![[BACK]](/image.pl?url=https%3a%2f%2fcvsweb.openbsd.org%2ficons%2fback.gif&f=jpg&w=240)
Up to[local] /src /sys /dev /usb
Request diff between arbitrary revisions
Default branch: MAIN
Revision1.67 / (download) -annotate -[select for diffs],Wed May 29 07:27:33 2024 UTC (20 months, 3 weeks ago) bystsp
Branch:MAIN
CVS Tags:OPENBSD_7_8_BASE,OPENBSD_7_8,OPENBSD_7_7_BASE,OPENBSD_7_7,OPENBSD_7_6_BASE,OPENBSD_7_6,HEAD
Changes since1.66: +11 -1 lines
Diff to previous1.66 (colored)
fix WEP onathn(4) USB hostapDeferring installation of software crypto keys to a task context isnot needed and results in race conditions that trigger the infamous"key not installed for sw crypto" panic.
Revision1.66 / (download) -annotate -[select for diffs],Thu May 23 03:21:08 2024 UTC (20 months, 4 weeks ago) byjsg
Branch:MAIN
Changes since1.65: +1 -5 lines
Diff to previous1.65 (colored)
remove unneeded includes; ok mpi@
Revision1.65 / (download) -annotate -[select for diffs],Sun Jul 10 21:13:41 2022 UTC (3 years, 7 months ago) bybluhm
Branch:MAIN
CVS Tags:OPENBSD_7_5_BASE,OPENBSD_7_5,OPENBSD_7_4_BASE,OPENBSD_7_4,OPENBSD_7_3_BASE,OPENBSD_7_3,OPENBSD_7_2_BASE,OPENBSD_7_2
Changes since1.64: +2 -2 lines
Diff to previous1.64 (colored)
Add _cb suffix to callback fields in struct ifmedia. Makes codeeasier to read and grep as ifm_status was used in both structsifmediareq and ifmedia with different meaning.OK mvs@
Revision1.63.2.1 / (download) -annotate -[select for diffs],Fri Apr 22 10:09:16 2022 UTC (3 years, 10 months ago) bybluhm
Branch:OPENBSD_7_1
Changes since1.63: +2 -2 lines
Diff to previous1.63 (colored) next main1.64 (colored)
Use memset() to initialize struct ieee80211_rxinfo properly.Sven Wolf noticed that scans onral(4) are buggy ever since I added a newfield to this struct. Turns out a lot of drivers were initializing fieldsone-by-one, leaving any newly added fields uninitialized by default.Affected drivers may report wrong channel numbers for received beacons.The net80211 stack will discard such beacons, assuming they were receivedon the wrong channel due to signal leakage. Scanning is broken as result.from stsp@; ok miod@this is errata/7.1/001_wifi.patch.sig
Revision1.64 / (download) -annotate -[select for diffs],Thu Apr 21 21:03:03 2022 UTC (3 years, 10 months ago) bystsp
Branch:MAIN
Changes since1.63: +2 -2 lines
Diff to previous1.63 (colored)
Use memset() to initialize struct ieee80211_rxinfo properly.Sven Wolf noticed that scans onral(4) are buggy ever since I added a newfield to this struct. Turns out a lot of drivers were initializing fieldsone-by-one, leaving any newly added fields uninitialized by default.Affected drivers may report wrong channel numbers for received beacons.The net80211 stack will discard such beacons, assuming they were receivedon the wrong channel due to signal leakage. Scanning is broken as result.ok miod@
Revision1.63 / (download) -annotate -[select for diffs],Mon Nov 22 10:17:14 2021 UTC (4 years, 2 months ago) bymglocker
Branch:MAIN
CVS Tags:OPENBSD_7_1_BASE
Branch point for:OPENBSD_7_1
Changes since1.62: +2 -2 lines
Diff to previous1.62 (colored)
Align memory allocation for USB device drivers and USB HC drivers:* USB device drivers use M_USBDEV instead of M_DEVBUF.* USB HC drivers use M_USBHC instead of M_DEVBUF.In a vanilla setup, this enlarges the USB memory pool.ok anton@
