summaryrefslogtreecommitdiffstats
path: root/sys/dev/usb (follow)
Commit message (Collapse)AuthorAgeFilesLines
...
* Add Ericsson F5521gw Mobile Broadband Modem.tobhe2020-06-221-1/+5
| | | | ok sthen@
* Fix broken HID descriptors of Elecom trackballs.bru2020-06-174-4/+99
| | | | | | | | | The report descriptors specify an invalid button count for models with 6 or 8 buttons. Thanks to Sivaram Gowkanapal, who provided the core of this patch. ok mpi@
* Release the rx node if we were unable to allocate a new rx buffer.jmatthew2020-06-161-2/+2
| | | | | | | | The node here is always ic_bss, for which the reference count isn't actually used (it's always freed when the interface detaches), so not releasing it in this case wasn't really a problem. ok stsp@
* When detaching uaudio, wait for device ref counter to drop to 0ratchov2020-06-111-7/+46
| | | | | | | To maintain the correct refs count, blocking calls to uaudio are wrapped in usbd_ref_incr() and usbd_ref_decr() calls. suggested by mpi@ and ok visa@
* Make use of hardware crypto for CCMP. The tricky bits had been lyingjmatthew2020-06-111-35/+123
| | | | | | | | dormant in the driver for years, we just needed to insert the IV before transmission and do packet number checks on received frames. tested by kevlo@ solene@ benno@ and me on a variety of hardware ok kevlo@ stsp@
* Do not call encap routines with the IFQ mutex held, as they mightgerhard2020-06-0911-49/+45
| | | | | | sleep in the stop routine in case of errors. Fix proposed by mpi@, ok patrick@
* add umstc(4) for Microsoft Surface Type Cover keyboardsjcs2020-05-312-1/+178
|
* regenjcs2020-05-312-4/+14
|
* add two Surface Go Type Cover modelsjcs2020-05-311-1/+3
|
* Fix UVIDEO_DEBUG buildfeinerer2020-05-301-1/+2
| | | | | | | | | | | | | | | | | | | | Include sys/proc.h to address following errors: /usr/src/sys/dev/usb/uvideo.c:2901:31: error: incomplete definition of type 'struct proc' vn_close(nd.ni_vp, FWRITE, p->p_ucred, p); ~^ /usr/src/sys/sys/types.h:223:8: note: forward declaration of 'struct proc' struct proc; ^ /usr/src/sys/dev/usb/uvideo.c:2925:40: error: incomplete definition of type 'struct proc' UIO_SYSSPACE, IO_APPEND|IO_UNIT, p->p_ucred, NULL, p); ~^ /usr/src/sys/sys/types.h:223:8: note: forward declaration of 'struct proc' struct proc; ^ 2 errors generated. ok mpi@
* dev/rndvar.h no longer has statistical interfaces (removed during variousderaadt2020-05-292-6/+2
| | | | | | conversion steps). it only contains kernel prototypes for 4 interfaces, all of which legitimately belong in sys/systm.h, which are already included by all enqueue_randomness() users.
* match on TP-Link TL-WN822N-EU v5jsg2020-05-261-2/+3
| | | | from Tero Koskinen
* regenjsg2020-05-262-5/+10
|
* add TP-Link TL-WN822N-EU v5jsg2020-05-261-2/+3
| | | | from Tero Koskinen
* regenpatrick2020-05-252-4/+16
|
* ThingM blink(1)patrick2020-05-251-1/+5
|
* change wsdisplay attribute type from long to uint32_tjsg2020-05-251-9/+9
| | | | | | | | miod explained it was initially a long as it was thought drivers may need to allocate storage but in practice they don't need more than 32 bits for an attribute. suggested and reviewed by miod@
* rename wsdisplay alloc_attr() to pack_attr()jsg2020-05-251-2/+2
| | | | | | | | Suggested by John Carmack. miod agrees a rename would make sense and explained it was initially thought drivers may need to allocate storage but in practice they don't need more than 32 bits for an attribute. ok mpi@
* Use seltrue_kqfilter() instead of hand-rolling it in multiple drivers.mpi2020-05-132-26/+6
| | | | ok visa@
* Remove redundant code testing if a terminal is a usb stream.ratchov2020-05-041-5/+1
| | | | | | Fixes coverity CID 1492830. ok sthen, deraadt
* Check v6 availablilty mask instead of v4.gerhard2020-05-041-2/+2
| | | | go for it deraadt@
* If there's only one input (output) name it simply "input" ("output")ratchov2020-04-301-4/+27
| | | | | | Besides making things simpler, this allows libsndio to figure out that this is a control affecting all inputs (outputs) that needs to be exposed.
* Use "inputs" class instead of "record"ratchov2020-04-301-11/+5
| | | | | | | There's no reliable way to decide whether a control must belong to "inputs" or to "record". In other words both classes were used for the same thing. Furthermore we've no "play" class which tends to make things more confusing.
* Use "dac" instead of "play" in mixer control namesratchov2020-04-301-2/+2
| | | | | All other drivers use "dac" and libsndio needs the control name to be "dac" for the control to be exposed.
