On some optimization levels we get warnings/errors about it being
potentially uninitialized (but not at -O2, hence this wasn't a
problem in nightly builds.)
The documentation comment says that only "caches starting from the top
cache to at least the cache the page lives in" are guaranteed to be locked.
So it should be safe to unlock all the lower caches that we didn't find
the page in here, and leave only the topmost cache locked, to reduce
contention.
(We can't do this in the CoW case, since there we may have to unmap
the read-only page when mapping in the new read-write page. But for
clean pages, since no lower cache had the page in question, we shouldn't
have anything mapped at all, and thus unlocking the lower caches ought
to be safe.)
We now will find its image struct using the standard iteration,
so just check for its address specifically.
Fixes commpage symbols printing in KDL after the recent refactor.
We insert pages from all over the underlying disk's area,
so in order to not trip asserts, the cache must be that large too.
(It's a "null" cache, so it doesn't have any commitment.)
Fixes #19373.
It checks whether the page or page's data is present in an underlying
"backing store", not whether the page is present in the cache itself.
No functional change intended.
The previous logic, which dates back to the NewOS kernel, inserted them
into the deepest cache instead if the fault was a read and not a write.
But this creates problems if the deepest cache is of a smaller size than
the topmost cache, because then we'll be inserting pages with offsets
past the cache's virtual_end. In the case there are multiple levels of
caches between the topmost and the deepest cache which all are smaller
than the topmost, then when/if the deepest cache is merged with the second-
deepest cache, those pages will get skipped and freed when they may still
be in-use, which was the cause of #19367.
To solve this, just insert new pages into the topmost cache always.
Inserting them into the deepest cache seems to have been an optimization
(and a savings of pages), and not a matter of correctness.
Move ColorListView and ColorItem to shared in BPrivate namespace.
ColorItem typedef to BPrivate::BColorItem for apps that are already
using this class.
Gravity screensaver:
* ColorItem => BColorItem.
* Use make_color() to set colors.
Make color drop message name agnostic, check for B_RGB_COLOR_TYPE.
Add be:sender and source to drag message.
Add _SetTermColors() convenience method to set colors on all tabs.
Change-Id: I5a9f55d3ab423ccaa341997cf444603373024553
Reviewed-on: https://review.haiku-os.org/c/haiku/+/8846
Reviewed-by: John Scipione <jscipione@gmail.com>
Added AlderLake-P and AlderLake-N variants to agp_gart bus
and intel_extreme driver. Tested and confirmed working.
Change-Id: I1ac1e04fbba222a2f9eef483d68575665561e837
Reviewed-on: https://review.haiku-os.org/c/haiku/+/8840
Reviewed-by: waddlesplash <waddlesplash@gmail.com>
AdoptSystemColors() tints document background color to match panel
color if uneditable.
Calling MakeEditable() will automatically apply or unapply uneditable
background tint if you have previously called AdoptSystemColors().
Parent BView::AdoptSystemColors() sets panel colors, we want document
colors here. Do not alter text color - only view, low and high colors
are changed.
Document AdoptSystemColors() and MakeEditable() in BTextView docs.
Change-Id: Ib215735f27bb01fc2f95fcf2fee0185e5fc83f70
Reviewed-on: https://review.haiku-os.org/c/haiku/+/8263
Haiku-Format: Haiku-format Bot <no-reply+haikuformatbot@haiku-os.org>
Reviewed-by: waddlesplash <waddlesplash@gmail.com>
Tested-by: Commit checker robot <no-reply+buildbot@haiku-os.org>
Also the bullet point color
Allow Usage Conditions window to be resized
Tint disabled color instead of hardcoding
Copyright 2024 Haiku, Inc.
Fixes the other half of #19067
Change-Id: Ibf74d9dd673807b0f107e56cefb5b30916baed96
Reviewed-on: https://review.haiku-os.org/c/haiku/+/8262
Reviewed-by: Adrien Destugues <pulkomandy@pulkomandy.tk>
Haiku-Format: Haiku-format Bot <no-reply+haikuformatbot@haiku-os.org>
Tested-by: Commit checker robot <no-reply+buildbot@haiku-os.org>
* Use a "null" cache and avoid committing memory unnecessarily.
* Block the 64-bit ranges also. On x86_64 the kernel is above
userspace, so we have to |= KERNEL_BASE here; but in case
something actually tries to access those pointers directly
they should get a GPE anyway (as the address should be outside
the canonical range.) If it's not, then SMAP should catch it.
We don't have many etc and non-etc syscall pairs in this file
(the semaphores seem to be a notable exception), so drop
wait_for_thread and just call wait_for_thread_etc instead.
This breaks syscall ABI, but we've already broken it since beta5
anyway.
They're still mapped into the team address space and with user addresses,
of course, and we reset all their protections later on anyway. This allows
a number of arch_cpu_{enable|disable}_user_access() calls to be dropped.
Also adjust the name of ro/text segments to "rx" to match what
runtime_loader now does.
Tested with SMAP enabled, still works.
If it's not set (and we aren't on pre-Haiku GCC2 ABI), then don't make
the region executable.
This mostly only makes a difference for Clang/LLD-linked binaries
that have a readable, non-writable, non-executable program segment.
GCC/LD-linked binaries just have all read-only data in the main
read+execute section by default.
... and B_DOCUMENT_TEXT_COLOR with the selected hand using
B_NAVIGATION_BASE_COLOR when dragging instead of blue
(well, it is still blue but customizable).
Analog clock colors adjusted for light and dark modes.
Change-Id: I2d5af04f10090258b0ef989498f8e43dbded8def
Reviewed-on: https://review.haiku-os.org/c/haiku/+/7940
Tested-by: Commit checker robot <no-reply+buildbot@haiku-os.org>
Haiku-Format: Haiku-format Bot <no-reply+haikuformatbot@haiku-os.org>
Reviewed-by: waddlesplash <waddlesplash@gmail.com>
If we don't have an Entry to attach/detach from, then we don't need
to acquire any locks at all. Now that we return multiple dirents
from read_dir, it's much more likely that we wind up with
EntryIterators in such a state, so this avoids some lock contention.
Like S_IFIFO. It doesn't support ones with "sub-vnodes", but FIFOs
and UNIX sockets don't seem to use those (does anything, actually?)
Fixes #19261.
Change-Id: I9eeac4efcbe74e9084653c77d883dbb44e6f1d2c
Reviewed-on: https://review.haiku-os.org/c/haiku/+/8830
Reviewed-by: waddlesplash <waddlesplash@gmail.com>
