| Age | Commit message (Collapse) | Author |
|---|
|
|
|
need to advertise the capability; tested by tsar at polarcap dot org
|
|
combination, found on the Tadpole SPARCbook 3 only (later model use the
Power9100 and are already supported).
Untested for lack of hardware and volunteers, but carefully crafted; if it
does not work out of the box, this should be a good starting point.
"worth having in the tree" millert@
|
|
and give be archs a better chance at getting the correct mac address as tested on the mac and hppa where mac has an additional bug w/ the phy bit still remaining and hppa seems fine; jason@ ok
|
|
|
|
a) Set xs->status rather than cmd_tables->status because there will be
no interrupt processing to move it from cmd_tables->status to
xs->status.
b) Set cmd_c.status to correct value (CMDST_SENSE_DONE) when an active
sense command is reset.
c) Don't put a reset command from the ready queue into the free_list
twice, once in siop_scsicmd_end() and once manually.
Condition a) meant that the scsi layer was seeing successfully
completed i/o's (xs->error == XS_NOERROR) when they were in fact reset
and should have had xs->error == XS_TIMEOUT or xs->error == XS_RESET.
This meant lost data on output, and random or zero'ed data on input.
Condition b) meant that the wrong bus_dmamap_sync() was called, though
the actual action was apparently identical.
Condition c) meant that the free_list could become corrupt.
The problem was discovered by pb@ on a heavily loaded server that
experienced timeouts. This fix was tested by pb@ and henric@ to prove
it did not affect normal processing. If nothing else it will provide
better error messages if the problem is ever encountered again.
Probably a good candidate for -stable if pb@ can successfully
reproduce his timeout problems and not have his server crash.
|
|
freebsd does as well
|
|
|
|
|
|
|
|
rescinded 22 July 1999. Proofed by myself and Theo.
|
|
|
|
|
|
random lockups; tested by brad@ on all models
|
|
cgthree. ok miod.
|
|
ok pb@
|
|
before we switch to wsdisplay, and this can clobber the font image in some
cases, especially lower (<= 1024x768) resolutions.
|
|
ok jason@
|
|
where required and modify existing ones to use a consistant delay(25).
From NetBSD with minor mods.
ok mickey@.
|
|
ok tdeval@ deraadt@ miod@
|
|
bug in the Momenco firmware on the Leopard-V, which now has one port
working.
|
|
of working on sparc64.
|
|
|
|
|
|
|
|
warn about no acb where applicable; krw@ ok
|
|
|
|
for hppa/osiop; krw@ ok
|
|
|
|
- remove a bunch of #if 0 stuff (most of which referred to dc_unit, grr)
- delay a bzero until necessary
|
|
|
|
|
|
ok mickey@
|
|
The lm driver provides support for the National Semiconductor LM series
hardware monitors and register compatible chips. It supports LM78,
LM78-J, LM79, Winbond W83697HF, W83627HF, W83781D and W83782D chips.
Tested and ok'ed by millert@ and henning@.
|
|
|
|
Fix an indent problem and an extra blank space.
ok mickey@
|
|
First, the only way to get there was if the timeout fired, in which
case timeout_del() is a noop. Second, it will be called in
osiop_scsidone() for every active command when osiop_reset() is called
in osiop_timeout().
From mickey@
|
|
|
|
|
|
Eliminate a couple of unused debug defines and variables.
ok mickey@
|
|
From mickey@
|
|
Fix setting of data buffer length when doing auto request sense.
Add a little general paranoia about setting data buffer length.
Eliminate $ifdef'ed field in acb structure, leaving diagnostic code
using it #ifdef'ed.
Mostly suggested by and ok mickey@.
|
|
|
|
In particular fix dma memory handling and as a result request sense
processing.
Much input/advice/testing from Mickey.
ok mickey@
|
|
allocated w/o cutting short onto the pmap layer
|
|
code, plus an extra ahc_flush_device_writes().
ok deraadt@ tdeval@
|
|
ok mjacob@.
|
|
1) Eliminate some magic numbers.
2) Eliminate some double semi-colons.
3) Rename iha_scsi_req_q to iha_scb.
|
|
1) Return XS_DRIVER_STUFFUP if request sense cannot be loaded.
2) Try to avoid bus_dmamap_unload()'ing maps that have already been
unloaded.
3) Try to avoid bus_dmamap_load()'ing maps that are loaded.
4) Be extra paranoid and ensure bus_dmamap_sync()'s are done before
a bus_dmamap_unload().
|
|
1) Actually map and sync scatter gather list that is also DMA'd.
2) Actually map and sync request sense buffer.
3) Actually POSTREAD/WRITE sync data buffer after I/O, rather than
skipping it because the I/O was successful and no chars were left to
xfer.
4) Eliminate extra fields, clarify some names, reorder fields to
clarify their use.
5) Put common scatter gather preparation logic into separate function.
6) If insertion of request sense command fails, complete active
command with XS_SENSE, but zero'd sense data.
7) Eliminate physical address arithmetic as much as possible,
centralizing remaining manipulations in code actually sending
addresses to adapter.
|
