down, the dc driver and receiver can fall out of sync with one another,
resulting in a condition where the chip continues to receive packets
but the driver never notices. Normally, the receive handler checks each
descriptor starting from the current producer index to see if the chip
has relinquished ownership, indicating that a packet has been received.
The driver hands the packet off to ether_input() and then prepares the
descriptor to receive another frame before moving on to the next
descriptor in the ring. But sometimes, the chip appears to skip a
descriptor. This leaves the driver testing the status word in a descriptor
that never gets updated. The driver still gets "RX done" interrupts but
never advances further into the RX ring, until the ring fills up and the
chip interrupts again to signal an error condition. Sometimes, the
driver will remain in this desynchronized state, resulting in spotty
performance until the interface is reset.
Fortunately, it's fairly simple to detect this condition: if we call
the rxeof routine but the number of received packets doesn't increase,
we suspect that there could be a problem. In this case, we call a new
routine called dc_rx_resync(), which scans ahead in the RX ring to see
if there's a frame waiting for us somewhere beyond that the driver thinks
is the current producer index. If it finds one, it bumps up the index
and calls the rxeof handler again to snarf up the packet and bring the
driver back in sync with the chip. (It may actually do this several times
in the event that there's more than one "hole" in the ring.)
So far the only card supported by if_dc which has exhibited this problem
is a LinkSys LNE100TX v2.0 (82c115 PNIC II), and it only seems to happen
on one particular system, however the fix is general enough and has low
enough overhead that we may as well apply it for all supported chipsets.
I also implemented the same fix for the 3Com xl driver, which is apparently
vulnerable to the same problem.
Problem originally noted and patch tested by: Matt Dillon
now, but we're getting interrupts!
o Add pcic_suspend/pcic_resume so we can detach our children on suspention
and fix the state of the pcic on resume.
o Remove some unused parts of softc.
o Centralize resource activation/deactivation for pcic bridge chip in
the stylistic pcic_activate/pcic_deactivate.
o Add bus_print_child method so we can see the pccard attachment.
o Add pcic_identify in an attempt to make it possible to automatically id
the pcic devices. This works great, but we cannot divine the irq to use
from this method, nor the memory hole. For the moment, KLUDGE irq to be
10 and memory hold to be 0xd0000.
o Loose the pnp probe stuff. This may be a big mistake, but it is easy
enough to add back later. I did this so the identify routines can do their
thing unmolested by pnp information. The whole identify thing may be a bad
idea to be ripped out later.
o change return type of pcic_intr to void, make it static and ripple
this through the code.
o Add explicit call to bus_generic_attach at the end of pcic_attach to
get any children probed/attached.
o add some comments about future directions/questionable things being
done at different layers, etc.
then invoke the children. As the value of HISR can be read
only once, pass the HISR to the children via struct
csa_bridgeinfo, stored in the ivars of them.
- Clear the contents of serial FIFO upon stopping the DMA for
playing. This may eliminate buzz on playing. Experimental.
(pci dev_id 0x21).
- Start the SCRIPTS processor without resetting the SCSI BUS
at initialization.
- Remove the "Host adapter CCB chain" (got useless given the
new queuing scheme).
- Display correctly the state of SCSI signals, when SCSI BUS
looks bad.
- Cosmetic changes in messages printed out at initialization.
- Notifications and messages on RESET conditions slightly
reworked.
- TEKRAM 24C16 NVRAM support fixed (also reported ok).
is an application space macro and the applications are supposed to be free
to use it as they please (but cannot). This is consistant with the other
BSD's who made this change quite some time ago. More commits to come.
5in HD 2 heads, 77 cylinders, 8 sectors/track, 1024 bytes/sector
5/3.5in DD 2 heads, 80 cylinders, 8 sectors/track, 512 bytes/sector
Meanings of the rogrammer-readeble fd name were explained by Brian
Fundakowski Feldman and Peter Wemm in hackers list and NOKUBI
Hirotaka.
Reviewed by: nyan
USB ethernet chip. Adapters that use this chip include the LinkSys
USB100TX. There are a few others, but I'm not certain of their
availability in the U.S. I used an ADMtek eval board for development.
Note that while the ADMtek chip is a 100Mbps device, you can't really
get 100Mbps speeds over USB. Regardless, this driver uses miibus to
allow speed and duplex mode selection as well as autonegotiation.
Building and kldloading the driver as a module is also supported.
Note that in order to make this driver work, I had to make what some
may consider an ugly hack to sys/dev/usb/usbdi.c. The usbd_transfer()
function will use tsleep() for synchronous transfers that don't complete
right away. This is a problem since there are times when we need to
do sync transfers from an interrupt context (i.e. when reading registers
from the MAC via the control endpoint), where tsleep() us a no-no.
My hack allows the driver to have the code poll for transfer completion
subject to the xfer->timeout timeout rather that calling tsleep().
This hack is controlled by a quirk entry and is only enabled for the
ADMtek device.
Now, I'm sure there are a few of you out there ready to jump on me
and suggest some other approach that doesn't involve a busy wait. The
only solution that might work is to handle the interrupts in a kernel
thread, where you may have something resembling a process context that
makes it okay to tsleep(). This is lovely, except we don't have any
mechanism like that now, and I'm not about to implement such a thing
myself since it's beyond the scope of driver development. (Translation:
I'll be damned if I know how to do it.) If FreeBSD ever aquires such
a mechanism, I'll be glad to revisit the driver to take advantage of
it. In the meantime, I settled for what I perceived to be the solution
that involved the least amount of code changes. In general, the hit
is pretty light.
