src - OpenBSD base system

Age	Commit message (Collapse)	Author
2007-11-16	Fix dmesg output. From form@.	Mike Belopuhov
	ok deraadt
2007-11-16	the cpu_feature/cpu_id stuff here is bullshit, XXX it so that it gets fixed ↵	Theo de Raadt
	one day soon
2007-11-16	fix the bus_space #define nightmare, so that amd64 and i386 are much more	Theo de Raadt
	uniform. as a result shared code like acpi needs less #ifdef's ok marco kettenis
2007-11-16	quotes around model name too	Theo de Raadt

2007-11-15	remove the #ifdef ACPI_ENABLE. the EC codepath is either safe enough (and	Theo de Raadt
	will be improved), or it is unsafe because AML assumes it must be there ok almost everyone
2007-11-15	Make acpicpu(4) attach to the existing CPUs only.	Mike Belopuhov
	Problem reported by form@. Fix idea by kettenis@. Lots of help from deraadt@. Tested by deraadt@, form@ and me. ok deraadt marco
2007-11-15	Nicer dmesg output.	Mike Belopuhov
	From Oleg Safiullin (form@) ok deraadt
2007-11-15	don't waste our time telling the aml that we are OpenBSD; it will never	Theo de Raadt
	make a good decision based on that; ok kettenis canacar
2007-11-14	shrink floppy build by knocking out some unused bits of code; seen by canacar	Theo de Raadt

2007-11-14	spacing	Theo de Raadt

2007-11-14	Fix handling of local AML variables for type conversion:	Can Erkin Acar
	Check whether a variable is local before and after dereferencing. Improve the check to make sure that the variable is in local in current scope. Tested by many, thanks a lot. ok marco@
2007-11-12	spacing	Theo de Raadt

2007-11-12	no need to print acpi name on attach	Theo de Raadt

2007-11-12	a little less excitement (every printf does not need to end in !)	Theo de Raadt

2007-11-12	correct : vs , in attach-time messages	Theo de Raadt

2007-11-12	spacing	Theo de Raadt

2007-11-12	only print bif info that is complete	Theo de Raadt

2007-11-12	missing :	Theo de Raadt

2007-11-10	do not check malloc return value against NULL, as M_WAITOK is used	Charles Longeau
	ok marco@
2007-11-08	Pass correct pointers to memcpy when concatanating strings/buffers	Can Erkin Acar
	ok weingart@, marco@
2007-11-08	convert the value to a string instead of using aml_val2str()	Can Erkin Acar
	which returns invalid pointers if the value is not a string, causing acpibat to panic, as reported by mbalmer@ and Alexey Suslikov. tested by Alexey Suslikov, ok marco@
2007-11-06	all ACPI_ENABLE kernels need acpi_interrupt()	Theo de Raadt

2007-11-06	move acpi_read_pmreg() and acpi_write_pmreg() out of SMALL_KERNEL so	Theo de Raadt
	that acpi ramdisks can compile (obvious solution)
2007-11-05	Establish the acpi interrupt earlier. Fixes problems with some	Can Erkin Acar
	GENERIC.MP kernels not getting acpi interrupts. This is a workaround until the real issue with late establishment is fixed. ok weingart@, beck@, gwk@
2007-11-05	Use the acpi_{acquire,release}_global_lock() to implement the global	Tobias Weingartner
	functionality for ACPI. Has not shown any regressions, and may solve some EC race conditions. Ok gwk@, canacar@, ckuethe@
2007-11-04	Check for non-zero address if mapping to the extended addresses.	Mike Belopuhov
	Patch from Oleg Safiullin (a.k.a form@) ok kettenis canacar weingart
2007-11-03	Notebook from ckuethe@ that notifies AC with 0x01, so add that in.	Can Erkin Acar
	cool ckuethe@
2007-11-03	ACPI allows the OS to identify itself in a couple of ways and may behave	Chris Kuethe
	differently based on what OS is running. This diff causes us to run the RedmondOS code path in hopes that it is "better" AML, or has better settings for machine that do not support _OSI. This is the same method used by other non-Redmond operating sytems. suggestions/ok deraadt, agonized whimpers from the others.
2007-11-03	Remove a useless printf, hit if acpicpu is disabled.	Gordon Willem Klok
	ok beck@
2007-11-03	Added support for displaying wakeup devices	Jordan Hargrave
	ok beck@,weingart@,gwk@
2007-11-03	Add missing else so that Concatanation of two buffers work.	Can Erkin Acar
	Noticed by robert@ while diagnosing a problem report by Johan Lindman. ok weingart@
2007-11-03	Fix setting values to buffers by obeying the conversion rules.	Can Erkin Acar
	The values are truncated or zero extended as required. The spec is somewhat ambigious on strings and integers requiring truncation "before copy" so we try to be safe and zero terminate the buffer in case the source is string. Reported by ckuethe@, diagnosis, initial diff and feedback by kettenis@ ok kettenis@, beck@, weingart@, robert@, gwk@, ckuethe@
2007-11-03	less colons	Theo de Raadt

