src - OpenBSD base system

Age	Commit message (Collapse)	Author
2013-11-15	Clean-up the HID environment.	Paul Irofti
	This set of drivers were very very dirty. i/ Clean-up the match/attach multi-casting hackjob - stop casting aux to every attach arg in existance - be consistent about casting it only to uhidev_attach_arg - fetch the usb_attach_arg from above where needed ii/ Sort out the activate routines - leave the deactivation to the parent (uhidev(4)) - ditch the sc_dying flag in favour of usbd_is_dying() iii/ Get closer to keeping all the usb hid information in the uhidev structure (one uhidev per reportID) - store the usbd_device in the uhidev - use it consistently instead of always peaking at the parent's soft state Okay mpi@
2013-11-15	Make vax stand build cross-build friendly.	Masao Uebayashi
	OK miod@
2013-11-15	Add missing parenthesis to make tick calculation work as intended. Should	Kenneth R Westerback
	fix excessive timeouts and 'Michael mic' errors. Problem pointed out by vigdis via bugs@ ok dlg@
2013-11-14	Clean up tabs and a backslash which seemed to have sneeked in in r1.29.	Patrick Wildt
	"clean that little thingy up" deraadt@
2013-11-14	Use (N * sizeof(struct klist)) instead of (N * sizeof(struct klist *))	Charles Longeau
	when malloc'ing struct klist *. Similar diff found upstream: http://svnweb.freebsd.org/base?view=revision&revision=197575 Found by LLVM/Clang Static Analyzer. ok tedu@ krw@ guenther@
2013-11-14	Improvements for address assignment and related issues in IPv6CP.	Stefan Sperling
	Move address assignment to process context. This uses a workq at present, conversion to taskq is on my todo list (discussed with dlg). Instead of rummaging around in sockaddr_in6 structs, use the proper netinet6 APIs to configure addresses. Deal with IFID collisions instead of ignoring them. The whole point of IPv6CP is to avoid IFID collisions. Use arc4random() during IFID generation. Assign destination address to /128 point-to-point links. tested by me and sthen with pppoe(4) ok sthen
2013-11-14	replace workqs with tasks for handling resume. state handling is	David Gwynne
	still in workqs. from kimberley manning
2013-11-14	replace workqs with tasks for handling resume	David Gwynne
	from kimberley manning
2013-11-14	replace workqs with tasks for handling resume	David Gwynne
	from kimberley manning
2013-11-14	replace workqs with tasks for handling resume	David Gwynne
	from kimberley manning
2013-11-14	replace workqs with tasks for handling resume	David Gwynne
	from kimberley manning
2013-11-14	replace workqs with tasks for handling resume	David Gwynne
	from kimberley manning
2013-11-14	replace workqs with tasks for handling resume	David Gwynne
	from kimberley manning
2013-11-14	replace workqs with tasks for handling resume	David Gwynne
	from kimberley manning
2013-11-14	replace workqs with tasks for handling resume	David Gwynne
	from kimberley manning
2013-11-14	replace workqs with tasks for handling resume	David Gwynne
	from kimberley manning
2013-11-14	replace workqs with tasks for handling resume	David Gwynne
	from kimberley manning
2013-11-14	improve maths and conditionals around ticks to cope with wraparound better.	David Gwynne

2013-11-13	Do not map frame buffer resources the kernel doesn't use; only keep the	Miod Vallat
	lowest bitmap plane, and a page per plane to be able to probe the frame buffer depth. Saves an insignificant amount of kernel memory (about 4KB), still worth doing. ok aoyama@
2013-11-13	add missing headers here too, forgotten after testing the diff.	Jasper Lievisse Adriaanse

2013-11-13	DIOCGETSRCNODES was leaking a little bit more kernel information	Theo de Raadt
	ok benno
2013-11-13	Make fusebuf.fh_err signed, it might store negative errno values; ok syl	Stefan Sperling

2013-11-13	Add missing includes.	Sylvestre Gallon
	Unbreak armv7 compilation. ok rapha@ "commit the fix" from deraadt@
2013-11-13	Polling is done per controller not per interface.	Paul Irofti
	The controllers are linked through device->bus so the iface is redundant. So fixing usbd_dopoll() to take the device as argument and making usbd_interface2device_handle() private (for now) inside usbdi_util. Tested and okay mpi@
2013-11-13	In in6_update_ifa(), don't set the TENTATIVE flag on an IPv6 address that	Stefan Sperling
	is marked as NODAD. Since we're not gonna do duplicate address detection for this address, the TENTATIVE flag won't get cleared, rendering the address unusable. The existing logic only checked whether DAD was enabled for the interface, but DAD can be configured on a per-address basis. ok sthen@ as part of a larger diff
2013-11-13	fix typo in last commit	Theo de Raadt

