[00:24:31] <CIA-109> bonefish * r42109 /haiku/branches/developer/bonefish/signals-merge/: Created merge branch for the signals branch based on trunk r42104.
[00:29:43] <CIA-109> bonefish * r42110 /haiku/branches/developer/bonefish/signals-merge/ (253 files in 81 dirs):
[00:29:43] <CIA-109> Merged signals branch. Ignoring tree conflict of
[00:29:43] <CIA-109> src/system/libroot/posix/pthread/pthread.c caused by r40665.
[00:47:35] <CIA-109> bonefish * r42111 /haiku/branches/developer/bonefish/signals-merge/ (3 files in 3 dirs):
[00:47:35] <CIA-109> Removed the __init_pthread() introduced in r40665. Directly initialize the
[00:47:35] <CIA-109> pthread_t::id for the main thread in initialize_before() instead.
[00:48:30] <CIA-109> bonefish * r42112 /haiku/branches/developer/bonefish/signals-merge/src/system/kernel/thread.cpp: Reverted r40701. There aren't sigaction handlers per thread anymore.
[00:54:53] <CIA-109> bonefish * r42113 /haiku/branches/developer/bonefish/signals-merge/src/system/libroot/posix/glibc/include/bits/ (types.h typesizes.h):
[00:54:53] <CIA-109> This is all quite a mess -- uncommented glibc's __timer_t and __TIMER_T_TYPE
[00:54:53] <CIA-109> types to avoid build issues (gcc 2).
[00:56:08] <CIA-109> bonefish * r42114 /haiku/branches/developer/bonefish/signals-merge/headers/private/kernel/UserTimer.h: Made UserTimer destructor virtual, as it should be. Spotted by gcc 2.
[01:01:51] <CIA-109> bonefish * r42115 /haiku/branches/developer/bonefish/signals-merge/src/system/kernel/signal.cpp:
[01:01:51] <CIA-109> Fixed access to Thread::debug_info::ignore_signals[_once]. It doesn't have to
[01:01:51] <CIA-109> be atomic and doesn't work that way anymore with gcc 2, since the fields are
[01:01:51] <CIA-109> now 64 bit wide.
[01:57:31] <CIA-109> bonefish * r42116 /haiku/trunk/ (258 files in 83 dirs): (log message trimmed)
[01:57:31] <CIA-109> Merged signals-merge branch into trunk with the following changes:
[01:57:31] <CIA-109> * Reorganized the kernel locking related to threads and teams.
[01:57:31] <CIA-109> * We now discriminate correctly between process and thread signals. Signal
[01:57:31] <CIA-109> handlers have been moved to teams. Fixes #5679.
[01:57:31] <CIA-109> * Implemented real-time signal support, including signal queuing, SA_SIGINFO
[01:57:32] <CIA-109> support, sigqueue(), sigwaitinfo(), sigtimedwait(), waitid(), and the addition
[02:35:12] <CIA-109> scottmc * r42117 /haiku/trunk/build/jam/OptionalPackages:
[02:35:12] <CIA-109> Updated KeymapSwitcher which fixes an issue in the 06-11 builds, from S.Zharski.
[02:35:12] <CIA-109> +alpha3
[03:53:36] <CIA-109> anevilyak * r42118 /haiku/trunk/src/system/boot/platform/bios_ia32/acpi.cpp: (log message trimmed)
[03:53:36] <CIA-109> Revert r42105. The rsdt_length field on the rsdp structure we have declared is
[03:53:36] <CIA-109> actually misleading, since it specifies the length of the xsdt rather than the
[03:53:36] <CIA-109> rsdt. Furthermore, it is only available if the BIOS reports ACPI version 2.0 or
[03:53:36] <CIA-109> greater. Since we don't currently support the XSDT, we can't rely on that field
[03:53:36] <CIA-109> and have to resort to the previous approach of mapping the RSDT header in order
[03:53:37] <CIA-109> to determine the length of the RSDT itself.
[04:09:13] <CIA-109> anevilyak * r42119 /haiku/trunk/headers/private/kernel/arch/x86/arch_acpi.h: Rename rsdt_length to xsdt_length to avoid future confusion.
[04:24:23] <CIA-109> anevilyak * r42120 /haiku/trunk/src/kits/interface/ListView.cpp:
[04:24:23] <CIA-109> Remove unnecessary calculations from BListView::ScrollToSelection(). At least I
[04:24:23] <CIA-109> see no useful purpose to the distinction made here.
[13:21:08] <CIA-109> anevilyak * r42121 /haiku/trunk/src/kits/interface/ListView.cpp: Revert r42120.
[14:33:45] <CIA-109> bonefish * r42122 /haiku/branches/developer/bonefish/signals/src/system/boot/loader/pager.cpp: Use gray background to avoid the annoying blinking.
[14:38:42] <CIA-109> bonefish * r42123 /haiku/branches/developer/bonefish/signals/src/system/boot/platform/bios_ia32/mmu.cpp:
[14:38:42] <CIA-109> * mmu_allocate_physical(): Fixed incorrect "base" parameter passed to
[14:38:42] <CIA-109> get_free_physical_address_range(). Fixes the debug syslog feature.
