imp [Thu, 27 Jan 2005 04:51:44 +0000 (04:51 +0000)]
Write cs_detach() and use it. This resolves the twin problems of the
cs1 interface linger on card eject, as well as the warnings about the
card still using resources. Ooops.
imp [Wed, 26 Jan 2005 21:18:58 +0000 (21:18 +0000)]
The tinderbox seems to have a different set of compiler options
enabled than my box, so the agp module gets different warnings than I
got when I built. WERROR= to compensate.
harti [Wed, 26 Jan 2005 18:19:39 +0000 (18:19 +0000)]
Fix quoting of the MAKEFLAGS environment variable by only quoting spaces
and tabs. This is still not correct for command line variable values
ending in a backslash because this would require a larger effort.
Document this limitation in the BUGS section of the man page. The
quoting is mostly compatible with that of gmake and smake.
Tested by: Max Okumoto and Joerg Sonnenberger from DragonFly BSD
Reviewed by: ru (man page, partly)
delphij [Wed, 26 Jan 2005 15:17:25 +0000 (15:17 +0000)]
Correct some style nits that I have mistakenly submitted as
suggestions which result in the last revision[*]:
- style(9) and sysexits(3) suggests that we use EX_*
as exit values, instead of some other values like
those returned from a system call as errno.
Additionally, follow Ruslan's suggestion about style(9) and
other style improvements:
- Since open(2) says that it returns -1 on errors,
explicitly determine whether it is returning -1
rather than whether the return value is less than
zero.
- Only set rval when there is no previous error.
This distinguishes the first error that occours.
- Use exit() in favor of return in main(), this is
suggested in old style(9), while the evolve has
fade the suggestion.
- Add some NOTREACHED comments.
- Add blank line after first { because no local variables
in usage()
Thanks to Ruslan for his tireless explaining of the code standards
and knowledge of the history of style(9).
[*] Pointy hat to: me
Submitted by: ru (with some minor changes)
Discussed with: ru, ssouhlal
rwatson [Wed, 26 Jan 2005 14:09:32 +0000 (14:09 +0000)]
Remove unused static declaration of ed_pccard_dl100xx(), which probably
accidentally snuck into a prior commit to if_ed, and prevented building
with -Werror.
kan [Wed, 26 Jan 2005 13:56:27 +0000 (13:56 +0000)]
Of all templates only lists need a special implmementation file compiled
to support __GLIBCXX_DEBUG. Add debug_list.cc to the list of files to
compile.
yar [Wed, 26 Jan 2005 13:44:47 +0000 (13:44 +0000)]
Respect the current setting of IFCAP_VLAN_HWTAGGING on
the interface when going to toggle VLAN support for
internal reasons. If the IFCAP_VLAN_HWTAGGING bit is
cleared, we should rely on the (re)init routine to turn
VLAN support off and never touch the relevant hardware bits.
This applies to other capability bits, too. The user
obviously has a reason for clearing a capability bit,
e.g., if his particular NIC is buggy and hangs if a
certain hardware capability is turned on even for a
fraction of a second.
The flag adapter->em_insert_vlan_header still is set or
reset irrespective of the IFCAP_VLAN_HWTAGGING setting,
as before, in order to handle the case when a user sets
promiscuous mode on an interface first and later turns
its IFCAP_VLAN_HWTAGGING bit on.
This change might look orthogonal to rev#1.85, but in fact
it is not. It introduces bugfixes that hopefully will make
implementing the general scheme mentioned in the commit
message of rev#1.85 easier.
rwatson [Wed, 26 Jan 2005 11:40:58 +0000 (11:40 +0000)]
Disable use of hardware VLAN tagging and stripping in if_em in the default
configuration: it appears to work properly in the non-promiscuous case, but
we've not yet implemented a more general solution that maintains full
functionality with promiscuous mode enabled. While my hope is that we can
get one implemented soon, this will improve functionality substantially in
the mean time.
imp [Wed, 26 Jan 2005 05:21:12 +0000 (05:21 +0000)]
Rework how we deal with the DL10019 and DL10022 cards (as well as the
AX88190 ones, but that one only minorly):
o don't set flags in the match routine. They appear to be cleared
when probe/attach is called. Before this change, they were
always treated as a simple ne2000, which would fail to get the
right NIC address.
o Lookup device again in the probe routine and probe based on the
cards that you see.
o Detect and report the DL10022 seprately from the DL10019 cards.