Revision1.62 / (download) -annotate -[select for diffs],Sun Oct 31 12:24:02 2021 UTC (4 years, 3 months ago) bystsp
Branch:MAIN
Changes since1.61: +3 -1 lines
Diff to previous1.61 (colored)
Makeathn(4) attach to the Sony UWA-BR100. Patch by martingot@protonmail
Revision1.61 / (download) -annotate -[select for diffs],Thu Apr 15 18:25:44 2021 UTC (4 years, 10 months ago) bystsp
Branch:MAIN
CVS Tags:OPENBSD_7_0_BASE,OPENBSD_7_0,OPENBSD_6_9_BASE,OPENBSD_6_9
Changes since1.60: +2 -2 lines
Diff to previous1.60 (colored)
Switchathn(4) 802.11n Tx rate adaptation from MiRA to RA.Tests:AR5418: Uwe WerlerAR9280: kn, jmatthew, Lauri Tirkkonen, Scott Bennett, Mikolaj KucharskiAR9285: kevlo, trondd, myselfok deraadt@
Revision1.60 / (download) -annotate -[select for diffs],Thu Feb 25 02:48:20 2021 UTC (4 years, 11 months ago) bydlg
Branch:MAIN
Changes since1.59: +3 -3 lines
Diff to previous1.59 (colored)
we don't have to cast to caddr_t when calling m_copydata anymore.the first cut of this diff was made with coccinelle using this spatch:@rule@type caddr_t;expression m, off, len, cp;@@-m_copydata(m, off, len, (caddr_t)cp)+m_copydata(m, off, len, cp)i had fix it's opinionated idea of formatting by hand though, soi'm not sure it was worth it.ok deraadt@ bluhm@
Revision1.59 / (download) -annotate -[select for diffs],Mon Nov 30 16:09:33 2020 UTC (5 years, 2 months ago) bykrw
Branch:MAIN
Changes since1.58: +11 -2 lines
Diff to previous1.58 (colored)
Fix deferred key tasks along the same lines asurtwn(4) andrun(4).athn(4) tested by stsp@, who points out thatotus(4) andrsu(4) don't yet invokethe code path to do their own key setting.ok stsp@
Revision1.58 / (download) -annotate -[select for diffs],Mon Jul 13 08:31:32 2020 UTC (5 years, 7 months ago) bystsp
Branch:MAIN
CVS Tags:OPENBSD_6_8_BASE,OPENBSD_6_8
Changes since1.57: +2 -2 lines
Diff to previous1.57 (colored)
Whenathn(4) is deciding how to decrypt an incoming frame, look into theieee80211_node data structure to find the WPA group cipher which wasnegotiated with the peer, rather than checking the wireless interface'sconfiguration data.Found while investigating a problem where WPA2+TKIP wasn't working with athn.Problem reported by Tim Chase.
Revision1.57 / (download) -annotate -[select for diffs],Fri Jul 10 13:22:21 2020 UTC (5 years, 7 months ago) bypatrick
Branch:MAIN
Changes since1.56: +2 -2 lines
Diff to previous1.56 (colored)
Change users of IFQ_DEQUEUE(), IFQ_ENQUEUE() and IFQ_LEN() to use the"new" API.ok dlg@ tobhe@
Revision1.56 / (download) -annotate -[select for diffs],Mon Apr 27 08:21:35 2020 UTC (5 years, 9 months ago) bystsp
Branch:MAIN
CVS Tags:OPENBSD_6_7_BASE,OPENBSD_6_7
Changes since1.55: +53 -5 lines
Diff to previous1.55 (colored)
Offload CCMP (WPA2) encryption and decryption toathn(4) hardware.This reduces CPU load during traffic bursts, which is especially noticeableon boards with relatively slow CPUs such as Alix and APU. For reference,my rcc-ve boards (Intel Atom C2558) now forward ~2000pps between Ethernetand wifi with <= 1% CPU interrupt time according to systat. Beforehand, theseboards became noticeably less responsive with fairly high interrupt and spin %.CCMP offload is used on both PCI and USB devices (thanks kevlo@ for catchingmy 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@
