* Separated the other stuff previously done in debug_init_post_vm() to the new
debug_init_post_settings().
* Removed superfluous status_t return codes - they are ignored, anyway, and if
there really is a show stopper in the init process, panicking would be the
thing one should do.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@36623 a95241bf-73f2-0310-859d-f6bbb57e9c96
improvements like a new Fullscreen mode, context menu's for downloads,
persistent storage of credentials...
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@36621 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Don't traverse symlinks. Besides that this is not how the BeOS
implementation behaves it could also lead to undesired recursion in case of
symlinks to directories.
* Only process regular files.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@36613 a95241bf-73f2-0310-859d-f6bbb57e9c96
about on the mailing list.
Stephan if you want to adjust the colors of the Touchpad or whatever else, feel
free.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@36611 a95241bf-73f2-0310-859d-f6bbb57e9c96
image was the only black wordmark with the shadow of the green leaf on the
bottom.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@36606 a95241bf-73f2-0310-859d-f6bbb57e9c96
added an echo, to confirm the success of installing all firmwares.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@36597 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Avoid showing read-only volumes as possible installation targets.
* Mark the menu item for the first suitable volume that was found.
* Removed no more needed debug output.
Fixes#5231. Thanks a lot!
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@36596 a95241bf-73f2-0310-859d-f6bbb57e9c96
I took the conservative approach and the penalty is really minimal if even not at all perceptible as i've measured. I decided to go this way as PoseView code is very fragile and depends on many of those "loose assertions" to be valid. This conservative approach also allows to confidently include the fix in alpha2:
+alphabranch
Besides those "wise" reasons, the thing is that i've wasted too many hours already this week and in the past due that VSPoseList/PoseList syncing fragility. I'm not willing to fix the optimisation of a broken design that i've already experimentally rewriten two times already, and that i (or any mentally sane person) will finally ditch sooner or later :)
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@36592 a95241bf-73f2-0310-859d-f6bbb57e9c96
Since we also force uncacheability via the PTEs, it is no longer necessary to
use MTRRs for uncacheable ranges. It is recommended, which is why we still do
it, if possible, but when we run out of MTRRs, the algorithm ignores
uncacheable ranges now -- iteratively it increases the minimum size of ranges
to drop until the available MTRRs suffice.
Should fix#5703 and #5865.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@36591 a95241bf-73f2-0310-859d-f6bbb57e9c96
memory. This avoids implementation defined behavior on Pentium Pro/II when
intersecting with an write-combining or write-protected MTRR range.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@36589 a95241bf-73f2-0310-859d-f6bbb57e9c96
* DefaultManager: added a lock around rescan thread start and exit:
this should fix the possible race condition spotted by Ingo.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@36588 a95241bf-73f2-0310-859d-f6bbb57e9c96
as described in the BeBook when oneLevelOnly is false - only
BListView::SortItems() does ignore the hierarchy. So the work-arounds for
this are completely unnecessary. Removed now unused compare_list_items().
* The BOutlineListView sorting methods have a different comparison signature
as the other comparison functions: they point to the items directly. The
compare_typed_list_items() function did not take this into account, and
dereferenced them anyway, leading to #5883.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@36587 a95241bf-73f2-0310-859d-f6bbb57e9c96
way.
* Use that newly exposed feature of taking the focus window to the new workspace
when using the Ctrl-Alt-Shift-Arrow shortcut.
* This fixes#5675.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@36585 a95241bf-73f2-0310-859d-f6bbb57e9c96
PAT support (i.e. uncacheable, write-through, and write-back). Has pretty
much no effect ATM, as the MTRRs restrict the types to what is actually
requested.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@36583 a95241bf-73f2-0310-859d-f6bbb57e9c96
implemented for any architecture yet.
* vm_set_area_memory_type(): Call VMTranslationMap::ProtectArea() to change the
memory type for the already mapped pages.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@36574 a95241bf-73f2-0310-859d-f6bbb57e9c96
* Don't set the VMArea's memory type in arch_vm_set_memory_type(), but let the
callers do that.
* vm_set_area_memory_type(): Does nothing, if the memory type doesn't change.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@36573 a95241bf-73f2-0310-859d-f6bbb57e9c96
of consistency.
* Moved the B_OVERCOMMITTING_AREA flag from B_KERNEL_AREA_FLAGS to
B_USER_AREA_FLAGS, since we really allow it to be passed from userland.
* Most VM syscalls check the provided protection against B_USER_AREA_FLAGS
instead of B_USER_PROTECTION, now. This way they allow for
B_OVERCOMMITTING_AREA as well.
* _user_map_file(), _user_set_memory_protection(): Check the protection like
the other syscalls do and use fix_protection() instead of doing that
manually.
git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@36572 a95241bf-73f2-0310-859d-f6bbb57e9c96
