Age | Commit message (Collapse) | Author |
|
|
|
DRTVWR-577 (#232)
|
|
|
|
|
|
enable github actions mac nightly builds
|
|
|
|
|
|
DRTVWR-559
|
|
The trouble with debugLoggingEnabled() is that it locked mutexes and searched
maps every time that call was reached. LL_DEBUGS() has the same functionality
(albeit with idiosyncratic syntax) but performs expensive lookups only once
per session, caching the result in a local static variable.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
# Conflicts:
# indra/cmake/CMakeLists.txt
# indra/newview/skins/default/xui/es/floater_tools.xml
|
|
# Conflicts:
# indra/cmake/Copy3rdPartyLibs.cmake
# indra/cmake/FindOpenJPEG.cmake
# indra/cmake/OpenJPEG.cmake
# indra/integration_tests/llui_libtest/CMakeLists.txt
# indra/newview/CMakeLists.txt
|
|
|
|
perfomance SL-18563
```
autobuild installables edit "tracy" url="https://automated-builds-secondlife-com.s3.amazonaws.com/ct2/110561/960415/tracy-v0.7.8.578230-darwin64-578230.tar.bz2" hash="70f31fa71ecb52bd092da52e27c3ff8c"
autobuild installables edit "tracy" url="https://automated-builds-secondlife-com.s3.amazonaws.com/ct2/110562/960424/tracy-v0.7.8.578230-windows-578230.tar.bz2" hash="1dc33422939adf015db85e96c5a8276e"
autobuild installables edit "tracy" url="https://automated-builds-secondlife-com.s3.amazonaws.com/ct2/110563/960429/tracy-v0.7.8.578230-windows64-578230.tar.bz2" hash="fcc6ecece2ecb65aa36500dfa9461fb3"
```
|
|
# Conflicts:
# indra/llcommon/llsdserialize.cpp
# indra/llcommon/llsdserialize.h
# indra/newview/llfilepicker.h
# indra/newview/llfilepicker_mac.h
# indra/newview/llfilepicker_mac.mm
|
|
# Conflicts:
# doc/contributions.txt
# indra/cmake/Copy3rdPartyLibs.cmake
# indra/cmake/FindOpenJPEG.cmake
# indra/cmake/OpenJPEG.cmake
# indra/integration_tests/llui_libtest/CMakeLists.txt
# indra/newview/CMakeLists.txt
|
|
|
|
speed matters. (#64)
This commit adds the HBXX64 and HBXX128 classes for use as a drop-in
replacement for the slow LLMD5 hashing class, where speed matters and
backward compatibility (with standard hashing algorithms) and/or
cryptographic hashing qualities are not required.
It also replaces LLMD5 with HBXX* in a few existing hot (well, ok, just
"warm" for some) paths meeting the above requirements, while paving the way for
future use cases, such as in the DRTVWR-559 and sibling branches where the slow
LLMD5 is used (e.g. to hash materials and vertex buffer cache entries), and
could be use such a (way) faster algorithm with very significant benefits and
no negative impact.
Here is the comment I added in indra/llcommon/hbxx.h:
// HBXXH* classes are to be used where speed matters and cryptographic quality
// is not required (no "one-way" guarantee, though they are likely not worst in
// this respect than MD5 which got busted and is now considered too weak). The
// xxHash code they are built upon is vectorized and about 50 times faster than
// MD5. A 64 bits hash class is also provided for when 128 bits of entropy are
// not needed. The hashes collision rate is similar to MD5's.
// See https://github.com/Cyan4973/xxHash#readme for details.
|
|
speed matters. (#64)
This commit adds the HBXX64 and HBXX128 classes for use as a drop-in
replacement for the slow LLMD5 hashing class, where speed matters and
backward compatibility (with standard hashing algorithms) and/or
cryptographic hashing qualities are not required.
It also replaces LLMD5 with HBXX* in a few existing hot (well, ok, just
"warm" for some) paths meeting the above requirements, while paving the way for
future use cases, such as in the DRTVWR-559 and sibling branches where the slow
LLMD5 is used (e.g. to hash materials and vertex buffer cache entries), and
could be use such a (way) faster algorithm with very significant benefits and
no negative impact.
Here is the comment I added in indra/llcommon/hbxx.h:
// HBXXH* classes are to be used where speed matters and cryptographic quality
// is not required (no "one-way" guarantee, though they are likely not worst in
// this respect than MD5 which got busted and is now considered too weak). The
// xxHash code they are built upon is vectorized and about 50 times faster than
// MD5. A 64 bits hash class is also provided for when 128 bits of entropy are
// not needed. The hashes collision rate is similar to MD5's.
// See https://github.com/Cyan4973/xxHash#readme for details.
|
|
|
|
|
|
|
|
|
|
Or rather, attempting to implicitly sign. On TeamCity we must explicitly sign
using viewer_manifest.py. On a developer system, without these changes, Xcode
produces many errors of the form:
error: An empty identity is not valid when signing a binary for the product
type 'Command-line Tool'. (in target 'INTEGRATION_TEST_lldir' from project
'SecondLife')
and refuses to compile anything at all.
Thanks to Rye Mutt and NickyD. Also thanks Geir Nøklebye for additional
settings to help tame Xcode 14.1 warnings.
|
|
|
|
This reverts commit 31917709d9f4d9d4742910ae7990009a1580b150.
|
|
|
|
|
|
enable partial bitstreams
Openjpeg was modified: "p_max_len -= l_nb_bytes_read;" was causing an overflow.
I'm not sure if I did something incorectly in opj_skip/opj_seek viewer side, but seems like openjpeg should have been checking remaining space in p_max_len either way.
P.S. Many thanks to Chafey and Neopallium for implementing openjpeg's partial bitstream support
|
|
Implied 'branch' is now two different version-control systems behind
|
|
|
|
|
|
|
|
|
|
|
|
Part of his change was omly taking Release builds into account, the other
part from me made the same mistake.
Use a generator expression with a custom command to get the symlink the way
we want.
|
|
|
|
The include was preveiously declared for all platforms, but the Mac
package does not even ship this include directory. With the new way
cmake works this will lead to an error (which is good, so we don't
end with bogus include dirs).
|
|
|
|
|
|
|
|
DRTVWR-568_cmake
|
|
generating tangents
|
|
|