and dsi drivers so I can't comment on that, but one >> thing I didn't like is that the DT compatible string has to be added to >> 2 different modules. >> >> As an example, a MI0283QT panel (ILI9341) supports these interface options: >> >> 1. SPI >> Panel setup/control and framebuffer upload over SPI >> >> 2. SPI + DPI >> Panel setup/control over SPI, framebuffer scanout over DPI >> >> 3. Parallel bus >> Panel setup/control and framebuffer upload over parallel bus > > To continue the configurations we should support: > - Panels where the chip can be configured to SPI, SPI+DPI, Parallel bus > (as detailed by Noralf above) > - Panels that supports only 6800 or 8080 - connected via GPIO pins or > memory mapped (maybe behind some special IP to support this) > Command set is often special. > > We will see a number of chips with many different types of displays. > So the drivers should be chip specific with configuration depending on > the connected display. > > What I hope we can find a solution for is a single file/driver that can > support all the relevant interface types for a chip. > So we would end up with a single file that included the necessary > support for ili9341 in all interface configurations with the necessary > support for the relevant displays. > > I do not know how far we are from this as I have not dived into the > details of any of the proposals. In an ideal world I would have liked to see the MIPI DBI parallel interface implemented using a new Linux parallel bus type. It could have drivers for gpio bitbanging and mmio in addition to other hw specific drivers. Now we could have a drm_mipi_dbi DRM driver that registers as a SPI client driver and a Parallel bus client driver. Or it can be a component driver for the existing DRM driver on the SoC. I had plans to do this and made a prototype, but dropped it since it would probably require a lot of work getting in a new Linux bus type. However if we're going to treat this parallel bus only as a MIPI DBI display interface but support gpio bitbanging and mmio as well, then we could add DRM drivers for each MIPI DBI bus (that don't have special parallel bus hw): - mipi-dbi-spi - mipi-dbi-gpio - mipi-dbi-mmio These drivers will register as a mipi_dsi_host adapted like Paul suggested. The panel drivers will be mipi_dsi_drivers. Now the panels should work regardless of bus type. They probably need to know about the bus type, at least whether the parallell bus is 8-bit or 16-bit wide. The current MIPI DBI SPI drivers (drm/tiny) will need to be treated specially to keep working with old Device Trees when moved over to drm/panel. Noralf. >> >> My suggestion is to have one panel driver module that can support all of >> these like this: > So I think we agree here. > >> >> For 1. and 2. a SPI driver is registered and if I understand your >> example correctly of_graph_get_port_by_id() can be used during probe to >> distinguish between the 2 options and register a full DRM driver for 1. >> and add a DRM panel for 2. >> >> For 3. a DSI driver is registered (adapted for DBI use like you're >> suggesting). >> >> Note that the interface part of the controller initialization will >> differ between these, the panel side init will be the same I assume. > > Sam >
On Tue, Jul 07, 2020 at 04:32:25PM +0200, Noralf Trønnes wrote: > (cc Dillon) > > Den 03.07.2020 19.26, skrev Sam Ravnborg: > > Hi Noralf/Paul. > > > > Trying to stir up this discussion again. > > > > On Sun, Jun 14, 2020 at 06:36:22PM +0200, Noralf Trønnes wrote: > >> > >> > >> Den 07.06.2020 15.38, skrev Paul Cercueil: > >>> Hi, > >>> > >>> Here's a follow-up on the previous discussion about the current state of > >>> DSI/DBI panel drivers, TinyDRM, and the need of a cleanup. > >>> > >>> This patchset introduces the following: > >>> * It slightly tweaks the MIPI DSI code so that it supports MIPI DBI over > >>> various buses. This patch has been tested with a non-upstream DRM > >>> panel driver for a ILI9331 DBI/8080 panel, written with the DSI > >>> framework (and doesn't include <drm/drm_mipi_dbi.h>), and non-upstream > >>> DSI/DBI host driver for the Ingenic SoCs. > >>> > >>> * A SPI DBI host driver, using the current MIPI DSI framework. It allows > >>> MIPI DSI/DBI drivers to be written with the DSI framework, even if > >>> they are connected over SPI, instead of registering as SPI device > >>> drivers. Since most of these panels can be connected over various > >>> buses, it permits to reuse the same driver independently of the bus > >>> used. > >>> > >>> * A TinyDRM driver for DSI/DBI panels, once again independent of the bus > >>> used; the only dependency (currently) being that the panel must > >>> understand DCS commands. > >>> > >>> * A DRM panel driver to test the stack. This driver controls Ilitek > >>> ILI9341 based DBI panels, like the Adafruit YX240QV29-T 320x240 2.4" > >>> TFT LCD panel. This panel was converted from > >>> drivers/gpu/drm/tiny/ili9341.c. > >>> > >>> I would like to emphasize that while it has been compile-tested, I did > >>> not test it with real hardware since I do not have any DBI panel > >>> connected over SPI. I did runtime-test the code, just without any panel > >>> connected. > >>> > >>> Another thing to note, is that it does not break Device Tree ABI. The > >>> display node stays the same: > >>> > >>> display@0 { > >>> compatible = "adafruit,yx240qv29", "ilitek,ili9341"; > >>> reg = <0>; > >>> spi-max-frequency = <32000000>; > >>> dc-gpios = <&gpio0 9 GPIO_ACTIVE_HIGH>; > >>> reset-gpios = <&gpio0 8 GPIO_ACTIVE_HIGH>; > >>> rotation = <270>; > >>> backlight = <&backlight>; > >>> }; > >>> > >>> The reason it works, is that the "adafruit,yx240qv29" device is probed > >>> on the SPI bus, so it will match with the SPI/DBI host driver. This will > >>> in turn register the very same node with the DSI bus, and the ILI9341 > >>> DRM panel driver will probe. The driver will detect that no controller > >>> is linked to the panel, and eventually register the DBI/DSI TinyDRM > >>> driver. > >>> > >>> I can't stress it enough that this is a RFC, so it still has very rough > >>> edges. > >>> > >> > >> I don't know bridge and dsi drivers so I can't comment on that, but one > >> thing I didn't like is that the DT compatible string has to be added to > >> 2 different modules. > >> > >> As an example, a MI0283QT panel (ILI9341) supports these interface options: > >> > >> 1. SPI > >> Panel setup/control and framebuffer upload over SPI > >> > >> 2. SPI + DPI > >> Panel setup/control over SPI, framebuffer scanout over DPI > >> > >> 3. Parallel bus > >> Panel setup/control and framebuffer upload over parallel bus > > > > To continue the configurations we should support: > > - Panels where the chip can be configured to SPI, SPI+DPI, Parallel bus > > (as detailed by Noralf above) > > - Panels that supports only 6800 or 8080 - connected via GPIO pins or > > memory mapped (maybe behind some special IP to support this) > > Command set is often special. > > > > We will see a number of chips with many different types of displays. > > So the drivers should be chip specific with configuration depending on > > the connected display. > > > > What I hope we can find a solution for is a single file/driver that can > > support all the relevant interface types for a chip. > > So we would end up with a single file that included the necessary > > support for ili9341 in all interface configurations with the necessary > > support for the relevant displays. > > > > I do not know how far we are from this as I have not dived into the > > details of any of the proposals. > > In an ideal world I would have liked to see the MIPI DBI parallel > interface implemented using a new Linux parallel bus type. It could have > drivers for gpio bitbanging and mmio in addition to other hw specific > drivers. Now we could have a drm_mipi_dbi DRM driver that registers as a > SPI client driver and a Parallel bus client driver. Or it can be a > component driver for the existing DRM driver on the SoC. > > I had plans to do this and made a prototype, but dropped it since it > would probably require a lot of work getting in a new Linux bus type. Channelling