Do a fallback on reading stuff from the fabric. Some devices/initiators
don't correctly register their type with the fabric nameserver. This
seems to be due to a misinterpretation of what TYPE should mean for a
CT_HDR. In any case, do a fallback to try and catch these misentered
entities.
Add some stuff for default framesize and throttle. Fix some buglets.
- implement SIMPLEQ_REMOVE(head, elm, type, field). whilst it's O(n),
this mirrors the functionality of SLIST_REMOVE() (the other
singly-linked list type) and FreeBSD's STAILQ_REMOVE()
- remove the unnecessary elm arg from SIMPLEQ_REMOVE_HEAD().
this mirrors the functionality of SLIST_REMOVE_HEAD() (the other
singly-linked list type) and FreeBSD's STAILQ_REMOVE_HEAD()
- remove notes about SIMPLEQ not supporting arbitrary element removal
- use SIMPLEQ_FOREACH() instead of home-grown for loops
- use SIMPLEQ_EMPTY() appropriately
- use SIMPLEQ_*() instead of accessing sqh_first,sqh_last,sqe_next directly
- reorder manual page; be consistent about how the types are listed
- other minor cleanups
PowerPC and provides a PCI-host bridge, interrupt controller, two com ports,
two IIC ports, a timer, and a DRAM controller.
The driver supports PCI, interrupts and com ports.
to check if we get a RQCS_DATA_UNDERRUN - if we're an FC card, we may
not have RQCS_RU set- if it isn't set, we just lost a DATA XFR IU in the
middle of the exchange. In this case, we have to bomb out the whole xfer.
We had been getting silent data corruption before.
Fix receiver lockups. When writing the RX Kick register, we need back up
on descriptor since we advanced one in the for loop. That will be the
last descriptor we haven't processed which is what we should write.
switch back to MII mode. Keep a sliding window for TX segments and when it
gets > 2/3 full, request a TX interrupt (window gets reset when the h/w TX
queue is empty). Add dv_xname to a few printfs. With the above changes,
the gem driver will now work on Macintoshes, even in GigE mode. On a 733
PowerMac G4 it gets ~355Mb/s TX and ~280Mb/s RX to/from an Alpha XP1000.
XXX mii autoselect is still flakey.
errors for narrow transfers after a reselect. FreeBSD sym driver has a comment
about this, but their workaround (disable SCSI gross error reports) doesn't
work for me. Instead dissallow disconnect if the target is not wide
(FreeBSD doens't allow disconnect until the target has been fully probed, which
is why they may not have noticed my problem).
particular, make sure that all the SIASTAT_ANS bits are in the right state
so we don't do something inane.
Still doesn't actually bring the link up properly, but at least it negotiates
most of the time, and does it a little faster.
1) Do not call tlp_sia_update_link() in Nway mode, and do not look at SIASTAT
in any other place that nway_status(), where we first check that it's valid.
In other places, look at IFM_ACTIVE after having call nway_status().
2) Eliminate stupid MII_MEDIACHG calls, and arrange for nway_service() to
update status on every call.
3) Nuke the synchronous case of nway_auto() from orbit.
4) Do not call nway_statchg() when using manual configuration; tlp_sia_set()
does everything we need.
1) Set OPMODE_TTM in the default tsti_opmode, so that nway_status doesn't
blow up and report the wrong media type when statically configured. (This
code is a hack.)
2) Do not set IFM_ACTIVE (i.e. ignore SIASTAT_LS*) when in auto-negotiation
mode and negotiation has not completed (per 21143 manual).
3) Do not clear auto-negotiation mode; otherwise the chip will not
renegotiate on a link failure.
With these changes, 10/100 selection is more stable, and auto-negotiation
comes up with the right status and detects link, but the link does not work
unless it's hardwired. More work is needed.
It is unclear if this realy is needed and if, on which type of cards. I
haven't run accross a card that needs it yet. This may have been just
a copy & pasto from the ISAC interrupt handler carried over to IPAC.
Remove the clear-the-irq-after-enabling it dance (which had bad side
effects on some cards). Instead disble the ISAC receiver when we have
interrupts disabled. Adjust the interrupt handler to properly deal with
subtle differences of the ISAC implementation in IPAC chips.
if error occurs after status is collected) race condition
when using the status byte to detect completed commands (a command descriptor
could be recycled before the device disconnected), and make the
interrupt routine handling completed commands more efficient (no need to
scan target * lun * tag array any more).
there may be tagged commands still running when we queue a request sense
command.