2007-11-03	Remove those words...	Chris Kuethe

2007-11-03	repair attach messages; ok gwk canacar	Theo de Raadt

2007-11-03	ACPI allows the OS to identify itself in a couple of ways and may behave	Chris Kuethe
	differently based on what OS is running. This diff causes us to run the WinNT code path in hopes that it is "better" AML, or has better settings for machine that do not support _OSI. This is the same method used by other non-windows operating sytems. ok gwk, beck, jordan
2007-11-03	Changed aml_parselength to use incremental length calc	Jordan Hargrave
	ok canacar@
2007-10-11	When walking the tree yo find _PRT methods, check whether the device is	Mark Kettenis
	actually enabled and functioning. Similar to what we do for _INI. ok weingart@, beck@
2007-10-08	More simple memset(,0,) -> M_ZERO changes. In this batch move to	Kenneth R Westerback
	size(*p) as the first malloc() parameter where p is declared locally and thus easy to check. Add M_ZERO to gpe_table allocation in acpi.c even though there is no obvious bzero or memset nearby.
2007-09-13	Remove unnecessary memset() (see aml_evalnode(9)) and a bit of KNF.	Michael Knudsen
	ok toby
2007-09-13	Implement ACPI 6.5.1 spec tree walk for _STA and _INI.	Tobias Weingartner
	Fixes mk's laptop. No regressions so far (thank you to the testers). ok gwk@, mk@, marco@
2007-09-08	Case ranges e.g. case 'A' ... 'Z': are a gcc-ism not valid C. Fix these, and	Gordon Willem Klok
	a bit of whitespace cleanup while here. ok otto, deraadt
2007-09-07	Fix code that figures out the PCI bus number for a _PRT configuration object.	Mark Kettenis
	This should get rid of the "multiple bus 0" problem seen on some machines, and fix PCI interrupt routing on them. ok marco@ (a while ago)
2007-06-23	Accept 0x00 as notify type since some notebooks (MSI-S260)	Can Erkin Acar
	incorrectly use it when signaling events. ok marco@
2007-06-15	Check that sc_notify actually points to something before attempting to call	Gordon Willem Klok
	it. Fixes marco@'s post c2k7 panics. ok marco
2007-05-31	Make powernow-k8 on amd64 and i386 use the _PSS object from acpi to	Gordon Willem Klok
	retreive p_state data as spelled out in the amd64 bios and kernel developers guide. This code is still a little rough around the edges but has been tested by myself on a tyan machine and by phessler at theapt dot org on an HP DL145. This diff also takes a first stab and cleaning up the acpicpu dmesg spam. ok tedu, marco
2007-05-28	fix build when ACPI_DEBUG is defined	Robert Nagy

2007-05-28	Eliminate unsightly blank lines in dmesg.	Kenneth R Westerback
	ok marco@
2007-05-06	at least always print a new line, even if NOPSS	Ted Unangst

2007-04-23	When receiving an undock request, walk the list of dependant devices	Michael Knudsen
	and pull their _EJ0 knobs before undocking. This ejects any dependant devices (if necessary) before actually undocking. Without this, the usb hub in my X6 gets upset when it's suddenly kicked off the system without warning, and upon redocking it attaches as uberry. Now it's behaving as it should.