Revision1.55 / (download) -annotate -[select for diffs],Mon Nov 25 11:32:17 2019 UTC (6 years, 2 months ago) bympi
Branch:MAIN
Changes since1.54: +6 -4 lines
Diff to previous1.54 (colored)
Convert totsleep_nsec(9) and use the correct timeout period in msec.Tested by kevlo@, ok stsp@
Revision1.54 / (download) -annotate -[select for diffs],Tue Nov 12 07:47:30 2019 UTC (6 years, 3 months ago) bympi
Branch:MAIN
Changes since1.53: +7 -4 lines
Diff to previous1.53 (colored)
Converttsleep(9) that wait for a multiple of `hz' totsleep_nsec(9).ok cheloha@
Revision1.53 / (download) -annotate -[select for diffs],Thu Sep 12 12:55:07 2019 UTC (6 years, 5 months ago) bystsp
Branch:MAIN
CVS Tags:OPENBSD_6_6_BASE,OPENBSD_6_6
Changes since1.52: +10 -6 lines
Diff to previous1.52 (colored)
Make wireless drivers call if_input() only once per interrupt.This reduces drops caused by the ifq pressure drop mechanism and henceincreases throughput. Such drops are visible with e.g. 'netstat -dnI iwm0'.Not all affected drivers have been tested yet but these changes are largelymechanical and should be safe. As usual, please report any regressions.With help from dlg@ and mpi@Problem found by robert@Tested by robert, jmc, Tracey Emer, Matthias Schmidt, florian, Björn Ketelaarsok mpi@
Revision1.52 / (download) -annotate -[select for diffs],Thu Dec 6 07:50:38 2018 UTC (7 years, 2 months ago) bystsp
Branch:MAIN
CVS Tags:OPENBSD_6_5_BASE,OPENBSD_6_5
Changes since1.51: +5 -10 lines
Diff to previous1.51 (colored)
Avoid unnecessary dynamic memory allocation in athn_usb_newauth().Patch by Benjamin Baierok ratchov@ mpi@
Revision1.51 / (download) -annotate -[select for diffs],Thu Sep 6 11:50:54 2018 UTC (7 years, 5 months ago) byjsg
Branch:MAIN
CVS Tags:OPENBSD_6_4_BASE,OPENBSD_6_4
Changes since1.50: +3 -3 lines
Diff to previous1.50 (colored)
fix whitespace
Revision1.50 / (download) -annotate -[select for diffs],Mon Feb 5 09:52:03 2018 UTC (8 years ago) bystsp
Branch:MAIN
CVS Tags:OPENBSD_6_3_BASE,OPENBSD_6_3
Changes since1.49: +2 -1 lines
Diff to previous1.49 (colored)
claim copyright
Revision1.49 / (download) -annotate -[select for diffs],Mon Feb 5 08:48:57 2018 UTC (8 years ago) bystsp
Branch:MAIN
Changes since1.48: +361 -67 lines
Diff to previous1.48 (colored)
Moveathn(4) USB devices to the open source atk9k HTC firmware.Also fix support for AR7010 devices, fix mode hostap by properlymanaging the firmware station table, and fix Tx rate reporting.Tested on AR7010 and AR9271 devices.ok kevlo@
Revision1.48 / (download) -annotate -[select for diffs],Thu Oct 26 15:00:28 2017 UTC (8 years, 3 months ago) bympi
Branch:MAIN
Changes since1.47: +1 -11 lines
Diff to previous1.47 (colored)
Move common code to add/remove multicast filters toieee80211_ioctl(9).ok jsg@, stsp@
Revision1.47 / (download) -annotate -[select for diffs],Sat Apr 8 02:57:25 2017 UTC (8 years, 10 months ago) byderaadt
Branch:MAIN
CVS Tags:OPENBSD_6_2_BASE,OPENBSD_6_2
Changes since1.46: +3 -2 lines
Diff to previous1.46 (colored)
A pile of sizes tofree(9). In test for a few days in snapshots.Errors will result in nice clean panic messages so we know what's wrong.Reviewed by dhill visa natano jsg.
