SuccessChanges

Summary

  1. Ghost image: further cleanups and fixes
  2. Ghost image: fix crash (issues #2796, #2783, #2878, #2885)
  3. features.tmpl edited online with Bitbucket: use id in th section heading to allow for direct linking via anchor
  4. Merged in edmac (pull request #886) EDMAC module: fixes for DIGIC 4, connection map
  5. Merged in adv_int (pull request #907) Advanced intervalometer updates
  6. Merged in memory-backend (pull request #906) Memory backend improvements
  7. Makefile: fix "make install_qemu CONFIG_QEMU=y" for minimal targets
  8. memset64/memcpy64: use versions from b831cb1; fix calls from Magic Zoom to ensure proper alignment
  9. Backout b831cb1 (breaks Magic Zoom and CONFIG_MARK_UNUSED_MEMORY_AT_STARTUP, as memset64 is not the same as memset)
  10. adv_int: ignore empty lines when loading keyframes
  11. adv_int: use strstr instead of my_strstr (fixes slow load)
  12. Include strstr in ML core
  13. adv_int: refactor keyframe time formatting
  14. adv_int: auto-load the saved sequence at startup
  15. adv_int: renamed menu entry to "Ramping options"
  16. Memory backend: drop timestamps from history; show last 1024 events
  17. Memory backend: show total allocations (number and amount), including untracked ones
  18. Memory backend: allow using shoot_malloc as a last resort, even for small blocks; allow small blocks (smaller than minimum_free_space / 64) to break the minimum_free_space limit (until minimum_free_space / 4) fine-tuned until the old set of Lua scripts were able to run on 1100D without umm_malloc
  19. Memory backend: a pointer to task names (rather than copying the full string) should be enough
  20. Memory backend: renamed memcheck_mallocbuf to memcheck_entries
  21. Moved memset64 and memcpy64 to bench.mo - simplified to accept only 32-bit aligned pointers - using regular memset for Magic Zoom - closing issue #1995
  22. Memory backend: if shoot_malloc fails, try SRM next (poor man's autodetection, since we don't know how to query the available size in advance)
  23. exmem: reworked shoot_malloc to handle large requests better - moved autodetection to dedicated functions (for both fragmented and continuous allocations) - autodetect the max available size for large requests (slower, but when it fails, it fails gracefully)
  24. exmem: removed the "relaxed" checks introduced in 04f02a4 (not sure why they were needed; we'll find out if it fails)
  25. exmem: speed-up shoot_malloc (1s timeout probably not needed)
  26. exmem: do not keep track of full shoot memory allocations (allows reserving some memory in advance and allocating all the remaining RAM afterwards)
  27. Memory backend: fix handling of untracked blocks (possible buffer overflow before this fix)
  28. Memory backend: figured out why AllocateContinuousMemoryResource may crash with many small blocks (todo: figure out where to keep 89e3b12 or not; adapt tests if not)
  29. Memory backend: when checking available memory, allocate both Shoot ans SRM at the same time (they may overlap)
  30. Memory backend: fix initialization
  31. mem.c: option to disable heap poisoning (useful when running in QEMU with -d memcheck - with this, overflowing reads will be caught as well)
  32. Memory backend: allow breaking minimum_free_space for small blocks (it was possible to fail the allocation for very small blocks that could not be covered by shoot_malloc)
  33. Memory backend: allow AllocateMemory as a last resort, even if free space is below limit (it was possible to fail the allocation for very small blocks that could not be covered by shoot_malloc)
  34. memory backend: reduce memcheck area (should reduce overhead a bit)
  35. memory backend: ignore null pointers passed to free (Lua does that)
  36. exmem: when autodetecting available memory, print all srm/shoot memory blocks to console
  37. edmac: make sure the log files don't have unnecessary newlines (channels with complex size configuration may get printed on multiple lines; that makes the log file harder to parse)
  38. edmac: LOG_INTERVAL can now be changed from menu (persistent setting)
  39. edmac: use numbered log file names
  40. edmac: fix behavior when the extra info buffer gets full
  41. edmac: use half-shutter as trigger for logging EDMAC channels; fine-tuned messages
  42. edmac: fix crash on models other than 5D3 (there was a hardcoded stub)
  43. edmac: fix crash when using "Log EDMAC usage" on D4 models
  44. Merged unified into edmac
  45. Added edmac.mo to nightly builds
  46. edmac-memcpy: check to make sure all models have AbortEDmac (to make sure it's correct: selftest.mo -> Stubs API test)
