Thread: [Lcd-linux] lcd over gpio
Brought to you by:
mjona
From: Claudio <cla...@gm...> - 2011-01-14 12:58:07
|
Hi all, I find this pretty good driver and I believe that a valuable feature that should be added is a gpio extension. On threat http://sourceforge.net/mailarchive/forum.php?thread_name=729F0B8B-978C-478B-A85F-B6AFE1F5208A%40cs.tu-berlin.de&forum_name=lcd-linux-users a arch specific improvement is proposed but using a gpio [1] implementation can be portable between different system. There is some intention to add it? 1. http://lxr.linux.no/#linux+v2.6.37/Documentation/gpio.txt |
From: Mattia Jona-L. <mat...@gm...> - 2011-01-18 08:01:59
|
Hi, the implementation of GPIO lines would certainly be a valuable feature to add to lcd-linux. Presently there is no plan to go in that direction but it can be put in the TODO list. Another feature would be to add an input line by using the input pins of the parallel port but in my opinion the most urgent feature is to move from a parallel port connection to something more modern and universal like USB. Unfortunately for this you need to connect the display to a controlling board. There is a wide choice of these boards but there is no standard on their implementation so it is hard to tell what can be supported or implemented. Best Mattia On Fri, Jan 14, 2011 at 1:57 PM, Claudio <cla...@gm...> wrote: > Hi all, > I find this pretty good driver and I believe that a valuable feature > that should be added is a gpio extension. > > On threat > http://sourceforge.net/mailarchive/forum.php?thread_name=729F0B8B-978C-478B-A85F-B6AFE1F5208A%40cs.tu-berlin.de&forum_name=lcd-linux-users > > a arch specific improvement is proposed but using a gpio [1] > implementation can be portable between different system. > > There is some intention to add it? > > > 1. http://lxr.linux.no/#linux+v2.6.37/Documentation/gpio.txt > > ------------------------------------------------------------------------------ > Protect Your Site and Customers from Malware Attacks > Learn about various malware tactics and how to avoid them. Understand > malware threats, the impact they can have on your business, and how you > can protect your company and customers by using code signing. > http://p.sf.net/sfu/oracle-sfdevnl > _______________________________________________ > Lcd-linux-users mailing list > Lcd...@li... > https://lists.sourceforge.net/lists/listinfo/lcd-linux-users > |
From: Claudio <cla...@gm...> - 2011-01-18 12:12:46
|
Dear mattia, let me introduce a different scenario. 2011/1/18 Mattia Jona-Lasinio <mat...@gm...>: > Hi, > > the implementation of GPIO lines would certainly be a valuable feature > to add to lcd-linux. Presently there is no plan to go in that > direction but it can be put in the TODO list. Another feature would be > to add an input line by using the input pins of the parallel port but > in my opinion the most urgent feature is to move from a parallel port > connection to something more modern and universal like USB. The USB device are really different each other and support for its could be difficult and inefficient, beside on embedded devices generally you want to use the USB port for something more useful. The gpio implementaion instread allow you to support a huge range of embedded devices like router, develop board as well some routerboard or industial board; moreover you can use a USB to GPIO adapter (like ftdi) and then connect any kind of lcd also on you modern laptop/pc (that typically don't have parport or gpio). The parport implementation is hacky and should be mainained just for history but it is de facto obsolete(how many device you can found with parport today in your home?). In this scenario the GPIO support is priority and such version of lcd-linux is, in imho, also more acceptable by mainline kernel. Best regards |
From: Mattia Jona-L. <mat...@gm...> - 2011-01-18 13:48:27
|
On Tue, Jan 18, 2011 at 1:12 PM, Claudio <cla...@gm...> wrote: > Dear mattia, > let me introduce a different scenario. > > The USB device are really different each other and support for its > could be difficult and inefficient, beside on embedded devices > generally you want to use the USB port for something more useful. Hmmmm, quite a lot of people might disagree with you as USB is now considered THE universal and preferred way to connect external devices to computers. Anyway, I see your point. > The gpio implementaion instread allow you to support a huge range of > embedded devices like router, develop board as well some routerboard > or industial board; moreover you can use a USB to GPIO adapter (like > ftdi) and then connect any kind of lcd also on you modern laptop/pc > (that typically don't have parport or gpio). I agree and this should in principle be very easy. If the GPIO implementation in the kernel is at a usable level then I'm more than happy to try to use it. In any case it will provide a sort of standard way to access the display. If this is the case then it is very easy as the only two functions in the HD44780 driver that need a rewrite are the two lowlevel read and write functions. If I write a patch, will you be able to test it and report on it? > The parport implementation is hacky and should be mainained just for > history but it is de facto obsolete(how many device you can found with > parport today in your home?). I agree and that's why I wanted to move to something