Revision1.46 / (download) -annotate -[select for diffs],Sun Mar 26 15:31:15 2017 UTC (8 years, 10 months ago) byderaadt
Branch:MAIN
CVS Tags:OPENBSD_6_1_BASE,OPENBSD_6_1
Changes since1.45: +6 -5 lines
Diff to previous1.45 (colored)
Add sizes to variousfree(9) calls. Fixing the simpler ones first.ok natano visa
Revision1.45 / (download) -annotate -[select for diffs],Sun Jan 22 10:17:39 2017 UTC (9 years, 1 month ago) bydlg
Branch:MAIN
Changes since1.44: +1 -2 lines
Diff to previous1.44 (colored)
move counting if_opackets next to counting if_obytes in if_enqueue.this means packets are consistently counted in one place, unlike themany and various ways that drivers thought they should do it.ok mpi@ deraadt@
Revision1.44 / (download) -annotate -[select for diffs],Thu Jan 12 16:32:28 2017 UTC (9 years, 1 month ago) bystsp
Branch:MAIN
Changes since1.43: +22 -21 lines
Diff to previous1.43 (colored)
Finish initial 11n support forathn(4).The heavy lifting was done by damien@ years ago. I didn't even haveto figure out what the hardware expects, the code was already there.This driver now supports MCS 0-15 in client and hostap mode.No Tx aggregation and no 40 MHz channels yet.tested by vgross@, bmercer@, tb@, jmc@, Vadim Vygonets, Peter Kayok bmercer@ tb@ phessler@
Revision1.43 / (download) -annotate -[select for diffs],Tue Nov 29 10:22:30 2016 UTC (9 years, 2 months ago) byjsg
Branch:MAIN
Changes since1.42: +2 -3 lines
Diff to previous1.42 (colored)
m_free() and m_freem() test for NULL. Simplify callers which had their ownNULL tests.ok mpi@
Revision1.42 / (download) -annotate -[select for diffs],Fri Dec 11 16:07:02 2015 UTC (10 years, 2 months ago) bympi
Branch:MAIN
CVS Tags:OPENBSD_6_0_BASE,OPENBSD_6_0,OPENBSD_5_9_BASE,OPENBSD_5_9
Changes since1.41: +5 -8 lines
Diff to previous1.41 (colored)
Replacemountroothook_establish(9) byconfig_mountroot(9) a narrower APIsimilar toconfig_defer(9).ok mikeb@, deraadt@
Revision1.41 / (download) -annotate -[select for diffs],Wed Nov 25 03:09:59 2015 UTC (10 years, 2 months ago) bydlg
Branch:MAIN
Changes since1.40: +8 -7 lines
Diff to previous1.40 (colored)
replace IFF_OACTIVE manipulation with mpsafe operations.there are two things shared between the network stack and driversin the send path: the send queue and the IFF_OACTIVE flag. the sendqueue is now protected by a mutex. this diff makes the oactivefunctionality mpsafe too.IFF_OACTIVE is part of if_flags. there are two problems with that.firstly, if_flags is a short and we dont have any MI atomic operationsto manipulate a short. secondly, while we could make the IFF_OACTIVEoperates mpsafe, all changes to other flags would have to be madesafe at the same time, otherwise a read-modify-write cycle on theirupdates could clobber the oactive change.instead, this moves the oactive mark into struct ifqueue and providesan API for changing it. there's ifq_set_oactive, ifq_clr_oactive,and ifq_is_oactive. these are modelled on ifsq_set_oactive,ifsq_clr_oactive, and ifsq_is_oactive in dragonflybsd.this diff includes changes to all the drivers manipulating IFF_OACTIVEto now use the ifsq_{set,clr_is}_oactive API too.ok kettenis@ mpi@ jmatthew@ deraadt@ Revision1.40 / (download) -annotate -[select for diffs],Tue Nov 24 13:45:06 2015 UTC (10 years, 2 months ago) bympi
Branch:MAIN
Changes since1.39: +1 -2 lines
Diff to previous1.39 (colored)
No need to include <net/if_arp.h>This header is only needed because <netinet/if_ether.h> declares astructure that needs it. But it turns out that <net/if.h> alreadyincludes it as workaround.A proper solution would be to stop declarting "struct ether_arp"there. But no driver should need this header.
Revision1.39 / (download) -annotate -[select for diffs],Tue Nov 24 13:33:18 2015 UTC (10 years, 2 months ago) bympi
Branch:MAIN
Changes since1.38: +1 -2 lines
Diff to previous1.38 (colored)
The only network driver needing <net/if_types.h> isupl(4) for IFT_OTHER.
Revision1.38 / (download) -annotate -[select for diffs],Wed Nov 4 12:12:00 2015 UTC (10 years, 3 months ago) bydlg
Branch:MAIN
Changes since1.37: +2 -2 lines
Diff to previous1.37 (colored)
replace the ifqueues in net80211 with mbuf_queues.the specific queues are ic_mgtq, ic_pwrsaveq, and ni_savedq. rtwhad its own queue for beacons.tested by mpi@ and jmc@ok mpi@
Revision1.37 / (download) -annotate -[select for diffs],Sun Oct 25 12:11:56 2015 UTC (10 years, 3 months ago) bympi
Branch:MAIN
Changes since1.36: +1 -5 lines
Diff to previous1.36 (colored)
arp_ifinit() is no longer needed.
