src - OpenBSD base system

Age	Commit message (Collapse)	Author
2016-08-26	The *_HEAD_INITIALIZER() macros are documented as taking the struct,	Philip Guenther
	not a pointer to it
2016-08-26	Pull in <sys/time.h> for clock_ymdhms bits	Philip Guenther
	ok deraadt@
2016-08-26	correct INTR_MSG_MASK to include INTR_MSG_1	Jonathan Gray
	ok dlg@
2016-08-25	Remove a check to expire received prefix.	Martin Pieuchot
	This check does not make sense since November 2000 when IPv6 autoconf address deletion has been made independent from prefix lifetimes. Fix a case when a route was added to the table but the corresponding address was not, leaving v6 unusable. Found the hardway by and ok sthen@
2016-08-25	Spliced TCP sockets become faster when the output part is running	Alexander Bluhm
	as its own task thread. This is inspired by userland copy where a process also has to go through the scheduler. This gives the socket buffer a chance to be filled up and tcp_output() is called less often and with bigger chunks. When two kernel tasks share all the workload, the current scheduler implementation will hang userland processes on single cpu machines. As a workaround put a yield() into the splicing thread after each task execution. This reduces the number of calls of tcp_output() even more. OK tedu@ mpi@
2016-08-25	Completely revert the M_WAIT change on the cluster allocation and	Alexander Bluhm
	bring back the behaviour of rev 1.72. Although allocating small mbufs when allocating an mbuf cluster fails seems suboptimal, this should not be changed as a side effect when introducing m_getuio(). OK claudio@
2016-08-25	Simplify ip6_getpmtu() to use a 'struct rtentry *' instead of two	Martin Pieuchot
	'struct route_in6 *'. This is another little step towards deprecating 'struct route{,_in6}' Inputs from and ok bluhm@
2016-08-25	Enable the UWXN bit in the SCTRL register when available. This should	Mark Kettenis
	prevent the kernel from accidentally executing userland pages that are writable. ok jsg@, patrick@
2016-08-25	pool_setipl for udf	David Gwynne
	ok phessler@ krw@
2016-08-25	pool_setipl	David Gwynne
	ok kettenis@
2016-08-25	pool_setipl	David Gwynne
	ok kettenis@
2016-08-24	Replace pmap_fault_fixup() with an access flag fault handler on armv7.	Mark Kettenis
	ok tom@
2016-08-24	pool_setipl for oce(4)	David Gwynne
	ok mikeb@
2016-08-24	Kill ip6_forward_rt reducing differences between v4 and v6.	Martin Pieuchot
	A single forwarding cache is not the answer. The answer is 42... err PF! ok bluhm@
2016-08-24	Use rtalloc(9) directly instead of in6_selectsr() in icmp6_reflect().	Martin Pieuchot
	This is another little step towards deprecating 'struct route{,_in6}'. ok bluhm@
2016-08-24	pool_setipl for cardbus and drm pools.	David Gwynne
	ok kettenis@
2016-08-23	rename nfiles to numfiles to avoid shadowing and stretch out the name.	Ted Unangst
	ok deraadt
2016-08-23	Deassert all reset signals for the controller. Add support for more recent	Mark Kettenis
	SoCs that have a separate clock for each PHY.
2016-08-23	Add functions to assert/deassert all reset signals for a device.	Mark Kettenis

2016-08-23	Add support for the usb clock on sun5i-a13 as well.	Mark Kettenis

2016-08-23	re-enable acpicbkbd by default now that acpiec won't try burst mode	Joshua Stein
	on the chrome ec
2016-08-23	don't enter burst mode for single-byte reads and writes.	Joshua Stein
	avoids problems on hardware with broken or unimplemented burst mode and isn't really necessary for such small transactions anyway. matches what linux and freebsd have done for a long time. tested in snaps ok deraadt kettenis
2016-08-23	The device trees for sun8i and sun9i no longer include an address in the	Mark Kettenis
	name of the /soc node. Leave it off, as the device path will still match on nodes that include it.
2016-08-23	Actually make fdt_find_node() return NULL if the node couldn't be found.	Mark Kettenis
	ok tom@
2016-08-23	Update the L2 content of a RTF_CACHED entry instead of going though	Martin Pieuchot
	a create/delete/insert cycle as such entry cannot be deleted when referenced. Regression reported by and ok bluhm@
2016-08-23	The TLB refill and XTLB refill exceptions use distinct exception vectors	Visa Hankala
	on Loongson 3A. Consequently, the kernel has to set up both vectors to prevent panics with virtual address references. On Loongson 2F, it is enough to set up the TLB refill vector because the XTLB refill exception uses the same vector address. ok miod@
2016-08-23	pool_setipl	David Gwynne

