|
Changeset
14989:301961548997
by alex@thinkpad:
mlv_snd: made mlv_rec_queue_block optional<br>- reduces the execution time of raw_rec_cbr_started in the vsync hook<br>- fixes memory leak (mlv_snd expects mlv_rec to free queued blocks)
|
 | modules/mlv_lite/mlv_lite.c (diff) |
| modules/mlv_snd/mlv_snd.c (diff) |
|
| modules/mlv_snd/mlv_snd.c (diff) |
Changeset
14987:7fcd17733b77
by alex@thinkpad:
mlv_snd: reworked to achieve better sync with mlv_lite:<br>- sound buffer allocation is done in raw_rec_cbr_starting (which must now be called after video memory allocation)<br>- raw_rec_cbr_started may optionally be called from the vsync hook (which runs from a task with very high priority)<br><br>mlv_lite: call raw_rec_cbr_started from the vsync hook, on first frame<br>- result: first audio frame is started ~ 2ms after the first video frame is fully captured<br><br>mlv_rec: moved raw_rec_cbr_starting after memory allocation<br>- required for mlv_snd to complete the audio memory allocation
|
| modules/mlv_lite/mlv_lite.c (diff) |
| modules/mlv_rec/mlv_rec.c (diff) |
| modules/mlv_snd/mlv_snd.c (diff) |
Changeset
14986:ecc32ff158ed
by alex@thinkpad:
mlv_snd: rather than trying to queue a WAVI block into the MLV recorder, provide a mlv_fill_wavi<br>this way, the MLV recording module can place this block wherever it wants (not necessarily in the queue)<br>previous method still kept for compatibility with mlv_rec<br><br>mlv_lite: refactored with mlv_fill_wavi and moved raw_rec_cbr_started right before the recording loop (where it should be)<br>this should fix audio sync issues when recording starts; not tested
|
| modules/mlv_lite/mlv_lite.c (diff) |
| modules/mlv_rec/mlv.h (diff) |
| modules/mlv_snd/mlv_snd.c (diff) |
|
Changeset
14984:d28868575447
by alex@thinkpad:
Intervalometer: attempt to get sligthly better timing accuracy, to be tested<br>http://www.magiclantern.fm/forum/index.php?topic=21617.msg197314#msg197314
|
| src/shoot.c (diff) |
Changeset
14983:bcbabb33f220
by alex@thinkpad:
Moved timer functions to timer.h (always included from dryos.h)<br>Renamed get_ms_clock_value / get_us_clock_value to get_ms_clock / get_us_clock<br>Replaced all hardcoded references to 0xC0242014 with GET_DIGIC_TIMER()<br>```<br>find . -not -path '*/\.*' -type f -name '*.c' -o -name '*.h' -exec sed -i 's/get_ms_clock_value/get_ms_clock/g' {} +<br>find . -not -path '*/\.*' -type f -name '*.c' -o -name '*.h' -exec sed -i 's/get_us_clock_value/get_us_clock/g' {} +<br>sed -i 's/\*(volatile uint32_t \*)0xC0242014/GET_DIGIC_TIMER()/' */*.c<br>...<br>```
|
| minimal/qemu-frsp/minimal.c (diff) |
| modules/bench/card_bench.c (diff) |
| modules/bench/mem_bench.c (diff) |
| modules/bench/mem_perf.c (diff) |
| modules/bulb_nd/bulb_nd.c (diff) |
| modules/edmac/edmac.c (diff) |
| modules/edmac/edmac_test.c (diff) |
| modules/file_man/file_man.c (diff) |
| modules/io_crypt/io_crypt.c (diff) |
| modules/lua/lua_dryos.c (diff) |
| modules/lua/lua_lens.c (diff) |
| modules/lv_rec/lv_rec.c (diff) |
| modules/mlv_lite/mlv_lite.c (diff) |
| modules/mlv_play/mlv_play.c (diff) |
| modules/mlv_rec/mlv.c (diff) |
| modules/mlv_rec/mlv_rec.c (diff) |
| modules/mlv_snd/mlv_snd.c (diff) |
| modules/selftest/selftest.c (diff) |
| modules/silent/silent.c (diff) |
| modules/trace/trace.c (diff) |
| src/chdk-dng.c (diff) |
| src/debug.c (diff) |
| src/dryos.h (diff) |
| src/fps-engio.c (diff) |
| src/gui-common.c (diff) |
| src/histogram.c (diff) |
| src/installer.c (diff) |
| src/lvinfo.c (diff) |
| src/mem.c (diff) |
| src/menu.c (diff) |
| src/ml_rpc.c (diff) |
| src/module.c (diff) |
| src/rand.c (diff) |
| src/raw.c (diff) |
| src/shoot.c (diff) |
| src/timer.h (diff) |
