[03:06:07] <CIA-98> mmlr * r41371 /haiku/trunk/src/system/kernel/arch/x86/ (irq_routing_table.cpp irq_routing_table.h): (log message trimmed)
[03:06:07] <CIA-98> * Implement the PCI address retrieval on the kernel side and remove the need
[03:06:07] <CIA-98> for the introduced get_pci_info() method in ACPI as it doesn't work reliably.
[03:06:07] <CIA-98> The new version should be more robust and efficient as it only resolves the
[03:06:07] <CIA-98> root bridge values once.
[03:06:07] <CIA-98> * Don't try to read and use the secondary bus register for the root bridge as
[03:06:08] <CIA-98> it isn't actually a normal bridge. We get the root bridge bus by using the
[03:14:07] <CIA-98> mmlr * r41372 /haiku/trunk/ (4 files in 2 dirs):
[03:14:07] <CIA-98> Basically revert r41365. Remove the get_pci_info() function of ACPI again as it
[03:14:07] <CIA-98> turns out to be flaky, is a little bit of a hack anyway and performs unnecessary
[03:14:07] <CIA-98> operations. The functionality is now handled kernel side which optimizes those
[03:14:08] <CIA-98> problems away.
[03:15:25] <CIA-98> mmlr * r41373 /haiku/trunk/src/system/kernel/arch/x86/arch_int.cpp: Print the routing table unconditionally for now.
[07:35:53] <CIA-98> kallisti5 * r41374 /haiku/trunk/src/add-ons/accelerants/radeon_hd/ (Jamfile accelerant.h mode.cpp mode.h): (log message trimmed)
[07:35:53] <CIA-98> * use common accelerant code
[07:35:53] <CIA-98> * remove manual sDisplayMode and make use of mode_list_area via Axel's common accelerant code using vesa as example
[07:35:53] <CIA-98> * move mask enabled write32 function to accelerant with write32 calls
[07:35:53] <CIA-98> * move register offsets into mode.h headers
[07:35:54] <CIA-98> * rename DxMode* to CardMode*
[07:35:55] <CIA-98> * make adjustments and implement radeon hd mode setting further
[09:44:51] <CIA-98> korli * r41375 /haiku/trunk/src/ (6 files in 5 dirs):
[09:44:51] <CIA-98> Made ieee80211_ratectl_tx_complete and ieee80211_ratectl_rate functions no op instead of not calling these functions in each wlan driver.
[09:44:51] <CIA-98> This should be iso functionality.
[10:24:02] <CIA-98> modeenf * r41376 /haiku/trunk/src/preferences/bluetooth/BluetoothSettingsView.cpp: Guess the constructor was not only used in the layout API ;)
[11:06:35] <CIA-98> laplace * r41377 /haiku/vendor/gutenprint/current/ (335 files in 34 dirs): Load gutenprint-5.2.7 into haiku/vendor/gutenprint/current.
[11:14:41] <CIA-98> laplace * r41378 /haiku/vendor/gutenprint/5.2.7/: * Tagging gutenprint-5.2.7
[12:12:03] <CIA-98> bonefish * r41379 /haiku/branches/developer/bonefish/signals/headers/private/kernel/user_debugger.h: Fixed typo.
[12:24:53] <CIA-98> bonefish * r41380 /haiku/branches/developer/bonefish/signals/ (9 files in 3 dirs): (log message trimmed)
[12:24:53] <CIA-98> * Eliminated gTeamSpinlock. Its tasks have been taken over by other locks:
[12:24:53] <CIA-98> locks.
[13:19:29] <CIA-98> laplace * r41381 /haiku/trunk/src/libs/print/libgutenprint/ (336 files in 34 dirs): * merging gutenprint-5.2.7 into the main branch.
[14:12:17] <CIA-98> mmlr * r41382 /haiku/trunk/ (4 files in 2 dirs):
[14:12:18] <CIA-98> Add set_current_resources() which will be used to configure device resources
[14:12:18] <CIA-98> from settings retrieved through get_{current|possible}_resources().