* Set terminal and clock names after all units are parsedratchov2020-04-301-13/+25
| | | | | Makes the code simpler and more flexible as the naming is done in a single place where the full units list is available.
* Use macros instead of hardcoded "play" and "record" string constantsratchov2020-04-301-4/+10
| | | | No object change.
* IPv6 is no longer on by default. It must be enabled with "inet6 eui64".gerhard2020-04-271-2/+3
| | | | Input from sthen@, ok claudio@
* Offload CCMP (WPA2) encryption and decryption to athn(4) hardware.stsp2020-04-272-6/+59
| | | | | | | | | | | | | | | | This reduces CPU load during traffic bursts, which is especially noticeable on boards with relatively slow CPUs such as Alix and APU. For reference, my rcc-ve boards (Intel Atom C2558) now forward ~2000pps between Ethernet and wifi with <= 1% CPU interrupt time according to systat. Beforehand, these boards became noticeably less responsive with fairly high interrupt and spin %. CCMP offload is used on both PCI and USB devices (thanks kevlo@ for catching my omission of USB devices in the first iteration of this change!) Tested by (at least) Kevin Chadwick, tracey@, kevlo@, kili@, Ted Patterson, David Dahlberg, and Scott Bennett. ok tracey@ kevlo@
* Skip "all-channels" controls if there are equivalend per-channel onesratchov2020-04-241-2/+28
| | | | | | Hardware may expose per-channel controls and equivalent ones affecting all channels, which is redundant. In this case expose only the per-channel ones as they are more flexible.
* Suffix per-channel non-numeric controls with channel numberratchov2020-04-241-3/+8
| | | | | | | | | The audio(9) mixer API can't represent such per-channel controls with a single entry, so we've to expose one mixer entry (with an unique name) per control. Fixes name clashes on usb audio devices with per-channel mute controls.
* Bump size of the buffer we use to fetch device tree properties. Thekettenis2020-04-081-2/+2
| | | | | | | | ethernet aliase paths in the new omap device trees are longer than 64 characters. Fixes the MAC address on Pandaboard-ES with the new device tree. ok jsg@
* Abstract the head of knote lists. This allows extending the lists,visa2020-04-072-6/+6
| | | | | | for example, with locking assertions. OK mpi@, anton@
* Move the responsibility of syncing the data buffers from the USBpatrick2020-04-036-36/+114
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | stack into the USB controller driver, since not all of the xfers actually use DMA and some invalidations might even be harmful. The important part is to sync before handing the buffer to the controller, and to sync on a successful transfer before handing it back to the USB stack. For isoc transfers it's easier to sync the complete length of the transfer, since the buffer to flush is not filled in a contiguous fashion. For dwc2 there's a common point which takes care of the start of transfers from or to devices, where we can make sure that our buffers are synced. On completion, there's a common point before handing it off to the USB stack. For ehci there are three places which take care of the start of transfers from or to devices, where one already does the sync, while the two other places still need the sync. There are two completion handler (isoc and non-isoc), where one already has a comment asking for the need of a sync. The done handler for intr xfers does a sync that is not needed anymore after adding the sync in the completion handler. For ohci there are three places which take care of the start of transfers from or to devices, where all of them were still in need of the sync. For completion, there is one place for isoc xfers and two places for handling successful generic xfers. For uhci there are two places which take care of the start of transfers from or to device, where all of them were still in need of the sync. For completion, there is one handler for both isoc and non-isoc xfers where syncs need to occur. For xhci there are three places which take care of the start of transfers from or to device, where all of them were still in need of the sync. For completion, there is one handler for isoc and one for non-isoc xfers where syncs need to occur. With this we can revert the workaround that implicitly allocated buffers are COHERENT, since now control transfers fulfilled by the driver code (instead of the controller doing DMA) are not flushed into oblivion anymore. Tested by Janne Johansson with dwc2 (octeon) Tested by kettenis@ with xhci(4) (octeon) Tested by patrick@ with ehci(4) on a Cubox-i (armv7) Tested by patrick@ with xhci(4) on an i.MX8MQ (arm64) Tested by tobhe@ with dwc2 on a rPi 3b (arm64) ok kettenis@
* Fix incoherencies of errors returned by various *kqfilter routines.mpi2020-04-032-4/+4
| | | | | | | | EOPNOTSUPP is returned when there's no "filterops" for a given fd. EINVAL when the requested filter isn't supported by the "filterops". ENXIO means the underlying device is no longer valid/present. ok visa@
* Fix use-after-free zyd(4) upon Tx error and device detach.stsp2020-04-021-4/+5