| src/tskmon.c (diff) |
| src/tweaks.c (diff) |
| src/zebra.c (diff) |
Changeset
14982:7cdf3152f685
by alex@thinkpad:
Reworked get_ms_clock_value / get_us_clock_value / get_seconds_clock<br>- only update the 64-bit us_clock on each call (overhead from ~ 4us reduced to less than 1us)<br>- lower-resolution clocks are always derived from the microsecond clock and computed only on request<br>- dropped get_ms_clock_value_fast
|
| modules/bench/mem_bench.c (diff) |
| src/chdk-dng.c (diff) |
| src/dryos.h (diff) |
| src/fps-engio.c (diff) |
| src/menu.c (diff) |
| src/shoot.c (diff) |
|
| src/imath.h (diff) |
|
|
|
|
|
| platform/600D.102/consts.h (diff) |
|
| platform/60D.111/consts.h (diff) |
|
| platform/1100D.105/consts.h (diff) |
| platform/500D.111/consts.h (diff) |
| platform/50D.109/consts.h (diff) |
| platform/5D2.212/consts.h (diff) |
| platform/6D.116/consts.h (diff) |
| platform/7D.203/consts.h (diff) |
|
| modules/silent/silent.c (diff) |
|
|
| src/lv-img-engio.c (diff) |
|
| src/module.c (diff) |
Changeset
14969:20a52d50f61b
by alex@thinkpad.localdomain:
Modules: fix enabling "long" modules when using config presets (issue #2789)<br>to reproduce: arkanoid.mo -> ML/SETTINGS/MENU.KEY/arkanoid. -> file create error on exfat<br>modules with shorter name: adv_int.mo -> ML/SETTINGS/MENU.KEY/adv_int.e -> it appears to work fine, but it's incorrect<br>worst case: arkanoid.mo -> ML/SETTINGS/MYPRESET.NAM/adv_i -> it appears to work fine, possible name clash with other modules
|
| src/module.c (diff) |
Changeset
14968:cb083db1c8c5
by alex@thinkpad:
shoot: attempt to make public photo capture functions thread safe, to some extent<br>a bit of a hack; some serious rework needed to fix this properly<br>at least it appears to fix http://www.magiclantern.fm/forum/index.php?topic=21590.0
|
| src/lens.c (diff) |
| src/shoot.c (diff) |
|
| src/shoot.c (diff) |
Changeset
14966:379b1afdee92
by alex@thinkpad:
shoot: make the so-called remote_shoot always take pictures, not movies<br>(moved movie functionality in lcdsensor/lightsensor - anything else broken?)
|
| src/lcdsensor.c (diff) |
| src/lightsensor.c (diff) |
| src/shoot.c (diff) |
Changeset
14965:5d7a200eb63c
by alex@thinkpad:
lens_take_picture: always wait until the image capture process starts (and optionally wait for it to finish)
|
| src/lens.c (diff) |
| src/lens.h (diff) |
|
| modules/lua/lua_camera.c (diff) |
Changeset
14963:25a4a032f7ea
by alex@thinkpad:
selftest: also test get_task_name_from_id, not just get_current_task_name<br>(they use different Canon APIs)
|
| modules/selftest/selftest.c (diff) |
Changeset
14962:867744137fd1
by alex@thinkpad:
Attempt to fix conflict between ML and Canon overlays on recent DIGIC 4 models<br>confirmed on 1200D, likely applies to 1300D as well<br>http://www.magiclantern.fm/forum/index.php?topic=12627.msg196304#msg196304
|
| src/propvalues.c (diff) |
|
| src/gui-common.c (diff) |
|
| src/tweaks.c (diff) |
Changeset
14959:4e2447037ef8
by alex@thinkpad:
Patch manager: nicer way to avoid calling _reapply_cache_patches when not present<br>(fixes compiling the installer target and removes the need for defining NO_CACHE_PATCHES)
|
| src/arm-mcr.h (diff) |
| src/reboot.c (diff) |
|
|
| src/arm-mcr.h (diff) |
| src/module.c (diff) |
| src/patch.c (diff) |
| src/patch.h (diff) |
| src/reboot.c (diff) |
| src/reloc.c (diff) |
|
Changeset
14955:db1d2c027f7a
by alex@thinkpad:
chdk-dng: fix saving uncompressed files after a lossless DNG<br>http://www.magiclantern.fm/forum/index.php?topic=19300.msg196806#msg196806
|
| src/chdk-dng.c (diff) |
![[Jenkins]](/jenkins/static/684c4793/images/jenkins-header-logo-v2.svg){kind=logo}
Changes