summaryrefslogtreecommitdiffstats
path: root/sys/dev
AgeCommit message (Collapse)AuthorFilesLines
2015-08-28Disable interrupts while the midi uart is not in use. Avoids generatingratchov2-11/+45
unused interrupts when a chatty peripheral is connected but is not used.
2015-08-28Fix compiling a kernel without NBPFILTER > 0.reyk1-2/+5
OK mikeb@
2015-08-28Fix support of M-Audio Delta 44 cards that use different GPIO pins,ratchov1-15/+24
than other Delta cards.
2015-08-28fairly simple sizes for free(); ok teduderaadt14-40/+40
2015-08-27size for free()deraadt1-2/+2
2015-08-27fairly simple size for free()deraadt1-2/+2
2015-08-27sizes for free()deraadt1-5/+5
2015-08-27simple sizes for free()deraadt1-5/+6
2015-08-27simple size for free()deraadt1-2/+2
2015-08-27simple sizes for free()deraadt1-6/+5
2015-08-26Simple size for free(). These are a pleasure to convert when thederaadt1-2/+2
allocation is clearly spelled out as ptr = malloc(sizeof(*ptr), ...)
2015-08-26two simple sizes to free()deraadt1-3/+3
2015-08-26trivial size to free()deraadt1-2/+2
2015-08-26trivial size to free()deraadt1-2/+2
2015-08-26Get rid if em_align. This approach used to make sense, but now that thekettenis2-69/+16
hardware rx mtu always gets set to the maximum supported value we will hit it for every received packet. Instead, use a larger mbuf cluster size on strict alignment architectures such that we can always m_adj to make sure the packets are properly aligned. This wastes some memory but simplifies things considerably. Hopefully we can reduce the spillage in the near future by taking advantage of recent improvements in the pool code. ok mpi@, mikeb@, dlg@
2015-08-25trivial size argument for free()deraadt1-2/+2
2015-08-25obvious sizes for free()deraadt1-3/+3
2015-08-25sizes for simple free() callsderaadt2-4/+4
2015-08-24Audio on the T400/T410/T510/T420/T520/X220/X220t docks needs a quirkjsg2-2/+38
for the pin configuration as well. From Dmitry Alenichev. ok mlarkin@
2015-08-24Add some additional urtwn ids found in the linux rtl8192cu driver.jsg1-1/+6
ok stsp@
2015-08-24regenjsg2-4/+29
2015-08-24Add some additional urtwn ids found in the linux rtl8192cu driver.jsg1-1/+6
ok stsp@
2015-08-23Make Atom S1200 UART work; Atanas Vladimirov has one.deraadt1-1/+8
2015-08-23fairly trivial sizes for free()deraadt1-7/+12
2015-08-22Make urtwn(4) attach to Netgear WNA1000Mv2. From Mark Willson.stsp1-1/+2
2015-08-22regenstsp2-4/+9
2015-08-22Add device ID for Netgear WNA1000Mv2. From Mark Willson via misc@stsp1-1/+2
2015-08-21Run the part of the interrupt handler that does rx completion without holdingkettenis2-9/+37
the kernel lock. ok mpi@, dlg@
2015-08-21enable beep and CD controls on ALC292jsg1-1/+2
2015-08-20Respect the access size when reading or writing to pci config space.kettenis2-14/+86
This version also makes sure writes are properly aligned, preventing the panics seen with the previous incarnation of this change. Fixes battery status passthrough in vmware, brightness keys on my x220 and who knows what else. ok mlarkin@, deraadt@
2015-08-20Enable audio on Thinkpad docks. Help from jsg@ for the *40 / *50 seriesmlarkin2-2/+74
docks. ok jsg@
2015-08-20remove unused vesafb code, had been commented out and not compiled in formlarkin9-1136/+2
7 years (for some parts, 9 years), and was only ever referenced in i386 anyway. ok jsg@, matthieu@, armani@
2015-08-19Stop passing around constants for metadata size and location. Justkrw1-12/+9
use the constants where needed. ok jsing@
2015-08-18Add missing trailing to error message:stsp1-2/+2
pckbc0: unable to establish interrupt for irq 12pckbd0 at pckbc0 (kbd slot) ok miod@
2015-08-17The 80's called and want their interface back. Remove last fragments ofkrw7-33/+26
ST-506 support. Remove redundant 'precomp' variables/fields and rename remaining few to 'features' to be consistant with the rest of the code. Consensus that no ST-506 disks (requiring ISA cards, <120MB, usually rife with bad sectors that can't be worked around) are still running OpenBSD.
