src - OpenBSD base system

Age	Commit message (Collapse)	Author
2011-06-26	kill mcd dead. ok krw matthew millert thib	Ted Unangst

2011-06-26	There are some problems with ppb msi interrupts. For instance, i386 in	Theo de Raadt
	some situations can run out of vectors. On amd64, there is at least one machine that fails to resume. Yes, those problems need to be fixed. But other problems should be found, too. If we leave ppb msi off, we won't learn anything new. So use an #ifdef to enable ppb msi for non-i386, temporarily. Kind of discussed this with kettenis.
2011-06-26	if_downall() cannot be done MI in sys_reboot() because vfs_shutdown()	Theo de Raadt
	might need network (ie. nfs). Move the call to the MD boot() routines. This cause for boot hangs diagnosed by kettenis.
2011-06-26	First stab at MSI support for vpci(4). Still a few rough edges, but seems	Mark Kettenis
	to work fine on a t1k and a t5120, so let's enable this right from the start.
2011-06-26	1.500 was a good revision, let's stay there a little longer. (revert ↵	Ted Unangst
	accidental commit)
2011-06-26	I see you hiding, scd	Ted Unangst

2011-06-26	remove remnants of scd device, discovered by Frederic Perrin	Ted Unangst

2011-06-26	In the shared interrupt handler, return the pil and ci_handled_intr_level	Theo de Raadt
	to the level they were at before we started splraise'ing for the various handlers. with jsing, ok kettenis
2011-06-26	constrain properly which machines get this firmware	Theo de Raadt

2011-06-26	Fix some inconsistencies in how we build 64-bit addresses from OF properties.	Mark Kettenis

2011-06-26	Implement pci_intr_map_msi().	Mark Kettenis

2011-06-26	Missed a parameter in the pci_msi_setmsiq hypervisor call.	Mark Kettenis

2011-06-25	Add MSI hypervisor calls.	Mark Kettenis

2011-06-25	Don't forget to print a newline at attach time.	Miod Vallat

2011-06-25	Remove irrelevant comments borrowed from loongson.	Miod Vallat

2011-06-25	ansi, no binary change	Jonathan Gray

2011-06-24	printf -> DPRINTF in the SIGKILL path	Damien Miller

2011-06-24	When converting endian to same-endian, the macros should cast to unsigned	Theo de Raadt
	so that signedness of the input is suppressed. ok matthew miod kettenis guenther
2011-06-24	machdep.kbdreset enables a shutdown by Ctrl-Alt-Del on amd64 and	Christian Weisgerber
	i386. Stop abusing it on other archs for controling a shutdown by pressing the soft power button: * Add a MI sysctl hw.allowpowerdown; if set to 1 (the default) it allows a power button shutdown. * Make acpi(4)/acpibtn(4) honor hw.allowpowerdown. * Switch the various power button intercepts on landisk, sgi, sparc64 and zaurus over to hw.allowpowerdown. * Garbage collect the machdep.kbdreset sysctl on all archs other than amd64 and i386. ok miod@
2011-06-24	We should only ack a shared interrupt once, after we've run all the handlers.	Mark Kettenis
	So let the intr_list_handler do this, and prevent the real handlers from doing it as well. tested by deraadt@
2011-06-24	When reading the value from a simple lock, we do not need to ldcws into a	Joel Sing
	__cpu_simple_lock_t - gcc most likely ignores the alignment for a stack variable anyway. Also remove unnecessary initialisation. ok kettenis@ miod@
2011-06-24	Avoid a possible null dereference.	Joel Sing
	ok marco@
2011-06-24	Should always include ukphy(4) in the kernel configs if using any devices	YASUOKA Masahiko
	using MII. This diff from brad@ Build test and ok yasuoka@
2011-06-24	Remove SET/CLR/ISSET macros from Octeon code. The kernel already provides	YASUOKA Masahiko
	these macros. This diff from brad@ built test and ok yasuoka@
2011-06-23	Back out inadvertently commited change.	Owain Ainsworth
	This one wanted some more comments before commit.
2011-06-23	ansi	Theo de Raadt

