**** BEGIN LOGGING AT Tue Jun 07 23:59:56 2005
Jun 08 03:46:01 <ep1220> [g2]: Hi
Jun 08 03:46:13 <[g2]> ep1220, Hi!
Jun 08 03:47:00 <ep1220> I first used the FTDI JTAG lib, but it has a nasty bug
Jun 08 03:47:10 <ep1220> So i write my own now.
Jun 08 03:47:16 <[g2]> really ?  what is it ?
Jun 08 03:47:34 <ep1220> in a few case they count the bits from the wrong direction.
Jun 08 03:48:08 <ep1220> reault. It is impossible to read out the ID of a chip
Jun 08 03:48:10 <[g2]> that's not good :(
Jun 08 03:48:25 <ep1220> Well, it is Win only, and  closed source anyway
Jun 08 03:48:36 <ep1220> just had saved some time upfront
Jun 08 03:49:24 <ep1220> And they are also missing functionality needed for JTAG debug
Jun 08 03:49:42 <ep1220> So a write-my-own was inevitable anyway
Jun 08 03:49:47 <[g2]> ep1220, is the MAN!
Jun 08 03:50:22 <[g2]> I got my 1st unit back last week
Jun 08 03:50:30 <[g2]> Memory and serial are good
Jun 08 03:50:41 <[g2]> I've been running with 64MB for many days now
Jun 08 03:51:23 <ep1220> NSLU2 or RV042 ?
Jun 08 03:54:24 <[g2]> nslu2
Jun 08 03:54:48 <ep1220> Nice. Did they add a JTAG ?
Jun 08 03:54:54 <[g2]> They did
Jun 08 03:55:14 <[g2]> I think that added the 0 ohm connector that's on TMS
Jun 08 03:55:26 <[g2]> I don't know if they added the pull-down 
Jun 08 03:55:51 <ep1220> there is a 0 Ohm at TRST and they should remove it
Jun 08 03:56:16 <ep1220> It is R135
Jun 08 03:56:59 <[g2]> that may be it
Jun 08 03:57:25 <ep1220> You maybe could add a pull-down directly at the JTAG connector.
Jun 08 03:57:35 <ep1220> if You do not add it:
Jun 08 03:57:49 <ep1220> You might have to reset the CPU after attaching a JTAG probe
Jun 08 03:58:44 <[g2]> yeah I need to double check that stuff 
Jun 08 03:58:55 <[g2]> and probably check with a hw guy :)
Jun 08 04:00:06 <ep1220> My SW is making good progress
Jun 08 04:00:23 <[g2]> excellent
Jun 08 04:00:49 <ep1220> Needs a few hours more, for testing.
Jun 08 04:01:12 <ep1220> There are some tricky shift bit into right place, read bit from right place situations in there.
Jun 08 04:01:27 <ep1220> Just writing a test-suite
Jun 08 04:01:40 <[g2]> that's a good plan
Jun 08 04:02:05 <ep1220> You remember we talked about writing a small program to out into the chache ?
Jun 08 04:02:12 <ep1220> s/out/put
Jun 08 04:03:19 <[g2]> ep1220, yes. I spoke with beewoolie about that
Jun 08 04:03:31 <[g2]> we didn't think it would be too difficult
Jun 08 04:03:46 <[g2]> I'm guessing you are ready now
Jun 08 04:04:06 <ep1220> tomorrow I will be
Jun 08 04:05:17 <[g2]> Ok... I'll try to have the LED program ready
Jun 08 04:05:36 <ep1220> Great.
Jun 08 04:05:58 <ep1220> using a gcc cross compiler ? which version ?