Solve this by using 2 DSA entry per LUN
- Now that we have the command DSA before select, we can load T/L/Q in
SCRATCHC. This makes the selection timeout handler simpler.
- Avoid a race condition when setting the free flag in the cmd ring (see
comment in the script)
- don't forget to update the ID in the head of LUN table after a sync/wide
negotiation. This fixes the command timeout at the first data command
after negotiation (the bus reset handler did update the ID properly,
so subsequent commands were OK).
- for DMA interrupts, clear fifo if it's not empty. Leaving the fifo dirty
would prevent subsequent interrupts from coming in.
- Various improvements in debug messages
- misc cleanups.
scheduler slot. This costs a few more instructions but divide the size of the
scheduler ring by 2, saving 1k of onboard RAM (a bus with 15 devices would
overflow the on-board RAM by 128 bytes).
- Add support for DT transfers (aka Ultra/160) in esiop
Note that DT transfers are not enabled for 53c1010-33 rev 0 yet; if I trust
FreeBSD it has a bug which prevent them to do DT properly.
From the same source there may be issues with some revs of 53c1010-66.
taken from OpenBSD. Test hardware kindly provided by Intel. This still needs
management bits, and doesn't support older controllers, but that shouldn't
be hard to fix.
For this add another indirecton: the DSA in the LUN table points to
a table of DSA indexed by the tag number when tagged command is in use.
For non tagged command, the LUN DSA still points to the tables describing the
xfer directly.
yet.
If is restricted to SIOP which implement the load/store instruction, and
has 10 scratch registers (basically, 825 and newer, possibly 770).
It implements a different interface between host and script, using a real
ring for command starts, and improved support for reconnect which will allow
256 tag per device. It uses interrupt on the fly to signal complete command,
which allows several commands to be serviced per interrupt and doesn't require
the script to stop to signal command completion.
3 bits are lun address modifiers.
Remove code that (incorrectly) thought it was asking the f/w to only
PLOGI if not already PLOGI'd. The current f/w documentation tells us
that we have this backwards.
"true" at the appropriate times for non-PCMCIA interfaces. This
means that the ENETRESET path in cs_ioctl() now runs, thus fixing
multicast (and IPv6) on my Shark. Yay.
* Simplify cs_hash_index(): Rather than taking the bottom 6 bits of
a big-endian CRC32 and reversing them, just take the top 6 bits of
a little-endian CRC32.
WI_RID_SCAN_APS in previous commit works for Intel Pro/Wirelss 2011
with firmware 2.51.1.
It seems that the firmware automatically updated after the card runs on
Windows 2000 with 2011_2011B_CD_3.0 in Intel web site.
The card is Type 1 CF card and it doesn't have firmware in.
So we need to download the firmware image into the card before
touching it.
XXX downloading code should be written in generic (bus independent),
but I don't have enough information for now.
- The version string doesn't need to start with 'V'.
- Symbol firmware also support CREATE_IBSS.
- But it doesn't support ROAMING_MODE nor MICROWAVE_OVEN.
into a strange failure mode if we do it with disabled interrupt. When
(re-)enabling interrupts reset transmitter and receiver and clear any
pending state.
not pre-load the chip's Tx buffer, but instead waits for the Tx Ready
interrupt to transmit the first chunk of data.
* On the IOP310, set COM_HW_NO_TXPRELOAD, rather than COM_HW_TXFIFO_DISABLE.
This solves the "UART hangs" problem on the Npwr in a nicer way (i.e. we
get to use the FIFO, whee). The COM_HW_NO_TXPRELOAD happens to match the
Linux 16550 driver's Tx algorithm, and the "UART hang" was never observed
on the Npwr running Linux.
Eventually, we might want to eliminate the COM_HW_NO_TXPRELOAD, and simply
always use its algorithm. But it should be tested on more 16x50 variants
before we do that.
Kudos to Valeriy Ushakov <uwe@netbsd.org> for pointing out this solution
(which also happens to fix the stray UART interrupt issue on the Krups
Javastation), and to Allen Briggs <briggs@netbsd.org> for experimenting
with various methods of fixing this.
are used- didn't make a difference, but hey...
Put in commented out GFF_ID code- for use in future attempts to search
the fabric- this probably has to go thru the management server path.
Don't whine about handles we can't find if these are aborted commands
(we know we can't find the handles because we destroy handles after
a successful mailbox abort- we don't wait for the F/W to decide whether
it wants to return a status IOCB after this happens).
the data transfer. This is mandatory for data out commands (although none are
used for now), and not forbiddend for data in commands. Also record if we
did transfers any data.
