summaryrefslogtreecommitdiffstats
path: root/sys/dev
AgeCommit message (Collapse)AuthorFilesLines
2013-11-05Instead of comparing the lower and higher addresses of all the multicastmpi5-34/+21
entries to decide if the IFF_ALLMULTI flag should be set, check if there is at least one real range between them. This should not introduce any behavior change but will help changing our representation of multicast entries.
2013-11-05Fix RAID levels 0, 4, 5, and 6 with partitions larger than 2TB.reyk3-8/+9
A 64bit bit operation with the 32bit strip size could overflow and result in ridiculously small volumes when using large partitions (eg. 4x 3TB in RAID 5 resulted in a ~2TB volume). It is fixed by casting the strip size to an unsigned 64bit value. ok tedu@ millert@ deraadt@
2013-11-04Enable locking of fields using the acpi global lock if required. Seems to fixkettenis1-3/+3
a couple of machines and so far caused no regressions. Pt this in the tree now to get wider testing. ok deraadt@
2013-11-04substantial namespace cleanup. Might go a little bit too far, but wederaadt4-82/+77
can expose some of the kernel structures with split .h files if need be. Discussed with various, including jsing.
2013-11-04remove iop(4) driver. it is entirely unmaintained and supports somethingderaadt12-5840/+1
which basically doesn't exist at all. reminded about it by brad
2013-11-04The radeondrm code that sets the DPMS mode may sleep, so it should not be runkettenis2-8/+29
from a timeout like our burner code does. Hand this off to a taskq. tracked down by claudio@ ok jsg@
2013-11-04Convert wskbd_set_mixervolume() to use a task internally instead ofmpi5-48/+60
being called in a workq. ok kettenis@
2013-11-04Use the async version of uhidev_set_report(), there's no need to sleepmpi1-2/+3
until the transfer is done when submitting a command since the answers are read from the interrupt and we cannot do much if it fails. ok sasano@
2013-11-04regenjasper2-4/+9
2013-11-04add vmware vendor idjasper1-1/+2
ok jsg@
2013-11-04Better defaults for the screen burner settings: all unblank actions enabled,miod2-5/+6
but burning still disabled by default. Setting display.screen_off to a nonzero value with wsconsctl will be enough to enable screen blanking after the given time.
2013-11-03Simplify jme_miibus_readreg() / jme_miibus_writereg() a bit by using thebrad1-16/+7
mii_attach() parameter to be able to specify the location of the PHY. From FreeBSD ok jsg@
2013-11-03Enable TX checksum offload.brad1-7/+3
ok naddy@
2013-11-02Replace drm_handle_create/delete with drm_gem_handle_create/delete and makekettenis7-117/+125
the necessary adjustments to reference counting of GEM objects. Matches what Linux does these days.
2013-11-02replace rc4 with ChaCha20 here, too; ok djm, tedu, deraadtmarkus1-70/+154
2013-11-02Fix reference leak in i915_gem_mmap_ioctl().kettenis1-6/+7
2013-11-02Remove a layer of abstraction for fetching a configuration descriptor.mpi3-31/+12
2013-11-02Begin conversion of USB stack to return int instead of usbd_status,jeremy4-19/+19
starting with usbd_endpoint_count and usbd_interface_count. usbd_status should only be used to indicate the status of a USB transfer, it should not be used for generic error codes. Idea from and much hand-holding by mpi@ OK mpi@
2013-11-01Spacing.mpi1-18/+15
2013-11-01Sprinkle (long long) casts where %lld is being used to print daddr_tkrw8-29/+37
variables. Some random whitespace/knf repairs encountered on the way. ok miod@ on inspection, feedback & more suggestions from millert@
2013-11-01Reduce the differences with ehci(4)'s transfer decriptors. This is ampi2-181/+147
first step to merge the code of our controllers that deal with usbd transfers. Tested with miod@
2013-11-01Do not abuse the stack of the current process to prepend a report IDmpi2-22/+28
to the report request, use malloc(9) with the appropriate size like it is done in the synchronous version. ok miod@
2013-11-01Move the abort_task storage to the generic USB xfer structure insteadmpi7-31/+26
of defining it in every controller specific structure.
2013-11-01move fd(4) from directly using disksort to using bufqs.dlg1-15/+18
tested by me and rapha@
2013-10-31Pipe don't have states. Kill an useless & commented out chunk of code.mpi1-8/+2
No functional change.
2013-10-31usbd_pipe2device_handle() is not used, should not be used and die.mpi2-9/+2
2013-10-31replace workqs with tasks. makes the code a bit simpler cos adding an addeddlg1-24/+17
task is a nop. tested by and ok jmatthew@
2013-10-30Fix jagged diagonal lines (kernel part)shadchin3-17/+11
Send WSCONS_EVENT_SYNC every time you call wsmouse_input(). Used to synchronize and separate events into packets of input data changes occurring at the same moment in time. For example, motion of a mouse may set the DELTA_X and DELTA_Y values for one motion, then emit a SYNC. ok matthieu@. tested edd@, Henri Kemppainen and Alf Schlichting.