2015-08-15Do not perform timeout maintainance on sc->poll_timeout if it has not beenmiod1-5/+6
initialized yet.
2015-08-15do the global tx free accounting in myx_start with a single atomic opdlg1-4/+8
instead of one per packet. seems to let me send packets a little faster.
2015-08-15rework the tx path to use a ring to keep track of dmamaps/mbufs.dlg1-212/+169
this removes the myx_buf structure and uses myx_slot instead. theyre the same expcet slots dont have list entry structures, so theyre smaller. this cuts out four mutex ops per packet out of the tx handling. just have to get rid of the atomic op per packet in myx_start now.
2015-08-14move to a per rx ring timeout for refilling empty rings.dlg1-81/+54
this lets me get rid of the locking around the refilling of the rx ring. the timeout only runs refill if the rx ring is empty. we know rxeof wont try and refill it in that situation because there's no packets on the ring so we wont get interrupts for it. therefore we dont need to lock between the timeout and rxeof cos they cant run at the same time.
2015-08-14rework how we track the packets on the rx rings.dlg1-231/+235
originally there were two mutex protected lists for rx packets, a list of free packets, and a list of packets that were on the ring. filling the ring popped packets off the free list, attached an mbuf and dmamapped it, and pushed it onto the list of active packets. the hw fills packets in order, so on rx completion we'd pop packets the active list, unmap the mbuf and shove it up the stack before putting the packet on the free list. the problem with the lists is that every rx ring operation resulted in two mutex ops. so 4 mutex ops per packet after you do both fill and rxeof. this replaces the mutexed lists with rings that shadow the hardware rings. filling the rx ring pushes a producer index along, while rxeof chases it with a consumer. because we know only one thing can do either of those tasks at a time, we can get away with not using atomic ops for them. there's more to be done, but this is a good first step.
2015-08-12Remove last two 'uses' of d_subtype in tree. No point in settingkrw1-2/+1
it to 0 immediately after a bzero() of the entire label. Nor in swapping endianness since nobody looks at it. Don't touch the actual field in disklabel. Yet. ok deraadt@ as part of larger diff that needs splitting up.
2015-08-12Merge two return branches in vmt_tclo_tick(). No functional changes.uebayasi1-3/+3
2015-08-12removes two mem leaks in ti (in error path).semarie1-5/+5
one leak reported by Maxime Villard initial patch from Christian Schulte (with spelling corrections in comments) second leak corrected by myself ok kettenis@
2015-08-12update author information of the copyright section.yuo1-2/+2
no binary update.
2015-08-12Use a linker script when building i386 kernels to allow section padding onmlarkin1-4/+7
PAGE_SIZE boundaries. This is required to enforce proper separation of sections when adding page protections (coming shortly). This was in snaps before release with no reported side effects. ok deraadt@
2015-08-12Check proper HPET period value during attach.mlarkin2-4/+9
Pointed out by patrick keshishian <pkeshish at gmail.com>, thanks.
2015-08-11Don't inline long functions as this tends to increase object sizeratchov1-4/+4
with no benefit. ok millert, "correct" deraadt
2015-08-11fix regressions introduced in my previous commmit, resulting injung1-10/+24
'unknown command' dmesg lines for TEMPer devices and a possible type command loop for unsupported TEMPerHUM devices, moreover, 2 whitespace fixes while here tested by myself and frederic cambus ok mpi@
2015-08-05Fix i217 PHY initializationsf2-2/+104
Add a workaround for a hardware bug were receiving of packets would stop until the battery of the laptop was removed. The problem could be triggered if the LAN cable was plugged in after the PHY had already been initialized. The workaround is already present in FreeBSD (e1000_ich8lan.c: e1000_lv_jumbo_workaround_ich8lan()) and Linux. The commit messages of b20a7744 and da1e2046 in Linux have some more details. Found and ported from FreeBSD by Valentin Dornauer. Tested by many. ok deraadt@
2015-08-04Skip C2 and C3 states from the FADT if the cpu doesn't have ARAT.guenther1-3/+8
problem reported by patrick keshishian (pkeshish (at) gmail.com) based on diff from mlarkin@, ok mlarkin@ deraadt@
Copyright © 1996 – 2026 Jason A. Donenfeld. All Rights Reverse Engineered.