my inner Greg KH: Please just create a new bus, it should be quite easy and boilerplate is manageable. Greg, did I get this right? Maybe any recommendations for a simple parallel bus with perhaps different register access paths depending upon how it's all wired up exactly? -Daniel > However if we're going to treat this parallel bus only as a MIPI DBI > display interface but support gpio bitbanging and mmio as well, then we > could add DRM drivers for each MIPI DBI bus (that don't have special > parallel bus hw): > - mipi-dbi-spi > - mipi-dbi-gpio > - mipi-dbi-mmio > > These drivers will register as a mipi_dsi_host adapted like Paul suggested. > > The panel drivers will be mipi_dsi_drivers. Now the panels should work > regardless of bus type. They probably need to know about the bus type, > at least whether the parallell bus is 8-bit or 16-bit wide. > > The current MIPI DBI SPI drivers (drm/tiny) will need to be treated > specially to keep working with old Device Trees when moved over to > drm/panel. > > Noralf. > > > >> > >> My suggestion is to have one panel driver module that can support all of > >> these like this: > > So I think we agree here. > > > >> > >> For 1. and 2. a SPI driver is registered and if I understand your > >> example correctly of_graph_get_port_by_id() can be used during probe to > >> distinguish between the 2 options and register a full DRM driver for 1. > >> and add a DRM panel for 2. > >> > >> For 3. a DSI driver is registered (adapted for DBI use like you're > >> suggesting). > >> > >> Note that the interface part of the controller initialization will > >> differ between these, the panel side init will be the same I assume. > > > > Sam > >
Hi Daniel, Le mer. 8 juil. 2020 à 9:23, Daniel Vetter <daniel@ffwll.ch> a écrit : > On Tue, Jul 07, 2020 at 04:32:25PM +0200, Noralf Trønnes wrote: >> (cc Dillon) >> >> Den 03.07.2020 19.26, skrev Sam Ravnborg: >> > Hi Noralf/Paul. >> > >> > Trying to stir up this discussion again. >> > >> > On Sun, Jun 14, 2020 at 06:36:22PM +0200, Noralf Trønnes wrote: >> >> >> >> >> >> Den 07.06.2020 15.38, skrev Paul Cercueil: >> >>> Hi, >> >>> >> >>> Here's a follow-up on the previous discussion about the current >> state of >> >>> DSI/DBI panel drivers, TinyDRM, and the need of a cleanup. >> >>> >> >>> This patchset introduces the following: >> >>> * It slightly tweaks the MIPI DSI code so that it supports MIPI >> DBI over >> >>> various buses. This patch has been tested with a non-upstream >> DRM >> >>> panel driver for a ILI9331 DBI/8080 panel, written with the >> DSI >> >>> framework (and doesn't include <drm/drm_mipi_dbi.h>), and >> non-upstream >> >>> DSI/DBI host driver for the Ingenic SoCs. >> >>> >> >>> * A SPI DBI host driver, using the current MIPI DSI framework. >> It allows >> >>> MIPI DSI/DBI drivers to be written with the DSI framework, >> even if >> >>> they are connected over SPI, instead of registering as SPI >> device >> >>> drivers. Since most of these panels can be connected over >> various >> >>> buses, it permits to reuse the same driver independently of >> the bus >> >>> used. >> >>> >> >>> * A TinyDRM driver for DSI/DBI panels, once again independent >> of the bus >> >>> used; the only dependency (currently) being that the panel >> must >> >>> understand DCS commands. >> >>> >> >>> * A DRM panel driver to test the stack. This driver controls >> Ilitek >> >>> ILI9341 based DBI panels, like the Adafruit YX240QV29-T >> 320x240 2.4" >> >>> TFT LCD panel. This panel was converted from >> >>> drivers/gpu/drm/tiny/ili9341.c. >> >>> >> >>> I would like to emphasize that while it has been >> compile-tested, I did >> >>> not test it with real hardware since I do not have any DBI panel >> >>> connected over SPI. I did runtime-test the code, just without >> any panel >> >>> connected. >> >>> >> >>> Another thing to note, is that it does not break Device Tree >> ABI. The >> >>> display node stays the same: >> >>> >> >>> display@0 { >> >>> compatible = "adafruit,yx240qv29", "ilitek,ili9341"; >> >>> reg = <0>; >> >>> spi-max-frequency = <32000000>; >> >>> dc-gpios = <&gpio0 9 GPIO_ACTIVE_HIGH>; >> >>> reset-gpios = <&gpio0 8 GPIO_ACTIVE_HIGH>; >> >>> rotation = <270>; >> >>> backlight = <&backlight>; >> >>> }; >> >>> >> >>> The reason it works, is that the "adafruit,yx240qv29" device is >> probed >> >>> on the SPI bus, so it will match with the SPI/DBI host driver. >> This will >> >>> in turn register the very same node with the DSI bus, and the >> ILI9341 >> >>> DRM panel driver will probe. The driver will detect that no >> controller >> >>> is linked to the panel, and eventually register the DBI/DSI >> TinyDRM >> >>> driver. >> >>> >> >>> I can't stress it enough that this is a RFC, so it still has >> very rough >> >>> edges. >> >>> >> >> >> >> I don't know bridge and dsi drivers so I can't comment on that, >> but one >> >> thing I didn't like is that the DT compatible string has to be >> added to >> >> 2 different modules. >> >> >> >> As an example, a MI0283QT panel (ILI9341) supports these >> interface options: >> >> >> >> 1. SPI >> >> Panel setup/control and framebuffer upload over SPI >> >> >> >> 2. SPI + DPI >> >> Panel setup/control over SPI, framebuffer scanout over DPI >> >> >> >> 3. Parallel bus >> >> Panel setup/control and framebuffer upload over parallel bus >> > >> > To continue the configurations we should support: >> > - Panels where the chip can be configured to SPI, SPI+DPI, >> Parallel bus >> > (as detailed by Noralf above) >> > - Panels that supports only 6800 or 8080 - connected via GPIO >> pins or >> > memory mapped (maybe behind some special IP to support this) >> > Command set is often special. >> > >> > We will see a number of chips with many different types of >> displays. >> > So the drivers should be chip specific with configuration >> depending on >> > the connected display. >> > >> > What I hope we can find a solution for is a single file/driver >> that can >> > support all the relevant interface types for a chip. >> > So we would end up with a single file that included the necessary >> > support for ili9341 in all interface configurations with the >> necessary >> > support for the relevant displays. >> > >> > I do not know how far we are from this as I have not dived into >> the >> > details of any of the proposals. >> >> In an ideal world I would have liked to see the MIPI DBI parallel >> interface implemented using a new Linux parallel bus type. It could >> have >> drivers for gpio bitbanging and mmio in addition to other hw >> specific >> drivers. Now we could have a drm_mipi_dbi DRM driver that registers >> as a >> SPI client driver and a Parallel bus client driver. Or it can be a >> component driver for the existing DRM driver on the SoC. >> >> I had plans to do this and made a prototype, but dropped it since it >> would probably require a lot of work getting in a new Linux bus >> type. > > Channelling my inner Greg KH: > > Please just create a new bus, it should be quite easy and boilerplate > is > manageable. The bus is already here, it's "mipi-dsi". DBI and DSI are basically the same thing, just that one is parallel and the other is serial. -Paul > Greg, did I get this right? Maybe any recommendations for a simple > parallel bus with perhaps different register access paths depending > upon > how it's all wired up exactly? > -Daniel > >> However if we're going to treat this parallel bus only as a MIPI DBI >> display interface but support gpio bitbanging and mmio as well, >> then we >> could add DRM drivers for each MIPI DBI bus (that don't have special >> parallel bus hw): >> - mipi-dbi-spi >> - mipi-dbi-gpio >> - mipi-dbi-mmio >> >> These drivers will register as a mipi_dsi_host adapted like Paul >> suggested. >> >> The panel drivers will be mipi_dsi_drivers. Now the panels should >> work >> regardless of bus type. They probably need to know about the bus >> type, >> at least whether the parallell bus is 8-bit or 16-bit wide. >> >> The current MIPI DBI SPI drivers (drm/tiny) will need to be treated >> specially to keep working with old Device Trees when moved over to >> drm/panel. >> >> Noralf. >> >> >> >> >> >> My