summaryrefslogtreecommitdiff
path: root/sys/dev
AgeCommit message (Collapse)Author
2013-07-09move AHCI_DEBUG stuff into ahci.c so we can build with it enabled againJonathan Matthew
ok dlg@
2013-07-09Add basic support for digitisers with pens in uhts(4).Edd Barrett
With help from matthieu@ and mpi@. OK mpi@
2013-07-09Fix a pair of aesthetic nits in dev/rnd.c:Bret Lambert
fix language (feed -> fed) in a comment 16 -> MD5_DIGEST_LENGTH for legibility No change to resulting object file. ok djm@
2013-07-08make use of udelay to reduce the diff to linuxJonathan Gray
2013-07-08add inline versions of udelay/mdelayJonathan Gray
2013-07-08add DRM_IOCTL_MODE_CURSOR2 definition from libdrm 2.4.46Jonathan Gray
2013-07-06Remove unsigned comparison < 0.Brad Smith
Pointed out by LLVM. ok kettenis@
2013-07-06Advertise utpms(4) as being a WSMOUSE_TYPE_USB because even if it couldMartin Pieuchot
be declared as PANEL or SYNAPTICS it doesn't implement GCALIBCOORDS for the moment. This makes ws(4) happier as it can now uses the touchpad directly wich in turn makes my touchpad usable if the bluetooth HID mouse is not detected or ums(4) is disable.
2013-07-06Use wsemuldisplaydevsubmatch.Mark Kettenis
2013-07-06s/wsemuldisplaysubmatch/wsemuldisplaydevsubmatch/gMark Kettenis
2013-07-05Add submatch function for use with config_found_sm(9).Mark Kettenis
ok miod@
2013-07-05Provide a forward declaration of struct device.Mark Kettenis
ok deraadt@
2013-07-05Correctly abort and free the pipe when detaching the device. Found theMartin Pieuchot
hard way by otto@
2013-07-05make use of the drm_i915_private macro to reduce the diff to linuxJonathan Gray
2013-07-05Use tmp_int to be able to pass back the -1 upon error and check for it properlyBrad Smith
but the value passed forward throughout the ioctl handler otherwise is an unsigned value. Pointed out by LLVM. bktr_core.c:1737:13: error: comparison of unsigned expression < 0 is always false bktr_core.c:1779:13: error: comparison of unsigned expression < 0 is always false bktr_core.c:2063:16: error: comparison of unsigned expression < 0 is always false ok krw@
2013-07-04Format string fix: %x instead of %p for intStefan Fritsch
ok deraadt@
2013-07-04Fall back to bit banging mode for DVO transmitter detectionJonathan Gray
from linux 3.8.13
2013-07-04Fixup Oops in the pipe config computationJonathan Gray
from linux 3.8.13
2013-07-04ensure single initialization and cleanup of backlight deviceJonathan Gray
from linux 3.8.13
2013-07-04set CPT FDI RX polarity bits based on VBTJonathan Gray
from linux 3.8.13
2013-07-04Workaround incoherence between fences and LLC across multiple CPUsJonathan Gray
from linux 3.8.13 this does not currently do the ipi to run wbinvd() on all processors
2013-07-04Fix SDVO connector and encoder get_hw_state functionsJonathan Gray
from linux 3.8.13
2013-07-04Add no-lvds quirk for Fujitsu Esprimo Q900Jonathan Gray
from linux 3.8.13
2013-07-04Fix sdvo connector get_hw_state functionJonathan Gray
from linux 3.8.13
2013-07-03fix: cast void * to char *, fix format args for (u)int64_t,Stefan Fritsch
tc_frequency is unsigned ok kettenis@
2013-07-03cast time_t to long long, time_t is signed, tv_nsec is longStefan Fritsch
ok deraadt@
2013-07-03The ipmi threshold value may be negative. Respect the sign bitAlexander Bluhm
when checking wether an ipmi sensor is critial. From Matthias Pitzl; OK deraadt@
2013-07-02Rearrange the inequalities.Brad Smith
Pointed out by LLVM. if_axe.c:1059:29: error: comparison of unsigned expression < 0 is always false if_smsc.c:1250:28: error: comparison of unsigned expression < 0 is always false ok jsg@ mikeb@
2013-07-02If after processing an SCI event the SCI_EVT is no longer sent, we're doneMark Kettenis
and shouldn't do another SCI query. Such a spurious SCI query blocks on certain HP laptops, which in turn blocks the acpi thread. tested by many ok mlarkin@
2013-07-02change splvm to splbio, missed in r1.4Stefan Fritsch
noticed by Alexey Suslikov
2013-07-02some trivial format string fixesStefan Fritsch
2013-07-02Lower IPL to IPL_BIO for the interrupt.Stefan Fritsch
There is no reason to use IPL_VM and it breaks with the recent IPL_MPSAFE changes. discussed with kettenis@
2013-07-01Pass ioctl() calls to parent uhidev() device first, to be able toMatthieu Herrb
get the HID descriptor with usbhidctl -r, among others. ok mpi@
2013-07-01The fast scrolling code causes random page table errors on older hardware.Mark Kettenis
Disable that code and use the write-only rasops code instead on the affected chips.