2013-10-30regenaoyama2-4/+9
2013-10-30Treat another OMRON UPS, BY35S, as ugen(4).aoyama2-2/+4
Now NUT (sysutils/nut in ports) works on this device, and may work on other OMRON UPSs, by: 1. Apply following diff to NUT source: https://gist.github.com/ao-kenji/7070817 2. Put following setting in /etc/nut/ups.conf. --- [by35s] driver = blazer_usb port = auto desc = "OMRON BY35S" vendorid = 0590 productid = 0080 subdriver = ippon --- Reference: http://sunnyone41.blogspot.jp/2011/05/omron-by50subuntu-1104.html (written in Japanese) ok deraadt@ mpi@
2013-10-30Replace workqs with tasks.mpi3-5/+18
ok dlg@
2013-10-30Add "handling" for dock/undock events for non-ACPI docks to acpithinkpad.mlarkin1-1/+5
The "handling" is just to consume the event so that no dmesg output is printed. Since these docks have no bays, there isn't much to do on these events anyway. ok jcs@
2013-10-30replace the workq bits to supply new tx pkt descriptors with a task.dlg2-14/+11
tested locally on a dell poweredge 2950
2013-10-30remove #include <sys/workq.h> cos this driver doesnt use it.dlg1-2/+1
2013-10-30replace workqs with tasks.dlg2-6/+6
ok mpi@
2013-10-30replace workq_add_task with the task api.dlg1-3/+5
this guarantees the reliability of the arc4_reinit task being run by not relying on a pool to give us memory in an interrupt context. ok mpi@
2013-10-30deprecate taskq_systq() and replace it with extern struct taskqdlg15-55/+55
*const systq defined in task.h this reduces the cost of using the system taskq and looks less ugly. requested by and ok kettenis@
2013-10-29Don't add sectors to blocks. Use DL_SECTOBLK() to change DL_GETPOFFSET()krw1-2/+5
sector value to 512-byte-blocks before adding to b_blkno. gcc tested on zaurus by matthieu@.
2013-10-29Kill sc_attached, attach and detach events for USB devices are triggeredmpi4-19/+4
by the same explore task and cannot happen in parallel.
2013-10-29Move most of the uses of workqs in drm to the new task/taskq api.jsg27-126/+137
Prevents unintended multiple additions to workqs that was causing hangs on radeon, and lets us remove tasks more closely matching the behaviour of the original linux code. ok kettenis@ cause of the ttm/radeon hangs debugged by claudio@ and kettenis@
2013-10-29implementing handling of the msg ring if its available. this lets us getdlg1-9/+156
hotplug events from the hypervisor, so you can add and remove disks at runtime. movement on the msg ring is detected in the interrupt handler and deferred to a thread via tasks.
2013-10-28Use the correct value for the Interframe Gap Time 0 bit in the transmitmpi1-3/+3
configuration register and fix a typo. From Paul A. Patience.
2013-10-28tedu FDDI support and the 3 flavors the driver for DEC devices, evenmpi10-3904/+3
miod@ cannot find two boards using the same media. With precious punctuation review from guenther@, thanks! ok deraadt@, henning@
2013-10-26Use the "activate" framework rather a shutdown hookderaadt1-6/+21
ok jmatthew dlg pelikan
2013-10-26Unstub i915_gem_dumb_destroy() and radeon_mode_dumb_destroy().kettenis4-31/+43
2013-10-25Fix/re-enable RX checksum offload for 8168C/8168CP revisions after revision rev 1.140.brad1-7/+6
The relevant code for enabling the HW's RX checksum offload support was checking for the TX feature flag to know when to enable RX checksum offload but that flag was removed/disabled with rev 1.140 which resulted in the RX checksum offload support inadvertently being disabled. ok naddy@
2013-10-25fix an off by one when calculating the length of an sgl segmentmikeb1-5/+4
that our chain scatter-gather element is pointing to. the bug was observed by pedro martelletto with some particular firmware doing raid 0. the fix wouldn't have been possible without extensive debugging and spec conformance verification done by pedro. tested by pedro, dlg and myself, ok dlg
2013-10-25Add support for Microsoft XBox 360 controller as a uhid. It doesn't usejeremy2-5/+120
the standard interface class and doesn't have a report descriptor, so use a manually created one. The controller shows up as multiple devices with different interface numbers, so only match the first one as a uhid. The controller shows up as both a uhid and a ugen. Rename ugraphire_rdesc.h to uhid_rdesc.h and add the manually created Xbox 360 report descriptor to the file. Add ifndef SMALL_KERNEL around the related blocks for the XBox 360 controller and Wacom Graphire, so this should actually shrink the ramdisks. Manual report descriptor taken from a 2005 FreeBSD commit. Much help from mpi@ for getting this suitable for commit to OpenBSD. OK mpi@
2013-10-25regenjeremy2-2/+7
2013-10-25Add Microsoft XBox 360 Controllerjeremy1-1/+2
OK mpi@
Copyright © 1996 – 2026 Jason A. Donenfeld. All Rights Reverse Engineered.