#816 Open

Bitbucket cannot automatically merge this request.

The commits that make up this pull request have been removed.

Bitbucket cannot automatically merge this request due to conflicts.

Review the conflicts on the Overview tab. You can then either decline the request or merge it manually on your local system using the following commands:

hg update 7a3b5fa3f4c6
hg pull -r update-to-5D3.134 https://bitbucket.org/daniel_fort/magic-lantern
# Note: This will create a new head!
hg merge update-to-5D3.134
hg commit -m 'Merged in daniel_fort/magic-lantern/update-to-5D3.134 (pull request #816)'
  1. Daniel Fort

Created 5D3.134 firmware update without destroying the well established 5D3.113 and easy to downgrade 5D3.123 versions.

The source code history was preserved by using 5D3.123 as a starting point:

hg copy 5D3.123 5D3.134

Comments (14)

    1. Alex

      That's a good sign.

      Both the stubs API and lua API have some extra tests on the lua_fix branch, so it's worth checking those as well.

      Unfortunately I'm away from the 5D3 until next week, and I'm not sure how much time I'll have after that (holidays will be over), but if these tests passed, I think it's OK for others to try and review.

      TODO: find out stubs/consts not covered by the self tests. That's another easy "coding" task, as it only requires looking up each of them and writing down what feature(s) must be tested to verify it. Adding tests for them is also a good idea.

      1. Daniel Fort author

        There's certainly no need to rush this out. I already got some issues to resolve and in fact we already have items that should be added to the stubs/consts list. Is there already a list started?

        • FRAME_CMOS_ISO_START - for dual_iso
        • dialog_refresh_timer_addr - for raw video recording
      1. Daniel Fort author

        Nice April 1st surprises. Looks like I've got some more work to do on this.

        Here's the reminder for the ML-SETUP.FIR.

  1. DeafEyeJedi

    Indeed it was easy to frowngrade (is that even a word?) from 134 to 113/123 via EOS Utility (Thanks @dfort for the instructions) and definitely what a nice 4/1/17 surprise, @a1ex!

    1. Daniel

      Did I miss any surprises? I only got a flickering screen (700D) and thought I had broken something. I tried to figure how to flash a backup ROM back, because I thought I had pulled the SD-Card to fast, then I remembered the date and nearly died laughing

      1. Daniel Fort author

        You posted a 700D comment on a 5D3 pull request--oops. In any case, ML is a hobby project and it is supposed to be fun but some people are taking this very very very seriously. Now that we're up to 4K on the 5D3 we can probably push some of the other cameras up to full HD.

        Back on topic -- it would be good if we get some more people doing relatively simple tasks like porting a minor Canon firmware update. I've got one for the 700D too.

  2. Daniel Fort author

    @a1ex Shameless bump--thought it was an appropriate day to resurrect this PR.

    Got 219 downloads of the 5D3.134 test build and users asking about the ML-SETUP.FIR file.

  3. Daniel Fort author

    A user PM'd me about getting Err 70 when recording raw video. That's the first I heard of it happening with 1.3.4 and there have now been 248 downloads of the test build. Suggested he clear all his settings and try it again. I also posted a new test build on my bitbucket download area that includes the merge with unified to eliminate the conflicts that showed up with the 5D3 1.2.3: current_task and current_interrupt stubs (commit ecaa37f).

  4. Daniel Fort author

    Found a few stubs that needed fixing. The Q button now works fine on the Canon menus.

    1. Daniel Fort author

      Well it is still missing a ML-SETUP.FIR file. That is keeping some users from trying it and giving feedback. I tested it when I first made the pull request but haven’t continued using it because Canon updated to firmware version 1.3.5. I ran into problems when doing the 1.3.5 port.



      Note that a1ex has some interesting ideas to get around the issues when downgrading the firmware to pre-1.3.3 versions.