2013-11-13	boot(): Don't forget calling if_downall() and uvm_shutdown() on arm ports.	Masao Uebayashi
	"Unification good" deraadt@
2013-11-13	ENOMEM is probably more appropriate than 0 in this error case.	Mike Larkin

2013-11-13	Use of uninitialized variable. There are obvious locking problems also	Mike Larkin
	present in this function, but I've been advised to walk away, and it have been this way forever (and this code is not even enabled in GENERIC by default anyway) original bug found by Maxime Villard, thanks.
2013-11-13	Use DL_GETPOFFSET() to get partition offset, except where we	Kenneth R Westerback
	explicitly check p_offseth before assigning p_offset to local variables. Add missing check for p_offseth.
2013-11-13	Honor RB_TIMEBAD on vax too as other ports.	Masao Uebayashi
	OK miod@
2013-11-12	Oops. Use %llu for DL_GETDSIZE() value.	Kenneth R Westerback

2013-11-12	two ioctl's were disclosing kernel pointers and such.	Theo de Raadt
	ok henning benno
2013-11-12	avoid math on void * pointers	Theo de Raadt
	(a few years ago, people went far too void * happy, it was like a drug or something)
2013-11-12	Implement sd card detection. Fixes the `sdmmc0: can't enable card' on bbb	Federico G. Schwindt
	when there is no card. Tested on pandaboard by patrick@, ok patrick@ and syl@
2013-11-12	Use daddr_t to hold calculated disk sector address. %d -> %lld to	Kenneth R Westerback
	match. Whitespace tweaks. All bringing this chunk into identity with same chunk in sparc64 fd.c.
2013-11-12	Tweak comment to explicitly mention that disk blocks are DEV_BSIZE	Kenneth R Westerback
	(a.k.a. 512-bytes) as far as ffs is concerned.
2013-11-12	Use %llu+DL_GETPSIZE() to show partition size. Replace %li+(long) with	Kenneth R Westerback
	%u for u_int32_t d_secsize.
2013-11-12	We always make symlinks to 'machine' and 'm88k' to compile.	Kenji Aoyama
	ok miod@
2013-11-12	Fix the bootloader random hang-up while counting down on LUNA-88K2.	Kenji Aoyama
	The same logic is also applied to luna88k/dev/timekeeper.c. suggested and ok miod@
2013-11-12	Use DL_GETDSIZE() to get disk size.	Kenneth R Westerback

2013-11-12	Use DL_[GET\|SET]POFFSET() and DL_[GET\|SET]PSIZE() to get and set	Kenneth R Westerback
	partition offsets and sizes.
2013-11-12	try bpf.c r1.84 again, this time without semantic changes to if statements.	David Gwynne
	cheers to sthen@ and krw@ for properly dealing with the fallout of my first commit.
2013-11-11	Revert bpf.c 1.84 / bpfdesc.h 1.19 for now, "panic: timeout_add: to_ticks (-1)	Stuart Henderson
	< 0" seen by RD Thrush, http://article.gmane.org/gmane.os.openbsd.bugs/20113 where he has a long-running process using bpf which is active at the time of panic. krw@ agrees with reverting for now.
2013-11-11	Replace sc_dying in favour of usbd_is_dying() and usbd_deactivate().	Paul Irofti
	Okay mpi@
2013-11-11	There is no fusebuf(9) man page anymore. Remove references to it from comments.	Stefan Sperling
	ok syl@
2013-11-11	Replace sc_dying in favour of usbd_is_dying() and usbd_deactivate().	Paul Irofti
	Okay mpi@
2013-11-11	Kill commented out code dealing with a non existent sc_dying variable.	Martin Pieuchot

2013-11-11	Clean-up the activate routines from the uoak*(4) drivers.	Paul Irofti
	Leave the deactivate bits to the parent. Simplify by making a single dying check in uhidev(4) interrupt routine instead of having each child doing it. Okay mpi@
2013-11-11	Replace most of our formating functions to convert IPv4/6 addresses from	Martin Pieuchot
	network to presentation format to inet_ntop(). The few remaining functions will be soon converted. ok mikeb@, deraadt@ and moral support from henning@