Revision1.36 / (download) -annotate -[select for diffs],Thu Jul 9 16:33:42 2015 UTC (10 years, 7 months ago) bystsp
Branch:MAIN
CVS Tags:OPENBSD_5_8_BASE,OPENBSD_5_8
Changes since1.35: +3 -4 lines
Diff to previous1.35 (colored)
Revert previous, doesn't fix anything. I managed to reproducethe problem on one of my machines and the patch doesn't help.Seehttp://marc.info/?l=openbsd-tech&m=143645936727569&w=2
Revision1.35 / (download) -annotate -[select for diffs],Thu Jul 9 13:48:54 2015 UTC (10 years, 7 months ago) bystsp
Branch:MAIN
Changes since1.34: +5 -4 lines
Diff to previous1.34 (colored)
Allow more time for USBathn(4) firmware boot. It seems people on daemonforumsare running into the previous 1 second timeout on some machines, which thedriver will treat as fatal. Not sure if this will really fix the issue butit won't hurt. Also reported in NetBSD land which inherited our driver:http://mail-index.netbsd.org/current-users/2014/05/06/msg024793.htmlok mpi@
Revision1.34 / (download) -annotate -[select for diffs],Fri Jun 12 15:47:31 2015 UTC (10 years, 8 months ago) bympi
Branch:MAIN
Changes since1.33: +4 -18 lines
Diff to previous1.33 (colored)
Only match devices with a valid configuration.Most of the WiFi/Ethernet USB adapter only have one configuration and alwaysuse its first interface. In order to improve USB descriptors parsing startby reducing the number of places where a configuration is set.Tests & ok stsp@
Revision1.33 / (download) -annotate -[select for diffs],Mon Mar 2 15:23:28 2015 UTC (10 years, 11 months ago) bystsp
Branch:MAIN
CVS Tags:OPENBSD_5_7_BASE,OPENBSD_5_7
Changes since1.32: +2 -2 lines
Diff to previous1.32 (colored)
Remove stray ')' fromathn(4) "firmware command timed out" error message.
Revision1.32 / (download) -annotate -[select for diffs],Mon Mar 2 15:18:37 2015 UTC (10 years, 11 months ago) bystsp
Branch:MAIN
Changes since1.31: +7 -1 lines
Diff to previous1.31 (colored)
athn(4) USB drops received frames larger than MCLBYTES.Count these dropped frames as input errors.
Revision1.31 / (download) -annotate -[select for diffs],Mon Mar 2 15:05:11 2015 UTC (10 years, 11 months ago) bystsp
Branch:MAIN
Changes since1.30: +23 -6 lines
Diff to previous1.30 (colored)
Fix re-associations withathn(4) USB devices in station mode.Sending the firmware a NODE_ADD command for our BSS doesn't seem to have thedesired effect in station mode. It made a new association after ifconfigdown/up impossible until the device was re-plugged. Leave the code in placebut limit it to hostap mode, just in case it is needed there (hard to tellsince hostap mode seems to be broken -- no beacons are sent by the device).Also, when leaving run state, reset the RX filer to allow all beacons again.ok mpi@
Revision1.30 / (download) -annotate -[select for diffs],Mon Mar 2 14:46:02 2015 UTC (10 years, 11 months ago) bystsp
Branch:MAIN
Changes since1.29: +54 -29 lines
Diff to previous1.29 (colored)
Reworkath(4) USB firmware command handling.The previous code was racy and could dead-lock the USB task thread when afirmware command timed out (e.g. because the user pulled out the device).Handle this condition by marking the device as dying as soon as the interrupthandler gets an IOERROR and making sleeping firmware command threads checkfor a dying device when waking up.Ensure that no thread will try to send a command before the previous onehas completed. There is only a single xfer data structure for firmwarecommand transfers and reusing an in-flight xfer will give the USB stackcurly toenails ("xfer not free").Allow up to ATHN_USB_HOST_CMD_RING_COUNT firmware commands to be enqueuedon the command ring, rather than just one. Use standard usdb_wait_task()when waiting for command ring completion instead of hand-rolled tsleep()s.discussed with and ok mpi@ Revision1.29 / (download) -annotate -[select for diffs],Mon Mar 2 13:47:08 2015 UTC (10 years, 11 months ago) bystsp