  47. edmac.mo: display connection map (experiment)
Changeset 12838:7a3b5fa3f4c6 by alex@thinkpad:
Ghost image: further cleanups and fixes
The file was modifiedsrc/zebra.c (diff)
Changeset 12837:4702ca0f7c61 by alex@thinkpad:
Ghost image: fix crash (issues #2796, #2783, #2878, #2885)
The file was modifiedsrc/zebra.c (diff)
Changeset 12836:94442d97e9fe by Erik Krause <erik.krause@gmx.de>:
features.tmpl edited online with Bitbucket: use id in th section heading to allow for direct linking via anchor
The file was modifiedfeatures/features.tmpl (diff)
Changeset 12835:5e4098760134 by alex@thinkpad:
Merged in edmac (pull request #886)<br><br>EDMAC module: fixes for DIGIC 4, connection map
Changeset 12834:b54a0c6bafcf by alex@thinkpad:
Merged in adv_int (pull request #907)<br><br>Advanced intervalometer updates
Changeset 12833:58e11ee4e47b by alex@thinkpad:
Merged in memory-backend (pull request #906)<br><br>Memory backend improvements
Changeset 12832:87dfe3c01e99 by alex@thinkpad:
Makefile: fix &quot;make install_qemu CONFIG_QEMU=y&quot; for minimal targets
The file was modifiedMakefile.inc (diff)
Changeset 12831:7c04c75d1098 by alex@thinkpad:
memset64/memcpy64: use versions from b831cb1; fix calls from Magic Zoom to ensure proper alignment
The file was modifiedsrc/stdio.c (diff)
The file was modifiedsrc/zebra.c (diff)
Changeset 12830:375fa4ee027e by alex@thinkpad:
Backout b831cb1 (breaks Magic Zoom and CONFIG_MARK_UNUSED_MEMORY_AT_STARTUP, as memset64 is not the same as memset)
The file was modifiedmodules/bench/mem_bench.c (diff)
The file was modifiedsrc/mem.h (diff)
The file was modifiedsrc/stdio.c (diff)
The file was modifiedsrc/zebra.c (diff)
Changeset 12829:287c785c3752 by alex@thinkpad:
adv_int: ignore empty lines when loading keyframes
The file was modifiedmodules/adv_int/adv_int.c (diff)
Changeset 12828:6239bc6d56d9 by alex@thinkpad:
adv_int: use strstr instead of my_strstr (fixes slow load)
The file was modifiedmodules/adv_int/adv_int.c (diff)
Changeset 12827:284656a24e27 by alex@thinkpad:
Include strstr in ML core
The file was modifiedsrc/Makefile.src (diff)
Changeset 12826:ef21d821a0d2 by alex@thinkpad:
adv_int: refactor keyframe time formatting
The file was modifiedmodules/adv_int/adv_int.c (diff)
Changeset 12825:5ca412b9c46f by alex@thinkpad:
adv_int: auto-load the saved sequence at startup
The file was modifiedmodules/adv_int/adv_int.c (diff)
Changeset 12824:a4869c838682 by alex@thinkpad:
adv_int: renamed menu entry to &quot;Ramping options&quot;
The file was modifiedmodules/adv_int/adv_int.c (diff)
Changeset 12823:86accb693515 by alex@thinkpad:
Memory backend: drop timestamps from history; show last 1024 events
The file was modifiedsrc/mem.c (diff)
Changeset 12822:d016ccfc6d49 by alex@thinkpad:
Memory backend: show total allocations (number and amount), including untracked ones
The file was modifiedsrc/mem.c (diff)
Changeset 12821:0e56fe7ef5c6 by alex@thinkpad:
Memory backend: allow using shoot_malloc as a last resort, even for small blocks;<br>allow small blocks (smaller than minimum_free_space / 64) to break the minimum_free_space limit (until minimum_free_space / 4)<br>fine-tuned until the old set of Lua scripts were able to run on 1100D without umm_malloc
The file was modifiedsrc/mem.c (diff)
Changeset 12820:28bf2592a855 by alex@thinkpad:
Memory backend: a pointer to task names (rather than copying the full string) should be enough
The file was modifiedsrc/mem.c (diff)
Changeset 12819:a67b3edd113a by alex@thinkpad:
Memory backend: renamed memcheck_mallocbuf to memcheck_entries
The file was modifiedsrc/mem.c (diff)
Changeset 12818:b831cb10605c by alex@thinkpad:
Moved memset64 and memcpy64 to bench.mo<br>- simplified to accept only 32-bit aligned pointers<br>- using regular memset for Magic Zoom<br>- closing issue #1995
The file was modifiedmodules/bench/mem_bench.c (diff)
The file was modifiedsrc/mem.h (diff)
The file was modifiedsrc/stdio.c (diff)
The file was modifiedsrc/zebra.c (diff)
Changeset 12817:d7866ee6d226 by alex@thinkpad:
Memory backend: if shoot_malloc fails, try SRM next<br>(poor man&#039;s autodetection, since we don&#039;t know how to query the available size in advance)
The file was modifiedsrc/mem.c (diff)
Changeset 12816:ee7abdbb1bd7 by alex@thinkpad:
exmem: reworked shoot_malloc to handle large requests better<br>- moved autodetection to dedicated functions (for both fragmented and continuous allocations)<br>- autodetect the max available size for large requests (slower, but when it fails, it fails gracefully)
The file was modifiedsrc/exmem.c (diff)
