that can be used by file systems.
* Changed the way the file cache works: instead of reading/writing to the
underlying device directly, it can now be used for any data source, ie.
also network file systems.
* As a result, the former pages_io() moved to the VFS layer, and can now be
called by a file system via {read|write}_file_io_vec_pages() (naming
suggestions are always welcomed :-)). It now gets an FD, and uses that to
communicate with the device (via its fs_{read|write}_pages() hooks).
* The file_cache_{read|write}() functions must now be called without holding
an I/O relevant file system lock. That allows the file cache to prepare the
pages without colliding with the page writer, IOW the "mayBlock" flag can
go into the attic again (yay!).
* This also results in a much better performance when the system does I/O and
is low on memory, as the page writer can now finally write back some pages,
and that even without maxing out the CPU :)
* The API changes put slightly more burden on the fs_{read|write}_pages()
hooks, but in combination with the file_map it's still pretty straight
forward. It just will have to dispatch the call to the underlying device
directly, usually it will just call its fs_{read|write}_pages() hooks
via the above mentioned calls.
* Ported BFS and FAT to the new API, the latter has not been tested, though.
* Also ported the API changes to the fs_shell. I also completely removed its
file cache level page handling - the downside is that device access is no
longer cached (ie. depends on the host OS now), the upside is that the code
is greatly simplified.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22886 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Don't generate an uninitialize job, when there's the partition wasn't
initialized anyway (the syscall would fail in this case).
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22883 a95241bf-73f2-0310-859d-f6bbb57e9c96
system by pretty name (not only module name) now.
* _user_initialize_partition() loads the disk system by pretty name.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22880 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Rewrote _AllocateEndpoint
* More work in the constructor
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22873 a95241bf-73f2-0310-859d-f6bbb57e9c96
source file. Added function to check the volume name.
* Removed bfs_validate_initialize(). This functionality is to be implemented
in a userland add-on.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22871 a95241bf-73f2-0310-859d-f6bbb57e9c96
systems, and possibly enable ports 7 and 8 in the future.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22869 a95241bf-73f2-0310-859d-f6bbb57e9c96
The Jmicron AHCI controller has a mode that combines IDE and AHCI functionality
into a single PCI device at function 0. This happens when the controller is set
in the BIOS to "basic" or "IDE" mode (but not in "RAID" or "AHCI" mode).
To avoid needing two drivers to handle a single PCI device, we switch to the
multifunction (split device) AHCI mode. This will set PCI device at function 0
to AHCI, and PCI device at function 1 to IDE controller.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22868 a95241bf-73f2-0310-859d-f6bbb57e9c96
did not propagate the "dontWait" argument, letting the page writer never wait
for its pages.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22860 a95241bf-73f2-0310-859d-f6bbb57e9c96
and write_to_cache() before, IOW the cache could reserve too few pages.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22859 a95241bf-73f2-0310-859d-f6bbb57e9c96
needed pages correctly, and would also not read/write enough in case the
offset didn't start at 0 resulting in undetected short reads/writes. It's
amazing how many bugs can be hidden in a few lines of code.
* Fixed a bug that might have been the cause for bug #1601: when the last part
of the write did not end on a page boundary, the last page had to be read
first, but that was done from the wrong offset. Also, if only parts of that
page could be read (because the file size didn't span over the whole page)
the remaining parts needed to be cleared.
* The cache_funcs were always called with the same value for numBytes and
bufferSize so I've eliminated the former.
* Large reads now also bypass the cache in case of low memory, large writes now
also only bypass the cache in that case, following Ingo's suggestion.
* Fixed compilation with debugging turned on.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22858 a95241bf-73f2-0310-859d-f6bbb57e9c96
than 64KB. Reads should probably get a similar logic, at least if memory is
tight.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22857 a95241bf-73f2-0310-859d-f6bbb57e9c96
* add partition creation menu items
* filter filesystems by their support for "initializing"
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22855 a95241bf-73f2-0310-859d-f6bbb57e9c96
the same pages over and over.
* Increased the priority of the page writer a bit, so that it is higher than
the one of the page daemon.
* Added a sModifiedTemporaryPages counter to let the page_writer decide how
many pages are there to write back (temporary pages would go to the swap file
and are only written back when memory is low).
* In case there are more than 1024 modified (non-temporary) pages around, the
page writer will constantly write out pages, not only when being pushed.
* The page writer now temporarily always leave out temporary pages (as long as
we don't have a swap file).
* Shuffled functions around a bit.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22852 a95241bf-73f2-0310-859d-f6bbb57e9c96
never worked correctly for any case which was very visible in Tracker (and
especially so if you had "sorting apps" turned on).
