| Age | Commit message (Collapse) | Author |
|---|
|
Retargeting PR #1496 to Maint B.
|
|
Introduce AlwaysReturn<void> specialization, which always discards any result
of calling the specified callable with specified args.
Derive new Windows_SEH_exception from LLException, not std::runtime_error.
Put the various SEH functions in LL::seh nested namespace, e.g.
LL::seh::catcher() as the primary API.
Break out more levels of Windows SEH handler to work around the restrictions on
functions containing __try/__except.
The triadic catcher() overload now does little save declare a std::string
stacktrace before forwarding the call to catcher_inner(), passing a reference
to stacktrace along with the trycode, filter and handler functions.
catcher_inner() accepts the stacktrace and the three function template
arguments. It contains the __try/__except logic. It calls a new filter_()
wrapper template, which calls fill_stacktrace() before forwarding the call to
the caller's filter function. fill_stacktrace(), in the .cpp file, contains
the logic to populate the stacktrace string -- unless the Structured Exception
is stack overflow, in which case it puts an explanatory string instead.
catcher_inner()'s __except clause passes not only the code, but also the
stacktrace string, to the caller's handler function. It wraps the caller's
handler function in always_return<rtype>(), where rtype is the type returned
by the trycode function. This allows a handler to return a value, while also
supporting the void handler case, e.g. one that throws a C++ exception. (This
is why we need AlwaysReturn<void>: some trycode() functions are themselves
void.)
For the dyadic catcher() overload, introduce common_filter() containing the
logic to distinguish a C++ exception from any other kind of Structured
Exception. The fact that the stacktrace is captured before the filter function
is called should permit capturing a stacktrace for a C++ exception as well as
for most other Structured Exceptions.
As before, the monadic catcher() overload supplies the rethrow() handler, in
the .cpp file.
Change existing calls from seh_catcher() to LL::seh::catcher().
|
|
|
|
Since August 2023, we've seen occasional GitHub Windows build test runs
terminate with 0xC00000FD: stack overflow. We've usually responded by bumping
up the default coroutine stack size.
On closer examination, it's always llleap_test.cpp that blows up that way --
and llleap_test.cpp doesn't appear to use coroutines at all. So apparently
we've been consuming more address space for ALL viewer coroutines without
actually addressing the problem.
Reset the default coroutine stack size to where it was before we started
bumping it up in response to these llleap_test.cpp stack overflow failures.
Note that LLCoros already catches and reports Windows structured exceptions,
underscoring that the observed stack overflow is not from within a coroutine.
While at it, restore the Windows llleap_test.cpp data volume to match Posix.
We think the problem that led to reducing that data volume was an APR bug,
which we hope has been fixed.
Equip test.cpp, the test driver program for all our TUT unit and integration
tests, with a Windows structured exception handler. Try to treat a Windows
structured exception as a test failure -- instead of silently terminating with
0xC00000FD. Moreover, when a structured exception occurs, output a stack trace
so we can try to track it down.
|
|
# Conflicts:
# autobuild.xml
# indra/cmake/ConfigurePkgConfig.cmake
# indra/cmake/ICU4C.cmake
# indra/media_plugins/gstreamer010/llmediaimplgstreamer_syms.cpp
# indra/media_plugins/gstreamer010/llmediaimplgstreamer_syms.h
# indra/media_plugins/gstreamer010/llmediaimplgstreamertriviallogging.h
# indra/media_plugins/gstreamer010/llmediaimplgstreamervidplug.cpp
# indra/media_plugins/gstreamer010/llmediaimplgstreamervidplug.h
# indra/media_plugins/gstreamer010/media_plugin_gstreamer010.cpp
# indra/newview/llappviewerlinux_api.h
# indra/newview/llappviewerlinux_api_dbus.cpp
# indra/newview/llappviewerlinux_api_dbus.h
# indra/newview/llfloateremojipicker.cpp
# indra/newview/lloutfitslist.cpp
|
|
Maintenance X
|
|
(cherry picked from commit dc0b3aed4782e4e4835fd6b9d59d1d70b78be4a7)
|
|
|
|
LF, and trim trailing whitespaces as needed
|
|
|
|
|
|
* Remove all GCC warning suppression pragmas.
* For Linux just just raise(SIGSEGV) as the crash driver. This has a much higher chance of the compiler understanding out intent and figuring out we end the program here.
* Remove -Wno-stringop-overflow and -Wno-stringop-truncation from GCC_WARNINGS. After calling raise(SIGSEGV) as the crash driver I saw no issue with those warnings anymore
After removing thoses GCC pragmas there is also no need for clang -Wno-unknown-warning-option anymore.
* Remove CMakePresets from this PR.
* Remove Lindens from comments :)
|
|
# Conflicts:
# indra/llimage/llimageworker.cpp
# indra/llimage/llimageworker.h
# indra/newview/llcontrolavatar.cpp
# indra/newview/llfloaterprofiletexture.cpp
# indra/newview/lloutfitslist.cpp
# indra/newview/lloutfitslist.h
# indra/newview/lltexturefetch.cpp
|
|
# Conflicts:
# autobuild.xml
# indra/llcommon/llsys.cpp
|
|
# Conflicts:
# indra/llrender/llgl.cpp
# indra/newview/lloutfitslist.cpp
|
|
* Enable CEF browser for Linux
* Disable the update for Linux, we don't have that one right now
* Update build_linux.yaml
We need libpulse-dev for volume_catcher Linux
* Add linux_volum_catcher* files
* Enable OpenAL for Linux-ReleaseOS
* Linux: Update OpenAL
* Update SDL2
* Add libsndio-dev to the dependencies.
