before trying to write to the frame buffer.
* QEMU can now be started without a standard vga output (-vga none).
Change-Id: Id46cd4d70ce16e2156e0b0668fb88f09112067a0
Reviewed-on: https://review.haiku-os.org/c/haiku/+/7667
Tested-by: Commit checker robot <[email protected]>
Reviewed-by: waddlesplash <[email protected]>
* like sigqueue() but for threads. was added recently to FreeBSD, long before to glibc.
* also include features.h in gnu/pthread.h
Change-Id: I73bca666e2c67d7ff05f341bf85d71df9ccccbe5
Reviewed-on: https://review.haiku-os.org/c/haiku/+/7659
Reviewed-by: waddlesplash <[email protected]>
Tested-by: Commit checker robot <[email protected]>
DT_RUNPATH is generated by the linker instead of DT_RPATH when using --enable-new-dtags
It seems to be the default when using ld.lld
Normally one difference is LD_LIBRARY_PATH is checked before DT_RUNPATH (and not with DT_RPATH),
but we don't check LD_LIBRARY_PATH at the moment. Checking LIBRARY_PATH isn't an option, because
runtime_loader dosesn't use default paths, test suites would define LIBRARY_PATH empty.
Tested with tcpdump build_matrix.sh script, with clang 17, local libpcap, which requires
--disable-new-dtags on r1beta4.
Change-Id: Iacccde8d20e25ad14c5c548dd8832ea32b67e228
Reviewed-on: https://review.haiku-os.org/c/haiku/+/7539
Reviewed-by: waddlesplash <[email protected]>
Tested-by: Commit checker robot <[email protected]>
BFont::LoadFont can now use an index (0-based) and an optional named-instance (1-based)
to select the font variant from a file.
BFont::LoadFont can also use an index when loading from memory.
Change-Id: I0ce3eb6cc77d32cf43847416561eafe3063ca693
Reviewed-on: https://review.haiku-os.org/c/haiku/+/7402
Reviewed-by: Jérôme Duval <[email protected]>
Reviewed-by: Máximo Castañeda <[email protected]>
Tested-by: Commit checker robot <[email protected]>
* our ip modules don't support connect()/send(). Just use sendto().
* ping6 disappears, ping supports -4 or -6 to force IPv4 or IPv6
Change-Id: I1e982e354cc75d3a314c5bbbfffa0373e8f4d9af
Reviewed-on: https://review.haiku-os.org/c/haiku/+/7427
Reviewed-by: waddlesplash <[email protected]>
by default, AdapterIO is initialized with a BMallocIO object, which will
be extended indefinitely. Flushing regularly is necessary to avoid
excessive memory usage. Tested in StreamRadio.
Change-Id: I9f3142c0a2300ad44dc54ccf6932d41c9526320b
Reviewed-on: https://review.haiku-os.org/c/haiku/+/7302
Reviewed-by: Adrien Destugues <[email protected]>
Tested-by: Commit checker robot <[email protected]>
some laptops decide to not charge a fast full battery to not consume a cycle.
In this discharging state, the current rate is zero, thus no time left can be computed.
We add a state "Not charging" to clearly differentiate and avoid user confusion.
PowerStatus was triggering low battery notifications, this is also fixed.
Change-Id: I4745c78eb0863ab01fe34cb065707d068cff0f0e
Reviewed-on: https://review.haiku-os.org/c/haiku/+/7300
Reviewed-by: Fredrik Holmqvist <[email protected]>
Tested-by: Commit checker robot <[email protected]>
means B_DEVICE_TYPE, B_DEVICE_SUB_TYPE, B_DEVICE_INTERFACE.
_CLS is rare, it means I don't own devices with this attribute.
from the spec: _CLS:
Class Code – supplies OSPM with the PCI-defined class, subclass and programming interface for a device. Optional.
Change-Id: I4f7b7ed66cbe6b4ff4511cb13df2af218350a5d8
Reviewed-on: https://review.haiku-os.org/c/haiku/+/7210
Tested-by: Commit checker robot <[email protected]>
Reviewed-by: Jérôme Duval <[email protected]>
* context->extradata seems to be set to fExtraData or fInputFormat.MetaData().