suggestion is to have one panel driver module that can >> support all of >> >> these like this: >> > So I think we agree here. >> > >> >> >> >> For 1. and 2. a SPI driver is registered and if I understand your >> >> example correctly of_graph_get_port_by_id() can be used during >> probe to >> >> distinguish between the 2 options and register a full DRM driver >> for 1. >> >> and add a DRM panel for 2. >> >> >> >> For 3. a DSI driver is registered (adapted for DBI use like >> you're >> >> suggesting). >> >> >> >> Note that the interface part of the controller initialization >> will >> >> differ between these, the panel side init will be the same I >> assume. >> > >> > Sam >> > > > -- > Daniel Vetter > Software Engineer, Intel Corporation > http://blog.ffwll.ch
Den 08.07.2020 14.49, skrev Paul Cercueil: > Hi Daniel, > > Le mer. 8 juil. 2020 à 9:23, Daniel Vetter <daniel@ffwll.ch> a écrit : >> On Tue, Jul 07, 2020 at 04:32:25PM +0200, Noralf Trønnes wrote: >>> (cc Dillon) >>> >>> Den 03.07.2020 19.26, skrev Sam Ravnborg: >>> > Hi Noralf/Paul. >>> > >>> > Trying to stir up this discussion again. >>> > >>> > On Sun, Jun 14, 2020 at 06:36:22PM +0200, Noralf Trønnes wrote: >>> >> >>> >> >>> >> Den 07.06.2020 15.38, skrev Paul Cercueil: >>> >>> Hi, >>> >>> >>> >>> Here's a follow-up on the previous discussion about the current >>> state of >>> >>> DSI/DBI panel drivers, TinyDRM, and the need of a cleanup. >>> >>> >>> >>> This patchset introduces the following: >>> >>> * It slightly tweaks the MIPI DSI code so that it supports MIPI >>> DBI over >>> >>> various buses. This patch has been tested with a non-upstream DRM >>> >>> panel driver for a ILI9331 DBI/8080 panel, written with the DSI >>> >>> framework (and doesn't include <drm/drm_mipi_dbi.h>), and >>> non-upstream >>> >>> DSI/DBI host driver for the Ingenic SoCs. >>> >>> >>> >>> * A SPI DBI host driver, using the current MIPI DSI framework. >>> It allows >>> >>> MIPI DSI/DBI drivers to be written with the DSI framework, >>> even if >>> >>> they are connected over SPI, instead of registering as SPI device >>> >>> drivers. Since most of these panels can be connected over various >>> >>> buses, it permits to reuse the same driver independently of >>> the bus >>> >>> used. >>> >>> >>> >>> * A TinyDRM driver for DSI/DBI panels, once again independent of >>> the bus >>> >>> used; the only dependency (currently) being that the panel must >>> >>> understand DCS commands. >>> >>> >>> >>> * A DRM panel driver to test the stack. This driver controls Ilitek >>> >>> ILI9341 based DBI panels, like the Adafruit YX240QV29-T >>> 320x240 2.4" >>> >>> TFT LCD panel. This panel was converted from >>> >>> drivers/gpu/drm/tiny/ili9341.c. >>> >>> >>> >>> I would like to emphasize that while it has been compile-tested, >>> I did >>> >>> not test it with real hardware since I do not have any DBI panel >>> >>> connected over SPI. I did runtime-test the code, just without >>> any panel >>> >>> connected. >>> >>> >>> >>> Another thing to note, is that it does not break Device Tree >>> ABI. The >>> >>> display node stays the same: >>> >>> >>> >>> display@0 { >>> >>> compatible = "adafruit,yx240qv29", "ilitek,ili9341"; >>> >>> reg = <0>; >>> >>> spi-max-frequency = <32000000>; >>> >>> dc-gpios = <&gpio0 9 GPIO_ACTIVE_HIGH>; >>> >>> reset-gpios = <&gpio0 8 GPIO_ACTIVE_HIGH>; >>> >>> rotation = <270>; >>> >>> backlight = <&backlight>; >>> >>> }; >>> >>> >>> >>> The reason it works, is that the "adafruit,yx240qv29" device is >>> probed >>> >>> on the SPI bus, so it will match with the SPI/DBI host driver. >>> This will >>> >>> in turn register the very same node with the DSI bus, and the >>> ILI9341 >>> >>> DRM panel driver will probe. The driver will detect that no >>> controller >>> >>> is linked to the panel, and eventually register the DBI/DSI TinyDRM >>> >>> driver. >>> >>> >>> >>> I can't stress it enough that this is a RFC, so it still has >>> very rough >>> >>> edges. >>> >>> >>> >> >>> >> I don't know bridge and dsi drivers so I can't comment on that, >>> but one >>> >> thing I didn't like