re-enabled with ASC_IOASIC_BOUNCE.
All DMA buffered processing is now done in the bus-specific DMA routines
in asc_ioasic.c and asc_tc.c.
Disable several informational messages dealing with non-empty FIFO conditions,
but allow enabling with ASC_DIAGNOSTIC for troubleshooting.
- It is now possible to handle devices that want "looping" DMA,
e.g. the SoundBlaster correctly. The WSS and SB drivers use this.
To do this several new methods were introduced in audio_hw_if.
- Different silence handling (forced by previous change).
- The audio driver can now be mmap()-ed, but due to problems in
the VM system only for writing for now.
- The OSS (Linux) audio emulation takes advantage of some of the
new features.
* Move clockvar.h (header file for generic clock code) to sys/dev/dec.
* Move if_le_dec.c with DEC padded LANCE-dma (pmax/pmax, ioasic, vax 3400)
to sys/dev/dec. Remove from sys/dev/tc.
* Declare attribute le_dec_subr in /sys/dev/dec/files.dec,
use if_le_dec.c when it's defined.
* Move IOCTL asic declaration from pmax and Alpha MD machine
files to sys/dev/tc/files.tc.
* move TurboChannel and ioasic if_le attachments from pmax and Alpha machine
config files to /sys/dev/tc/files.tc. Add le_dec_subr attribute.
* Add if_le_dec attribute to if_le_ibus (pmax ds3100 pmax) LANCE attachment.
zero. We use this later when the COMPLETE message comes in to set the
final residual count to zero. The flag is reset if the target resumes
a data phase. This obsoletes the `AUTOSAVE' quirk (for this driver).
Also, avoid overwriting the residual count if a SENSE was appended to
the current transaction.
renamed AUDIO_ENCODING_SLINEAR and AUDIO_ENCODING_LINEAR reverts to the
NetBSD 1.2 sematics. A kernel with COMPAT_12 defined will accept
AUDIO_ENCODING_LINEAR and treat it as before, without COMPAT_12 it
will be rejected.
'inline' as well as static. mark prototypes for static inline functions
as possibly unused (with __attribute__ ((unused))), to avoid generating
warnings when compiling without optimization but with most ports'
default warning flags. Clean up prototype list spacing, and make it more
consistent.
connect to their 1-wire bus like the DS2404 "EconoRAM Time Chip"
- and -
* Interface function definitions for this kind of chips. Currently only
low-level byte_read and _write, implenented as inline functions.
This functions take a struct ds_handle * (also defined here), which contains
pointers to bit-read/write and reset functions.
Eventually, prototypes for memory-access functions should go here, and the
1-wire bus should be made a BSD auto-configuration bus.
(a) The interrupt is a RESEL interrupt, and
(b) our state is SELECTING.
This condition can occur in perfectly normal operation if we are using
DMA to select the target and we are interrupted by another target
reselecting us. Per discussion with Paul Krannenburg.
Every ccb locks 64k of memory for dma buffers.
Instead of AHA_CCB_MAX ccbs using 1MByte only sc_link.openings ccbs
per device are allocated. Thus we now use only 128KByte per device present.
- If the partition is already open, skip the open/close step. (Sync
with other disk drivers.)
- foosize()'s return value is in DEV_BSIZE units; adjust the size obtained
from the disklabel accordingly.
- Pass correct arguments to ofdopen() and ofdclose().
- If the partition was previously open, don't do the open/close steps.
(Sync with other disk drivers.)