Both are managed by us, so set context->extradata to NULL before freeing.
* avcodec_close() shouldn't use according to the documentation:
"Use avcodec_free_context() to destroy a codec context (either open or closed).
Opening and closing a codec context multiple times is not supported anymore – use multiple codec contexts instead."
* fix#18713
Change-Id: I820deefcffea52a39fcb7587153d40dc03c85024
Reviewed-on: https://review.haiku-os.org/c/haiku/+/7208
Reviewed-by: waddlesplash <[email protected]>
Tested-by: Commit checker robot <[email protected]>
after a0131eaae2884fdced27158c3d34732d1656aca9 mxcsr was possibly also incorrect.
fpu control and mxcsr will be restored with fxrstor/xrstor.
no need to clear pending exceptions on #MF
fix#18656 (and #18624 after reverting).
Change-Id: I7dd5e2e4610747c5b82abd6c67e302d264b4be92
Reviewed-on: https://review.haiku-os.org/c/haiku/+/7104
Tested-by: Commit checker robot <[email protected]>
Reviewed-by: waddlesplash <[email protected]>
possibly help with #15789:
I was testing virtual_block devices with dd, and noticed it didn't work with higher block sizes.
Indeed the driver was passing a list of single pages, very inefficient. To be remarked is that
the physical pages were reversed in memory, causing get_memory_map to provide a map of single pages.
What happens: dd fills up the buffer from 0 until the block size, causing soft faults for still unmapped
pages. Each fault maps a cleared physical page out of the clear page queue. This means that some page runs
are actually put reversed in the clear page queue, which this commit fixes.
Change-Id: I36395849c7f13086b44a1b284cc0c858d8c29f75
Reviewed-on: https://review.haiku-os.org/c/haiku/+/7081
Tested-by: Commit checker robot <[email protected]>
Reviewed-by: waddlesplash <[email protected]>
Reviewed-by: Axel Dörfler <[email protected]>
* the target inode would still exist, unreferenced.
* also remove from entry_cache.
* this fixes fsck complaining about unreferenced inodes.
Change-Id: I4d48d6bf89bdd37d366da0ae4795e42bcc0532d9
Reviewed-on: https://review.haiku-os.org/c/haiku/+/7078
Tested-by: Commit checker robot <[email protected]>
Reviewed-by: waddlesplash <[email protected]>
* the problem appears after the directory has been splitted.
* this fixes fsck complaining about an invalid directory entry.
Change-Id: I1566a279fc6438cb735439be002c634882e5a83b
Reviewed-on: https://review.haiku-os.org/c/haiku/+/7077
Tested-by: Commit checker robot <[email protected]>
Reviewed-by: waddlesplash <[email protected]>
on Pentium Silver, loading the microcode on the boot cpu also updates the other ones.
Change-Id: Ifbd767e7d73fdbc8ae2bf0740fcce523e500de1b
Reviewed-on: https://review.haiku-os.org/c/haiku/+/6806
Reviewed-by: waddlesplash <[email protected]>
registering new devices requires the device_manager lock. The explore thread would wait
that the initial device scan is over to register devices.
Change-Id: I46529ab0926f349023f06ada3411979c0950dfcf
Reviewed-on: https://review.haiku-os.org/c/haiku/+/6475
Tested-by: Commit checker robot <[email protected]>
Reviewed-by: Adrien Destugues <[email protected]>
Reviewed-by: waddlesplash <[email protected]>
ReadData() should return the current error on closed state, not always not connected.
Change-Id: I286ac1dd9ded127e8658ceb61088783b9993eacf
Reviewed-on: https://review.haiku-os.org/c/haiku/+/6459
Reviewed-by: Rene Gollent <[email protected]>
* FIONREAD shouldn't report errors, only EINVAL when listening
* read available data in closing and closing-wait states
* fix#18327
Change-Id: Idd53a043a72ef6c7b282ea669895ff2e947adbb4
Reviewed-on: https://review.haiku-os.org/c/haiku/+/6420
Reviewed-by: waddlesplash <[email protected]>
'tail' selects read events on its write-only file descriptors,
so let's use B_SELECT_READ only for read-write file descriptors.
