Development

This is hrev49245.

The Network preference panel should include the last remaining settings files from /system/settings/network: hostname and hosts. Not sure if both should go into the "DNS" view, as I'm not that versed in all things netwok related...
Maybe, since the hosts entries could become quite a list for some, that should go into its own add-on/view.

fixed:

Fixed with hrev49245.

* Remove the BBox. * Don't resize buttons when horizontally resizing the window. * Compute the server address field width correctly by taking the label into account. * Remove the BBox. * Don't resize buttons when horizontally resizing the window. * Compute the server address field width correctly by taking the label into account.

264 MB

265 MB

266 MB

fixed:

Fixed in hrev49244.

* There was an infinite exchange of messages between the main window and the BackgroundView. * Fixes #11996. * There was an infinite exchange of messages between the main window and the BackgroundView. * Fixes #11996.

Otherwise the child process's main thread will inherit the thread id of its parent and pthread operations will be directed at the wrong thread. Otherwise the child process's main thread will inherit the thread id of its parent and pthread operations will be directed at the wrong thread.

In order to be able to write the MBR boot code when creating an Intel partition map on non-x86 Haiku installations, we first need to rewrite it to Intel syntax so that we can use NASM to compile it, as the gcc assembler won't understand x86 assembly for non-x86 builds (e.g. an ARM build).

The MBR boot code was previously written in Intel syntax and compiled with NASM; however, this was changed in hrev49240 in order to implement #4028.

Logs attached. Apparently something in ICU didn't build correctly.

fixed:

Implemented in hrev49242!

invalid:

Reported at the wrong place, sorry, it won't do.

* with #4028 implemented, we can now use makebootable on GPT formatted disks as well. * with #4028 implemented, we can now use makebootable on GPT formatted disks as well.

Just tried to build ncurses for x86 and failed due to this. It was fixed for x86_gcc2 in ticket #11097 but the others weren't changed.

duplicate:

Closing as a dupe of #12083 which has a reproducible testcase

duplicate:

Most likely a dupe of #3011.

* Appears $(TARGET_ARCH) is the correct variable to use here. * Appears $(TARGET_ARCH) is the correct variable to use here.

* There is no need for the Intel partitioning system to write MBR bootcode on non-x86 systems * The MBR_HEADER define is used to prevent Jam from making MBR.h a dependency; the dependency is then specified explicitly when building for x86/x86_64 platforms * There is no need for the Intel partitioning system to write MBR bootcode on non-x86 systems * The MBR_HEADER define is used to prevent Jam from making MBR.h a dependency; the dependency is then specified explicitly when building for x86/x86_64 platforms

This is the wrong place to report but since I can't access haikuports site with either WebPositive or BeZilla it will have to do. (Ref: ticket #12013)

A few days ago I updated haikuporter and haikuports git repositories and a major change occurred. All packages built with the _x86 extension (eg GCC4 packages) were moved to haikuports/packages/.obsolete and I can no longer build for this target.

This is severely hampering development of some fairly complex stuff I'm working on, so I'd like the change reverted or a REALLY good reason why it was changed. Thanks.