Branch:MAIN
Changes since1.28: +32 -5 lines
Diff to previous1.28 (colored)
Fix scanning withathn(4) USB devices.The driver artificially limited the number of slots for nodes in its net80211node cache to 8, the rationale being the device firmware can only handle 8associated nodes at a time. But that's not a good reason to only cache 8 ofpotentially many APs in the area.Instead of crippling upper layer data structures, impose the limit locallyby raising an error if more than 8 nodes are trying to associate (which canonly happen in hostap mode, and hostap mode seems to be broken anyway).This matches what Linux ath9k does.Also, use a USB-specific scan timeout callback to we can do USB refcountingwhile scanning.ok mpi@
Revision1.28 / (download) -annotate -[select for diffs],Mon Mar 2 13:13:51 2015 UTC (10 years, 11 months ago) bystsp
Branch:MAIN
Changes since1.27: +14 -6 lines
Diff to previous1.27 (colored)
Inathn(4), set USB pipe pointers to NULL after closing pipes. Not reallyrequired since this code runs during device detach but looks prettier.Also remove an unneeded call to usbd_abort_pipe() which is implied byusbd_close_pipe().ok mpi@
Revision1.27 / (download) -annotate -[select for diffs],Mon Mar 2 13:08:18 2015 UTC (10 years, 11 months ago) bystsp
Branch:MAIN
Changes since1.26: +18 -2 lines
Diff to previous1.26 (colored)
Teachathn(4) about USB reference counting to prevent it from causing majorconfusion in the USB stack. Also add some usbd_is_dying checks.suggested by + ok mpi@
Revision1.26 / (download) -annotate -[select for diffs],Tue Feb 10 23:25:46 2015 UTC (11 years ago) bympi
Branch:MAIN
Changes since1.25: +1 -2 lines
Diff to previous1.25 (colored)
Wireless drivers call if_input() via ieee80211_input() which set `rcvif'on every received mbuf, so there's no need to initialize this pointer inthe drivers.Tested by and ok phessler@
Revision1.25 / (download) -annotate -[select for diffs],Mon Dec 22 02:28:52 2014 UTC (11 years, 2 months ago) bytedu
Branch:MAIN
Changes since1.24: +1 -3 lines
Diff to previous1.24 (colored)
unifdef INET
Revision1.24 / (download) -annotate -[select for diffs],Fri Dec 19 22:44:59 2014 UTC (11 years, 2 months ago) byguenther
Branch:MAIN
Changes since1.23: +2 -2 lines
Diff to previous1.23 (colored)
Use <sys/endian.h> instead of <machine/endian.h>ok dlg@ mpi@ bcook@ millert@ miod@
Revision1.23 / (download) -annotate -[select for diffs],Fri Oct 31 21:19:14 2014 UTC (11 years, 3 months ago) bytedu
Branch:MAIN
Changes since1.22: +3 -1 lines
Diff to previous1.22 (colored)
atheros ub94 support, from Matt Markfort
Revision1.22 / (download) -annotate -[select for diffs],Sun Jul 13 15:52:49 2014 UTC (11 years, 7 months ago) bympi
Branch:MAIN
CVS Tags:OPENBSD_5_6_BASE,OPENBSD_5_6
Changes since1.21: +1 -3 lines
Diff to previous1.21 (colored)
No need for <netinet/in_systm.h> here.
Revision1.21 / (download) -annotate -[select for diffs],Sat Jul 12 18:48:52 2014 UTC (11 years, 7 months ago) bytedu
Branch:MAIN
Changes since1.20: +4 -4 lines
Diff to previous1.20 (colored)
add a size argument to free. will be used soon, but for now default to 0.after discussions with beck deraadt kettenis.