While I'm here:
o remove a bad printf
o change another bad printf to device_printf.
o minor style(9) formatting tweaks.
# note: a lot of OEM entries are in the ed_pccard_products such that we can
# likely remove, or collapse, many of them.
This makes all of my DL100xx cards at least probe the ethernet address
correctly, which it wasn't doing before. I can't seem to locate my
AX88xxx based cards, so those haven't been tested, but they were
busted before the change so they can't be any worse now...
imp [Wed, 26 Jan 2005 05:12:16 +0000 (05:12 +0000)]
Repair probe messages a bit. Previously, we'd print the ethernet
address, and additional information. Then the printing of the
ethernet address was moved into ether_attach, and so we were printing
orphaned information about the card. Now the probe message is
prefixed by edX:. Prepare for it to move under bootverbose, but don't
move it there yet (the || 1 trick).
sobomax [Wed, 26 Jan 2005 00:46:36 +0000 (00:46 +0000)]
Split out kernel side of msgctl(2) into two parts: the first that pops data
from the userland and pushes results back and the second which does
actual processing. Use the latter to eliminate stackgap in the linux wrapper
of that syscall.
imp [Tue, 25 Jan 2005 23:15:20 +0000 (23:15 +0000)]
Turns out that Digital's DEPCM-BA and I-O Data PCLATE are both OEM
versions of the Racore PC Card Ethernet card. Rearrange to reflect
this reality. This ejects IODATA from 0x1bf, which belongs to Racore.
Thanks to Wilko for providing me with a dumpcis for the DEPCM card.
sobomax [Tue, 25 Jan 2005 21:28:28 +0000 (21:28 +0000)]
Split out kernel side of {get,set}itimer(2) into two parts: the first that
pops data from the userland and pushes results back and the second which does
actual processing. Use the latter to eliminate stackgap in the linux wrappers
of those syscalls.
imp [Tue, 25 Jan 2005 18:31:45 +0000 (18:31 +0000)]
Unbreak the Digital DEPCM-BA. Since it doesn't seem to have a proper
manufacturer ID, we need to use -1 rather than 0x100 when matching.
Do this by defining and using DIGITAL2 .
wpaul [Tue, 25 Jan 2005 17:00:54 +0000 (17:00 +0000)]
Apparently, the Intel icc compiler doesn't like it when you use
attributes in casts (i.e. foo = (__stdcall sometype)bar). This only
happens in two places where we need to set up function pointers, so
work around the problem with some void pointer magic.
jeff [Tue, 25 Jan 2005 16:06:05 +0000 (16:06 +0000)]
- Include LK_INTERLOCK in LK_EXTFLG_MASK so that it makes its way into
acquire.
- Correct the condition that causes us to skip apause() to only require
the presence of LK_INTERLOCK.
jeff [Tue, 25 Jan 2005 10:15:26 +0000 (10:15 +0000)]
- Make lf_print static and move its prototype into kern_lockf.c
- Protect all of the advlock code with Giant as some filesystems
may not be entering with Giant held now.
mr [Tue, 25 Jan 2005 08:59:06 +0000 (08:59 +0000)]
Add quirk for easyRAID ER5 alias areca ARC-6010.
It reports itself as SCSI-3 but doesnt like getting probed on high luns
because it hangs hard after finding itself again on lun 32...
delphij [Tue, 25 Jan 2005 08:14:00 +0000 (08:14 +0000)]
The kernel specified in main() of reboot(8) will be initialized
with -k option and never be used without kflag. This confuses
gcc because we set "kflag" at the same time with "kernel", but
the logic is not that apparant for gcc.
Since we can initialize "kernel" to NULL then know if "k" option
is set through determining whether it is still NULL, don't try
to have gcc to guess why we are connecting "kflag" with "kernel"
and use "kernel" directly in place of kflag.
imp [Tue, 25 Jan 2005 04:45:25 +0000 (04:45 +0000)]
Remove obsolete comment from head this time. The RELENG_5 commit was
accidental, but harmless and correct so I'll not be reverting it unless
there's some hidden damage I can't see right now...