summaryrefslogtreecommitdiff
path: root/sys/dev
AgeCommit message (Collapse)Author
2002-12-30Add $OpenBSD$ tag.Thierry Deval
Thanks to Clarie Wouter (rimshot at pandora point be)
2002-12-30Update.Thierry Deval
2002-12-30Some cleanup.Thierry Deval
Re-work the SBP2 data manipulation to support concurrent node accesses. (That data manipulation MUST go into SBP2 code, eventually)
2002-12-30Implement the reconnect callback. This will be called whenever an SBP2 node'sThierry Deval
node_id changes, following a BusReset (happens when you add/remove a device to/from the FireWire chain). Re-work the ORB management to support concurrent node accesses.
2002-12-30As we still don't have a data_page_dir support yet, transfers of more thanThierry Deval
65535 bytes blocks silently broke. So limit ourself to 32k blocks, for now. That mistake gave us the false impression that huge transfers were fast. They usually resulted of 0-byte transfers. Fast indeed :-( Now, we have real data going through.
2002-12-30Some devices have bad ConfigROM crc, but contain valid Unit entries.Thierry Deval
So just ignore, and continue the auto-configuration.
2002-12-30Make the callbacks entry non-constant.Thierry Deval
2002-12-30Re-design fwohci_hadler_set() to use a new key3.Thierry Deval
That new key will be used for Request handlers to discriminate the requests by nodes. Key3 will also get the lenght field specifier... (may still change) Add an implementation for a BusReset callback that will be called whenever a node's node_id changes. This will allow us to work with more than one device at the same time...
2002-12-30proper HOSTAP_FLAGS_BITS; from merith@vantronix.netMichael Shalayeff
2002-12-23Repair minor indentation problem.Miod Vallat
2002-12-23dev/ata/ata.c should depends on wdc_base since it uses some callsAlexander Yurchenko
from dev/ic/wdc.c; from NetBSD. ok miod@ deraadt@
2002-12-23Fix complete botch in handling of 'openings'.Kenneth R Westerback
Assign a fixed value (SIOP_NTAGS) to the openings field in the adapter's template sc_link, rather than incrementing the value as cbd's are allocated. The template value is the one copied into each device's sc_link structure as it is created. Incrementing the value meant that each new device got a larger value for openings. The total number of openings claimed by devices on a bus soon exceeded the number of cbd's available. e.g. after 5 devices there would be 132 allocated cbd's, but the total number of openings claimed by devices would be 300. A heavy i/o load on an adapter with multiple devices could have caused the upper scsi layer to try to queue more i/o's than the driver had cbd's to store them in. Such i/o's would fail with EIO if they were started with SCSI_NOSLEEP (e.g. sdstart()) or were not queued within the specified retry limit. I/o's for devices 'later' on the bus would be more likely to trigger this behaviour, due to their inflated openings values. This is good candidate for -stable.
2002-12-22Define atagettrace_t type and use it instead of struct atagettrace.Alexander Yurchenko
ok costa@
2002-12-22No more magic wdc events types, give them names.Alexander Yurchenko
ok costa@
2002-12-20no need for these to sit in their own dirs, there not gonna be any sources ↵Michael Shalayeff
there anyway, all code is in the particular bus adapter's source
2002-12-19regenHenning Brauer
2002-12-19more O2Micro PCI-Cardbus; FreeBSD via pamifer at terra.esHenning Brauer
2002-12-19Protect WDCDEBUG_PRINT macro with do {} while (0)Alexander Yurchenko
ok costa@
2002-12-19fix a debugging printf on cmd wait timeoutMichael Shalayeff
2002-12-19proper barrier call in ie_ackMichael Shalayeff
2002-12-19comparam() does not need an spltty() for it, already called there.Michael Shalayeff
time-bound loops in com_common_cnputc() and lower to spltty(), which i guess is left from times when timeouts were processed at splhigh(). jason@ tested and ok, art@ ok
2002-12-18fix memory leak when downing an interface.Nathan Binkert
from Patrik Lindergren <patrik@lindergren.com>
2002-12-17Cleaning.Thierry Deval
2002-12-16Major KNF. Incentive from TeduThierry Deval
2002-12-15Fix/Cleanup some SCSI #defines.Kenneth R Westerback
1) Correctly define MODE_SENSE_BIG as 0x5a, not 0x54. 2) Delete duplicate #define's of some opcodes in scsi_disk.h. 3) Delete multiple #define's for same opcode in scsi_disk.h. 4) Replace uses of deleted opcodes in umass.c with remaining ones. ok millert@ tdeval@
2002-12-15and one more typoHenning Brauer
2002-12-15more typos fixed by torhHenning Brauer
2002-12-15typos; torh at bogus.netHenning Brauer
2002-12-14Using quirks that make more sense.Thierry Deval
(iPod is happy with this)
2002-12-13Some more cleaning...Thierry Deval
2002-12-13Limit our Data transfers only to the link speed imposed one.Thierry Deval
Writes are now on a par with reads (-:
2002-12-13Only WRITE_REQUEST_DATABLOCK are limited to the device's max_receive.Thierry Deval
Other writes are only limited by the link speed. Idea from NetBSD.
2002-12-13Define IEEE1394_MAX_ASYNC(), which gives the maximum (link speed dependent)Thierry Deval
asynchronous packet size. Idea from NetBSD.
2002-12-13Fix the memory leak introduced with the use of fwohci_block_handler_set().Thierry Deval
2002-12-13Malloc debug cleaning (MPRINTF).Thierry Deval
Uses option FW_MALLOC_DEBUG.
2002-12-13Enable support for SCSI over FireWire.Thierry Deval
2002-12-13Add preliminary support for the Serial Bus Protocol II (SBP-2) standard.Thierry Deval
As well as a first rough implementation of a SCSI over FireWire support, following the SBP-2 standard.
2002-12-13"Documentation" update.Thierry Deval
2002-12-13Better memory balancing (between alloc/free),Thierry Deval
some enhancements, lots of debugging, KNF...
2002-12-13Detach fwnodes first.Thierry Deval
Solves panics when removing the adapter while leaving the devices connected.
2002-12-13KNFThierry Deval
2002-12-12Always use queue(3) macros.Alexander Yurchenko
ok chris@ gluk@
2002-12-12statement after label.Artur Grabowski
2002-12-11smaller delays, unneeded splsMichael Shalayeff
2002-12-11send-pr -> sendbug here as well; Peter WernerHenning Brauer
2002-12-11sendbug(1), not send-pr; Peter WernerHenning Brauer
2002-12-10RegenMiod Vallat
2002-12-10Dec Pixelvision card, unfortunately not tga compatible enough.Miod Vallat
2002-12-10Translate mickeyspeak into english.Miod Vallat
2002-12-10Add support for VIA VT8231; from NetBSD, untested.Alexander Yurchenko
ok gluk@