May solve kern/16159 by making the probe more robust in face of fake identify.
for the same purpose (ignoring invalid interrupts).
For cards that are not able to stop all interrupts (or we don't know a way
to do that in software, at least) run the clearirq callback even when
ignoring an interrupt because we are not enabled. Otherwise the card would
stop interrupting.
Reserve a driver specific callout handle and an int value in the generic
isic_softc to allow card drivers to implement fancy blinkenlights.
It improves playing/recording quality greatly
and it was almost done by Yosuke Sugahara <penta@fuchu.or.jp>.
Thanks a lot!
Add support of slinear8, slinear16_le, slinear16_be.
pcmcia cards. Now that pcmcia attachements properly handle the activate
callback, this is no longer needed (and is suspect to cause completely
unrelated problems.)
count them when reading the NIC counters - it doubles the count. Read the
NIC counters to prevent counter overflow interrupts, but don't add them to
the interface counters. Don't bother reading the upper counts because they
are just latched when reading the totals.
Fixes final part of PR#11549.
on the following PRO/100 chips:
* i82558 step A4
* i82558 step B0
* i82559 step A0
* i82559S step A
* i82550
* i82550 step C
The interrupt delay is configurable on all microcodable chips. The
maximum "bundle" size (packet count) is configurable on all but the
i82558.
The microcode is enabled by setting IFF_LINK0 on the interface.
Derived from code in FreeBSD.
the interface was wired to full-duplex mode. Duh.
Also, add OPMODE_TTM to OPMODE_MEDIA_BITS, to insure that it is changed when
we switch between 10 and 100.
to be trying to wriggle out of supporting this well. Instead, use
GID_FT to get a list of Port IDs and then use GPN_ID/GNN_ID to find the
port and node wwn. This should make working on fabrics a bit cleaner and
more stable.
This also caused some cleanup of SNS subcommand canonicalization so that
we can actually check for FS_ACC and FS_RJT, and if we get an FS_RJT,
print out the reason and explanation codes.
We'll keep the old GA_NXT method around if people want to uncomment a
controlling definition in ispvar.h.
This also had us clean up ISPASYNC_FABRICDEV to use a local lportdb argument
and to have the caller explicitly say that a device is at the end of the
fabric list.
The ICH on-board Ethernet and some i82559 chips have a bug which
will cause a PCI protocol violation if the chip receives a CU_RESUME
command as it is entering the IDLE state by deasserting #CLKRUN.
(This is the so-called "resume bug" that we previously had an incomplete
work-around for on ICH chipsets.)
The work-around is to disable Dynamic Standby Mode, such that the
chip will never deasert #CLKRUN. Dynamic Standby Mode is disabled
by clearing a bit in the EEPROM and updating the EEPROM (and EEPROM
checksum).
Unfortunately, the chip will only consult the EEPROM setting after
a PCI bus reset, so a system reboot is required once the EEPROM
has been updated (the EEPROM update only needs to happen once,
and the driver usses a warning instructing the user to reboot the
system once the work-around has been applied).
Issue pointed out by David Brownlee, and code more-or-less lifted
from FreeBSD.
for bus_dmamap_sync calls. They'd been blindly ported from Solaris which
had *one* dma map for the entire control space, so offset was incremented
for the Request, Response and FC Scratch spaces. Tsk. There are three maps
in NetBSD. I should probably make them one anyway.
1. Fix setting of nominal fan speeds with ENVSYS_STREINFO.
2. Treat Winbond 83781D specially because it has a programmable divisor for
FAN3 unlike generic devices.
3. Set nominal RPMs.
4. Fix a typo in the code for setting FAN3's divisor for W83782 type
hardware monitors.
possible pending interrupt. This should avoid the interrupt loop described
in PR kern/15841.
It is safe to read status here, because if we're not waiting for an interrupt
we have no command pending, so the device should be idle.
(de)activate for pcmcia cards.
Implement detach/(de)activate for PCI cards.
Clean up internal state (free call-descriptors) if a controller is
detached while it has open connections.
B-channel and D-channel drivers separately) split the Fritz!PCI card
driver out of the isic driver.
The new device is called "ifpci" and uses the same D-channel driver as the
isic devices, but has it's own B-channel driver.
Make the ISA probe actually (silently) fail if no card is found.
