Messages

I don't have any issue with mlv_rec.
ML still working even after recording a clip.

Yes, it works, I see now. I was wrong.

A really silly mistake on my part, because I never use my camera for video.  I disabled global draw to avoid errors while recording.  And don't remembered it when finished.

After restarting the camera and not seeing the usual ML text on screen ... I concluded to soon ML was not loading  Sorry 

So everything works fine.

mlv_rec, mlv_play work 
Extra Hacks? The easiest way, disable this in the menu.

Ok, disabling Extra hacks, the error message does not appear anymore.

But there is an annoying issue with mlv_rec  : after recording a clip, Magic Lantern does not boot anymore (the camara works Ok, but without Magic Lantern).
I need to put the sd card on my computer, erase ML folder, and launch "make install" from platorm/500D directory to have it working again.

No problem for me, I don't use video recording at all. In photo mode the build seems to work well.

This one is OK, I hope:

https://bitbucket.org/hudson/magic-lantern/pull-request/531/added-missing-stubs-for-500d-gui-timers-an/diff

I think something is wrong. I accepted to do a pull request at the same time I was doing the commit.

But the pull request appears only on my own repository, not in ML main repository  .

Allready done 

https://bitbucket.org/josepvm/ml_500d_josepvm/pull-request/1/added-missing-stubs-for-500d-gui-timers-an/diff

Many thanks to Greg, Audionut, A1ex and Pelican for assisting me in my first contribution to this awesome project. 

Thanks, Audionut.   

This makes it much easier for me.

Done again 

Code Select Expand


NSTUB(0xFF0696FC,  msg_queue_count)


I hope this was the last one, or my brain will explode 

mlv_rec and mlv_play are working now.  Only tested at minimum resolution, global draw disabled.  An error message appears when starting or ending the recording, but the movie is saved,  mlv_play successfully plays it.

And Stubs API test completes with no errors.

I will try to commit and push request my stubs.S file tomorrow, I have not done this before.

Done 

Code Select Expand


/** ResLock **/
NSTUB(0xFF19384C,  CreateResLockEntry)
NSTUB(0xFF194190,  LockEngineResources)
NSTUB(0xFF193ED0,  UnLockEngineResources)


Now it builds , and works nicely  in my 500D

mlv_rec and mlv_play modules do not build, but the ones I use most (Dual-ISO, ETTR) work well.

There is any specific functionality I should test, to be sure the new stubs are working ?

Ok, thanks, I will look for these ones.

Hi,

Since the last stubs reformat one month ago, Nightly Builds for 500D are broken.

Following A1ex tutorial, I have finally found the missings stubs for 500D:

FIO_SeekSkipFile    0xFF1967D4
SetTimerAfter       0xFF070500
CancelTimer         0xFF070700
SetHPTimerAfterNow  0xFF068268
SetHPTimerNextTick  0xFF0682E8

I have put this values in stubs.S for 500D, and tried to build.
But the build still fails:

Code Select Expand

[ CC       ]   chdk-dng.o
[ CC       ]   edmac-memcpy.o
[ AR       ]   strrchr.o
[ AR       ]   dietlibc.a
[ AR       ]   lib_a-setjmp.o
[ AR       ]   newlib-libc.a
[ CP       ]   newlib-libm.a
[ CP       ]   gcc-libgcc.a
[ LD       ]   magiclantern
edmac-memcpy.o: In function `edmac_memcpy_init':
edmac-memcpy.c:(.text+0x78): undefined reference to `CreateResLockEntry'
edmac-memcpy.o: In function `edmac_memcpy_res_lock':
edmac-memcpy.c:(.text+0xe0): undefined reference to `LockEngineResources'
edmac-memcpy.o: In function `edmac_memcpy_res_unlock':
edmac-memcpy.c:(.text+0x114): undefined reference to `UnLockEngineResources'
edmac-memcpy.o: In function `edmac_memcpy_start':
edmac-memcpy.c:(.text+0x628): undefined reference to `LockEngineResources'
make[1]: *** [magiclantern] Error 1
make[1]: Leaving directory `/home/jv/ML_Unified/magic-lantern/platform/500D.111'
make: *** [500D] Error 2


These errors are also present in build output for failed builds on the Nightly builds download page.

So it seems there is still something missing, and not only GUI timer stubs   

Any chance to see builds for 500D working again ?

Sorry.  I'm still trying to find the missing stubs for 500D, in order to have nightly builds working again for this camera.

I've posted new data here: https://bitbucket.org/hudson/magic-lantern/commits/03bb4ef73a77a5bf4b4dbb05c8b473f789439bb2?at=gui-timers

Please, if a developer could take a look at it, I will appreciate very much your help 

Hi, I'm new to this forum, but I'm very happy as a Magic Lantern user, and I would like to help having nightly builds for 500D working again.

I'm very outdated as a developer.  Last time I wrote a C program was 23 years ago, using Borland Turbo C 2.0. And in assembler, things are even worse, last time was using a ZX Spectrum, storing my code in a cassette tape 

But I've read Alex's tutorial for finding stubs, and I have disassembled my 500D firmware using "disassemble.pl"

I've found that FIO_SeekSkipFile is located at FF1967D4

That follows the 0xD0 offset rule from FIO_SeekFile (FF1968A4 for 500D) as observed in other cameras. And the stub has the structure described in the first example of the tutorial: starts with a "push" instruction, the stub is the destination for several calls from other sections in the firmware. with "bl" instructions, including a call from FIO_SeekFile.

I've also found a similar structure starting with "push" at FF3A96EC, that could be the "CancelTimer" stub.

It's called from 4 places, and the routine includes the string "DlgPlayMovieControlGuide.c CancelTimer" after an "add" instruction.

This is the first time I do this. So please, if a more experienced ML contributor could look at this adresses, would be great.

And many thanks to all the developers for your great work.
-----------------------------------------------------------------------------------------
  Edit:  I've posted this info as a comment in the related bitbucket commits page https://bitbucket.org/hudson/magic-lantern/commits/03bb4ef73a77a5bf4b4dbb05c8b473f789439bb2?at=gui-timers , I supose that is a better place.