newer. LCD-Linux was intended at the beginning to provide a cheap display output to old and obsolete hardware. These days this goal does not make any sense as an embedded device can run as a router much more efficiently than an old computer. > In this scenario the GPIO support is priority and such version of > lcd-linux is, in imho, also more acceptable by mainline kernel. LCD-Linux was already proposed for inclusion in the kernel mainline but some issues prevented this. The main reason being that LCD-Linux duplicates in some sense the console emulation and kernel developers don't like duplicate code (why reinventing the wheel?). To make a long story short LCD-Linux wasn't rejected but it was asked to merge some parts of it with the existing kernel code, which is not always easy. There was some traffic about this on the main linux kernel maling list in july 2010. Best Mattia > > Best regards > |
From: Claudio <cla...@gm...> - 2011-01-18 19:13:21
|
2011/1/18 Mattia Jona-Lasinio <mat...@gm...>: > On Tue, Jan 18, 2011 at 1:12 PM, Claudio <cla...@gm...> wrote: >> Dear mattia, >> let me introduce a different scenario. >> >> The USB device are really different each other and support for its >> could be difficult and inefficient, beside on embedded devices >> generally you want to use the USB port for something more useful. > > Hmmmm, quite a lot of people might disagree with you as USB is now > considered THE universal and preferred way to connect external devices > to computers. Anyway, I see your point. > >> The gpio implementaion instread allow you to support a huge range of >> embedded devices like router, develop board as well some routerboard >> or industial board; moreover you can use a USB to GPIO adapter (like >> ftdi) and then connect any kind of lcd also on you modern laptop/pc >> (that typically don't have parport or gpio). > > I agree and this should in principle be very easy. If the GPIO > implementation in the kernel is at a usable level then I'm more than > happy to try to use it. In any case it will provide a sort of standard I can tell you that a lot of interface are yet build on top of gpio layer for example i2c or spi bus, as well as led, buttons (for input) i2c http://lxr.linux.no/#linux+v2.6.37/drivers/i2c/busses/i2c-gpio.c spi http://lxr.linux.no/#linux+v2.6.37/drivers/spi/spi_gpio.c buttons http://lxr.linux.no/#linux+v2.6.37/include/linux/gpio_keys.h#L17 > way to access the display. If this is the case then it is very easy as > the only two functions in the HD44780 driver that need a rewrite are > the two lowlevel read and write functions. If I write a patch, will > you be able to test it and report on it? > Yes of course! |
From: Claudio <cla...@gm...> - 2011-03-16 17:03:17
|
Hello Mattia, did you have some news on this topic? I took a couple of hitachi compatible LCD and a linux gpio enabled board. 2011/1/18 Mattia Jona-Lasinio <mat...@gm...>: > On Tue, Jan 18, 2011 at 1:12 PM, Claudio <cla...@gm...> wrote: >> Dear mattia, >> let me introduce a different scenario. >> >> The USB device are really different each other and support for its >> could be difficult and inefficient, beside on embedded devices >> generally you want to use the USB port for something more useful. > > Hmmmm, quite a lot of people might disagree with you as USB is now > considered THE universal and preferred way to connect external devices > to computers. Anyway, I see your point. > >> The gpio implementaion instread allow you to support a huge range of >> embedded devices like router, develop board as well some routerboard >> or industial board; moreover you can use a USB to GPIO adapter (like >> ftdi) and then connect any kind of lcd also on you modern laptop/pc >> (that typically don't have parport or gpio). > > I agree and this should in principle be very easy. If the GPIO > implementation in the kernel is at a usable level then I'm more than > happy to try to use it. In any case it will provide a sort of standard > way to access the display. If this is the case then it is very easy as > the only two functions in the HD44780 driver that need a rewrite are > the two lowlevel read and write functions. If I write a patch, will > you be able to test it and report on it? > >> The parport implementation is hacky and should be mainained just for >> history but it is de facto obsolete(how many device you can found with >> parport today in your home?). > > I agree and that's why I wanted to move to something newer. LCD-Linux > was intended at the beginning to provide a cheap display output to old > and obsolete hardware. These days this goal does not make any sense as > an embedded device can run as a router much more efficiently than an > old computer. > >> In this scenario the GPIO support is priority and such version of >> lcd-linux is, in imho, also more acceptable by mainline kernel. > > LCD-Linux was already proposed for inclusion in the kernel mainline > but some issues prevented this. The main reason being that LCD-Linux > duplicates in some sense the console emulation and kernel developers > don't like duplicate code (why reinventing the wheel?). To make a long > story short LCD-Linux wasn't rejected but it was asked to merge some > parts of it with the existing kernel code, which is not always easy. > There was some traffic about this on the main linux kernel maling list > in july 2010. > > Best > > Mattia > > >> >> |