2011-06-23	Don't leak the ctrl_data in uvideo_queryctrl, uvideo_s_ctrl and uvideo_g_ctrl.	Owain Ainsworth
	If one of the usb calls we did here failed we'd return immediately and not free our buffer. ok miod@
2011-06-23	make malloc calls in isp_pci_mbxdma NOWAIT to match the bus_dma calls that are	Owain Ainsworth
	all nowait. While here, plug a memory leak if one of the bus_dma calls near the bottom of the function fails because it would fail to free the dmamap, unmap the memory and free the memory. make that part of the code use the standard list of goto labels for error handling that most callers of bus_dmamem_alloc; bus_dmamem_map; bus_dmamap_creat; bus_dmamap_load() use to clean up. ok miod@
2011-06-23	free(null) ist verboten in the kernel.	Owain Ainsworth
	The addition of M_CANFAIL here made it possible (if rather unlikely) that we'd fail and goto done, where we would free(NULL). protect the free with a null check to prevent this. ok miod@
2011-06-23	Check for the correct flag when checking to see if the page is part of an aobj.	Owain Ainsworth
	This is no function change since aobjs never actually hit this path. (also it is my bug from a while ago). ok ariane@
2011-06-23	Don't bother checking for an empty queue before calling uvm_pglistfree.	Owain Ainsworth
	It will handle an empty list just fine (there's a small optimisation possible here to avoid grabbing the fpageqlock if no pages need freeing, but that is definitely another diff) ok ariane@
2011-06-23	Replace handrolled version of uvmfault_unlockmaps with uvmfault_unlockmaps.	Owain Ainsworth
	ok ariane@
2011-06-23	Move uvm_pglistalloc and uvm_pglistfree to uvm_page.c and garbage	Owain Ainsworth
	college uvm_pglist.c uvm_pglistalloc and free are just thin wrappers around pmemrange these days and don't really need their own file. ok ariane@
2011-06-23	Make mbufs and dma_alloc be contig allocations.	Ariane van der Steldt
	Requested by dlg@ ok oga@
2011-06-23	Fix the error path in bus_dmamem_map.	Ariane van der Steldt
	As discussed on icb: remove the comment, remove pmap_remove (uvm_km_free does that for us). ok oga@, deraadt@
2011-06-23	Set pd->af very early in pf_setup_pdesc() since the AF is used in	Claudio Jeker
	the pflog code and not setting it will cause a panic in pf_setup_pdesc(). Fixes dhill's crashes with funky v6 packets. OK bluhm@
2011-06-23	Simplify softraid(4) to behave like other HBAs by using a single	Matthew Dempsky
	scsibus(4) for all attached disks, rather than one scsibus(4) per disk. This takes advantage of recent improvements in the SCSI midlayer to better support hotplug devices, and also decouples softraid(4) from scsibus(4) device unit numbers. ok jsing@, dlg@, krw@, deraadt@; marco@'s still unconvinced
2011-06-23	Eliminate ramdisk.c and ramdisk.h now that they're unused.	Matthew Dempsky

2011-06-23	Complete rewrite for rd(4) to eliminate unnecessary functionality and	Matthew Dempsky
	better follow current disk driver conventions. Prompted by oga@. "go ahead" deraadt@
2011-06-23	a bit more ansi; ok tedu	Theo de Raadt

2011-06-23	ansify a few files. ok deraadt dlg krw matthew	Ted Unangst

2011-06-23	cope with empty rx rings by scheduling a timeout to keep trying until it	David Gwynne
	gets some packets onto the rings. also annoying, but the hardware doesnt report empty rings, we have to handle it ourselves.
2011-06-23	this chip has an annoying "feature" where it cannot report the link	David Gwynne
	state unless the chip is up and handling packets. while its down it does not report the link state, so it is unknown. this tweaks the link state handling, in particular it adds code to myx_down so it moves the link state to unknown, ie, it correctly reflects reality. stupidity pointed out by deraadt
2011-06-22	reset the tx_count on UP, since it may have been advanced from non-zero	Theo de Raadt
	by a previous use ok claudio
2011-06-22	kill a few more casts that aren't helpful. ok krw miod	Ted Unangst

2011-06-22	Print correct function name in debug statements. ata_get_params, not	Kenneth R Westerback
	ata_get_parms or wdc_ata_get_parms.
2011-06-22	msi support. this is a complicated one...	David Gwynne
	ok kettenis@
2011-06-22	another myri10ge device matched by freebsd/linux drivers	Jonathan Gray
	ok dlg@
2011-06-22	regen	Jonathan Gray

2011-06-22	add a few more devices	Jonathan Gray