summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2007-08-24On the older devices, apparently early configuration (and doing it againTheo de Raadt
later) removes the need for forced device disconnection. And now they do not bounce back a few times, but go to 500mA on the 2nd connection.
2007-08-23Add commented-out entry for siop@gsc, the onboard Fast-Wide SCSI on theMark Kettenis
735 and 755.
2007-08-23First stab at a driver for the onboard Fast-Wide SCSI on the 735 and 755.Mark Kettenis
Only partly tested; works fine if there are no devices on the SCSI bus.
2007-08-23Look for STI firmware at the additional address given by the firmware. ThisMark Kettenis
removes the need for the hardcoded address for the internal Visualize EG on the B180L, and makes it work on the C180. Tested on a handful of other models too. ok miod@
2007-08-23add a new ieee80211_save_ie() function to make a copy of anDamien Bergamini
information element in a management frame. the authenticator has to make a copy of the RSN/WPA1 IE from (Re)Association requests so that it can be compared with the one included in message 2/4 of the 4-way handshake.
2007-08-23add ieee80211_get_hdrlen() to compute the size of the 802.11 headerDamien Bergamini
of a management or data frame (check if the i_qos, i_ht or i_addr4 are present).
2007-08-23the authenticator can't differentiate 4-way handshake message 2 fromDamien Bergamini
message 4 based on the EAPOL-Key frame secure bit. the only thing that differentiate these messages is the presence of the RSN/WPA1 IE in the EAPOL-Key frame data field. only message 2/4 contains this IE.
2007-08-23add pseudo-header definition for TKIP MIC computation.Damien Bergamini
2007-08-23fix ieee80211_map_ptk() for TKIP.Damien Bergamini
2007-08-23implement the completion half of the tx path. the driver keeps track ofDavid Gwynne
mbufs we've put on the hardware with the nxe_pkt struct. this struct contains a unique id for each pkt which is used the hardwares tx descriptor so we could identify which pkts its finished transmitting. turns out the hardware doesnt return ids though, it tells us which slots in the tx ring its up to. so now i stash the slot a pkt went into in the pkt_id field.
2007-08-23configure the NIU to rx packets that will fit into the mbufs we willDavid Gwynne
allocate for it, rather than the default of 1600.
2007-08-23support ArgumentX - unbreaks more gnu cvs client things.Joris Vink
2007-08-23missing 'u' option for update caused gnu cvs client to not functionJoris Vink
with our server mode. Unfortunatly we dont really support rcs diffs for update just yet.
2007-08-23wrap writes to the doorbell register.David Gwynne
2007-08-23the first read or write in the crb space after you switch the windowsDavid Gwynne
tends to be junk, except in the pci regs for some reason. read the crb window register back after setting it to force the junk to go away. this fixes some really weird issues i had.
2007-08-23program the niu with the mac address in the right order.David Gwynne
2007-08-23tweak the register macros a bit. add macros for the NIUs port registersDavid Gwynne
which are useful to see whats happening on the wire.
2007-08-23allow RSTs with th_seq == seqlo +- 1, reduces the amount of 'loose state'Daniel Hartmeier
messages, add 'dir=' part to 'loose state' message, ok henning@, markus@
2007-08-23write out the object url in the error message if that's an object error andMarc Espie
not a basic listing repository issue.
2007-08-23fix v* order in keep_most_recent.Marc Espie
2007-08-23Check fgets for failure and do a proper newline strip.Ray Lai
Initial diff from Charles Longeau. OK otto and moritz.
