summaryrefslogtreecommitdiff
path: root/sys/dev
AgeCommit message (Collapse)Author
2019-06-24Add glkgpio(4), a driver for the GPIO controllers found on Intel's GeminiMark Kettenis
Lake SoC. From James Hastings.
2019-06-24Don't retry to get the usb descriptor in usbd_new_device() if the callClaudio Jeker
timed out. If usbd_get_desc() returns USBD_TIMEOUT the chance is high that the next call will do the same and it slows down attachement of devices a lot. Makes reattaching USB at resume a lot faster with my umb(4) card (which likes to time out). "move fast and fix things" deraadt@
2019-06-22Set if_link_state based on the PAOS port register, and update it whenJonathan Matthew
we get a port change event. ok dlg@
2019-06-17Add aplgpio(4) a driver for the gpio controllers found on Intel's ApolloPatrick Wildt
Lake SoC. From James Hastings ok kettenis@
2019-06-17Split out the code that assigns addresses to BARs into its own function.Mark Kettenis
ok patrick@
2019-06-17When reloading station address via TWSI is timed out, try to reloadKevin Lo
station address from EEPROM. From Brad.
2019-06-16drm/amd/display: Use plane->color_space for dpp if specifiedJonathan Gray
From Nicholas Kazlauskas 668440f6ee3f553252cb52da488d3ac3ba404382 in linux 4.19.y/4.19.51 a1e07ba89d49581471d64c48152dbe03b42bd025 in mainline linux
2019-06-15Fix previous commit and unbreak the tree.Mark Kettenis
2019-06-14timeout_add(9)+hz(9) -> timeout_add_sec(9); ok patrick@cheloha
2019-06-14Use timeout_add_msec(9)kn
UPGT_LED_ACTION_TMP_DUR is 100ms, avoid converting to Hz and back again. OK kevlo
2019-06-14Use timeout_add_msec(9)kn
The timeval is used to represent 100ms, which are converted to Hz so they can be converted back - simplify this by using the new ms interface directly. Tested and OK kevlo
2019-06-14match on Intel Apollo Lake idsJonathan Gray
from James Hastings
2019-06-14regenJonathan Gray
2019-06-14more Intel Apollo Lake idsJonathan Gray
from James Hastings
2019-06-14Only add and remove flow table entries for multicast while the interface isJonathan Matthew
up, and re-add flow table entries for all existing multicast groups when bringing the interface up.
2019-06-14Add a barrier after arming a completion queue. If multiple cqs are usingJonathan Matthew
the same uar page, these writes could be combined, which would result in a queue not being armed.
2019-06-13Do not consider the pipe as halted if the device is gone.Martin Pieuchot
Analysed by and ok claudio@
2019-06-13the rx drops field is 8 bits (and a 24 bit flow tag), not 32 bits.David Gwynne
sympathy from jmatthew@
2019-06-12Set interrupt moderation on the completion queue for around 10000Jonathan Matthew
interrupts per second, or 90% of the cq size. ok dlg@
2019-06-11drm: don't block fb changes for async plane updatesJonathan Gray
From Helen Koike fbb7e114e6e690c46f170dedd6fd2fb22f241519 in linux 4.19.y/4.19.50 89a4aac0ab0e6f5eea10d7bf4869dd15c3de2cd4 in mainline linux
2019-06-11drm/i915: Maintain consistent documentation subsection orderingJonathan Gray
From Jonathan Corbet 6600ec2600d63868efa8c6641375451677469b24 in linux 4.19.y/4.19.50 551bd3368a7b3cfef01edaade8970948d178d40a in mainline linux
2019-06-11drm/i915/fbc: disable framebuffer compression on GeminiLakeJonathan Gray
From Daniel Drake 360e00e290a355219425b5ea0ca62d9df3261fa1 in linux 4.19.y/4.19.50 396dd8143bdd94bd1c358a228a631c8c895a1126 in mainline linux
2019-06-11drm/amdgpu: remove ATPX_DGPU_REQ_POWER_FOR_DISPLAYS check when hotplug-inJonathan Gray
From Aaron Liu f3dcc88d531fd21af0c4261896eb23b455499366 in linux 4.19.y/4.19.50 bdb1ccb080dafc1b4224873a5b759ff85a7d1c10 in mainline linux
2019-06-11drm/radeon: prefer lower reference dividersJonathan Gray
From Christian Koenig 84c82ab8f133f1c1e27f6b1bafe3f7d8835acc4c in linux 4.19.y/4.19.50 2e26ccb119bde03584be53406bbd22e711b0d6e6 in mainline linux
2019-06-11drm/amdgpu/psp: move psp version specific function pointers to early_initJonathan Gray
From Alex Deucher 748a97ec6c1588e3a49e09fe9ca56dcdb7f976b5 in linux 4.19.y/4.19.50 9d6fea5744d6798353f37ac42a8a653a2607ca69 in mainline linux
2019-06-11drm: add non-desktop quirks to Sensics and OSVR headsets.Jonathan Gray
From Ryan Pavlik 98a8cb0282ab22d0533de34412f2118bf21147e4 in linux 4.19.y/4.19.50 29054230f3e11ea818eccfa7bb4e4b3e89544164 in mainline linux
2019-06-11drm: add non-desktop quirk for Valve HMDsJonathan Gray
From Andres Rodriguez 490290b0415f069eb156db50d165ca6a407d47bc in linux 4.19.y/4.19.50 30d62d4453e49f85dd17b2ba60bbb68b6593dba0 in mainline linux
2019-06-11Use km_alloc() instead of mallocarray() when allocating ttm dma segmentsJonathan Gray
to avoid running out of kmem space. Suggested by and ok kettenis@
2019-06-11pass what pci_intr_established returns to intr_barrierDavid Gwynne
passing a pointer to a pci_intr_handle_t doesnt work too well on sparc64, where it faults in the guts of the scheduler. dunno why amd64 is fine. ok jmatthew@
2019-06-11if ifiq_input says there's too much pressure, tell the rxr we're livelockedDavid Gwynne
ok jmatthew@
2019-06-11bump up the ring sizes for dlg@Jonathan Matthew
2019-06-11records a flowid in the mbuf if the rx cqe provides an rx hash value.David Gwynne
jmatthew@ says we need to configure the hw to populate that, but i can put this in now in preparation for it.