2016-08-23	Use rtalloc(9) directly instead of in6_selectsr() in NS/NA output	Martin Pieuchot
	routines. This is another little step towards deprecating 'struct route{,_in6}'. ok bluhm@
2016-08-23	Do not use a single global struct route_in6 to cache route lookups.	Martin Pieuchot
	This is a little step towards deprecating 'struct route{,_in6}'. ok jca@, claudio@
2016-08-23	Implement support for DT_INIT_ARRAY, DT_FINI_ARRAY and DT_PREINIT_ARRAY.	Mark Kettenis
	Don't skip DT_INIT and DT_FINI for the main executable. This matches what Linux and Solaris do. ok guenther@
2016-08-23	pool_setipl for tmpfs.	David Gwynne

2016-08-23	remove duplicated test for em_82574	Jonathan Gray

2016-08-23	Convert %q to %ll in format strings	Philip Guenther
	ok natano@ krw@
2016-08-22	Fix two small bugs in the new reset API code.	Mark Kettenis

2016-08-22	Remove unused code and definitions.	Mark Kettenis

2016-08-22	Use new clock API.	Mark Kettenis

2016-08-22	Use new clock API to manipulate clocks.	Mark Kettenis

2016-08-22	Add support for the gmac clock.	Mark Kettenis

2016-08-22	Add an API to set the clock frequency.	Mark Kettenis

2016-08-22	Move all the platform-specific code into a new function sxiehci_attach_phy()	Mark Kettenis
	making use of pinctrl, clock and reset APIs where appropriate.
2016-08-22	Add support for the usb clock. Also implements reset signal support.	Mark Kettenis

2016-08-22	Add a reset signal API alongside the clock API.	Mark Kettenis

2016-08-22	Use rtalloc(9) instead of ifa_ifwithnet() to find an interface	Martin Pieuchot
	when adding a route to gateway to ensure a most specific match. This makes "# route add" coherent to "# route get" even with p2p interfaces. Fix a problem reported by Mart TÃµnso. This also fix rttest20 after the introduction of RTF_CACHED. ok vgross@
2016-08-22	Make the ``rt_gwroute'' pointer of RTF_GATEWAY entries immutable.	Martin Pieuchot
	This means that no protection is needed to guarantee that the next hop route wont be modified by CPU1 while CPU0 is dereferencing it in a L2 resolution functions. While here also fix an ``ifa'' leak resulting in RTF_GATEWAY being always invalid. dlg@ likes it, inputs and ok bluhm@
2016-08-22	Do not dereference ``rt->rt_ifa'' after calling rtfree(9).	Martin Pieuchot
	This could result in a use after free if the route entry was holding the last reference of the address descriptor. ok jca@, bluhm@, claudio@
2016-08-22	Use the new clock API.	Mark Kettenis

2016-08-22	Implement interfaces to disable clocks and add interfaces that enable or	Mark Kettenis
	disable all clocks for a device. The latter interfaces are useful for devices that have multiple clocks that don't have specific names/purposes such as sxiahci(4).
2016-08-22	Call csignal() and selwakeup() from a KERNEL_LOCK'd task.	Martin Pieuchot
	This will allow us make bpf_tap() KERNEL_LOCK() free. Discussed with dlg@ and input from guenther@
2016-08-22	Sizes for free(9) from David Hill.	Martin Pieuchot

2016-08-22	Refactor the uio to mbuf code out of sosend and start to make use of	Claudio Jeker
	MCLGETI and large mbuf clusters. This should speed up local connections a fair bit. OK dlg@ and bluhm@ (after reverting the M_WAIT change on the cluster allocation)