is that the DT compatible string has to be >>> added to >>> >> 2 different modules. >>> >> >>> >> As an example, a MI0283QT panel (ILI9341) supports these >>> interface options: >>> >> >>> >> 1. SPI >>> >> Panel setup/control and framebuffer upload over SPI >>> >> >>> >> 2. SPI + DPI >>> >> Panel setup/control over SPI, framebuffer scanout over DPI >>> >> >>> >> 3. Parallel bus >>> >> Panel setup/control and framebuffer upload over parallel bus >>> > >>> > To continue the configurations we should support: >>> > - Panels where the chip can be configured to SPI, SPI+DPI, >>> Parallel bus >>> > (as detailed by Noralf above) >>> > - Panels that supports only 6800 or 8080 - connected via GPIO pins or >>> > memory mapped (maybe behind some special IP to support this) >>> > Command set is often special. >>> > >>> > We will see a number of chips with many different types of displays. >>> > So the drivers should be chip specific with configuration >>> depending on >>> > the connected display. >>> > >>> > What I hope we can find a solution for is a single file/driver >>> that can >>> > support all the relevant interface types for a chip. >>> > So we would end up with a single file that included the necessary >>> > support for ili9341 in all interface configurations with the >>> necessary >>> > support for the relevant displays. >>> > >>> > I do not know how far we are from this as I have not dived into the >>> > details of any of the proposals. >>> >>> In an ideal world I would have liked to see the MIPI DBI parallel >>> interface implemented using a new Linux parallel bus type. It could >>> have >>> drivers for gpio bitbanging and mmio in addition to other hw specific >>> drivers. Now we could have a drm_mipi_dbi DRM driver that registers >>> as a >>> SPI client driver and a Parallel bus client driver. Or it can be a >>> component driver for the existing DRM driver on the SoC. I have realised that there is a problem here. A drm_mipi_drm driver would require a device and this device would be virtual since the actual device is the panel itself. IIRC we can't have virtual devices in the Device Tree. So this won't work I think. The same problem applies to my mipi-dbi-spi case below, it is also a virtual device. >>> >>> I had plans to do this and made a prototype, but dropped it since it >>> would probably require a lot of work getting in a new Linux bus type. >> >> Channelling my inner Greg KH: >> >> Please just create a new bus, it should be quite easy and boilerplate is >> manageable. I'm not sure I agree with that. Wouldn't this be something akin to SPI or I2C? There's quite some code in those subsystems. > > The bus is already here, it's "mipi-dsi". DBI and DSI are basically the > same thing, just that one is parallel and the other is serial. Can the parallel interface on your hw be used outside the display domain as say connected to an ADC? If not then a new Linux bus type won't help in your case. My concern is tying a parallel interface to the DRM subsystem if it can be used as a generic interface. The Raspberry Pi has a SMI (Secondary Memory Interface) parallel interface that can be used for DBI and I believe the Beaglebone Black can have a hw bus through the realtime unit. But I have no experience with either of these. Noralf. > > -Paul > >> Greg, did I get this right? Maybe any recommendations for a simple >> parallel bus with perhaps different register access paths depending upon >> how it's all wired up exactly? >> -Daniel >> >>> However if we're going to treat this parallel bus only as a MIPI DBI >>> display interface but support gpio bitbanging and mmio as well, then we >>> could add DRM drivers for each MIPI DBI bus (that don't have special >>> parallel bus hw): >>> - mipi-dbi-spi >>> - mipi-dbi-gpio >>> - mipi-dbi-mmio >>> >>> These drivers will register as a mipi_dsi_host adapted like Paul >>> suggested. >>> >>> The panel drivers will be mipi_dsi_drivers. Now the panels should work >>> regardless of bus type. They probably need to know about the bus type, >>> at least whether the parallell bus is 8-bit or 