1.4 development planning thread - page 16 - General Discussion and Assistance - CHDK Forum

1.4 development planning thread

  • 195 Replies
  • 32287 Views
*

Offline srsa_4c

  • ******
  • 4060
Re: 1.4 development planning thread
« Reply #150 on: 06 / September / 2015, 19:46:15 »
Advertisements
I'm fine with any solution that lets me do a full batch build without significant time penalty.
I'm only using batch builds to test stuff I might be working on (a change in CHDK source that affects all ports or a change in a makefile/script). As I'm also working on the sigfinders from time to time, I need a working parallel build script too that doesn't do redundant work. So, as long as these are available, any change is ok for me.

*

Offline reyalp

  • ******
  • 12218
Re: 1.4 development planning thread
« Reply #151 on: 07 / September / 2015, 01:29:15 »
Quote
* benchmark module ui
I accept any ideas... The default settings and behaviour is the same as the old benchmark's.
Attached patch adds a start button and checkboxes for the log and SD card tests. Left/right switch, set activates.

I think that's good enough. Strings aren't lang strings in the patch but they can be (though length is limited)
Don't forget what the H stands for.

*

Offline srsa_4c

  • ******
  • 4060
Re: 1.4 development planning thread
« Reply #152 on: 07 / September / 2015, 13:50:31 »
Attached patch adds a start button and checkboxes for the log and SD card tests. Left/right switch, set activates.

I think that's good enough. Strings aren't lang strings in the patch but they can be (though length is limited)
Looks okay to me too, thanks.

*

Offline reyalp

  • ******
  • 12218
Re: 1.4 development planning thread
« Reply #153 on: 07 / September / 2015, 15:26:40 »
Looks okay to me too, thanks.
Checked in, with strings in lang files. I took out the "benchmark" title since it wouldn't fit with some of the existing translations. I figure it isn't needed since you just clicked benchmark to get there.

I realized I don't really what the best practice is for removing strings from the language file. I replaced the existing id 177 "Camera Benchmark   Press [SET] to start" with the CPU label and and removed the existing translations from non-english files.

Feel free to check in the cpuinfo changes for d6. That module mostly exists as a place to keep track of the code, since it really only needs to be run once per port at most.
Don't forget what the H stands for.


*

Offline srsa_4c

  • ******
  • 4060
Re: 1.4 development planning thread
« Reply #154 on: 07 / September / 2015, 19:42:15 »
Feel free to check in the cpuinfo changes for d6. That module mostly exists as a place to keep track of the code, since it really only needs to be run once per port at most.
Done.

Quote
I realized I don't really what the best practice is for removing strings from the language file. I replaced the existing id 177 "Camera Benchmark   Press [SET] to start" with the CPU label and and removed the existing translations from non-english files.
I don't know the impact of removed string IDs (should probably check), but reusing old IDs has a temporary impact on users who only upgrade the most necessary parts (core+modules). Since the benchmark is not a really frequently used part of CHDK, it probably won't hurt much.

*

Online philmoz

  • *****
  • 3138
    • Photos
Re: 1.4 development planning thread
« Reply #155 on: 11 / September / 2015, 05:26:37 »
Attached patch is for the alternate method of handling the case where the same firmware exists for multiple cameras. It builds; but as I don't have any of the affected cameras I can't test it properly.

Changes:
- remove the PID, and Source Camera columns from camera_list.csv
- update auto_build.sh to cater for the removed columns
- fix Makefile when batch building all cameras with EABI
- add the necessary directories and files for SX270 & SX275 (these are the only affected cameras at the moment).

This is probably simpler than my previous version (with the extra columns in camera_list.csv).

This could also be applied to the cases where we currently copy one firmware version to another .zip file for a specific camera (e.g. a1100 100b & 100c). This would further simplify camera_list.csv and the build stuff, these versions could then also be built with a 'make firsub' command. The downside would be additional batch build time.

Phil.

Updated patch - this adds 'sub' folders and makefile.inc files for all the versions that are 'cloned' from other source versions by copying the zip files.

With this it might also be possible to get CHDK-Shell to work by just ignoring the 'SOURCE_FIRMWARE' column in camera_list.csv - all cameras should build with a 'make firsub' operation.

