summaryrefslogtreecommitdiff
path: root/sys
AgeCommit message (Collapse)Author
2003-03-03Enabled inclusion of ISP SBus f/w for GENERIC kernels. This tendsmjacob
to make people's days so much better.
2003-03-03Argh. Put back mickey's change here and also follow Ted's suggestionmjacob
wrt bogus identity assignments.
2003-03-03Add some A64 defines.mjacob
Fix egregious botch where ICBXOPT_ZIO is suppose to be bit #5, not the value '5'.
2003-03-03Get rid of ISP_DMA_ADDR_T and uses ISP_DAC_SUPPORTED to define themjacob
width of a dma address type- this also allows us to define macros that set upper words of a dma address. Remove STRNCAT. Add (untested) 10160 support.
2003-03-03spelling fixesmjacob
2003-03-03In isp_putback_atio, propagate iid && rxid for FC ATIOs.mjacob
LINT fixes for printing out the lun for recevied messages. Spelling fix.
2003-03-03Remove STRNCAT define.mjacob
2003-03-03For type 3 requests, the count is ISP_RQDSEG_T3, not ISP_RQDSEG_T2, that we needmjacob
to set up.
2003-03-03Add (untested, no h/w) 10160 support. Propagate some spelling fixes.mjacob
Add back in merged 'mhz' change. Use A64 variants to initialize ULTRA2 or better && 1240 cards. If PAE is ever implemented for *BSD, we'll be closer to being ready to go. Add in a commented out example for how to adjust idelaytimer on 2300s. If we have ISP_FW_CRASH_DUMP defined, then don't reinit the isp in ISP_CONN_FATAL case- that's up to the platform outer layer to do after it finishes saving a f/w crashdump. Replaces STRNCAT with SNPRINTF.
2003-03-03march through the trapframes on the stack trace until the very lastMichael Shalayeff
2003-03-03Use new pci_matchbyid routine. Add (unable to test, no h/w) 10160 support.mjacob
Use correct size to zero sdparam area for 12160. Compile tested: sparc64, sparc run tested: sparc64 (2200, 1040), sparc (sbus isp)
2003-03-03Make "pfctl -ss" output easier to parse. NO TRAFFIC -> NO_TRAFFIC.Cedric Berger
ok dhartmei@ henning@
2003-03-03set n->m_len, somehow the line have lost. thanks danielJun-ichiro itojun Hagino
2003-03-03use M_READONLY instead of defining M_SHAREDCLUSTER(). the change should haveJun-ichiro itojun Hagino
been made when M_READONLY was introduced. sync w/kame
2003-03-03Provide a functional d_psize() function for ramdisk block devices.Miod Vallat
ok deraadt@
2003-03-02Use priority queue for TCP ACKs that have no payload. Very useful onDaniel Hartmeier
asymmetric links. ok henning@, cedric@
2003-03-02only assign a new queue id if a->qid is 0. otherwise, check a->qid forHenning Brauer
validity and take that. validity checks provided by kjc@ ok dhartmei@ kjc@
2003-03-02The gem driver will sometimes unmap bus_dma mappings that are stillHenric Jungheim
in use by the hardware. Since the sparc64 has IOMMUs, an access by the hardware after the unmap operation will cause a fault. This replaces "disable" calls by "reset" calls. This increases the time between the command to disable DMA and when the maps are actually unloaded.
2003-03-01- only declare and provide pmap_map() for mac68k, until it bites the dust.Miod Vallat
Other m68k-based arches do not need it. - do not wait to allocate struct pv_page. Either pmap_enter() has been invoked with PMAP_CANFAIL and can live with this, or we are close to fandagoland anyways.
2003-03-01Provide real ieee fp*() functions; from NetBSD.Miod Vallat
ok millert@
2003-02-28intrcnt's are now int's to fit with kern.intrcnt assumptions; miod okJason Wright
2003-02-28Based on several comments from tedu:Jason Wright
- two variables 'err' and 'error', whacked - missing initialization in the error path for the case where an SA expired while off in crypto land. - a small bit of knf.
2003-02-28update lasttick in clockintr() so that time doesn't run away with usJason Wright
(based on a larger diff from henric).
2003-02-28do not entice people into playing with broken fsTed Unangst
ok deraadt@
2003-02-28make intrcnt[n] an int to fit with the kern.intrcnt changes in JanuaryJason Wright
2003-02-28Update firmware to latest levels- fixes several nasty bugs on the 2300 at least.mjacob
2003-02-28Revert previous change. This driver is shared with 4 other platforms.mjacob
Baloney gratuitous changes to comments w/o even a headsup to the author will be reverted.
2003-02-28Increase IDENTIFY command timeout to 3 sec, as required by someAlexander Yurchenko
drives; from NetBSD
2003-02-28take too. patch ahs to screw up once in a whileMichael Shalayeff
2003-02-28remove non-typo changes. sorry folksMichael Shalayeff
2003-02-28typo; from weissmanndudeMichael Shalayeff
2003-02-28Fix more evil typos.Miod Vallat
2003-02-28fix comments, from weissmanndude.Michael Shalayeff
make RAMDISK not include sys/conf/GENERIC
2003-02-28splsoftnet() around rn_lookup() which is not thread-safe.Cedric Berger
ok dhartmei@ henning@
2003-02-28add support for 3cr990 (not 3cr990a,b,c, etc) and 3cr990-fx; from linux driverJason Wright
2003-02-28more txp's; from linux driverJason Wright
2003-02-27regenAlexander Yurchenko
2003-02-27Add ServerWorks CSB6 ATA chip, found in dmesg from niklas@Alexander Yurchenko
2003-02-27regenAlexander Yurchenko
2003-02-27Add some more Promise chips (including SATA ones) and a littleAlexander Yurchenko
cleanup this section
2003-02-27print out the ethernet address in the dmesg.Nathan Binkert
from David Krause <openbsd@davidkrause.com>
2003-02-27make packet classification for altq work in the IPv6 caseHenning Brauer
ok dhartmei@ cedric@ + "looks good" mcbride@
2003-02-27Repair IPv6 support for tables.Cedric Berger
ok dhartmei@ henning@
2003-02-27regenMichael Shalayeff
2003-02-27add ENE cb1410 and sort out the EFFICIENTNETSMichael Shalayeff
2003-02-26Let this compile if no DDBMiod Vallat
2003-02-26Fix symbol clash that would prevent a kernel configured with both ncr5380Miod Vallat
drivers to link.
2003-02-26minor fix in a debug printfMichael Shalayeff
2003-02-26Remove an unnecessary structure copy from useage of setfault(), callDale Rahn
by reference, not by value, ok matthieu#, miod@
2003-02-26Better support for the ps/2-style HIL keyboards (those with a real numericMiod Vallat
pad and a few extra keys).