* Restore missing definitions of std::nothrow and mynothrow that are
required for the debug build.
* Additionally, cleanup function overrides provided by kernel_cpp,
such that any exceptions in kernel, bootloader or runtime_loader will
trigger a panic.
Same fix as hrev42151 for usb_serial:
When determining the bit to be set for RTS/DTR there was a missing comparison
to op which would've caused a wrong line state to be set.
* The given void pointers are pointers to the actual list items, which
are pointers themselves, so there was an indirection missing
* Fix inverted sort order
This could have caused spurious crashes related to initialization of
the locale kit, most notably this was responsible for triggering a
crashing bug in the plaintext add-on that caused the x86 image
not being able to boot.
I will continue with trying to find the actual crash, too.
The 160x120 preview size is a bad choice, because it makes XRoger crash
in an infinite recursion when trying to draw bezier arcs with custom
code (instead of using the BeAPI for that).
Use 120x90, just like BeOS, to maximise compatibility and avoid such
problems.
Fixes#10601.
* Use a BGroupLayout with insets to make space for the monitor drawing
* Don't put the "no preview available" inside the view passed to
screensavers, this could confuse them
* Have the actual preview area (not including borders) be 160x120
This reverts 0fb019bbf2. Making the window
a BDirectWindow isn't a good idea, especially if you want things other
than a direct rendering view in it. Moreover, this is completely
unneeded as GL works just as well in non-direct mode.
DirectDraw is called only in direct drawing mode (BDirectWindow). Our GL
kit also works fine (but slower) in indirect mode (standard BWindow, GL
stuff rendered offscreen then blitted). However, for this to work we
need the screensavers to actually do something. Use the
BScreenSaver::Draw method, which is called in either case.
This makes it possible to use indirect mode for the preview of GL
screensavers. The performance isn't really a problem on the very small
preview (160x200 pixels).
Fixes GLife and Gravity. Flurry still crashes Mesa.
- When walking through and querying volumes for a missing source file,
store the error and report it back in the completion message rather
than exiting the query worker. Otherwise, if something went wrong
with the queries, it'd potentially occur that we'd never prompt the
user to locate the file, as the UI would never be notified that
the background work was complete.
* Use integers for x and y coordinates in the draw loop, too
* Simplify _GetCell by using the return value of vterm_screen_get_cell
to detect out of bounds access, instead of testing for that manually.
* From now on, the gcc-specific system libraries (libgcc, libsupc++ and
libstdc++) are provided by separate packages built along with gcc:
- gcc_syslibs contains the shared libraries (libgcc_s.so, libsupc++.so and
libstdc++.so)
- gcc_syslibs_devel contains the static libraries and both c++ and gcc
headers
The shared libraries now make proper use of symbol versioning and there
are version-specific symlinks
* The buildsystem has been adjusted to no longer use the libraries and
headers from the cross-compiler, but use the ones provided by the
above-mentioned packages. The only exception is that the 32-bit libraries
required for the bootloader of the x86_64 architecture are still taken
from the cross-compiler.
* This stage builds the gcc packages to get the shared syslibs, which
only requires the Haiku glue code.
* Add separate declaration section for stage0 packages to
HaikuPortsCross repository files.
* For the bootstrap_stage0 platform, fall back to the gcc headers
provided by the cross-compiler.
* No longer apply somewhat crude mechanism for converting a package name
into a corresponding port name - haikuporter by now supports package
names directly.
This fixes a problem when trying to pass 'gcc_syslibs_devel' to
haikuporter, which only ever saw 'gcc_syslibs'.