- foosize()'s return value is in DEV_BSIZE units; adjust the size obtained
from the disklabel accordingly.
greater flexibility in its use. Additionally, add support for "geometry
emulation". This allows the "geometry" of the "disk" to be specified
at config time, providing near-perfect emulation of disklabel-less floppies,
CD-ROMs, etc., including non-512-byte sectors. If a geometry is not
specified at config time, a default based on 1M cylinders will be used.
clock rate for this board on Alpha/PCI systems. Under x86/PCI, the
board f/w will correctly tell you "I'm running at 60Mhz", so the code
that preserved that across a board reset (which would drop the chip
back to 40Mhz) worked fine. On the 8200, the chip was saying "I'm 40Mhz"-
which wasn't true. This turned out to be okay as long as you didn't have
any FAST or UltraFast targets- In fact, setting the chip to 40Mhz allowed
you to run up to 8Mhz SCSI. Unfortunately you die bigtime on the devices
that go faster than that. The fix here is to only use what the chip tells
you the clock rate is in the cases you don't really know (sbus is the
only case where this could be different, although with 66Mhz PCI coming up,
this may change).
call for the board's memory space to be PCI_MAPREG_MEM_TYPE_32BIT_1M or
PCI_MAPREG_MEM_TYPE_32BIT depending on the board ID. Also, remove a
bogus extra argument to an interrupt-establishment-error printf. Problems
pointed out by Jarkko Torppa <torppa@cute.fi> in PR 3753, but fixed slightly
differently than he suggested.
(1) fix a printf format (%x to print int, not %lx).
(2) fix probe of 4th chip/16th channel (used to tell whether or not the
board is a 16- or 32-port board) by removing an incorrect offset so
that the code matched its comments. (!!!)
(3) fix storage of chip number in per-channel structure so that it actually
stores the chip number, rather than the chip offset. This allows the
driver to work with more than the first four channels (i.e. with chips
other than chip number 0, which happens to have an offset of zero). (!!!)
MESSAGE_REJECT in response to SDTR or WDTR. Because of this, the
printfs that indicate refusal of sync/wide negotiation are unneeded
in normal operation. In the __NetBSD__ case, disable them by default.
They, like the other extra-verbose ahc driver boot messages, may be
reenabled with "options DEBUG". The behavior in the !__NetBSD__ case
is unchanged.
after synch negotation during device probes, the 53C94 stops with the target
reqesting more command data. The fifo appears to still contain a couple of
bytes, so transfer them. If the target still wants command data after that,
do a transfer pad. The target seems to function normally after this. A
much better solution thant simply rebooting!
Also fix a problem when compiling with DIAGNOSTIC - a mismatch between using
DIAGNOSTIC and MACH_DIAGNOSTIC caused the compile to fail.
Try to start another target and a device disconnects.
- Add NetBSD autoconfiguration support.
- Rearrange code slightly to minimize the number of #ifdefs.
- Don't use a structure to access CSRs. Use macros that DTRT for
the NetBSD and FreeBSD cases.
- Deal with alignment contraint on Alpha - add 2-byte padding at the
beginning of the RFA, so that the data will be 4-byte aligned, after
the 14-byte Ethernet header.
Thanks to Matthias Drochner for the testing, and David Greenman for
the feedback on the changes.
cast if it isn't. (These casts aren't necessarily safe, because of
alignment issues, but they allow the code to compile with prototyped versions
of bus_space_{read,write}_multi_2().)
if VOP_BMAP() does not produce a translation.
IO_SYNC is used to prevent dirty file cache buffers. On a ffs filesystem,
once a hole is filled, subsequent vnd accesses find will find valid
VOP_BMAP() translations.
Concerns:
* is the assumed semantics correct for all filesystems?
* do we actually want the automagic extension on the VND
backing store..
initiated by vnd_strategy(). This allows for more natural error handling
and solves two bugs:
* vnd can disk_unbusy without disk_busy (PR#2657)
* b_resid is set correctly on the external at the end
of a transfer in case of an error.
mark mulaw as emulated (it's done via a mapping table)
use GUS query encoding routine, even on GUS max, since the encoding
stuff is handled by the GF1.
XXX might not be right for recording, but it's correct for playback.
Set the encoding parameters slightly differently.
Remove the SW encoding/decodinf functions from this interface
and move them to the audio_parameter struct; this is both more efficient
and flexible.