[14:38:42] <CIA-109> * mmu_init_for_kernel(): Use the B_PRI* macros for printing the memory ranges.
[16:50:07] <CIA-109> bonefish * r42124 /haiku/trunk/src/system/boot/ (loader/pager.cpp platform/bios_ia32/mmu.cpp):
[16:50:07] <CIA-109> I accidentally committed r42122 and r42123 to the signals branch. Merged into
[16:50:07] <CIA-109> trunk. The interesting part of r42122 was already fixed in the trunk, though.
[17:22:00] <CIA-109> axeld * r42125 /haiku/trunk/src/add-ons/ (4 files in 2 dirs):
[17:22:00] <CIA-109> * Let checkfs collect stats of how the data streams are used.
[17:22:00] <CIA-109> * Would be nice to only show this when an extra argument had been passed to
[17:22:00] <CIA-109> checkfs, but I'm afraid the API is somewhat limited, not to say pretty much
[17:22:00] <CIA-109> unusable for the task.
[19:12:49] <CIA-109> bonefish * r42126 /haiku/trunk/ (12 files in 9 dirs): (log message trimmed)
[19:12:49] <CIA-109> * Added arch_debug_blue_screen_try_getchar() and arch_debug_serial_try_getchar()
[19:12:49] <CIA-109> which don't wait for a character, but return -1 when no character is
[19:12:49] <CIA-109> available ATM. Implemented correctly for x86 only.
[19:12:49] <CIA-109> * Changed the semantics of the debugger_module_info::debugger_getchar() hook.
[19:12:49] <CIA-109> It is supposed to return immediately now.
[19:12:50] <CIA-109> * Adjusted usb_keyboard accordingly. Hacked UHCI's debug_process_transfer() to
[20:00:37] <CIA-109> humdinger * r42127 /haiku/trunk/docs/ (175 files in 33 dirs): Update user guide and welcome pages and their translations. Thanks everyone. +alpha
[20:33:06] <CIA-109> humdinger * r42128 /haiku/trunk/data/catalogs/ (133 files in 84 dirs): (log message trimmed)
[20:33:06] <CIA-109> * Updated all catkeys from HTA. Thanks all translators!
[20:33:06] <CIA-109> * Since the updated Time prefs didn't make it into alpha3, I kept the older
[20:33:06] <CIA-109> catkeys in SVN.
[20:33:06] <CIA-109> * Renamed "Application name" to "System name" manually in the catkeys of
[20:33:07] <CIA-109> CodyCam and Poorman. For some reason HTA didn't get these changes in the
[20:33:08] <CIA-109> code...
[22:14:20] <CIA-109> anevilyak * r42129 /haiku/trunk/ (5 files in 4 dirs):
[22:14:20] <CIA-109> Add private get_memory_properties() syscall which allows one to retrieve the
[22:14:20] <CIA-109> address protection bits as well as the wiring flags for an arbitrary address
[22:14:20] <CIA-109> in a team's address space. Will be used in the debugger for the purposes
[22:14:20] <CIA-109> of the memory inspector/editor, in order to determine whether it can in fact
[22:14:21] <CIA-109> allow editing for the currently inspected address range.
[22:15:05] <CIA-109> anevilyak * r42130 /haiku/trunk/headers/private/kernel/vm/vm.h: Should have been part of previous commit.
[22:37:42] <CIA-109> mmlr * r42131 /haiku/trunk/src/tests/kits/device/ (Jamfile bserial/ bserial/Jamfile bserial/SerialIO.cpp):
[22:37:44] <CIA-109> Add serial_io, a simplistic BSerialPort based command line tool that prints what
[22:37:45] <CIA-109> it reads from the specified port and writes what gets input on stdin. It
[22:38:38] <CIA-109> currently hardcodes the basic setup of the serial port.
[22:48:33] <CIA-109> anevilyak * r42132 /haiku/trunk/src/apps/debugger/ (6 files in 2 dirs):
[22:48:33] <CIA-109> * TeamMemoryBlock: Now stores whether or not the block it represents is writable.
[22:48:33] <CIA-109> * RetrieveMemoryBlockJob: Make use of get_memory_properties() to retrieve the
[22:48:33] <CIA-109> block's protection bits and mark it appropriately.
[23:58:49] <CIA-109> anevilyak * r42133 /haiku/trunk/ (2 files in 2 dirs): (log message trimmed)
[23:58:49] <CIA-109> * Add validation of the RSDP checksums.
[23:58:49] <CIA-109> * If we detect ACPI 2.0 or higher, the spec says we should use the XSDT rather
[23:58:49] <CIA-109> than the RSDT. Attempt to do so, falling back to the RSDT if the former fails
[23:58:49] <CIA-109> to be mapped/validated.
[23:58:51] <CIA-109> * Refactored acpi_find_table into a templated version to account for the fact
[23:58:51] <CIA-109> that the XSDT exports different pointer widths for its links to other tables





top