| | | | | | | | | | | | | | | If a transmit error occurs in zyd_tx() we must reset tx data's ni pointer to NULL since the node reference will be released in zyd_start(). A stale node pointer could potentially be released again which would mess up the node's reference count and potentially result in a use-after-free. Upon detach, ic->ic_bss was freed in ieee80211_ifdetach(), and afterwards ic->ic_bss->ni_refcnt was decremented in zyd_free_tx_list(). Change the order of operations during detach to prevent this use-after-free. Reported on bugs@ by Raf Czlonka ok mpi@
* USB control requests to the root hub differ from those to devices. Forpatrick2020-04-011-2/+3
| | | | | | | | | | | | | | | | | devices we have proper DMA transfers, so that the DMA syncs before and after transfers are necessary. For the root hub, we seem to fulfill those requests ourselves, e.g. by using memcpy. The POSTREAD DMA sync on a read will invalidate the caches for the buffer, and unless our root hub code explicitly flushed the data to memory, it's possible that our memcpy got removed from history. Until a better solution is implemented, like moving the DMA syncs from the USB subsystem into the controller driver, allocate all buffers that are not explicitly allocated using USB_DMA_COHERENT. The IO buffers continue to be allocated as non-coherent. Regression found by jsg@ ok kettenis@
* No point in checking the loop variable for NULL inside a LIST_FOREACH.krw2020-03-301-3/+1
| | | | CID 980279
* If samples per frame is larger than the device usb packet size,bluhm2020-03-231-1/+5
| | | | | | | fail early in uaudio(4). This prevents negative safe_blksz and integer overflow in nframes_max in case of a bad call to uaudio_stream_open() or a hardware bug. OK ratchov@
* If an USB audio device supports less rates for recording than forbluhm2020-03-231-2/+8
| | | | | | | | playing, the uaudio(4) driver selected a wrong rate. Then recording failed with "block size too large". The v1_rates field of the parameters contains the correct intersection of valid rates. Use it for v1 devices. OK ratchov@
* Instead of passing the flags as part of a struct that's supposed to bepatrick2020-03-2110-54/+39
| | | | | | | filled by the callee, change the usb_allocmem() API to take another argument for the flags. ok mpi@
* Switch USB to use non-coherent buffers for data transfers. Sincepatrick2020-03-199-16/+46
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | the import in '99 all buffers allocated using usb_allocmem() have been mapped COHERENT. On some ARM SoCs, where the USB controller is not coherent with the caches, this means the buffers were mapped uncached. This drastically reduces the performance, especially on reads. We already added cache syncs before and after USB transfers, but so far those have essentially been no-ops. Since other drivers make use of the same allocation code, and those haven't been proven to have correct syncs, we can't just remove the COHERENT flag. This splits the allocation into coherent and non-coherent blocks. All drivers who call usb_allocmem() themselves now pass a flag to show they require coherent blocks. Onced verified that they also work fine without coherent, or once they have been refactored, we can remove this again. On a ure(4) connected to an i.MX8MQ, the receive performance is 10x as fast as before. The Raspberry Pi's ethernet receive speed doubled. Debugged using dt(4) Original diff from Marius Strobl Feedback from kettenis@ Tested on an i.MX8MQ (arm64) by patrick@ Tested on a Raspberry Pi (arm64) by tobhe@ Tested on an ERL (octeon) by Janne Johansson ok mpi@ gerhard@ tobhe@
* Prevent out-of-bound read if `rate' doesn't match known values.mpi2020-03-171-9/+9
| | | | | | CID 1453258 ok kettenis@
* Prevent a NULL derefence in alloc_all_endpoints_fixed_ep()jasper2020-03-161-1/+8
| | | | | | | | | | when no IO endpoints were found. Coverity CID 1453263, 1453156 Check return value of usbd_interface2endpoint_descriptor() in alloc_all_endpoints_yamaha() as it may return NULL which ends up being dereferenced by the UE_GET_XFERTYPE() macro. Coverity CID 271356 ok ratchov@
* Remove unused function argument.mpi2020-03-161-5/+5
| | | | CID 1453335.
* Remove dead code which is actually duplicated a few lines abovejasper2020-03-111-6/+1
| | | | | | | | right after err is set. Coverity CID 975917 ok kettenis@ kn@
* synckevlo2020-03-102-4/+9
|
* Match on Lenovo OneLink Plus Dock Ethernet.kevlo2020-03-102-2/+4
| | | | Tested by David Demelier
* Rework ucom(4) a bit. The logic of RTS follows DTR should only be doneclaudio2020-03-081-6/+16
| | | | | | | | when the port is initialized (with a tcsetattr() call) but not on every changed of DTR (especially those caused by other tty ioctls). Fixes firmware upload on various microcontroller boards that use DTR and RTS as signaling lines to reset the device and enter the bootloader. Tested by tracey@, OK deraadt@
* Fix use of uninitilized variable in case of emty AC descriptor.ratchov2020-03-071-2/+2
| | | | | | | | USB audio devices with empty AC descriptor (i.e. emtpy devices) don't exist as they would be of no use. So this fix is mostly to properly skip broken hardware. Found by coverity.
* Fix endian swapping of trb_flags and trb_status in xhci(4).visa2020-03-021-9/+10
| | | | | | | | | This corrects computation of xfer->actlen, allowing xhci(4) to work again on octeon and other big endian architectures. Reported by Allen Smith OK patrick@
Copyright © 1996 – 2023 Jason A. Donenfeld. All Rights Reverse Engineered.