Change-Id: I5bb3837f82a7b06ec367fe871c15d67beb2d412c
Reviewed-on: https://review.haiku-os.org/c/haiku/+/6417
Reviewed-by: waddlesplash <[email protected]>
* the packet capture expects frames, so loopback_frame should actually apply framing and deframing
for loopback packets.
* datalink: set the address for host routes
* device_interfaces: call the deframing if any in device_enqueue_buffer()
tested with ping 127.0.0.1 and ping6 ::1, while removing the local flag on both routes.
Change-Id: I2085735bdac3bb85908189a2e1acb2540818c7bd
Reviewed-on: https://review.haiku-os.org/c/haiku/+/6451
Reviewed-by: Fredrik Holmqvist <[email protected]>
Tested-by: Commit checker robot <[email protected]>
Reviewed-by: waddlesplash <[email protected]>
tcp_segment_header.advertised_window is 16 bits.
Previously, instead of using the maximum window, zero would be sent, thus
the partner wouldn't send anything.
fix#18337
Change-Id: Ibff98ee58b84bdf52527a7821648a5faf20c5589
Reviewed-on: https://review.haiku-os.org/c/haiku/+/6359
Tested-by: Commit checker robot <[email protected]>
Reviewed-by: Adrien Destugues <[email protected]>
Some laptops like G713QM-HX121 or FX705DY have an ACPI WMI DSL, which evaluates five arguments
for all method IDs similar to
https://github.com/torvalds/linux/commit/98e865a522983f2afde075648ec9d15ea4bb9194
but with 5 arguments:
Method (WMNB, 3, Serialized)
{
P8XH (Zero, 0x11)
CreateDWordField (Arg2, Zero, IIA0)
CreateDWordField (Arg2, 0x04, IIA1)
CreateDWordField (Arg2, 0x08, IIA2)
CreateDWordField (Arg2, 0x0C, IIA3)
CreateDWordField (Arg2, 0x10, IIA4)
Local0 = (Arg1 & 0xFFFFFFFF)
Patch to fix proposed by Reijo F from russian haiku os chat on #18189
Change-Id: I4536029600e6e9608aceafa45b36af8f0ffe35b3
Reviewed-on: https://review.haiku-os.org/c/haiku/+/5977
Reviewed-by: Jérôme Duval <[email protected]>
Reviewed-by: Fredrik Holmqvist <[email protected]>
* aborted transfers will release the notify semaphore when the cancel is notified.
* the allocated buffer would be freed on return, while the usb stack eventually copied
data in the buffer in our back, leading to KDL crashes, because the freed buffer would
be right reallocated for some kernel team structures.
* regression introduced by hrev55806, the transfers didn't need to be cancelled before.
Change-Id: Ifb6e941f71d05c37c36f878059c33883bb72a67c
Reviewed-on: https://review.haiku-os.org/c/haiku/+/5905
Reviewed-by: Adrien Destugues <[email protected]>
Reviewed-by: waddlesplash <[email protected]>
Tested-by: Commit checker robot <[email protected]>
we detect basically the cpu info before loading the microcode,
to be able to detect the vendor, and avoid any update on hypervisor.
I couldn't test because my cpu doesn't have any update available.
Change-Id: I6aea830158423b3ee13b640be8a788fc9041e23c
Reviewed-on: https://review.haiku-os.org/c/haiku/+/5859
Tested-by: Commit checker robot <[email protected]>
Reviewed-by: Jérôme Duval <[email protected]>
the entropy source is read every second and pushed to the PRNG.
the PCI device is tested, not the ACPI.
Change-Id: I9bb6b21c7189b28a1d8a624d83b33ff6682152dc
Reviewed-on: https://review.haiku-os.org/c/haiku/+/5825
Tested-by: Commit checker robot <[email protected]>
Reviewed-by: Jérôme Duval <[email protected]>
Jérôme Duval