Revision1.20 / (download) -annotate -[select for diffs],Sat Jul 12 07:59:23 2014 UTC (11 years, 7 months ago) bympi
Branch:MAIN
Changes since1.19: +2 -17 lines
Diff to previous1.19 (colored)
Kill a bunch of DVACT_DEACTIVATE handlers whose only purpose were toset the 'dying' flag of a device. Such handlers are useless now thatusbd_detach() already set this flag.Even if the purpose of this flag is questionnable on OpenBSD becauseDVACT_DEACTIVATE acts as a pre-detach event, this is just a first smallstep towards the simplification of theautoconf(9) device states.This cleaning is now possible thanks to the work of pirofti@ to convertall the USB drivers to properly use usbd_is_dying().Discussed many times with deraadt@
Revision1.19 / (download) -annotate -[select for diffs],Wed Mar 19 10:09:19 2014 UTC (11 years, 11 months ago) bympi
Branch:MAIN
Changes since1.18: +2 -2 lines
Diff to previous1.18 (colored)
Stop abusing the rcvif pointer to pass wireless nodes down to thedriver start routines. Instead add & use a pointer in the pkthdrsince we don't want the overhead of using ambuf_tags(9).claudio@ pointed out that other subsystems might want to use thispointer too, so here's a new cookie!ok claudio@, mikeb@, deraadt@
Revision1.18 / (download) -annotate -[select for diffs],Wed Aug 7 01:06:41 2013 UTC (12 years, 6 months ago) bybluhm
Branch:MAIN
CVS Tags:OPENBSD_5_5_BASE,OPENBSD_5_5
Changes since1.17: +1 -2 lines
Diff to previous1.17 (colored)
Most network drivers include netinet/in_var.h, but apparently theydon't have to. Just remove these include lines.Compiled on amd64 i386 sparc64; OK henning@ mikeb@
Revision1.17 / (download) -annotate -[select for diffs],Tue Jun 11 18:45:25 2013 UTC (12 years, 8 months ago) byderaadt
Branch:MAIN
CVS Tags:OPENBSD_5_4_BASE,OPENBSD_5_4
Changes since1.16: +2 -2 lines
Diff to previous1.16 (colored)
Replace more ovbcopy with memmove; swap the src and dst arguments too
Revision1.16 / (download) -annotate -[select for diffs],Mon Apr 15 09:23:01 2013 UTC (12 years, 10 months ago) bymglocker
Branch:MAIN
Changes since1.15: +11 -11 lines
Diff to previous1.15 (colored)
Get rid of various 'typedef struct' definitions and use plain structuredefinitions instead. We don't change usb.h for now to stay compatiblewith userland.Tested by mpi@ on macppc and myself on i386.ok mpi@
Revision1.15 / (download) -annotate -[select for diffs],Fri Apr 12 12:58:39 2013 UTC (12 years, 10 months ago) bympi
Branch:MAIN
Changes since1.14: +4 -3 lines
Diff to previous1.14 (colored)
Instead of using a wrapper around usbd_transfer() when submitting atransfer synchronously just pass the USBD_SYNCHRONOUS flag like anyother flags when creating a transfer.ok sthen@, mglocker@
Revision1.14 / (download) -annotate -[select for diffs],Fri Apr 5 05:39:46 2013 UTC (12 years, 10 months ago) bybrad
Branch:MAIN
Changes since1.13: +3 -1 lines
Diff to previous1.13 (colored)
Fix USBathn(4) to be able to build on the RAMDISK. RAMDISK has SMALL_KERNELso there is no HostAP. athn_usb_newassoc() would return if not in HostAP modeso just stub out the function and don't build the callback function forthe RAMDISK.ok deraadt@
Revision1.13 / (download) -annotate -[select for diffs],Thu Mar 28 04:12:12 2013 UTC (12 years, 10 months ago) byderaadt
Branch:MAIN
Changes since1.12: +1 -2 lines
Diff to previous1.12 (colored)
removed unneccessary .h file
Revision1.12 / (download) -annotate -[select for diffs],Mon Jan 14 09:50:31 2013 UTC (13 years, 1 month ago) byjsing
Branch:MAIN
CVS Tags:OPENBSD_5_3_BASE,OPENBSD_5_3
Changes since1.11: +2 -2 lines
Diff to previous1.11 (colored)
Fix the sotfware.
Revision1.11 / (download) -annotate -[select for diffs],Mon Jan 14 09:48:24 2013 UTC (13 years, 1 month ago) byjsing
Branch:MAIN
Changes since1.10: +1 -1 lines
Diff to previous1.10 (colored)
Call the correct callback on node disassociation.
