gigabeat screen rotation patch

[Mad Cow]

Haven't tried it yet, but SWEET!  Thanks for your work!  And again, thanks to MadCow and EvilG for reviving this effort.

*** EDIT ***
Just tried your bootloader:
* patched, and loaded patched build.  
   * I have .rockbox and .rockbox_1 directories.  
   * Made sure I copied the fonts directory over to the .rockbox_1 dir.
* Powered down.
* Turned the battery off and on
* Powered on.
*Hit/held the up "button."

I received some print out of text that I can't really see because it flashes too quickly.  I believe it looks like the info you get when you don't have the bootloader correctly installed or something.  Is there anyway to capture that message?

After error, it loads in portrait mode.  Everything works fine in portrait mode.

First of all, you don't need to hold the "up" button, just hold any button on the touchpad, preferably the middle, and start holding it while you're still in Toshiba's bootloader with the loading bar. Also, make sure there's a "rockbox_1.gigabeat" in your ".rockbox_1" dir. The best way to capture that message would probably to take a video of it loading and find the frame(s) with the error message.

[Nate!]

AaaHaa!  I needed to rename the gigabeat.rockbox in the .rockbox_1 dir.  Thanks.  You guys are the greatest.  For Real!

[Mad Cow]

AaaHaa!  I needed to rename the gigabeat.rockbox in the .rockbox_1 dir.  Thanks.  You guys are the greatest.  For Real!

Why did you have to rename it? Did you download my rotated build?

[Nate!]

Oh no, I didn't download your build.  I only really use the Album Art patch.  Maybe I'll add scrolling margins occasionally when it syncs.  Aside from those two and the  Rotate patch, I don't really use any other patches.  And as far as games, I don't really play anything in the airport or where ever except what comes with Rockbox.  (Mostly just Bubbles, Brickmania, and Xobox.)

So that's why I had to rename.  Everything works great now.  I had to of course update a couple WPSs and config files, but everything is great!

I do appreciate you finally getting this to work.  Again, my sincerest thanks!

[lights0ut]

Mad Cow, this is awesome news. Thanks for taking the time to get this code to work, I'm glad that all my headaches were worth someting . Guess this means I can make a build for that can be used on the X series, and with all the fancy code improvements like those neet button lights.

[lights0ut]

Nate has let me know that this patch needs a resync due to recent commits for LCD optimizations.
We no longer use DMA for the LCD so the code has been changed a bit. I've been unsuccessfully trying to sync this, so I thought I'd just get the word out.
The file that is out of sync is:
/firmware/target/arm/s3c2440/gigabeat-fx/lcd-meg-fx.c
which has 3 hunks fail. I think what needs to be done is modify the variables in the new functions (LCD_WIDTH changes to R_LCD_WIDTH) but it may be more.

[lights0ut]

I've managed to sync this patch so no worries, but thanks to anyone who even considered lending a hand

[Mad Cow]

I've managed to sync this patch so no worries, but thanks to anyone who even considered lending a hand

What did you do? I'm glad to hear that though, because I'm terrible with patches and I didn't really want to do it.

[lights0ut]

I'll attach the .rej so you can see what wasn't patching, and try to explain as best I can:
-the first hunk I needed to just copy and paste, as well as the second.
-The next hunk dealt with DMA, which was removed from SVN as we do not usDMA to access the
  Gigabeat's LCD. By looking at the diffs from the recent SVN commits I figured that this stuff was
  unneccessary, so I continued to the next hunk
-the hunk after that deals with the memcpy function (which I assume is used to draw the LCD) and appears to
  be the same, so again i just copied the the intended code and replaced it.

At first, I thought I had to find all instances of LCD_WIDTH and replace them with R_LCD_WIDTH etc. but after running into problems I decided to take some closer looks at the patch file, the .orig, the .rej, and the SVN diffs (the SVN diffs were the real key to this). After reading them, I came up with the solution I just explained, and after some 'massaging' it worked. I didn't quite copy and paste as I was supposed to, an extra } character here or there, and I think one of the parts I needed I forgot to paste in the first time. That stuff I figured out from gcc's error messages.

[attachment deleted by admin for age]

[Mad Cow]

I'll attach the .rej so you can see what wasn't patching, and try to explain as best I can:
-the first hunk I needed to just copy and paste, as well as the second.
-The next hunk dealt with DMA, which was removed from SVN as we do not usDMA to access the
  Gigabeat's LCD. By looking at the diffs from the recent SVN commits I figured that this stuff was
  unneccessary, so I continued to the next hunk
-the hunk after that deals with the memcpy function (which I assume is used to draw the LCD) and appears to
  be the same, so again i just copied the the intended code and replaced it.

At first, I thought I had to find all instances of LCD_WIDTH and replace them with R_LCD_WIDTH etc. but after running into problems I decided to take some closer looks at the patch file, the .orig, the .rej, and the SVN diffs (the SVN diffs were the real key to this). After reading them, I came up with the solution I just explained, and after some 'massaging' it worked. I didn't quite copy and paste as I was supposed to, an extra } character here or there, and I think one of the parts I needed I forgot to paste in the first time. That stuff I figured out from gcc's error messages.

I have no idea what you just said, but good job.

[Nate!]

I have no idea what you just said, but good job.

I thought it was just me.  Thanks again EvilG for the work.  

[lights0ut]

I'm not joking, this makes complete sense to me the way I've explained it.

[Mad Cow]

I'm not joking, this makes complete sense to me the way I've explained it.

Maybe it's because I'm tired. I'll check it out tomorrow, maybe then I'll get it.