The elfedit tool doesn't exist with the old binutils, so configure was
failing when it tried to get the path to it. Only try to search for it
if building GCC 4.
This reduces perceived latency. One day when I add a tab close undo stack, this
will be even better. But generally using the middle mouse button is not
accidental.
GCC 2 built for OS X 10.7/10.8 was broken, the Haiku build would fail
with some strange errors. Forcing compilation of GCC 2 in 32-bit mode
results in a working GCC.
The USER_BASE_ANY definition exists to specify where to start searching
for B_ANY_ADDRESS allocations, but this was not being used correctly.
On x86_64, this was causing the runtime loader's heap to be allocated
at address 0 so NULL pointer accesses were not getting caught.
- When loading an image, we now check if it has a .gnu_debuglink section.
If so, we try to load the linked file for our symbolic debugging information.
The file is searched for with the following rules:
1) If self-contained app, app image is looked for in ./debug while
library/add-on images are looked for in ../debug/.
2) If the image in question is in one of the global locations
(/boot/{common,system,home/config}/{lib,add-ons}), we search
in develop/debug for the corresponding global dir.
Implements #8135.
The cookie is used to store the base address of the area that was just
visited. On 64-bit systems, int32 is not sufficient. Therefore, changed
to ssize_t which retains compatibility on x86 while expanding to a
sufficient size on x86_64.
Refactor the icon scaling code in IconUtils.cpp to avoid code
duplication. Basically create and delete the temp bitmap to
convert from B_CMAP8 to B_RGBA32 for scale2x/scale3x/scale4x
just one time instead of 3.
There was an off-by-one error in Deskbar which was causing
it to scale up the 16x16 Bitmap icon to 32x32 instead of just
using the 32x32 icon. This only affected BeOS bitmap-based
icons, not Haiku HVIF icons.
* The input loop can now wait on abstract events, which other threads
(or even the input loop thread itself) can signal.
* Use the new mechanism in QuitSession().
* Also (with the exception of the SIGINT part) implement
WaitForThreadOrUser().
* Added x86_64 linker script and relocation code.
* Some 64-bit safety fixes to the heap code.
* Added runtime_loader, libroot and bash to the x86_64 image. The boot
script will be launched, but will panic shortly after because fork
is broken.
These have been pulled from glibc 2.10, the last version before a
bunch of changes were made that would have made porting more difficult.
None of this has been tested yet, it is not currently possible to do
so: I'm just trying to get libroot compiling so that I can work on the
runtime loader. I will test them when I am able to.
* Move the libedit interface there and provide nicer to use methods.
* Also start adding utility methods for the input loop. It is going to
manage all interactions of the input loop with outside events.
* Fix the "quit" command. The user is now prompted what to do with the
debugged team and the input loop thread avoids reentering the input
loop.
* UserInterface::SynchronouslyAskUser() is now allowed to return -1 to
indicate that the user cannot be asked at this point for whatever
reason. The caller needs to handle that case.
* UserInterfaceListener::UserInterfaceQuitRequested(): Add new parameter
"quitOption" to specify what is supposed to happen. The previous
behavior (ask user) is only one of the options. The others are to kill
the debugged team or to resume it.
