This is a temporary solution, until moving to a global kernel export data struct
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@9819 a95241bf-73f2-0310-859d-f6bbb57e9c96
will do this if this function is called by anyone other than the root
user.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@9804 a95241bf-73f2-0310-859d-f6bbb57e9c96
for temporary memory anymore, it's the store's responsibility to do that
correctly now, and that functionality is reached via the vm_cache using
vm_cache_set_minimal_commitment().
Therefore, the vm_store commit() function now returns a status instead of
the size that could be commited.
Replaced the max_commit mechanism with one that cares about the available
memory, stores can reserve and unreserve such memory. The anonymous_commit()
will now fail in case it could not reserve the needed amount of memory.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@9777 a95241bf-73f2-0310-859d-f6bbb57e9c96
off to enable bright background colors. It's now always done which enabled
me to improve the slider colors a bit.
Therefore, console_set_color() no longer clears bit 8 from the background color.
Now no longer tries to change the palette in non-8-bit modes; that fixes
the strange colors I got in high and true color modes (VESA palette changing
would fail and the fallback VGA palette changing was responsible for the effect).
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@9772 a95241bf-73f2-0310-859d-f6bbb57e9c96
with an empty queue, it crashed. sem_info::latest_holder is obviously not
set correctly, and should be fixed.
Some cosmetics.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@9771 a95241bf-73f2-0310-859d-f6bbb57e9c96
serial_init() is now called from start().
Changed the way the serial debug output is served: instead of having
the console node to dump everything to serial, too, only dprintf()
triggers serial output now.
dprintf() is now silent unless in debug mode; serial output could
be enabled separetely, though (currently only at build time).
There is no need to disable serial output while the menu is running.
Removed unnecessary grist from the Jamfile.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@9756 a95241bf-73f2-0310-859d-f6bbb57e9c96
some stuff blinked and had wrong colors), we now have a new color scheme
in the boot loader. Not perfect, but looks okay and the same on Bochs and
real machines.
Made changing the colors again easier by moving most colors to constants.
We now have 3 lines for help texts.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@9754 a95241bf-73f2-0310-859d-f6bbb57e9c96
Now clears the screen and scrolls using the current colors (which allows
for other background colors than black).
console_set_color() no longer allows to set the 8 bit of the background
color which indicates blinking.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@9753 a95241bf-73f2-0310-859d-f6bbb57e9c96
Added comments about which colors can be used for what (unlike Bochs,
real systems support blinking modes, and therefore only 7 bits for
the background color).
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@9752 a95241bf-73f2-0310-859d-f6bbb57e9c96
or I am messing things up (it will now use good old VGA palette programming).
This fixes the strange colors on my system, should investigate on how to
properly use the VESA stuff here.
Debug output now goes through dprintf().
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@9751 a95241bf-73f2-0310-859d-f6bbb57e9c96
Now correctly removes the partitioning system from a device when it is no longer
needed. Also resets the parent pointer of its children.
Debug output now goes through dprintf() rather than printf().
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@9748 a95241bf-73f2-0310-859d-f6bbb57e9c96
the partition was deleted, but never removed from the partition list.
No longer prints out the root directory contents of all file systems.
Debug output now uses dprintf() rather than printf().
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@9747 a95241bf-73f2-0310-859d-f6bbb57e9c96
in the wrong direction for some time now, corrupting the argument area (user
apps would crash).
B_SYSTEM_TEAM is now reserved, so that no team can be created there.
Fixed some warnings when debug output is enabled.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@9719 a95241bf-73f2-0310-859d-f6bbb57e9c96
Renamed console_reader() function to keyboard_reader().
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@9718 a95241bf-73f2-0310-859d-f6bbb57e9c96
The inode now keeps track of the number of readers/writers.
Inode::BytesInChain() is now only used protected by the request lock.
Implemented some missing POSIX demands:
- when the last writer is closed, all waiting read requests are aborted
- pipefs_write() now returns EPIPE and signals the current thread with
SIGPIPE in case there are no readers left
- pipefs_read() will now return 0 in case there are no writers and no
unread buffers left in the pipe.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@9717 a95241bf-73f2-0310-859d-f6bbb57e9c96
Correctly implemented devfs_create() - it will now open the an already
existing file if O_EXCL is not set, and no longer fail with EROFS.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@9716 a95241bf-73f2-0310-859d-f6bbb57e9c96
divided into several parts, it could happen that overwrite the whole data
portion with data beyond the part that should have been read.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@9706 a95241bf-73f2-0310-859d-f6bbb57e9c96
was running. Thanks a lot to Jerome Duval who found this!
Since TTY ECHO mode doesn't look too well with the old shell, it will directly
start "sh" now, which should be the bash.
No longer prints out the TTY used when started, but now sets (as usually
done in BeOS) the TTY environment variable.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@9697 a95241bf-73f2-0310-859d-f6bbb57e9c96
Removed the fs_initialize_volume() function, as it's probably not going
to be needed.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@9695 a95241bf-73f2-0310-859d-f6bbb57e9c96
absolutely no reason to move them to B_SYSTEM_TEAM (additionally, during
startup, set_sem_owner() may crash).
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@9693 a95241bf-73f2-0310-859d-f6bbb57e9c96
Changed the last argument of _kern_mount() to be a string rather than a void pointer.
Greatly reduced the stack usage of _user_mount(); it now uses KPath instead
of on stack paths. It now also copies the parameter argument on the heap.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@9692 a95241bf-73f2-0310-859d-f6bbb57e9c96
devfs_read_link() and rootfs_read_link() could write beyond the buffer
size passed in (off by one).
The comment for _kern_read_link() didn't fit to what the function really
did.
common_read_link() no longer null terminates the link - it's the file
system's responsibility to do that.
fs_read_link() is not supposed to return the length of the link anymore,
but only B_OK for success.
Note, we deviate slightly from POSIX here, where even a buffer too small
would be filled, and no terminating null byte has to be written at all.
We always return an error in case the buffer is too small, and the link
is not partially copied into the buffer.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@9652 a95241bf-73f2-0310-859d-f6bbb57e9c96
Made the FS internal API a bit more consistent.
Removed devfs_unlink(), and devfs_rename() (which was just returning EROFS
anyway) - the devfs now appears completely read-only to the user. All
changes are triggered by kernel internal APIs (most usually through either
the devfs itself (old style drivers), the disk_device_manager, or the
standard device_manager).
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@9645 a95241bf-73f2-0310-859d-f6bbb57e9c96
Also implemented the now needed devfs_read_link() function.
git-svn-id: file:///srv/svn/repos/haiku/trunk/current@9644 a95241bf-73f2-0310-859d-f6bbb57e9c96