This allows the front end to override the default DCR (byte-wide DMA,
x86 byte order, 8-byte FIFO) with different transfer size, byte order, DMA
parameters, and FIFO threshhold. If the loopback select bit is not set for
normal operation, the default is used instead.
Inspired by thoughts from Bernd Ernesti.
* epinit() had both explicit xcvr selection for 3c589 and a call
to epsetmedia(). The first is redundant; delete it, and EP_COAX_DEFAULT.
* Update comments to reflect 3c589 and 3c509B fixes. Fix typos.
- Determine and remember if we are a floppy.
- Workaround for what is apparently a firmware bug - ignore the sector
size returned by the device. On my Firepower's floppy, block-size
is the same as max-transfer, which causes Lossage.
- Don't read the disklabel on a floppy; do what the ISA floppy driver
does, which is assign the entire disk to each "partition", although
we do not deal with the density stuff.
- FIREPOWERBUGS -> FIRMWORKSBUGS
- Some general cleanup.
by default if it's usable, but falling back to I/O space if mem isn't usable.
If NCR_IOMAPPED is defined (default on the x86), prefer I/O space
then fall back to mem. Also, clean up the various memory consistency checks
so that they can deal with run-time determination of whether or not the
device is to be memory- or I/O-mapped.
by default if it's usable, but falling back to I/O space if mem isn't usable.
If TULIP_IOMAPPED is defined (default on the x86), prefer I/O space
then fall back to mem.
mapping register, maps it, and returns all of the relevant information.
deprecate use of pci_{io,mem}_find(), but leave them around (for a while)
for backward compatibility with third-party drivers.
arguments, so that a device can tell if its memory and I/O spaces are
enabled. The flags are cleared, depending on the contents of devices CSR
registers, in the machine-independent PCI bus code.
this code makes equal sense for memory and I/O space, prefer to map
the PCI front end via memory space (conditionalized on a patchable kernel
variable), and do a bit of other random NetBSD-specific cleanup. (These
changes were sent to Justin Gibbs on March 28.)
This can be disabled (to save a bit of space) with the NO_KERNEL_RCSIDS
options, which is present but commented out in the ALPHA config file.
In ELF-format kernels, these strings are present in the kernel binary but
are not loaded into memory. (In ECOFF-format kernels, there's no easy way
to keep them from being loaded, so they _are_ loaded into memory.)
* Don't print DMA_OUT message in the `normal' case of a 16-byte residual.
* ioasic-connected 53c94 devices are always clocked at 25MHz even on
machines with 12.5MHz TurboChannel.
in reset. If none there, try and get from the bus/platform specific code.
If a nonzero value for either, set the clock rate. This is why the PCI
card versions weren't working- they need to be set at 60MHZ, rather than
the default 40MHZ (which worked fine for the internal ISP chips on the
Alpha 8X00).
B) If a isp_poll returns failure (command never completed) to the caller
and no error is set in the xs struct, set XS_SELTIMEOUT. And then call...
C) Added isp_lostcmd function to try and ask the ISP chip about it's current
state as well as the state of commands for a particular target/lun. This is
going along to try and figure out why the very first command to the ISP always
seems to get swallowed up.
byte comparisons. Compare the ethernet addresses backwards on the
assumption that address number byte 6 has the most random distribution,
so packets not for us spend the least time in here.
is called *after* the driver `done' routine. This fixes disk I/O statistics
on SCSI devices.
Also, calling the `done' routine with a `complete' argument of 0 and actually
having it do anything meaningful loses in at least 3 ways, so just nuke the
argument altogether and don't call it this way. If the driver needs to do
some error handling, that's what `err_handler' is for.
we return from the driver expecting to come back due to an
interrupt, because the interrupt might not happen...
Do the untimeout in ncr53c9x_done instead of just before
almost every call to ncr53c9x_done as was done previously.
Make ncr53c9x_sense schedule its own timeout for the new
command it is starting (request sense), separate from the
timeout for the command that just completed.
