[00:00:33] <CIA-34> korli * r41425 /haiku/trunk/src/libs/compat/freebsd_wlan/net80211/ (ieee80211_haiku.cpp ieee80211_haiku.h): enable KASSERT in net80211 and fixes wrong uses of the macro.
[02:51:26] <CIA-34> mmlr * r41426 /haiku/trunk/ (7 files in 2 dirs):
[02:51:26] <CIA-34> * Move the legacy PIC and the IO-APIC code into their own source file. No
[02:51:26] <CIA-34> functional change intended.
[03:30:08] <CIA-34> mmlr * r41427 /haiku/trunk/headers/private/kernel/arch/x86/arch_int.h:
[03:30:09] <CIA-34> Quick build fix for the FreeBSD compatibility layer. Maybe the definition should
[03:30:09] <CIA-34> be moved to a (even) more private header?
[04:49:00] <CIA-34> mmlr * r41428 /haiku/trunk/src/system/kernel/arch/x86/ioapic.cpp:
[04:49:00] <CIA-34> * Prepared everything to eventually support multiple IO-APICs.
[04:49:00] <CIA-34> * Added some TODOs regarding register write order that might have bad side
[10:20:39] <CIA-34> mmlr * r41429 /haiku/trunk/ (4 files in 2 dirs): Provide a function to gain ACPI table access.
[10:25:33] <CIA-34> mmlr * r41430 /haiku/trunk/src/system/kernel/arch/x86/ioapic.cpp:
[10:25:33] <CIA-34> Enumerate IO-APICs using the ACPI Multiple APIC Descriptor Table (MADT). We do
[10:25:33] <CIA-34> the same for the first IO-APIC in the bootloader (also using the MADT), but
[10:46:56] <CIA-34> mmlr * r41431 /haiku/trunk/src/system/kernel/arch/x86/ (ioapic.cpp irq_routing_table.cpp irq_routing_table.h):
[10:46:56] <CIA-34> Provide an interrupt availability check callback instead of a maximum IRQ
[10:46:56] <CIA-34> number. This accounts for possible gaps in the IO-APIC GSI mappings. Since most
[11:00:20] <CIA-34> mmlr * r41432 /haiku/trunk/src/system/kernel/arch/x86/ioapic.cpp:
[11:00:20] <CIA-34> When enumerating IO-APICs failed, try to still get the routing done with how far
[11:00:20] <CIA-34> we got. If the routing is possible with the limited IO-APICs everythings good,
[11:04:01] <CIA-34> mmlr * r41433 /haiku/trunk/ (3 files in 2 dirs): Make the signature argument to get_table() const. Should fix the GCC4 build.
[11:11:42] <CIA-34> mmlr * r41434 /haiku/trunk/src/system/kernel/arch/x86/ioapic.cpp:
[11:11:42] <CIA-34> Only make GSI 0 into an ExtInt and GSIs 1-15 ISA interrupts. When having
[11:11:42] <CIA-34> multiple IO-APICs every IO-APIC was initialized to the standard mapping before.
[11:48:15] <CIA-34> mmlr * r41435 /haiku/trunk/src/system/kernel/arch/x86/ioapic.cpp: Obviously we want to use the GSI as interrupt vector as well...
[13:14:37] <CIA-34> mmlr * r41436 /haiku/trunk/src/system/kernel/arch/x86/ioapic.cpp: Extended and fix some of the debug output.
[13:22:43] <CIA-34> mmlr * r41437 /haiku/trunk/src/system/kernel/arch/x86/ioapic.cpp: Fix previous commit and add printing APIC id fields.
[15:16:16] <CIA-34> bonefish * r41438 /haiku/branches/developer/bonefish/signals/ (19 files in 8 dirs): (log message trimmed)
[15:16:17] <CIA-34> Work in progress towards replacing gThreadSpinlock:
[15:16:17] <CIA-34> * Introduced a global spinlock gSchedulerLock. It is supposed to take over the
[15:31:37] <CIA-34> bonefish * r41439 /haiku/branches/developer/bonefish/signals/src/system/kernel/sem.cpp: Replaced uses of gThreadSpinlock by gSchedulerLock.
[15:40:09] <CIA-34> bonefish * r41440 /haiku/branches/developer/bonefish/signals/ (7 files in 4 dirs):
[15:40:10] <CIA-34> * Renamed B_TIMER_ACQUIRE_THREAD_LOCK to B_TIMER_ACQUIRE_SCHEDULER_LOCK.
[15:40:10] <CIA-34> * Adjusted the timer code to actually acquire the scheduler lock.
[16:48:23] <CIA-34> bonefish * r41441 /haiku/branches/developer/bonefish/signals/src/system/kernel/ (6 files in 4 dirs):
[16:48:23] <CIA-34> Replaced more occurrences of gThreadSpinlock by gSchedulerLock where used as
[16:48:23] <CIA-34> scheduler lock.
[18:11:21] <CIA-34> bonefish * r41442 /haiku/branches/developer/bonefish/signals/headers/private/kernel/util/AutoLock.h: Added ThreadLocker AutoLocker typedef.
[19:03:03] <CIA-34> kallisti5 * r41443 /haiku/trunk/src/add-ons/accelerants/radeon_hd/mode.cpp: remove (now) un-needed modeline examples; no functional change
[19:12:44] <CIA-34> bonefish * r41444 /haiku/branches/developer/bonefish/signals/ (3 files in 2 dirs):
[19:12:44] <CIA-34> * Added Thread::GetAndLock().
[19:12:44] <CIA-34> * Changed and clarified locking semantics for Thread::user_thread. It's now
[20:03:38] <CIA-34> mmlr * r41445 /haiku/trunk/ (6 files in 2 dirs): (log message trimmed)
[20:03:38] <CIA-34> * When initializing MSI support, don't assume a single 24 entry IO-APIC. Instead
[20:03:38] <CIA-34> mark the ISA interrupts as unusable and then use ioapic_is_interrupt_available
[21:08:44] <CIA-34> bonefish * r41446 /haiku/trunk/src/system/boot/platform/bios_ia32/mmu.cpp:
[21:08:44] <CIA-34> mmu_allocate_physical(): The wrong base argument was passed to
[21:08:44] <CIA-34> get_free_physical_address_range(). Fixes the debug syslog feature.
[21:14:07] *** CIA-34 has quit IRC





top