summaryrefslogtreecommitdiff
path: root/sys
AgeCommit message (Collapse)Author
2004-11-09Do not map empty mbufs (m_len == 0) in bus_dmamap_load_mbuf() as these mappingsClaudio Jeker
may disturb the dma as seen in ipw(4). Emtpy mbufs are at the beginning of the mbuf chain and are as example a "side-effect" of a previous m_adj() call. OK miod@ mickey@ jason@ markus@
2004-11-09Fix single-stepping.Mark Kettenis
ok miod@
2004-11-09Kill more unused stuff.Miod Vallat
2004-11-09Luna88k does not have an iomap, so kill all iomap related defines, andMiod Vallat
move the mapiodev()/unmapiodev() code back to mvme88k.
2004-11-09remove useless bpf code. removes dependancy on code from an(4).David Gwynne
2004-11-09Always free memory on failure in c{e,s}4231_alloc(); from Patrick Lafiti.Miod Vallat
2004-11-09big KNF spanking, no change in object file, ecstatic ok from claudioHenning Brauer
2004-11-09use NULL instead of 0 for poiter comparisions,Henning Brauer
from "Alexey E. Suslikov" <cruel@texnika.com.ua> with a little help from itojun
2004-11-09Split {get,set,raise}ipl() into per-board implementations.Miod Vallat
2004-11-08let it actually compileTheo de Raadt
2004-11-08commented out entry for atuDavid Gwynne
"yup" by deraadt@
2004-11-08Atmel AT76c503 / AT76c503a / AT76c505 / AT76c505a USB WLAN driverDavid Gwynne
originally the atuwi driver from Daan Vreeken, but with enough changes by deraadt@ and me to warrant a name change. there are still plenty of issues to resolve, but at least i can get packets over it now. ok deraadt@
2004-11-08ugly ugly hack to allow a usb device to do a full reset of the port duringDavid Gwynne
attach. this is required by atu (atmel based usb wireless) devices. something better will be done, but we need it for now to work on atu. ok deraadt@
2004-11-08Simple cleaning of sgivol, with no functional change (yet):Miod Vallat
- remove hardcoded constants - use strtol() and check for errors instead of using atoi() - use err()/errx()/warnx() when applicable - use bounded string operations whenever possible - use strncpy() instead of a home-grown construct Plus you get the "/usr/mdec/sgivol -p" instant coredump fixed for free.
2004-11-08put more debug stuff inside #ifdef, to make the driver a wee bit smallerTheo de Raadt
for the alpha
2004-11-08TypoMiod Vallat
2004-11-08Show in the dmesg as mongoose, to matche the manpage and kernel configurationMiod Vallat
file.
2004-11-08TypoMiod Vallat
2004-11-08use __packed instead of __attribute__ ((__packed__)) (ok miod@), checkReyk Floeter
possible NULL mac in the keycache function.
2004-11-08Remove mvme88k leftovers.Miod Vallat
2004-11-08Kill struct md_p, which was really only necessary for mvme88k; onMiod Vallat
luna88k it disappears completely, while mvme88k keeps 3 global variables, one of them scheduled to disappear very soon.
2004-11-07add a new message: inject. this allows the tracer to to inject buffers intomarius eriksen
the stackgap. this in turn allows for argument replacement with indirection. for example replacing an entire envp or struct pollfd. ok provos@ millert@
2004-11-07enable ath, iwi, and ipwTheo de Raadt
2004-11-07For RST generated due to state mismatch during handshake, don't setDaniel Hartmeier
th_flags TH_ACK and leave th_ack 0, just like the RST generated by the stack in this case. Fixes the Raptor workaround. ok beck@, markus@
2004-11-07struct ucode.length should be in number of elements (u_int32_t), notDaniel Hartmeier
number of bytes. reported by Pyun YongHyeon, patch from mickey@, ok deraadt@
2004-11-06fixed types in ieee80211_regdomain (thanks to Joerg Sonnenberger). now itReyk Floeter
also works with gcc3 (ok naddy@).
2004-11-06disable changing softdep linkcnt (last commit). it can cause a loop thatTed Unangst
locks the vnode twice, and we can't fix it right now. discussed with pedro.
2004-11-06sync pci idsReyk Floeter
2004-11-06regenReyk Floeter
2004-11-06added new ar5212 device idsReyk Floeter
2004-11-05we want ukphy on the media; miod okTheo de Raadt
2004-11-05Better granularity between wsmouse support and wsmoused support, as oneMiod Vallat
could use wsmoused with a serial mouse; wsmoused support will now be compiled regardless of wsmouse attachments, unless option SMALL_KERNEL. Thanks to mickey@ for pointing out my mistake.
2004-11-04Only compile the mouse pasting code if wsmouse is configured in the kernel;Miod Vallat
shaves a few dozens of KB on installation media.
2004-11-04fix memory leak on deleting a screen; from janjaap@stack.nl via pr3977Michael Shalayeff
2004-11-04implememtation -> implementationPedro Martelletto
2004-11-04Stricter rules for isadma_bounce, shaves 4KB off a RAMDISKB kernel.Miod Vallat
2004-11-03Use IFM_IEEE80211_IBSS for IBSS mode instead of IFM_IEEE80211_ADHOC.Claudio Jeker
With this ipw(4) works in IBSS mode. ok millert@ aaron@ damien@
2004-11-03register typoReyk Floeter
2004-11-03oops missed to set devtype here rather than in sti_sgc.c:1.23Michael Shalayeff
2004-11-03Call ipw_stop() at the beginning of ipw_init(): this fixes a memory leakDamien Bergamini
when ipw_init() is called on ENETRESET. Call ipw_stop() when the radio transmitter is turned off. ok claudio@
2004-11-03Return 1 instead of 0 in ipw_intr() if the interrupt is handled by us.Damien Bergamini
Ignore invalid 0xffffffff interrupts. ok claudio@
2004-11-03IFM_IEEE80211_ADHOC -> IFM_IEEE80211_IBSS in ipw_media_status().Damien Bergamini
ok claudio@ aaron@
2004-11-03Calling ipwcontrol -k without downloading the firmware first lead to aDamien Bergamini
panic. Fix the ipw_free_firmware() function itself instead of fixing the callers. ok claudio@
2004-11-03Fix calls to bus_dmamap_create().Damien Bergamini
2004-11-03knf, use positive error values (suggested by miod@)Reyk Floeter
2004-11-03Enable pcscp; from Martin Reindl <mreindl at catai.org>Xavier Santolaria
ok miod@.
2004-11-03Revert to 3.6-current code since IBM onboard mpt's act up. Found by pval & ↵Marco Peereboom
deraadt no cookie for me...
2004-11-02fix getting strings off a device which broke with the removal of theDavid Gwynne
UQ_NO_STRINGS quirk.
2004-11-02we do not just delete lines but rather comment 'em outMichael Shalayeff
2004-11-02Remove __HAVE_NWSCONS and related remnants of pre-wscons days; no functionalMiod Vallat
change.