16-bit wide. >>> >>> The current MIPI DBI SPI drivers (drm/tiny) will need to be treated >>> specially to keep working with old Device Trees when moved over to >>> drm/panel. >>> >>> Noralf. >>> >>> >>> >> >>> >> My suggestion is to have one panel driver module that can support >>> all of >>> >> these like this: >>> > So I think we agree here. >>> > >>> >> >>> >> For 1. and 2. a SPI driver is registered and if I understand your >>> >> example correctly of_graph_get_port_by_id() can be used during >>> probe to >>> >> distinguish between the 2 options and register a full DRM driver >>> for 1. >>> >> and add a DRM panel for 2. >>> >> >>> >> For 3. a DSI driver is registered (adapted for DBI use like you're >>> >> suggesting). >>> >> >>> >> Note that the interface part of the controller initialization will >>> >> differ between these, the panel side init will be the same I assume. >>> > >>> > Sam >>> > >> >> -- >> Daniel Vetter >> Software Engineer, Intel Corporation >> http://blog.ffwll.ch > > >
Hi, Here's a follow-up on the previous discussion about the current state of DSI/DBI panel drivers, TinyDRM, and the need of a cleanup. This patchset introduces the following: * It slightly tweaks the MIPI DSI code so that it supports MIPI DBI over various buses. This patch has been tested with a non-upstream DRM panel driver for a ILI9331 DBI/8080 panel, written with the DSI framework (and doesn't include <drm/drm_mipi_dbi.h>), and non-upstream DSI/DBI host driver for the Ingenic SoCs. * A SPI DBI host driver, using the current MIPI DSI framework. It allows MIPI DSI/DBI drivers to be written with the DSI framework, even if they are connected over SPI, instead of registering as SPI device drivers. Since most of these panels can be connected over various buses, it permits to reuse the same driver independently of the bus used. * A TinyDRM driver for DSI/DBI panels, once again independent of the bus used; the only dependency (currently) being that the panel must understand DCS commands. * A DRM panel driver to test the stack. This driver controls Ilitek ILI9341 based DBI panels, like the Adafruit YX240QV29-T 320x240 2.4" TFT LCD panel. This panel was converted from drivers/gpu/drm/tiny/ili9341.c. I would like to emphasize that while it has been compile-tested, I did not test it with real hardware since I do not have any DBI panel connected over SPI. I did runtime-test the code, just without any panel connected. Another thing to note, is that it does not break Device Tree ABI. The display node stays the same: display@0 { compatible = "adafruit,yx240qv29", "ilitek,ili9341"; reg = <0>; spi-max-frequency = <32000000>; dc-gpios = <&gpio0 9 GPIO_ACTIVE_HIGH>; reset-gpios = <&gpio0 8 GPIO_ACTIVE_HIGH>; rotation = <270>; backlight = <&backlight>; }; The reason it works, is that the "adafruit,yx240qv29" device is probed on the SPI bus, so it will match with the SPI/DBI host driver. This will in turn register the very same node with the DSI bus, and the ILI9341 DRM panel driver will probe. The driver will detect that no controller is linked to the panel, and eventually register the DBI/DSI TinyDRM driver. I can't stress it enough that this is a RFC, so it still has very rough edges. Cheers, -Paul Paul Cercueil (4): gpu/drm: dsi: Let host and device specify supported bus gpu/drm: Add SPI DBI host driver gpu/drm: Add TinyDRM for DSI/DBI panels gpu/drm: Add Ilitek ILI9341 DBI panel driver drivers/gpu/drm/bridge/Kconfig | 8 + drivers/gpu/drm/bridge/Makefile | 1 + drivers/gpu/drm/bridge/dbi-spi.c | 159 +++++++++ drivers/gpu/drm/drm_mipi_dsi.c | 9 + drivers/gpu/drm/panel/Kconfig | 9 + drivers/gpu/drm/panel/Makefile | 1 + drivers/gpu/drm/panel/panel-ilitek-ili9341.c | 347 +++++++++++++++++++ drivers/gpu/drm/tiny/Kconfig | 8 + drivers/gpu/drm/tiny/Makefile | 1 + drivers/gpu/drm/tiny/tiny-dsi.c | 262 ++++++++++++++ include/drm/drm_mipi_dsi.h | 31 ++ 11 files changed, 836 insertions(+) create mode 100644 drivers/gpu/drm/bridge/dbi-spi.c create mode 100644 drivers/gpu/drm/panel/panel-ilitek-ili9341.c create mode 100644 drivers/gpu/drm/tiny/tiny-dsi.c