* Removed superfluous white space at the end of lines.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22850 a95241bf-73f2-0310-859d-f6bbb57e9c96
think it wasn't a really good idea. The alignment of shortcut
charachters and submenu symbols could use some more work, though.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22849 a95241bf-73f2-0310-859d-f6bbb57e9c96
errno to be set to a positive value on error, but since BeOS error codes are
negative numbers, we can't comply. Changed the tests accordingly to check for
the expected error code (EINVAL) instead.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22845 a95241bf-73f2-0310-859d-f6bbb57e9c96
actually forgot to commit the changes I made to the ide_adapter...
* the IDE bus master command/status stuff is now used via flags; it's no
bitfield anymore.
* Changed a few constants to upper case.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22842 a95241bf-73f2-0310-859d-f6bbb57e9c96
on pages are recorded in a history and can be printed via the
"page_allocations" command.
* Fixed a problem in the page scrubber. It temporarily removed pages
from the free list, which could have been reserved by someone else.
When actually allocating the reserved pages, that someone could find
free and clear lists empty and would therefore rightfully panic
("Had reserved page, but there is none!").
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22841 a95241bf-73f2-0310-859d-f6bbb57e9c96
ideally, we would only have one file and no copies - later.
* That also changed how READ/WRITE_6 commands are translated to READ/WRITE_10
commands, but it was obviously correct before, at least I can still reproduce
bug #1577 (so it has to be something else).
* Spotted a few suspicious uses of transform_6_10() vs. transform_cmd_6_10();
at least the FreeBSD driver seem to work differently here, anyway (it always
uses 12 byte commands for most of these).
* Cleaned up the file.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22838 a95241bf-73f2-0310-859d-f6bbb57e9c96
that same nick, not a different one)
* performance improvement in Draw() (using ItemFrame() in the inner loop scales
really bad)
* fixed scroll range for when BListItem::Height() is non-integer (scrolling
all the way to the bottom in Visions nick list works now)
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22835 a95241bf-73f2-0310-859d-f6bbb57e9c96
non-reserve case.
* vm_page_allocate_page() could also lose a stolen page in case more free pages were available
after steal_pages() was called.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22834 a95241bf-73f2-0310-859d-f6bbb57e9c96
K{Disk,File,Partitioning}System writing methods. It is now required
that the caller has marked the concerned partitions busy, hence we can
(read-)access them without needing a lock. The module interfaces will
will be changed to take advantage of the fact as well. The methods take a
disk_job_id instead of a KDiskDeviceJob* now, though I haven't quite
decided, whether we need it at all or just want to add a special
handling in the cases where notifications during the operation make
sense.
* Reimplemented the disk device write support syscalls (save
_user_move_partition() for which other module hooks are needed). They
call the KDiskSystem methods, now.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22832 a95241bf-73f2-0310-859d-f6bbb57e9c96
be initialized when probe(0 is called. The driver now loads and works for a
while until it hangs. Looking into it.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22826 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Made output much more readable and useful (especially with 80 characters).
* Removed useless -u option when built for Haiku.
* Added a new option -t that shows the tunables.
* Removed the hack to let be_app run - it works fine without on BeOS.
* Improved parameter parsing (now uses getopt_long()), also added long names
for all options.
* I'm all for removing BeOS/Dano support completely from this app.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22825 a95241bf-73f2-0310-859d-f6bbb57e9c96
device's attach() method twice - this would let every network device that found a MII PHY fall into
an endless loop on start. In other words 3com and VIA Rhine should now work fine.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22815 a95241bf-73f2-0310-859d-f6bbb57e9c96
reason the MII bus did not work as mediainit was never called.
* Got rid of the delayed MII bus scanning again; not only did it not solve the problem, it was
also completely unnecessary.
* Minor cleanup.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22812 a95241bf-73f2-0310-859d-f6bbb57e9c96
probing/attaching the MII bus should come later, though.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22806 a95241bf-73f2-0310-859d-f6bbb57e9c96
pmap_kextract() not vice versa, just like it is done in FreeBSD itself.
* Added missing contigmalloc() definition for kernel drivers.
* Fixed the "*_devclass defined but not used" warning.
* Fixed warnings.
* Added/adjusted (to our style guide) the license headers for the files I
touched.
* Minor cleanup.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22802 a95241bf-73f2-0310-859d-f6bbb57e9c96
synchronous scanning is supported.
* Removed the disk device job support from the disk device manager.
* K{Disk,File,Partitioning}System:
- Remove querying and validation methods.
- Commented out the modification methods until their fate is decided.