- new MD variable: sc_initmodemedia, 79c960 cards can select
between two ports, but this needs to be done in the MI init mode page
- new defines for PCnet-ISA cards (79c960)
- new MD variable: sc_initmodemedia, 79c960 cards can select
between two ports, but this needs to be done in the MI init mode page
- new defines for PCnet-ISA cards (79c960)
- new MD variable: sc_initmodemedia, 79c960 cards can select
between two ports, but this needs to be done in the MI init mode page
- new defines for PCnet-ISA cards (79c960)
particularly older ones, don't do a SAVE DATA POINTER before disconnecting.
When the driver reconnects, it does an implicit RESTORE POINTERS, and
restores the data pointer from *before* the last transfer. If the driver
calculates the residual using the data pointer and doesn't account for
this, the residual will be wrong.
What we do is test for the SDEV_AUTOSAVE quirk when we see a disconnect
message, falling through to do a SAVE DATA POINTER if the quirk applies
to this target.
The workaround was suggested by Mycroft, who also provided the explanation
of the problem. Actual code was (mostly) ripped off from the 53C9x driver.
Anne Hutton <hutton@isi.edu>]:
- add support for Adaptec 155 PCI ATM cards (e.g. ANA-5940)
- add sc->is_adaptec to handle differences between cards.
- break out MID_MK_TXQ/MID_MK_RXQ seperate macros to handle
the new Adaptec format TXQ/RXQ.
- adjust en_dqneed to return 1 on ADP (since the Adaptec can
DMA anything in one DRQ/DTQ!)
- add hook for a bus specific reset function (adaptec has
a seperate reset register that needs to be hit when
resettting the midway).
- adjust DMA test to not worry about burst sizes on the
adaptec (since it handles it all for us!) and to handle
the new DTQ/DRQ format.
- add Adaptec DMA support to en_txlaunch() and en_service()
BUG FIXES:
- fixed receiver panic under heavy load ("lost mbuf in slot 0!").
when the reassembly buffer overflows, the T-bit is set in
the RDB and the data field is empty. en_service() sets up
a 4-byte (RDB size) dummy DMA without IF_ENQUEUE. but the
recv intr handling in en_intr() always does IF_DEQUEUE.
as a result, a successive recv intr loses its mbuf and
leads to a panic. the solution is to only IF_DEQUEUE if
the interrupt has non-zero length (indicating that there
is an mbuf to get). in order for this to work, EN_DQ_MK
must always be non-zero. we do this by or'ing in an unused
bit (0x80000).
reported by: Kenjiro Cho <kjc@csl.sony.co.jp>
- fix setting of transmit channel when txspeed[] is non-zero
(e.g. traffic shaping). the old scheme didn't work
properly (it allowed the same VCI to use multiple tx channels
thus defeating the txspeed[] parameter). the new scheme
statically assigns a VC to a channel when txspeed[] is set.
[note that the code to set txspeed[] isn't in the driver right
now since a MI interface to do this hasn't been made yet]
we add sc->txvc2slot[] and sc->txslot[n].nref for this.
reported by: Kenjiro Cho <kjc@csl.sony.co.jp>,
Milind M Buddihikot <milind@ccrc.wustl.edu>,
Dong Lin <dong@eecs.harvard.edu>
- when doing SRAM copies, be sure to round up the length to the next
largest word (otherwise the driver will try to do a byte clean
up DMA and then get an ID error interrupt).
MINOR CLEANUPS:
- clean up loops in DMA test
contributed by: Kenjiro Cho <kjc@csl.sony.co.jp>
- restructure and cleanup of en_read/en_write macros/inlines
- clean up some byte ordering stuff so that we are consistant throughout
the driver
correspond to the ANA numbers... certainly includes ANA-5940.
- add Efficient Nets product 0x0000. according to linux pci.h
the 0x0000 is an FPGA version of the midway card and the 0x0002
is the ASIC version.
pk