Phil.
CHDK ports:
  sx30is (1.00c, 1.00h, 1.00l, 1.00n & 1.00p)
  g12 (1.00c, 1.00e, 1.00f & 1.00g)
  sx130is (1.01d & 1.01f)
  ixus310hs (1.00a & 1.01a)
  sx40hs (1.00d, 1.00g & 1.00i)
  g1x (1.00e, 1.00f & 1.00g)
  g5x (1.00c, 1.01a, 1.01b)

*

Offline reyalp

  • ******
  • 12218
Re: 1.4 development planning thread
« Reply #156 on: 12 / September / 2015, 16:08:27 »
Updated patch - this adds 'sub' folders and makefile.inc files for all the versions that are 'cloned' from other source versions by copying the zip files.
I like this approach.

It's slightly weird that the copy sub firmware can be either built or copied, but I think it's good enough.
Don't forget what the H stands for.

*

Offline srsa_4c

  • ******
  • 4060
Re: 1.4 development planning thread
« Reply #157 on: 13 / September / 2015, 14:25:05 »
Updated patch - this adds 'sub' folders and makefile.inc files for all the versions that are 'cloned' from other source versions by copying the zip files.
Look okay to me too. Found one c&p error, the sx275 files are currently duplicates of the sx270 files (wrong notes and P-ID). I'm also not sure if the content of platform/camera/makefile.inc should be also present in platform/camera/sub/fw/makefile.inc (if building breaks otherwise, that will need to be fixed too).

If the patch fixes CHDK-Shell, then it should probably be committed.

Independently of this, some (not frequently used) build actions will need fixing.


*

Offline reyalp

  • ******
  • 12218
Re: 1.4 development planning thread
« Reply #158 on: 13 / September / 2015, 17:00:28 »
I'm also not sure if the content of platform/camera/makefile.inc should be also present in platform/camera/sub/fw/makefile.inc (if building breaks otherwise, that will need to be fixed too).
The autobuild.sh not2 logic assume THUMB_FW is in the sub/fw/makefile.inc. This doesn't have a significant impact now since all the d6 cameras are SKIP_AUTOBUILD anyway.
edit: my mistake, it looks in the platform/cam/makefile.inc only.

grep'ing the makefile.inc files seems a bit suspect, but the only alternative I see is putting a thumb2 indicator both in camera_list.csv and the makefiles, which isn't ideal either.

Quote
If the patch fixes CHDK-Shell, then it should probably be committed.
CHDKShell is confused by the copies (for example D10 shows 100a, 100a, 100b) but it can actually be used to compile rather than just giving a fatal error when source tree is selected. I'd expect that fixing this up should be relatively simple.

In the interest of keeping things moving, I'm inclined to check this in (with the SX275 PID issue fixed, good catch) and fixing whatever remaining issues as the show up.

Edit:
Another minor thing,
override OPT_GEN_STUBS=
was added to the sx280 sub/<fw>/makefile.inc, would probably be better as a conditional on THUMB_FW in one of the top-level makefiles.
« Last Edit: 13 / September / 2015, 17:28:31 by reyalp »
Don't forget what the H stands for.

*

Online philmoz

  • *****
  • 3138
    • Photos
Re: 1.4 development planning thread
« Reply #159 on: 13 / September / 2015, 17:52:52 »
Update patch. This should fix the sx275 c&p error, and auto_build_parallel.sh, also simplified auto_build.sh slightly.

I've left the OPT_GEN_STUBS override in the sx280 makefile.inc file.
My reasoning is that development of a finsig version for thumb2 would most likely begin with a single camera version. Having it in the makefile.inc means it can be turned on/off for individual versions for testing rather than being a global setting.

Phil.
CHDK ports:
  sx30is (1.00c, 1.00h, 1.00l, 1.00n & 1.00p)
  g12 (1.00c, 1.00e, 1.00f & 1.00g)
  sx130is (1.01d & 1.01f)
  ixus310hs (1.00a & 1.01a)
  sx40hs (1.00d, 1.00g & 1.00i)
  g1x (1.00e, 1.00f & 1.00g)
  g5x (1.00c, 1.01a, 1.01b)

 

Related Topics