[01:10:20] *** echelog-2 has joined #haiku-jp
[01:14:03] <CIA-111> bonefish * r42026 /haiku/branches/developer/bonefish/signals/ (5 files in 2 dirs): (log message trimmed)
[01:14:03] <CIA-111> * Introduced new private flag B_TIMER_REAL_TIME_BASE. When specified add_timer()
[01:14:03] <CIA-111> interprets the given time as a real-time value for absolute timers.
[01:14:03] <CIA-111> * Added per_cpu_timer_data::real_time_offset field, which stores the current
[01:14:04] <CIA-111> real-time offset. Initialized by the new timer_init_post_rtc().
[01:14:04] <CIA-111> * Added timer_real_time_clock_changed() which is called when the real-time clock
[01:14:05] <CIA-111> has been set. It updates and reschedules all absolute real-time timers.
[02:41:29] *** echelog-2 has joined #haiku-jp
[02:42:10] <CIA-111> bonefish * r42037 /haiku/branches/developer/bonefish/signals/src/tests/system/libroot/posix/posixtestsuite/conformance/interfaces/clock_getres/ (5-1.c 6-1.c):
[02:42:10] <CIA-111> Fixed broken tests: Instead of using an arbitrary clock ID, use the CPU clock
[02:42:10] <CIA-111> ID of an already dead child. This has at least a better chance of working.
[02:56:05] <CIA-111> kallisti5 * r42038 /haiku/trunk/src/add-ons/kernel/file_systems/ramfs/kernel_interface.cpp: change legacy fs_cookie to void**; tweak a few node function parameters
[03:37:35] *** echelog-2 has joined #haiku-jp
[03:54:42] <CIA-111> bonefish * r42039 /haiku/branches/developer/bonefish/signals/ (2 files in 2 dirs):
[03:54:42] <CIA-111> * Moved fNextTime and fInterval to base class UserTimer and also use them for
[03:54:42] <CIA-111> {System,Real}TimeUserTimer. This makes them independent from whatever the
[03:54:42] <CIA-111> kernel timer code does with the timer fields.
[03:54:42] <CIA-111> * Be more overflow and underflow conscious, and avoid passing a negative time to
[03:54:42] <CIA-111> add_timer(). This would cause the timer never to be scheduled, while we would
[03:54:43] <CIA-111> actually want to fire it immediately.
[04:14:22] *** echelog-2 has joined #haiku-jp
[04:14:22] -asimov.freenode.net- [freenode-info] if you're at a conference and other people are having trouble connecting, please mention it to staff: http://freenode.net/faq.shtml#gettinghelp
[04:14:47] <CIA-109> bonefish * r42041 /haiku/branches/developer/bonefish/signals/src/system/kernel/thread.cpp:
[04:14:47] <CIA-109> _user_snooze_etc(): When converting a relative timeout to an absolute one,
[04:14:47] <CIA-109> convert the time base from real-time to system time before getting the time. We
[04:14:47] <CIA-109> did compute an absolute real-time value, but used that with system timebase
[04:14:47] <CIA-109> afterwards.
[04:22:57] <CIA-109> kallisti5 * r42042 /haiku/trunk/src/add-ons/kernel/file_systems/ramfs/ (Volume.cpp Volume.h kernel_interface.cpp): tweak the last of the nodes; break apart old style file_system_module_info; add fVolume to Volume to replicate bfs style
[04:44:19] <CIA-109> bonefish * r42043 /haiku/branches/developer/bonefish/signals/src/tests/system/libroot/posix/posixtestsuite/conformance/interfaces/clock_settime/17-2.c:
[04:44:19] <CIA-109> Fixed the test: Don't pass arbitrary clock IDs to clock_settime(). Try a CPU clock from a
[04:44:19] <CIA-109> dead child instead.
[04:54:08] <CIA-109> bonefish * r42044 /haiku/branches/developer/bonefish/signals/ (2 files in 2 dirs): Disallow negative {timespec,timeval}::tv_sec values.
[05:11:14] <CIA-109> kallisti5 * r42045 /haiku/trunk/src/add-ons/kernel/file_systems/ramfs/ (Volume.cpp Volume.h kernel_interface.cpp): begin reorganizing ramfs callback parameters to match new filesystem driver standard; correct incorrect style fix as per Ingo; correct layout of fs operation mapping struct
[05:32:38] <CIA-109> kallisti5 * r42046 /haiku/trunk/src/add-ons/kernel/file_systems/ramfs/ (OpenHashTable.h kernel_interface.cpp): correct cookie pointers to match filesystem operations; new to new.h; fix parameters of ramfs_read_vnode
[05:39:02] <CIA-109> kallisti5 * r42047 /haiku/trunk/src/add-ons/kernel/file_systems/ramfs/Query.cpp: fix gcc4 warning on missing parentheses on &
[05:47:41] <CIA-109> kallisti5 * r42048 /haiku/trunk/src/add-ons/kernel/file_systems/ramfs/Query.cpp: fix deprecated non-const char
[14:21:59] <CIA-109> bonefish * r42049 /haiku/branches/developer/bonefish/signals/ (2 files in 2 dirs): (log message trimmed)
[14:21:59] <CIA-109> * Initialize fNextTime, fInterval, and fOverrunCount in the UserTimer
[14:21:59] <CIA-109> constructor.