Revision1.10 / (download) -annotate -[select for diffs],Mon Nov 12 22:38:26 2012 UTC (13 years, 3 months ago) bymikeb
Branch:MAIN
Changes since1.9: +5 -1 lines
Diff to previous1.9 (colored)
protect rxeof path with an splnet; ok kettenis
Revision1.9 / (download) -annotate -[select for diffs],Sat Nov 10 14:35:06 2012 UTC (13 years, 3 months ago) bymikeb
Branch:MAIN
Changes since1.8: +4 -2 lines
Diff to previous1.8 (colored)
add a flag indicating whether shared code was initialized so that wecan prevent calling athn_detach if it wasn't; suggestion and ok stsp
Revision1.8 / (download) -annotate -[select for diffs],Sun Jul 3 15:47:17 2011 UTC (14 years, 7 months ago) bymatthew
Branch:MAIN
CVS Tags:OPENBSD_5_2_BASE,OPENBSD_5_2,OPENBSD_5_1_BASE,OPENBSD_5_1,OPENBSD_5_0_BASE,OPENBSD_5_0
Changes since1.7: +1 -3 lines
Diff to previous1.7 (colored)
Remove config_activate() and DVACT_ACTIVATE. PCMCIA's the only thingthat's ever used it, and it's long since been changed to useDVACT_{QUIESCE,SUSPEND,RESUME} instead.ok deraadt@, dlg@; miod@ also agreed with this idea when I brought itup a few weeks ago Revision1.7 / (download) -annotate -[select for diffs],Tue Jan 25 20:03:35 2011 UTC (15 years ago) byjakemsr
Branch:MAIN
CVS Tags:OPENBSD_4_9_BASE,OPENBSD_4_9
Changes since1.6: +1 -4 lines
Diff to previous1.6 (colored)
garbage collect "usb events". without /dev/usb there is no way to accessthem from userland, and nothing in the kernel uses them.ok krw@, miod@
Revision1.6 / (download) -annotate -[select for diffs],Sat Jan 8 15:18:01 2011 UTC (15 years, 1 month ago) bydamien
Branch:MAIN
Changes since1.5: +9 -7 lines
Diff to previous1.5 (colored)
reserve one xfer for beacons during init.
Revision1.5 / (download) -annotate -[select for diffs],Sat Jan 8 15:05:24 2011 UTC (15 years, 1 month ago) bydamien
Branch:MAIN
Changes since1.4: +6 -7 lines
Diff to previous1.4 (colored)
fix LEDs handling.do not read back the LED GPIO pin since it is configured in outputonly mode.makes the link LED blink on the WiFiStation EXT when scanning (theLEDs indicating the signal strength are not working yet).
Revision1.4 / (download) -annotate -[select for diffs],Thu Jan 6 19:20:54 2011 UTC (15 years, 1 month ago) bydamien
Branch:MAIN
Changes since1.3: +32 -28 lines
Diff to previous1.3 (colored)
cleanup athn_usb_rx_frame().remove padding after 802.11 header if any.
Revision1.3 / (download) -annotate -[select for diffs],Thu Jan 6 19:03:12 2011 UTC (15 years, 1 month ago) bydamien
Branch:MAIN
Changes since1.2: +26 -16 lines
Diff to previous1.2 (colored)
implement fast channel change.this dramatically reduces the number of register reads/writes andthus the number of context switches when changing channel.makes scanning much faster.
Revision1.2 / (download) -annotate -[select for diffs],Thu Jan 6 17:45:36 2011 UTC (15 years, 1 month ago) bydamien
Branch:MAIN
Changes since1.1: +34 -11 lines
Diff to previous1.1 (colored)
create STA entry for our BSS upon association.update Rx filter after association.move RECV_START command outside of rx_enable since rx_enable callbackis not allowed to fail.
Revision1.1 / (download) -annotate -[select for diffs],Thu Jan 6 07:27:15 2011 UTC (15 years, 1 month ago) bydamien
Branch:MAIN
"athn* at uhub? port ?"this adds preliminary support for the Atheros AR9271 chipset andprobably the AR9280+AR7010 and AR9287+AR7010 too though those werenot tested.scanning still takes a very long time (~1 sec per channel) butotherwise, operation in STA mode seems stable.will implement fast channel change soon.committed over the Ubiquiti WifiStation EXT (AR9271) on i386 with WPA.requires firmware (see man page for details)ok deraadt@ (who checked the .h files)