* Update CEF to an official LL version
* Remove dupe of emoji_shortcodes
* Reording autobuild does because it can and wants to
* Linux: Disable NDOF for the time being. After updating the ndof 3P needs to be rebuilt and we do not have a fresh one from LL yet.
Forcefully undefine LIB_NDOF, it gets defined in the build variables no matter if it is safe to define.
* Remove wrestling with mutliarch and LIBGL_DRIVERS_PATH
* Remove tcmalloc snippet, tcmalloc is a very faint bad dream of the past
* Putting out a warning this viewer ran on a x64 arch and then suggesting to install 32 bit compat packages makes no sense at all
* CEF resources need to be in lib
* It;'s okay to warn about missing plugins
* Linux: CEF keyboard handling
* Remove old gstreamer 0.10 implementation
* Linux DSO loading always had been very peculiar due to macro magic.
At least now it is peculiar shared magic with only one implementation.
* Remove -fPIC. We get that one from LL_BUILD
* /proc/cpuinfo is not reliable to detrmine the max CPU clock. Try to determine this by reading "/sys/devices/system/cpu/cpu0/cpufreq/cpuinfo_max_freq".
Only if this fails go back to /proc/cpuinfo
* Cleanup
* Cleanup common linker and compiler flags, make it more obvious which flags are for which OS/compiler
* Switch to correct plugin file
* Install libpulse-dev for volume catcher.
* And the runner needs libsndio-dev as well.
* check for runner.os=='linux'. matrix.os is the full name of the image (limux-large).
|
|
* compile flag cleanup for linux
* rollback indra/llcommon/llprofiler.h
* use 3p fltk package
* fix build gcc 13 (dangling-pointer)
|
|
Co-authored-by: AiraYumi <aira.youme@airanyumi.net>
|
|
|
|
|
|
looks like file that was being parced got corrupted 'in progress'
|
|
|
|
# Conflicts:
# indra/llui/lltransutil.cpp
# indra/newview/app_settings/settings.xml
# indra/newview/llfloaterenvironmentadjust.cpp
# indra/newview/llpaneleditwater.cpp
# indra/newview/llpanelface.cpp
# indra/newview/lltexturectrl.cpp
# indra/newview/lltexturectrl.h
|
|
# Conflicts:
# .github/workflows/build.yaml
|
|
Closing window correctly caused a significant amount of logout freezes
with no known reproes. Temporarily returning to old behavior were thread
was killes without closing window and will reenable in later maints to
hopefully get a scenario or at least more data of what is causing the
freeze.
|
|
fallback fonts.
With the emojis support, a new font was added, which not only provides emojis
but also fancy colorful replacements for UTF-8 characters that used to be
supported by our fallback (monochrome) fonts: this causes discrepancies and
unwanted/undesired changes in scripted objects menus (e.g. an empty circle or
square may render as a black, full one, a heart may render red instead of white),
not to mention the larger font size used by the emoji characters...
This patch restores the aspect of such menus/dialogs/UI elements with UTF-8
characters that *are* supported by the usual fallback fonts (fonts which may
also vary from one viewer to another, and from one OS to another), so that
everything keeps working/rendering as it always did so far, while not impairing
the use of new colorful emojis.
This second proposal ensures that:
- "genuine" emojis (in the 0x1f000-0x1ffff range), will *always* be rendered
using the new emojis font (this solves, for example, the monochrome "yellow
faces" issue seen with some characters in my first proposal).
- Special UTF-8 characters (in the 0x2000-0x32FF range) which have been used by
scripters so far, will render as they used to, using the monochrome fallback
fonts (this repairs scripted dialogs menus).
- Remaining special characters, that do not have a corresponding glyph in the
monochrome font, but do have one in the emojis font, will use the latter font
to render.
It also got the nice side-effect of removing the dependency on the ICU4C library.
Note however that the recent commit:
https://github.com/secondlife/viewer/commit/326055ba82c22fedde186c6a56bafd4fe87e613a
will need to be reverted to allow this patch to actually fix scripted dialogs.
Also, some cleanup might be needed in skins/default/xui/*/emoji_characters.xml to
remove from it the special UTF-8 characters that will no longer be rendered with
fanciful colors, but instead with the monochrome font glyphs.
|
|
Under debug LL_ERRS will show a message as well, but release won't show
anything and will quit silently so show a notification when applicable.
|
|
# Conflicts:
# indra/llcommon/llstring.cpp
# indra/llcommon/llstring.h
|
|
|
|
# Conflicts:
# indra/newview/llinventorygallery.cpp
|
|
|
|
|
|
Note that crash happened when setting LLProgressView::setMessage
|
|
|
|
|
|
|
|
|
|
|
|
# Conflicts:
# indra/newview/llchiclet.h
|
|
|
|
UIImgInvisibleUUID doesn't exist
Default normal for material is 'null'
|
|
1. After window closes viewer still takes some time to shut down, so
added splash screen to not confuse users (and to see if something gets
stuck)
2. Having two identical mWindowHandle caused confusion for me, so I
split them. It looks like there might have been issues with thread being
stuck because thread's handle wasn't cleaned up.
3. Made region clean mCacheMap immediately instead of spending time
making copies on shutdown
|
|
|
|
a preset...' option of the 'Preferences' floater
|
|
|
|
|
|
coroutines).
|
|
|
|
|
|
Under debug LL_ERRS will show a message as well, but release won't show
anything and will quit silently so show a notification when applicable.
|