Also note that my only USB test box has a UHCI controller: I haven't
I don't have a machine with an OHCI controller available.
Highlights:
- Updated usb_quirks.* to add UQ_NO_TSLEEP quirk for ADMtek part.
- Updated usbdevs and regenerated generated files
- Updated HARDWARE.TXT and RELNOTES.TXT files
- Updated sysinstall/device.c and userconfig.c
- Updated kernel configs -- device aue0 is commented out by default
- Updated /sys/conf/files
- Added new kld module directory
devices.
o Return ENXIO from sn_isa_probe
o Fix SN_DEBUG printf
o Use IFQ_MAXLEN rather than 8
I'll fix the isa probe when I get access to a real isa attachment
device to test against here in a few days.
Overly agressive snagging behavior noticed by: phk
- In uhci_intr() check to see if sc->sc_bus.bdev is NULL, and if it is,
ack any pending interrupts and disable them, then return. It is possible
for interrupts to be delivered the moment a handler is set up at attach
time in uhci_pci.c, particularly when attempting to kldload the usb.ko
module after the system is already up. However the driver isn't ready
to field interrupts at that time and certain pointers in the softc
struct aren't initialized yet, and we invariably end up falling off
the end of one of them. The effect is that kldloading the usb module
will panic the system in uhci_intr(). This added sanity check stops
this from happening: I can now kldload the usb.ko module without any
problems and load/attach other USB drivers after it.
Of course the uhci driver has no detach method, but that's another
problem.
- In uhci_run(), set the UHCI_CMD_MAXP bit in the command register to
allow 64-byte packets to be used for full speed bandwidth reclamation.
Certain high speed devices (in this case the ADMtek USB ethernet
adapter) require this bit to be set, otherwise babble errors occur
at the end of large (between 1100 and 1500 byte) transfers. This
should not affect other devices, although supposedly it is less efficient
than the 32-byte setting. Unfortunately, this is a per-bus setting,
not a per-device setting, so we can't just enable it for certain
devices on the USB bus.
Add support, kinda, for megaheartz xjack nic cards. This support
works well for one machine per ethernet segment because it hard codes
the MAC address. The pccardd in -current doesn't have support to
parse the ethernet address from the CIS in the funky way that the
megaheartz card does things (it includes it in the info tuple, as
ascii, which is non-standard). I'd rather kludge this for the moment
and work to read the CIS from the kernel rather than mess with
pccardd.
The isa attachment is untested. The pccard attachment is known to
work since I'm committing over it.
Card Obtained from: Chris D. Faulhaber <jedger@fxp.org>
- Don't keep private copies of some of the data fields from the
ENQUIRY and ENQUIRY2 commands. Instead, standardise on the ENQUIRY2
command for initial adapter information, and keep a copy of the entire
structure. Refer to it where appropriate.
- Move all of the controller description functionality into a new
function. Print lots more controller data if bootverbose is set.
Add knowledge of the DAC960 PR, PT, PTL0 and PRL controllers, rename
the 960PTL -> PTL0 and 1100P -> 1100PVX.
- Correctly terminate an error message.
The controller interface procedures have been reviewed against the
Mylex-supplied documentation; no changes appear necessary at this
time.
on all combinations (I hope)...
Add DMA support for the AMD 756 chip (K7 chipset) this is actually the
same as the VIA 82C686 chip (the ATA part that is).
Treat the intel MX chipset PIIX as a PIIX4
Allow UDMA on all disks that say they can handle it.
Cleanup probe printf's a bit
Remove alot of the old #ifdef DEBUG crap.
going to a lot of trouble to identify it and set the tag and then not use
it. Convert the pnp id matching to the preferred table based system.
@@@0001 (CMI8330 ldn 0) is a mss, not a SB.
of these are bound to have a PNP05xx compatid, but there's no easy way to
tell. Since it's just an ID list and uses the pnp header's description
strings rather than encoding strings here, it doesn't seem to be too
expensive to err on the safe side.
misdetecting FIFO capabilities, at least on my girlfriend's Thinkpad 755,
the driver doesn't work using the FIFO.
While i was at it, i (partially) fixed option FCC_YE since it would no
longer have compiled at all under -current. I've also made an attempt
to document the device driver flags value (ab-)used internally by this
option.
RELENG_3 candidate, but with a slightly different patch there (will go
to jkh in email).
or less. This is a kludge because there is no support for delayed
creation of threads early in the boot process.
NEWCARD new boots and the thread process starts and is visible from
ps.
Now, on to dev/pccard.
the input fifo to be returned as successful and frozen. Most, if not
all, peripheral drivers do not check the qfrozen bit for successfully
completed commands, so the result would not only be lost commands, but
devices locked out from receiving commands. This was a bad bug that
crept in two or three months ago during some target mode work.
config_intrhook_establish doesn't work. Children aren't yet attached
properly, but that's ok because pccard would likely panic in its
current shape.
o Save dev of pcic early in attach process
o save dev in pcic_handle for use in pcic_create_event_thread.
o Remove direct attachments of pccard children for now
o move establishment of pcic kthread to config_intrhook.