Changeset 12815:bd7a1ddab55b by alex@thinkpad:
exmem: removed the &quot;relaxed&quot; checks introduced in 04f02a4<br>(not sure why they were needed; we&#039;ll find out if it fails)
The file was modifiedsrc/exmem.c (diff)
Changeset 12814:766ffbb4be33 by alex@thinkpad:
exmem: speed-up shoot_malloc<br>(1s timeout probably not needed)
The file was modifiedsrc/exmem.c (diff)
Changeset 12813:51f6129949d9 by alex@thinkpad:
exmem: do not keep track of full shoot memory allocations<br>(allows reserving some memory in advance and allocating all the remaining RAM afterwards)
The file was modifiedsrc/exmem.c (diff)
Changeset 12812:f12eba5ec94a by alex@thinkpad:
Memory backend: fix handling of untracked blocks<br>(possible buffer overflow before this fix)
The file was modifiedsrc/mem.c (diff)
Changeset 12811:f9f3358c78fd by alex@thinkpad:
Memory backend: figured out why AllocateContinuousMemoryResource may crash with many small blocks<br><br>(todo: figure out where to keep 89e3b12 or not; adapt tests if not)
The file was modifiedsrc/mem.c (diff)
Changeset 12810:e4fd29d03af9 by alex@thinkpad:
Memory backend: when checking available memory, allocate both Shoot ans SRM at the same time<br>(they may overlap)
The file was modifiedsrc/mem.c (diff)
Changeset 12809:1162fce78981 by alex@thinkpad:
Memory backend: fix initialization
The file was modifiedsrc/boot-hack.c (diff)
The file was modifiedsrc/mem.c (diff)
The file was modifiedsrc/mem.h (diff)
Changeset 12808:67f582f58d2e by alex@thinkpad:
mem.c: option to disable heap poisoning<br>(useful when running in QEMU with -d memcheck - with this, overflowing reads will be caught as well)
The file was modifiedsrc/mem.c (diff)
Changeset 12807:26f16479aace by alex@thinkpad:
Memory backend: allow breaking minimum_free_space for small blocks<br>(it was possible to fail the allocation for very small blocks that could not be covered by shoot_malloc)
The file was modifiedsrc/mem.c (diff)
Changeset 12806:c6d66993f4dd by alex@thinkpad:
Memory backend: allow AllocateMemory as a last resort, even if free space is below limit<br>(it was possible to fail the allocation for very small blocks that could not be covered by shoot_malloc)
The file was modifiedsrc/mem.c (diff)
Changeset 12805:1e4985e7b889 by alex@thinkpad:
memory backend: reduce memcheck area (should reduce overhead a bit)
The file was modifiedsrc/mem.c (diff)
Changeset 12804:4e59a7474a04 by alex@thinkpad:
memory backend: ignore null pointers passed to free (Lua does that)
The file was modifiedsrc/mem.c (diff)
Changeset 12803:2ade4a5001dc by alex@thinkpad:
exmem: when autodetecting available memory, print all srm/shoot memory blocks to console
The file was modifiedsrc/exmem.c (diff)
The file was modifiedsrc/mem.c (diff)
Changeset 12802:fee1ee2bf3b3 by alex@thinkpad:
edmac: make sure the log files don&#039;t have unnecessary newlines<br>(channels with complex size configuration may get printed on multiple lines; that makes the log file harder to parse)
The file was modifiedmodules/edmac/edmac.c (diff)
Changeset 12801:1508a5468098 by alex@thinkpad:
edmac: LOG_INTERVAL can now be changed from menu (persistent setting)
The file was modifiedmodules/edmac/edmac.c (diff)
Changeset 12800:564b0c5a164b by alex@thinkpad:
edmac: use numbered log file names
The file was modifiedmodules/edmac/edmac.c (diff)
Changeset 12799:edcb52f6d37c by alex@thinkpad:
edmac: fix behavior when the extra info buffer gets full
The file was modifiedmodules/edmac/edmac.c (diff)
Changeset 12798:056e56ee215a by alex@thinkpad:
edmac: use half-shutter as trigger for logging EDMAC channels;<br>fine-tuned messages
The file was modifiedmodules/edmac/edmac.c (diff)
Changeset 12797:2eddc9f9347b by alex@thinkpad:
edmac: fix crash on models other than 5D3<br>(there was a hardcoded stub)
The file was modifiedmodules/edmac/edmac.c (diff)
Changeset 12796:4b161ce83f0f by alex@thinkpad:
edmac: fix crash when using &quot;Log EDMAC usage&quot; on D4 models
The file was modifiedmodules/edmac/edmac.c (diff)
Changeset 12795:bc3bd0dea55d by alex@thinkpad:
Merged unified into edmac
Changeset 12794:b4a8fadef200 by alex@thinkpad:
Added edmac.mo to nightly builds
The file was modifiedmodules/Makefile.modules.default (diff)
Changeset 12793:20619228f9a6 by alex@thinkpad:
edmac-memcpy: check to make sure all models have AbortEDmac<br>(to make sure it&#039;s correct: selftest.mo -&gt; Stubs API test)
The file was modifiedsrc/edmac-memcpy.c (diff)
Changeset 12792:fff0aeb9343b by alex@thinkpad:
edmac.mo: display connection map (experiment)
The file was modifiedmodules/edmac/edmac.c (diff)