2007-08-23Execute ProxyCommands with $SHELL rather than /bin/sh unconditionallyDamien Miller
2007-08-23Support "Banner=none" to disable displaying of the pre-login banner;Damien Miller
ok dtucker@ deraadt@
2007-08-23login_cap.h doesn't belong hereDamien Miller
2007-08-23missed include bits from last commitDamien Miller
2007-08-23unifdef HAVE_LOGIN_CAP; ok deraadt@ millert@Damien Miller
2007-08-23"correctly" activate high-power charging on (hopefully) all devices,Theo de Raadt
based on work in barry/bcharge; tested on a few units. still not tested on pearl-style units... if someone has one, contact me please
2007-08-23syncTheo de Raadt
2007-08-23more berry device typeTheo de Raadt
2007-08-23syncTheo de Raadt
2007-08-23blackberry pearlTheo de Raadt
2007-08-22Add bus_space_vaddr to i386 arch.Marco Peereboom
Comments and ok kettenis, art & miod
2007-08-22Allow ntpd to log sensor offsets and adjtime calls to syslog at LOG_DEBUGChris Kuethe
priority. ok gwk, mbalmer, weingart "explicit non-ok from" henning
2007-08-22add a ieee80211_get_txkey() function to determine the key to use forDamien Bergamini
transmitting a frame to a given node. change ieee80211_encrypt() so that it now takes the key as parameter. this change is required because drivers doing hardware crypto need to know what key is being used.
2007-08-22- add k_rxmic and k_txmic fields to struct ieee80211_key to store theDamien Bergamini
Tx/Rx MIC for TKIP. - add two functions to map a PTK and a GTK to an IEEE 802.11 key and use them in ieee80211_input.c instead of duplicating the same code. properly set Tx/Rx MIC in the IEEE 802.11 key in the case of TKIP. - add ic_psk to struct ieee80211com to store the pre-shared key. - fix setting of the SECURE bit in outgoing EAPOL-Key frames. - when receiving msg 2 of the 4-way handshake, deauthenticate the station if the RSN IE does not match that of the (Re)Association request. - before parsing an RSN or WPA IE, check that there's enough room for the version field (2 bytes) which is mandatory. - various tweaks while i'm here.
2007-08-22when running out of rx buffers, allocate a new mbuf and copy the rxDamien Bergamini
buffer into it using m_copym2() instead of just dropping the received frame. otherwise, it may happen that all rx buffers are queued by tcp_input() waiting for a frame that will never arrive because we are dropping it. when it happens, rx stop working and it is hard to recover from this. the problem is not visible when using WEP because WEP decryption is currently done by software and that rx buffers are copied into mbufs by the software decryption code. while i'm here, increase the number of rx buffers allocated by wpi(4) from 80 to 96 to limit the number of copies. problem and solution found by bluhm@ - thanks.
2007-08-22Subsequent address ranges start at index 1, not index 0.Mark Kettenis
2007-08-22Remove ``#define SDMMC_DEBUG''. Avoids redefinition error when compilingMarcus Glocker
a kernel and setting SDMMC_DEBUG option. ok deraadt@
2007-08-22Update to tzcode2007f grom elsie.nci.nih.gov; no actual code changesTodd C. Miller
2007-08-22err msg looks better with a \nMarc Espie
2007-08-22plural of ID is IDs, not ID's, ok mk jmcHenning Brauer
2007-08-22update to tzdata2007g from elsie.nci.nih.govTodd C. Miller
2007-08-22fix compilation with KDGBJasper Lievisse Adriaanse
from Nicholas Marriott ok deraadt@
2007-08-22kill trailing whitespace;Jason McIntyre
2007-08-22sort usage();Jason McIntyre
2007-08-21No need to keep RLOGIN_PORT and RSH_PORT defines hanging around since theirKenneth R Westerback
uses were eliminated five years ago. Spotted by Diego Casati. ok deraadt@ hshoexer@
2007-08-21Remove two out-of-date comments.Mark Kettenis
2007-08-21add AUTHORS sectionJasper Lievisse Adriaanse
no objections from kettenis@
2007-08-21Fix up the -s description, with help from tom@Todd C. Miller
2007-08-21http://openssl.org/news/patch-CVE-2007-3108.txt; ok pval, deraadtMarkus Friedl