2019-06-10support hardware timestamping for received packets.David Gwynne
the chip has a cycle counter or something, and the value of the counter when the packet is sent to the host is stored in the completion queue entry. the driver periodically checks the cycle counter and records the time at that point, so the rx path can do some maths to figure out what the clock time is for the cycle counter. if the driver is calibrated with the chips counter, the rx packets are timestamped. this is based on the changes made to freebsd, but with some simplifications.
2019-06-10Print proper ACPI version number.Mark Kettenis
ok mlarkin@
2019-06-10drm/lease: Make sure implicit planes are leasedJonathan Gray
From Daniel Vetter 390a0fd31b02f2f7086af2f3bdb8ae576b9a4b84 in linux 4.19.y/4.19.49 204f640da6914844b3270b41b29c84f6e3b74083 in mainline linux
2019-06-09Let drm(4) allocate memory without constraints if the hardware supportsMark Kettenis
64-bit DMA. Should reduce the pressure on DMA-reachable memory, which is important since there are still cases where the pagedaemon ends up being triggered continuously if we run out of DMA-reachable memory but have plenty of memory left. ok jsg@
2019-06-08When establishing the mapping between ACPI device nodes and PCI devices,Mark Kettenis
skip PCI host bridges and devices that aren't there. ok patrick@
2019-06-08disambiguate memory extent conflicts for membars and rom addressesDavid Gwynne
there's some more work to do in this space, but this helps us start. ok kettenis@
2019-06-08Update Spleen kernel fonts to the latest released version, bringingFrederic Cambus
the following improvements: - Shift the middle bar of the upper case 'G' one pixel down in the 12x24 version - Shift lower case 'k' character right, for better alignment in the 12x24, 16x32, and 32x64 versions - Make upper case 'X' thicker in the 16x32 and 32x64 versions - Make upper case 'V' thicker in the 32x64 version - Make lower case 'g' character smoother in the 16x32 and 32x64 versions - Add some artefacts on each side of the lower case 'i' characters OK kettenis@, naddy@, deraadt@
2019-06-07add ukspan(4), a driver for the Keyspan USA19HS USB serial adapterJoshua Stein
written by Cody Cutler <ccutler at csail.mit.edu>
2019-06-07regenJoshua Stein
2019-06-07add Keyspan USA-19HSJoshua Stein
2019-06-07Introduce acpi_getsta() and use it to eliminate some duplicated code.Mark Kettenis
ok patrick@
2019-06-07make struct mcx_cq_entry a bit closer to realityDavid Gwynne
this puts the timestamp, rx drops, and hash fields in the right place. ok jmatthew@
2019-06-07have mcx_process_rx return the number of slots it made freeDavid Gwynne
this is instead of passing a pointer to the counter. while here use byte swapping loads and stores, which is mostly a nop cos i dont think we have an LE arch with swapping memory operations. ok jmatthew@
2019-06-06ConnectX-5 cards work tooJonathan Matthew
2019-06-06Set up the interrupt handler before creating the event queue.Jonathan Matthew
Apparently the boot rom on some cards leaves an interrupt vector set up, which will be run on completion of the first command after we set up the event queue, causing kernel crashes. chris@ reported this a while ago.
2019-06-05We know which flow table entries are actually in use, so we can justJonathan Matthew
delete those rather than blindly deleting everything, which also means we don't need to check against a specific syndrome code to ignore errors we get deleting entries that aren't there.
2019-06-05this can support 64BIT DVA for ioDavid Gwynne
2019-06-04Move a function used as a callback out of a header so there will only beJonathan Gray
one function with a single address.
generated by cgit v1.2.3 (git 2.25.1) at 2025-01-09 00:22:31 +0000