ansi

If you use sys/param.h, you don't need sys/types.h

add a size argument to free. will be used soon, but for now default to 0.
after discussions with beck deraadt kettenis.

Remove unsigned comparison < 0.

Pointed out by LLVM.

ok fgsch@

Off-by-two in the length check for CISTPL_CONFIG tuples.

Use pcmcia_cis_read_1() to read cis data to be checksummed, instead of
trying to roll our inline version of it and fail. Fixes `checksum error'
warnings on all the cards with checksum tuples I could

Use pcmcia_cis_read_1() to read cis data to be checksummed, instead of
trying to roll our inline version of it and fail. Fixes `checksum error'
warnings on all the cards with checksum tuples I could find (admittedly not
many), and probably fixes all of them since this code has never been correct
in the first place.

show more ...

Add support for indirect CIS data (CISTPL_INDIRECT tuple).

KNF

prompted and "much better" by marco@, ok pyr@

sprinkle some M_ZERO love.

CISTPL_CONFIG min length should be 3, not 5. audiovox GPRS works now.
deraadt@ ok.

some additional CIS sanity checks.

From NetBSD

typos from Tom Cosgrove;

range check for the CIS table scan
normally, the CIS table is scanned until we hit the end tuple (CISTPL_END). I
have at least one card with an invalid CIS table without end tuple - while
this card i

range check for the CIS table scan
normally, the CIS table is scanned until we hit the end tuple (CISTPL_END). I
have at least one card with an invalid CIS table without end tuple - while
this card is in error, this should not bring us to ddb. thus stop scanning
when we reach PCMCIA_CIS_SIZE

ok millert@

show more ...

Use queue.h macros instead of using the structure names directly.

move initialization a bit later.

First round of __P removal in sys

Check result from malloc(9) when using M_NOWAIT. fgsch@ ok

* correct some length checkings based on the specs; still a few to fix.
* add 1 to number of memspaces since MS_COUNT it's minus 1.
* print the correct function type in case of IOBUS.
* clean cfe->fl

* correct some length checkings based on the specs; still a few to fix.
* add 1 to number of memspaces since MS_COUNT it's minus 1.
* print the correct function type in case of IOBUS.
* clean cfe->flags if intface is present

show more ...

OR the values in misc within cftable entry; in iospace, irq and misc
clean the flags before checking for particular capabilities. this is
needed because we use a template that might end messing the f

OR the values in misc within cftable entry; in iospace, irq and misc
clean the flags before checking for particular capabilities. this is
needed because we use a template that might end messing the flags.

show more ...

Treat CIS functions a bit different as Xircom devices order the tuples a
bit backwards

Add support for:
- 3COM 3CXEM556 (Ethernet/Modem)
- SVEC Combo and Lancard
- Corega PCC-T
Add pcmcia_check_cis_quirks to fix some cards whose CIS flat-out lies.
Add more products to pcmciadevs.
Co

Add support for:
- 3COM 3CXEM556 (Ethernet/Modem)
- SVEC Combo and Lancard
- Corega PCC-T
Add pcmcia_check_cis_quirks to fix some cards whose CIS flat-out lies.
Add more products to pcmciadevs.
Correctly read 1 controller if CL-PD6729.
Remove some unused vars.

show more ...

PCMCIA code ported from NetBSD.
Support for aic, ep, pccom and sm.