Jun 08 04:06:36 <[g2]> well for the led code I'd imagine it's just a dozen or so instructions
Jun 08 04:07:06 <ep1220> likely
Jun 08 04:07:07 <[g2]> so wanted binary and a .c equivalent
Jun 08 04:07:13 <[g2]> plus source 
Jun 08 04:07:33 <[g2]> We're currently using gcc 3.4.4 with OpenSlug
Jun 08 04:08:54 <ep1220> source + an object/binary or the hexcodes in a .c (array of int or similar)
Jun 08 04:09:06 <ep1220> source only would be helpfull as well 
Jun 08 04:09:25 <ep1220> Just as You guys find time
Jun 08 10:08:15 <prpplague> ~seen beewoolie
Jun 08 10:08:16 <purl> beewoolie <~beewoolie@florence.buici.com> was last seen on IRC in channel #openjtag, 3d 23m 40s ago, saying: 'jacques: ping.'.
Jun 08 11:04:59 <beewoolie-afk> prpplague: yo
Jun 08 11:05:16 <prpplague> beewoolie: hey
Jun 08 11:05:23 <prpplague> pl110 driver
Jun 08 11:05:34 <prpplague> beewoolie: i've had hell with getting my display working
Jun 08 11:05:43 <beewoolie> What sort of panel.
Jun 08 11:05:58 <prpplague> beewoolie: 160x160 4bit monochrome
Jun 08 11:06:21 <beewoolie> Did you get RMKs patch for grayscape?
Jun 08 11:06:28 <beewoolie> s/grascape/grayscale/
Jun 08 11:06:39 <prpplague> negative, didn't know there was one
Jun 08 11:06:49 <prpplague> url?
Jun 08 11:06:52 <beewoolie> It was posted to the arm list a few weeks back.
Jun 08 11:06:57 <beewoolie> Search marc.
Jun 08 11:07:04 <prpplague> that was for 2.6 though right?
Jun 08 11:07:08 <beewoolie> Yeah.
Jun 08 11:07:13 <beewoolie> Are you still in 2.4 land?
Jun 08 11:08:27 <prpplague> yea
Jun 08 11:08:28 <prpplague> legacy stuff
Jun 08 11:08:30 <prpplague> i can't seem to find any good examples of using a grayscale with the pl110
Jun 08 11:09:02 <beewoolie> Not much I can offer for 2.4.  Based on my previous experience, I'm skeptical that it works with anything not already supported.
Jun 08 11:09:04 <prpplague> all my settings appear correct, but its like nothing is going from the sdram fb to the lcd controler
Jun 08 11:09:19 <beewoolie> did you try sram?
Jun 08 11:09:44 <prpplague> yea, i tried mapping to sram, but i don't know if it was correct
Jun 08 11:10:00 <prpplague> i'm kinda suspecting that my lcd controler is toast at the momment
Jun 08 11:10:42 <prpplague> beewoolie: i'll have a look at rmk patch
Jun 08 11:10:51 <prpplague> beewoolie: any other suggestions?
Jun 08 11:12:31 <beewoolie> prpplague: Nothing I can think of.  I've not spent much time with LCD stuff.  In fact, it's always worked straight away for me.  Even when I adapted a custom panel.
Jun 08 11:12:51 <beewoolie> prpplague: Can you measure the frequency of the pixel clock?
Jun 08 11:14:12 <prpplague> beewoolie: yea, we have a scope on it
Jun 08 11:14:20 <prpplague> beewoolie: all the timing looks good
Jun 08 11:14:23 <beewoolie> So, that's it's ok.
Jun 08 11:14:35 <prpplague> beewoolie: just don't get anything when we send data to the the fb
Jun 08 11:14:52 <beewoolie> I'd start with SRAM.
Jun 08 11:15:10 <[g2]> beewoolie, how goes ?
Jun 08 11:15:22 <beewoolie> I'd also try programming the controller outside of the kernel.
Jun 08 11:15:26 <beewoolie> [g2]: hey man
Jun 08 11:15:35 <[g2]> did you get my latest e-mail ?
Jun 08 11:15:51 <beewoolie> prpplague: Taking the MMU out of the loop may help illuminate problems.