* Removed obsolete _user_get_partitionable_spaces().
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22799 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Added sigignore() tests.
* Renamed names of the test executables to avoid clashes in the build system.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22792 a95241bf-73f2-0310-859d-f6bbb57e9c96
* fixed +/- 1 bug in _BTextGapBuffer_::GetString()
* used the correct text and offsets in BTextView whenever the visible
text is to be used
* when copying to the clipboard, copy the bullets
* when dragging the text, drag the bullets
TODO:
* test more with UTF8 chars in the original text (I am unsure if fSelStart
and so on is really in bytes rather than glyphs)
* test with multiple font styles
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22787 a95241bf-73f2-0310-859d-f6bbb57e9c96
does not drop you into the debugger on R5. Should fix#1598.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22784 a95241bf-73f2-0310-859d-f6bbb57e9c96
- use movem
- fix return value: the convention is to return in d0:d1 (MSB:LSB)
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22783 a95241bf-73f2-0310-859d-f6bbb57e9c96
(_kern_{supports,validate}_*(), etc.).
* Adjusted the prototypes of the disk device modification syscalls.
Commented out their implementations for the time -- they'll mostly
have to be rewritten completely.
* Implemented the userland disk device jobs.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22781 a95241bf-73f2-0310-859d-f6bbb57e9c96
acceleration works fine, but overlay doesn't - that's next on my list.
Turns out the i965 differentiates between RGB-32 and RGB-32-alpha, and
didn't like trying to use the latter as display mode (the i865 didn't
care at all)... finding that took me *way* too long, though.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22780 a95241bf-73f2-0310-859d-f6bbb57e9c96
the timing of the original mode into account (I haven't tested if the computed
score makes that much sense, though).
* GetPreferredMode() now also works correctly for EDID data): it now returns
the correct status, and also constructs the mode correctly. Furthermore, it
will now search for the mode in the mode list, instead of using that mode
directly (so that it can take hardware limits into account).
* Minor cleanup.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22778 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Turned off debug output for interrupts.
* Temporarily turned on IFF_PROMISC mode so that the driver can at least be used a bit.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22775 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Calmed down the FreeBSD compatibility layer considerably.
* Now also sets the arpcom's ac_ifp field for completeness.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22772 a95241bf-73f2-0310-859d-f6bbb57e9c96
turning off interrupts on the device doesn't seem to work, and when one only acknowledges the
interrupts, they are also removed from XL_STATUS register.
The driver now starts to work: it can send and receive packets, though it currently only seem
to receive broadcast packets.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22771 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Fixed device_set_driver(): it needs to clear the allocated buffer (it's
allocated with M_ZERO in FreeBSD).
* The first mbuf buffer needs to have the M_PKTHDR flag set.
* With these changes, the 3com driver now fails where it should have failed
before, that is in the missing interrupt handling.
* Minor cleanup.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22769 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Added missing job classes (they don't do anything yet, though) and
completed the implementation of the job generator.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22767 a95241bf-73f2-0310-859d-f6bbb57e9c96
to be expanded for other types, but for now, we'll probably only get IFT_ETHER
anyway.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22766 a95241bf-73f2-0310-859d-f6bbb57e9c96
* reformat long lines so that they don't exceed 80 chars
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22751 a95241bf-73f2-0310-859d-f6bbb57e9c96
I started this from the amiga code but it's closer to the DOS layout.
TODO: rewrite from the DOS code to support extended partitions.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22749 a95241bf-73f2-0310-859d-f6bbb57e9c96
Lure the preprocessor to not replace #n by "123" for syscalls...
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22739 a95241bf-73f2-0310-859d-f6bbb57e9c96
this information in the fValidClipping flag... which should prevent
any drawing code to run if Painter is not attached properly
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22736 a95241bf-73f2-0310-859d-f6bbb57e9c96
sequential access.
* This is now used to let the cache free some pages (or schedule them to be
written back) before vm_reserve_pages() is called, but only if that cache
is not mapped, and there is a memory shortage.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22731 a95241bf-73f2-0310-859d-f6bbb57e9c96
doing something. If the signal is not valid, set errno and return -1.
Mini style changes by myself.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22728 a95241bf-73f2-0310-859d-f6bbb57e9c96
for this reason, we don't add such a node in the list of nodes to be scanned
this fixes the bug #1578
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22726 a95241bf-73f2-0310-859d-f6bbb57e9c96
it can happen that a view's background region is outside the
framebuffer... for some weird reason, it doesn't get triggered
though.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@22725 a95241bf-73f2-0310-859d-f6bbb57e9c96