XXX - need to aquire the SBUS variant of this card some day or have
Jaromir find an MCA one.
and move them in their proper places.
Move the BRI registry from layer 2 (duh!) to layer 4, so active cards
(which don't have layer 3 or layer 2 in their driver). Remove all remaining
hard coded controller and driver types. Remove any arbitrary hard coded
limits, at least those that show up in the internal API.
This fixes PR 15950.
the variant name is printed.
This fixes a problem that the card would be treated as ESP100 in
ncr53c9x_reset(), but not on couple other places (pointed out by
Andy Doran in private e-mail).
g/c now redundant case entry in ncr53c9x_reset()
Unconst pointer to f/w in the ispdv structure. Too many compilers get
unhappy over our walking the array. Make casts as appropriate so that
initialization in structure is still happy.
Limit length of fabric to 256. This will all go away soon.
Do a cleaner case of keeping multiple CPUs/threads from reading the
same response queue entries.
the generic layer 4 and layer 3 management system.
This should make the layer 4 driver API LKM clean - finaly.
Make the Fritz!PCI driver work again after resent changes (oops!),
noted by Frank Kardel (PR 15948) and Matthias Scheeler.
indicating an unhandled "command". ERESTART is -1, which can lead to
confusion. ERESTART has been moved to -3 and EPASSTHROUGH has been
placed at -4. No ioctl code should now return -1 anywhere. The
ioctl() system call is now properly restartable.
use "cp" instead of "ncp" as a temporary pointer into the SROM.
we still need the value of ncp here, it points to the next media block.
(update the copy of this code in tlp_21142_reset() just 'cause.)
this makes the BNC port on the Adaptec ANA-6911A work.
Add an ugly hack, #ifdef WI_AT_BIGENDIAN_BUS_HACK, to make the wi driver
access the underlying bus in big endian byte order.
This makes it possible to use this driver (probably one of the most
interesting ones) in the pcmcia slot of a stp4020 (nell) adapter at sbus. The
sparc ports bus_space_{read,write}_* macros do not provide a way to do this
cleanly now as they ignore the bus_space_tag_t parameter.
XXX - make bus_space_{read,write}_* on sparc do the right thing.
Audio-related stuff is left almost intact.
* support audiocs at ebus playback and capture
tested on krups and u5 (thanks, martin)
* make first attempt at supporting audiocs at sbus capture
* nb: full-duplex is not tested
* while here, fix CSAUDIO_MONITOR_MUTE to be of CSAUDIO_MONITOR_CLASS
i.e. outputs.monitor.mute -> monitor.monitor.mute
Ok by pk, eeh.
deal with shortages of the VM maps where the backing pages are mapped
(usually kmem_map). Try to deal with this:
* Group all information about the backend allocator for a pool in a
separate structure. The pool references this structure, rather than
the individual fields.
* Change the pool_init() API accordingly, and adjust all callers.
* Link all pools using the same backend allocator on a list.
* The backend allocator is responsible for waiting for physical memory
to become available, but will still fail if it cannot callocate KVA
space for the pages. If this happens, carefully drain all pools using
the same backend allocator, so that some KVA space can be freed.
* Change pool_reclaim() to indicate if it actually succeeded in freeing
some pages, and use that information to make draining easier and more
efficient.
* Get rid of PR_URGENT. There was only one use of it, and it could be
dealt with by the caller.
From art@openbsd.org.
it worked- but I ran into a case with a 2204 where commands were being lost
right and left. Best be safe.
For target mode, or things called if we call isp_handle_other response- note
that we might have dropped locks by changing the output pointer so we bail
from the loop. It's the responsibility of the entity dropping the lock to
make sure that we let the f/w know we've read thus far into the response
queue (else we begin processing the same entries again- blech!).
Allow SIOCSWAVELAN with WI_RID_IFACE_STATS to request
an immediate statistics update.
We might in the future want to have SIOCGWAVELAN with
WI_RID_IFACE_STATS return an error if a statistics gathering is in
progress. Then the caller could request fresh statistics and gather
them when they are done. Instead, we currently just return whatever
statistics are available, which may be up to 60 seconds old.
this partially addresses pr14385.
there is still a byte swapping bug retrieving the
stats structure. Since both the type and the length
are incorrectly swapped, we would timeout while reading
an extra long section (caused by the byteswapped length)
of garbage data.
turning on WI_DEBUG now just reports the unknown info type
which can be seen to be obviously incorrectly swapped
on macppc
mjacob