Jun 08 11:16:01 <beewoolie> prpplague: There is LCD initialization code in APEX for this purpose.
Jun 08 11:16:05 <beewoolie> [g2]: ...loooking
Jun 08 11:16:19 <[g2]> sorry... I thought you were done with prpplague 
Jun 08 11:16:25 <beewoolie> [g2]: I got one that told me prpplague was looking for me.
Jun 08 11:16:26 <prpplague> [g2]: np
Jun 08 11:16:47 <prpplague> beewoolie: is the code in apex using the sram?
Jun 08 11:17:15 <[g2]> nod that's the one
Jun 08 11:17:40 <beewoolie> prpplague: No.  It uses sdram.  But you can change the pointer easily.
Jun 08 11:18:10 <beewoolie> [g2]: did you get my reply?
Jun 08 11:18:48 <[g2]> beewoolie, nod.
Jun 08 11:23:09 <prpplague> beewoolie: have you used the pl110 with a 4bit grayscale before?
Jun 08 11:23:58 <beewoolie> prpplague: No.  I've only used a couple of color panels.
Jun 08 11:24:39 <prpplague> any 4bit?
Jun 08 11:25:13 <beewoolie> I think maybe.  I don't remember.  I've used the 3.5" standard display and another wierd custom display.
Jun 08 11:25:31 <beewoolie> I don't recall the details on the custom panel.  It had a really wierd output format.
Jun 08 11:25:53 <prpplague> hmm, i've got rmk's original pl110 patch, but i'm not seeing a grayscale patch
Jun 08 11:26:14 <beewoolie> prpplague: Let me see if I can find the email.
Jun 08 11:32:33 <[g2]> beewoolie, lennert stopped by in nslu2-linux :)
Jun 08 11:34:21 <beewoolie> [g2]: Who is that?
Jun 08 11:36:24 <[g2]> lennert from the arm-linux ml
Jun 08 11:36:31 <beewoolie> http://lists.arm.linux.org.uk/pipermail/linux-arm-kernel/2005-April/028564.html
Jun 08 11:36:38 <beewoolie> prpplague: I think that this is the patch.
Jun 08 11:39:43 * prpplague looks
Jun 08 11:44:40 <prpplague> beewoolie: ok thanks, that looks like some of the issues i've already corrected, but i look further
Jun 08 12:23:03 <[g2]> beewoolie, what's the latest status with jffs2 and loading the kernel ?
Jun 08 13:07:41 <beewoolie> [g2]: It's waiting for me to have a chance to look at it again.  I understand the scope of the problem and have a solution.  It's a matter of implementation and test.
Jun 08 13:08:04 <[g2]> beewoolie, thx
Jun 08 13:08:44 <[g2]> jbowler has a tool used in the turnup proceedue that can replace the jffs2 and kernel from within OpenSlug
Jun 08 13:12:10 <beewoolie> Noce
Jun 08 13:12:12 <beewoolie> Nice
Jun 08 13:12:24 <beewoolie> I may be able to do some work on this today.
Jun 08 13:13:01 <[g2]> I reflash jffs2 from the /dev/sda1 today and rebooted
Jun 08 13:13:06 <[g2]> it's nice :)
Jun 08 13:13:47 <[g2]> I think for me I probably won't be reloading redboot for quiet a while
Jun 08 13:27:15 * beewoolie celebrates
Jun 08 13:27:52 <beewoolie> [g2]: I'm clearing my plate as we speak.  It looks like I can carve out a couple of hours to work on the JFFS2 driver.
Jun 08 13:27:53 <[g2]> beewoolie, what's the celebration ? not using redboot ?
Jun 08 13:28:01 <beewoolie> Uh huh.
Jun 08 14:24:23 <prpplague> jeeze just when i had a question
**** ENDING LOGGING AT Wed Jun 08 23:59:56 2005