[14:55:16] <CIA-98> mmlr * r41383 /haiku/trunk/src/system/kernel/arch/x86/ (irq_routing_table.cpp irq_routing_table.h): (log message trimmed)
[14:55:16] <CIA-98> Reimplement the IRQ setting function. Based on the template we get from current
[14:55:16] <CIA-98> resources we input the desired configuration and then set them. Note that the
[14:55:16] <CIA-98> set_current_resources() ("_SRS") method is pretty picky (on the low-level ACPI
[14:55:16] <CIA-98> side) and needs the configuration in exactly the form get_current_resources()
[14:55:17] <CIA-98> ("_CRS") returns, so that's why we need to utilize that buffer as a template.
[14:55:18] <CIA-98> Looking at FreeBSD there seem to be systems that don't actually provide a
[15:13:53] <CIA-98> mmlr * r41384 /haiku/trunk/src/system/kernel/arch/x86/irq_routing_table.cpp:
[15:13:53] <CIA-98> Implement a very naive configure_link_device() function that just blindly sets
[15:13:53] <CIA-98> the first possible configuration. This will probably result in a "everything
[15:13:53] <CIA-98> mapped to a single IRQ" situation, but should get systems that make use of link
[15:13:53] <CIA-98> devices in APIC mode at least running.
[15:13:53] <CIA-98> To make more sensible decisions the interface has to be changed next as it
[15:13:54] <CIA-98> currently doesn't allow to get anything but the very first possible config.
[20:25:14] <CIA-98> scottmc * r41385 /haiku/trunk/data/system/boot/SetupEnvironment: Added setting of LC_CTYPE to SetupEnviroment, this fixes #7412
[21:15:02] <CIA-98> korli * r41386 /haiku/trunk/src/add-ons/kernel/drivers/common/usb_modeswitch.cpp: style fix
[21:16:51] <CIA-98> korli * r41387 /haiku/trunk/src/add-ons/kernel/drivers/graphics/intel_extreme/intel_extreme.cpp: "intel_extreme" was printed twice
[21:35:08] <CIA-98> korli * r41388 /haiku/trunk/src/add-ons/kernel/drivers/audio/hda/hda_controller.cpp: spaces -> tab
[22:01:05] <CIA-98> zooey * r41389 /haiku/trunk/ (7 files in 6 dirs):
[22:01:05] <CIA-98> Switch haiku-revision from uint32 to string, as that's going to be required soon, no matter if we switch to Git or Mercurial
[22:01:05] <CIA-98> * increase _SYS_NAMELEN defined in sys/utsname.h to 128 to allow long(ish) revisions
[22:01:05] <CIA-98> * sHaikuRevision is now a static character array (in both libroot and kernel)
[22:01:05] <CIA-98> * adjust build tool set_haiku_revision to write the revision as string
[22:12:28] <CIA-98> korli * r41390 /haiku/trunk/src/libs/compat/freebsd_network/compat/sys/haiku-module.h: NO_HAIKU_FIRMWARE_NAME_MAP defined an element {NULL} whereas the array length is zero
[22:41:17] <CIA-98> scottmc * r41391 /haiku/trunk/build/jam/OptionalPackages: Updated expat to fix its directories.
[23:50:22] <CIA-98> zooey * r41392 /haiku/trunk/headers/posix/sys/utsname.h:
[23:50:23] <CIA-98> Revert part of r41389 that injured binary compatibility (rightly pointed out by Axel)
[23:50:23] <CIA-98> * length of utsname members must stay at length of 32
[23:50:23] <CIA-98> This will work just the same for now. When using a DVCS yields longer revisions,
[23:50:23] <CIA-98> we need to adjust AboutSystem to fetch the full revision directly from the ELF
[23:50:23] <CIA-98> section '_haiku_revision' (as found in libroot.so or libbe.so).





top