2013-07-01When an I/O error occurs on a softraid chunk, only take it offline if theJoel Sing
discipline supports redundancy. In the non-redundant case, there is little to gain my failing the chunk, in fact it just makes any form of data recovery significantly harder. ok krw@ todd@
2013-07-01Remove vaddr_t casts in bktr_mmap().Brad Smith
Pointed out by LLVM. bktr_os.c:478:22: error: comparison of unsigned expression < 0 is always false ok krw@ kettenis@
2013-07-01Within ns_to_timeval() change the rem field type from unsigned to signed.Brad Smith
Pointed out by LLVM. drm_irq.c:154:10: error: comparison of unsigned expression < 0 is always false kettenis@ says it should be signed and this is what the equivalent Linux code does. ok jsg@
2013-06-28Add support for write-only framebuffers. This makes the code that copiesMark Kettenis
rows and columns redraw the destination area instead of reading pixel values from the framebuffer. This can be a considerable performance win. Only works together with RI_VCONS. ok miod@
2013-06-28send proper z value to elantech_send_input in previous v4 commitjoshua stein
pointed out by Owain G. Ainsworth
2013-06-28Turns out that BGE_STATFLAG_UPDATED bit in the status block doesn't getMike Belopuhov
properly updated by the newer hardware (seen in the TX completion case). This leads to very poor transmit performance in the beginning of a TCP connection. Linux and FreeBSD don't rely on BGE_STATFLAG_UPDATED bit since they enable MSI and tagged status for 5717+. Doing the same does indeed fix an issue. Change was tested by David Imhoff on 5719, 5720 and 5721/5750, Hrvoje Popovski on 5704 B0, sthen@ on 5723/5784, benno@ on 5704 A3, and me on 5719, 5720 adn 5714/5715. No objections from kettenis@ and dlg@.
2013-06-28Rewrite the media / link state handling and have itBrad Smith
handled within sis_miibus_statchg() instead of calling sis_init(). Based on the FreeBSD sis(4) driver. ok mikeb@ sthen@
2013-06-26Remove some lies in comments.Mark Kettenis
2013-06-26Flash the LEDs only if the output report size is non null.Martin Pieuchot
This prevent some Logitech Mice, at least G500 and G9x, that contain various HID devices including a keyboard without LEDs to keep disconnecting right after being attached. Problem reported by abieber@ and Damien Couderc, ok miod@
2013-06-25Do not dereference sc before checking if it exists, pointed by otto@.Martin Pieuchot
2013-06-25Repair flow control broken in the rev1.329 and make sure that a simpleMike Belopuhov
ifconfig done by a user won't alter our negotiated flow control settings. Both problems were identified by David Imhoff <dimhoff_devel @ xs4all !nl> Tested by David on 5719, 5720, 5721, Hrvoje Popovski on 5704 B0, sthen@ on 5723/5784, naddy@ and jmatthew@ on 5702/5703, benno@ on 5704 A3 and me on 5715 and 5719.
2013-06-25Bring ohci(4) and uhci(4) in sync with ehci(4) by ensuring that aMartin Pieuchot
transfer is submitted when a zero-length bulk or interrupt transfer is requested. This is the missing part of FreeBSD's svn r159024 that should have been committed with ehci's r1.57. ok uebayasi@
2013-06-25Remove unsigned comparison < 0.Brad Smith
Pointed out by LLVM. ok mikeb@
2013-06-23Remove unsigned comparison < 0.Brad Smith
Pointed out by LLVM. ok ratchov@
2013-06-22Prefix all functions. Pointed out by brad@.Masao Uebayashi
2013-06-21De-static.Masao Uebayashi
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-28 00:54:00 +0000