- "out of resource" errors cause receive buffer chain corruption
- resets can confuse the interrupt handler
- multi-cast setup causes receive buffer chain corruption
- shared memory setup incomplete
* Enhance effiency by avoiding unnecessary shared memory access,
improved handling of receive frame & buffer descriptors, and
introducing an `asynchronous' option when issuing 82586 commands.
* Exclusively use offsets relative to the bus handle representing the shared
memory area to formulate accesses to the chip's data-structures. The
front-ends provide glue functions that cater to the chip's endian-
sensitivity, to perform the actual device access (note: single-byte
accesses are done here using `bus_space_{read,write}_1()').
This concludes the transformation into a bus-independent driver module.
* Abolish C structures to access chip data-structures; instead use macros
that take indices and offsets relative to the bus handle representing
the chip's resources.
* Include the old version of this file wholesale, until all drivers
have been updated to use the MI 82586 code.
- allow get/set of enums SUNAUDIO_SOURCE and SUNAUDIO_OUTPUT rather
than returning EINVAL
- add missing SUNAUDIO_MONITOR_CLASS case in query_devinfo
- convert SUNAUDIO_MONITOR case from a MIXER_CLASS to a MIXER_VALUE
like it is supposed to be
- the labels for outputs/record class were swapped: fix it
this patch allows "mixerctl" to work properly on a sparc
- Closing the audio device should stop recording.
- Manipulating the pause function could accidentally start playing or recording.
- AUDIO_FLUSH could accidentally start playing or recording.
- guard against synchronous I/O completion
- avoid race conditions
- use bgetvp/brelvp to properly maintain the vnode holdcount
and clean/dirty buffer lists.
the historical link? flags for media select)
XXX No pullup request for this. But if sysinst supports explicit media
XXX selection in 1.3, this should be in 1.3 as well.
does a "restore data pointers" when reselected after disconnecting in
the middle of a DMA transfer). The driver needs a different way to know
which script to continue the DMA transfer. The message-in for the "restore
data pointers" loses the original "resume" script, and the driver would
attempt to continue the DMA transfer at the beginning of the current DMA
chunk, rather than at the point the disconnect occured. The result was a
spurious console message, and a trashed filesystem.
clobbers my SMC. Now I can really use a generic kernel with my SMC @0x300.
This change has been tested on various machines with ne2000 and 3c509 baords.
XXX All these probes sould be reordered after the release, with invasive
XXX probes at the end.
to be reloaded every time it is checked. This avoids a condition where
it can be cached in a register in such a way that updates to the flags in
an interrupt handler to not be noticed, which in turn causes the process
doing the i/o to sleep forever. Bug report and suggested fix from
Hiroshi HORIMOTO <horimoto@cs-yuugao.cs.sist.ac.jp>, PR $4460.
nullbuf (used to pad packets < ETHER_MIN_SIZE) is used for all tl
interfaces. Allocates only once, and never deallocate it (as we can't say
if another instance of the driver is interface is using it).
i/o-mapped space to always be used), we discover that at least one
ThunderLAN interface can't read the EEPROM properly if memory-mapped
access is used. Kludge around this for now by "prefering" i/o space.
to look specifically at the address it was provided *only*, since the
scan isn't safe (it can stomp on cards that will be probed later, like
NE2000 clones).
- Do PCI space configuration like the other drivers. In particular,
don't _disable_ the space we're not using because some lame firmware
implementations might not reenable it on warm boot. Also, prefer
memory space always.
- Make match and info-gathering in attach table-driven.
- Rearrange things a bit to be a bit more visually pleasing during boot.
Also, fixup some #include problems.
- clean up debug code
- Don't check ATAPI signature when probing ATA drives, ATAPI devices were
detected before
- Reset controller after disks probes. The probe, with some combinations of
ATA/ATAPI device keep it in a mostly working, but strange state (with busy
led on)
- The WDCF_IRQ_WAIT flag is now cleared by wdc_ata_intr and wdc_atapi_intr
when appropriate (helps recover from failure conditions)
- In wdcunwedge, send ATAPI_SOFT_RESET to non-ata drives
(helps recover from failure conditions)
- in wdctimeout be a bit more verbose when we missed an interrupt
- Always Increment xfer->c_skip where it should be
- Set the ITSDONE flag when a polled command completed.
for 12-bit addresses. This causes PCMCIA cards mapped at 0x400 and higher
to not function properly. However, the range 0x300-0x3ff causes some
laptops (e.g. the NEC Versa 6030X) to hang if cards are mapped there.
So, after some experimentation, we compromise. If the probe discovers
a 12-bit address bus, don't trust it. Instead, use the range 0x330-0x3ff.
This has been shown to work on the laptops that the 0x300-0x3ff range is
known to fail on.
supports generic SMC91cxx-based ISA cards and the built-in SMC
Ethernet found in some laptop docking stations.
Thanks much to Andrew Gillham <gillhaa@ghost.whirlpool.com> for
making this work!
more robust in resource shortage situations, basically identical to
code I added to the "ahc" driver some time ago.
Thanks to Brad Spencer for the testing help.
* Make the ring buffer size and water marks patchable, and allocate the buffer
separately.
* Do the ttymalloc() at attach time.
* Reorganize the receive buffer so the status and data pair are next to each
other. This is slightly faster.
* Make sure we actually do turn off interrupts in comclose() if we have DDB
configured and it's not the console. (D'oh!!!!)
* When we exhaust the current transmit run, turn off transmit interrupts in
comintr(), so we're fairly sure we don't get another one.
* Nuke the silly lsrmap[] idea; it's slower in the normal case.
* Cache the l_rint pointer in the soft interrupt routine.
* Carrier detect (TS_CARR_ON) is based on the actual DCD bit, even if it's
being ignored.
* Set TS_WOPEN early on in zsopen().
* Don't disable interrupts on the console during close if we have DDB.
Inert changes:
* Don't handle ZS_HWFLAG_NO_DCD here; the frontend does it.
* Deprecate `register'.
* Use SET(), CLR(), and ISSET().
More performance changes:
* Rototill receive handling; use a backpressure mechanism to prevent livelock.
* Output silo/ibuf overflow warnings at most once per minute, from a callout.
* When we exhaust the current transmit run, turn off transmit interrupts in
zstty_txint(), so we're fairly sure we don't get another one.
* Make the ring buffer size and water marks patchable, and allocate the buffer
separately.
* Do the ttymalloc() at attach time.
* Reorganize the receive buffer so the status and data pair are next to each
other. This is slightly faster.
* Make sure we actually do turn off interrupts in comclose() if we have DDB
configured and it's not the console. (D'oh!!!!)
* When we exhaust the current transmit run, turn off transmit interrupts in
comintr(), so we're fairly sure we don't get another one.
* Nuke the silly lsrmap[] idea; it's slower in the normal case.
* Cache the l_rint pointer in the soft interrupt routine.
drochner