[14:21:59] <CIA-109> * Don't reset fOverrunCount in UserTimer::HandleTimer() after successful event
[14:21:59] <CIA-109> delivery. It's not specified that this should happen.
[14:21:59] <CIA-109> * Reset fOverrunCount in the Schedule() methods. It's not specified that this
[14:22:00] <CIA-109> should happen either, but that's what Linux does and it kind of makes sense.
[15:04:11] *** Clex has joined #haiku-jp
[15:04:15] *** Clex has left #haiku-jp
[15:14:14] <CIA-109> bonefish * r42050 /haiku/branches/developer/bonefish/signals/src/system/libroot/posix/sys/times.c:
[15:14:14] <CIA-109> * Use CLOCKS_PER_SEC instead of CLK_TCK.
[15:14:14] <CIA-109> * Replaced local variable by a macro.
[15:24:34] <CIA-109> bonefish * r42051 /haiku/branches/developer/bonefish/signals/src/system/libroot/posix/time/clock.c: Use the CLOCKS_PER_SEC macro instead of hard-coding the conversion factor.
[15:29:59] <CIA-109> bonefish * r42052 /haiku/branches/developer/bonefish/signals/headers/ (posix/time.h private/system/user_timer_defs.h): Corrected the type of the clock ID constants.
[15:32:38] <CIA-109> bonefish * r42053 /haiku/branches/developer/bonefish/signals/src/system/kernel/lib/Jamfile: Use a subdir based path to include kernel_lib.h. Fixes the build.
[16:00:21] <CIA-109> bonefish * r42054 /haiku/branches/developer/bonefish/signals/ (13 files in 5 dirs):
[16:00:21] <CIA-109> * Changed CLOCKS_PER_SEC (and CLK_TCK) to 1000000, as required by POSIX.
[16:00:21] <CIA-109> * Respectively introduced versioned symbols for clock(), times(), and sysconf().
[19:20:07] <CIA-109> axeld * r42055 /haiku/trunk/src/add-ons/mail_daemon/inbound_protocols/pop3/ (Jamfile MessageIO.cpp MessageIO.h pop3.cpp):
[19:20:07] <CIA-109> * Removed support for building on BeOS R5.
[19:20:07] <CIA-109> * Made MessageIO comply to our coding style.
[19:20:07] <CIA-109> * No other functional change (intended).
[19:37:02] <CIA-109> scottmc * r42056 /haiku/trunk/build/jam/OptionalPackages:
[19:37:02] <CIA-109> Disabled BeHappy as an OptionalPackage as it is currently broken. If/when it gets fixed, it will be easy to re-add it.
[19:37:02] <CIA-109> +alpha3
[20:26:14] <CIA-109> axeld * r42057 /haiku/trunk/src/add-ons/kernel/file_systems/bfs/BlockAllocator.cpp: * Remove the entry from the entry cache as well when removing a bad inode.
[20:42:37] <CIA-109> kallisti5 * r42058 /haiku/trunk/src/add-ons/kernel/drivers/disk/usb/usb_disk/usb_disk.cpp: (log message trimmed)
[20:42:37] <CIA-109> * change overly large uint32 to uint8 as we only count to 12 max
[20:42:37] <CIA-109> * add additional tracing to usb_disk lun inquiry
[20:42:37] <CIA-109> * cd and other optical disk devices can take some time to spin up
[20:42:37] <CIA-109> when plugged. Now we give them some time to do so to ensure we
[20:42:38] <CIA-109> can detect the media type properly if there is a cd in the drive.
[20:42:38] <CIA-109> Things non-mechanical (flash drives) will continue almost immediately
[20:54:16] <CIA-109> kallisti5 * r42059 /haiku/trunk/src/add-ons/kernel/drivers/disk/usb/usb_disk/usb_disk.cpp: small change, use non-posix style int typedef
[21:24:54] <CIA-109> kallisti5 * r42060 /haiku/trunk/src/add-ons/kernel/drivers/disk/usb/usb_disk/usb_disk.cpp: change to non-posix int introduced a small bug, sorry about that
[22:12:20] <CIA-109> pulkomandy * r42061 /haiku/branches/releases/r1alpha3/ (39 files in 28 dirs): Merge 41910 and 41971-42056 from trunk





top