Commit Graph

41055 Commits

Author SHA1 Message Date
Matthew Dillon
67ddfcaf69 More removals of vnode->v_lastr, replaced by preexisting seqcount
heuristic to detect sequential operation.

    VM-related forced clustering code removed from ufs in preparation for a
    commit to vm/vm_fault.c that does it more generally.

Reviewed by: David Greenman <dg@root.com>, Alan Cox <alc@cs.rice.edu>
1999-09-20 23:27:58 +00:00
Matthew Dillon
8411bf657d Fix handling of a device EOF that occurs in the middle of a block. The
transfer size calculation was incorrect resulting in the last read being
    potentially larger then the actual extent of the device.

    EOF and write handling has not yet been fixed.

Reviewed by:	Tor.Egge@fast.no
1999-09-20 23:17:47 +00:00
Marcel Moolenaar
85fce0e478 When bcopying the program name into the ktrace header, make sure we include
the terminating zero by copying MAXCOMLEN + 1 bytes. This fixes the garbage
that occasionally appeared behind the programname when it is at least MAXCOMLEN
bytes long (such as communicator-4.61-bin).
1999-09-20 21:53:17 +00:00
Poul-Henning Kamp
faad302913 Fix a harmless bug I introduced, simplify a bit more while here. 1999-09-20 21:14:43 +00:00
Mark Murray
1f687ff1e7 Minor (but important) comment change. 1999-09-20 20:27:09 +00:00
Bill Paul
3d927b9b51 Close PR . I managed to figure out the problem, no thanks to the
submitter, who *still* hasn't bothered to answer me back.

The thing which the submitter completely failed to mention is that
his 3c900B-TPO card has the transceiver selection in the EEPROM set
to "auto." You can tweak the setting using the 3C90XCFG.EXE utility
that 3Com provides with the card. I'm not sure if it's supposed to
default to auto or if the user fiddled with it. Currently, the xl
driver only does autoselection for 10/100 NICs (i.e. those with NWAY
autonegotiation capabilities). For the 10baseT, 10base5, 10base2,
10baseFL and 100baseFX cards, the driver sets the default media to
whatever the EEPROM transceiver selector says. The problem is that
the "auto" selection is mistakenly identified as "10/100 NWAY
autoselection mode" and this is not handled correctly: the default
media ends up being chosen as 100baseTX, which doesn't work because
we've only added 10baseT media types to the ifmedia word. This leads
to a panic in ifmedia_set() (something else which the submitter never
bothered to mention).

A workaround for this is to re-run the 3C90XCFG.EXE utility and change
the transceiver selection to something besides "auto." I have also
patched the driver to watch for the "auto" setting in the non-miibus
case and select a reasonable default based on the card type instead of
falling through to 100baseTX and exploding.

PR:		misc/13665
1999-09-20 20:26:14 +00:00
Poul-Henning Kamp
2ab8cd46e9 Set si_iosize_max rather than d_maxio.
Register devsw in *attach instead of a SYSINIT.
1999-09-20 19:58:38 +00:00
Poul-Henning Kamp
fae03f66d1 Step one of replacing devsw->d_maxio with si_bsize_max.
Rename dev->si_bsize_max to si_iosize_max and set it in spec_open
if the device didn't.

Set vp->v_maxio from dev->si_bsize_max in spec_open rather than
in ufs_bmap.c
1999-09-20 19:57:28 +00:00
Poul-Henning Kamp
552f337f1f Initialize vp->v_maxio to its default in getnetvnode() rather than
four different places in vfs_cluster.c
1999-09-20 19:53:23 +00:00
Matthew Dillon
8fdd2461b3 Add comment to clarify a commit rpc optimization already being performed. 1999-09-20 19:10:28 +00:00
Matthew Dillon
4cc712004c Fix bug in pipe code relating to writes of mmap'd but illegal address
spaces which cross a segment boundry in the page table.  pmap_kextract()
    is not designed for access to the user space portion of the page
    table and cannot handle the null-page-directory-entry case.

    The fix is to have vm_fault_quick() return a success or failure which
    is then used to avoid calling pmap_kextract().
1999-09-20 19:08:48 +00:00
Bill Paul
0355003f26 Un-do the changes to the DRIVER_MODULE() declarations in these drivers.
This whole idea isn't going to work until somebody makes the bus/kld
code smarter. The idea here is to change the module's internal name
from "foo" to "if_foo" so that ifconfig can tell a network driver from
a non-network one. However doing this doesn't work correctly no matter
how you slice it. For everything to work, you have to change the name
in both the driver_t struct and the DRIVER_MODULE() declaration. The
problems are:

- If you change the name in both places, then the kernel thinks that
  the device's name is now "if_foo", so you get things like:

if_foo0: <FOO ethernet> irq foo at device foo on pcifoo
if_foo0: Ethernet address: foo:foo:foo:foo:foo:foo

  This is bogus. Now the device name doesn't agree with the logical
  interface name. There's no reason for this, and it violates the
  principle of least astonishment.

- If you leave the name in the driver_t struct as "foo" and only
  change the names in the DRIVER_MODULE() declaration to "if_foo" then
  attaching drivers to child devices doesn't work because the names don't
  agree. This breaks miibus: drivers that need to have miibuses and PHY
  drivers attached never get them.

In other words: damned if you do, damned if you don't.

This needs to be thought through some more. Since the drivers that
use miibus are broken, I have to change these all back in order to
make them work again. Yes this will stop ifconfig from being able
to demand load driver modules. On the whole, I'd rather have that
than having the drivers not work at all.
1999-09-20 19:06:45 +00:00
Justin T. Gibbs
d8a4660643 Clear the SELINGO bit after a selection timeout occurs. SELINGO is
usually cleared by a successful selection, but there is no guarantee
that a future successful selection will ever occur (e.g. empty bus).
The driver never looks at SELINGO, but the busy LED does, so this
change has the cosmetic effect of fixing the rare instance where the
busy LED was left on, confusing the user.
1999-09-20 19:04:22 +00:00
Justin T. Gibbs
55bad6b5c3 Work around a defect in the FIFOEMP status bit of Ultra2 class
aic7xxx parts.  This problem could result in data corruption
during periods of my PCI bus load by busmasters other than the
aic7xxx.

Many thanks to Andrew Gallatin <gallatin@cs.duke.edu> for characterizing
the symptoms of this problem and testing this fix.
1999-09-20 18:57:04 +00:00
Wes Peters
236cb8163d Fixed the description of when and why aio_suspend returns.
Also spelled out the return values and conditions a little
better.

Noticed by:	John Polstra <jdp@polstra.com>
1999-09-20 18:30:55 +00:00
Matthew Dillon
5b4cc8ee79 Fix the situation where an I/O crosses the EOF point. We need to properly
adjust the bp->b_bcount the same way for unlabeled VN access as dscheck
    does for labeled access.

    We also increase the block size to at least 8K for efficiency.

Reviewed by:	 Tor.Egge@fast.no
1999-09-20 18:06:17 +00:00
Poul-Henning Kamp
f7d9e545c0 Trying to set BAD144 handling on a CD is sufficiently "Dont Do That
then" that we don't need to give the ioctl special treatment.
1999-09-20 17:06:29 +00:00
Poul-Henning Kamp
81f7c76af5 Add a DSO_BAD144 flag which indicates that the driver actually understand
BAD144 handling.

Reject DIOCSBAD and labels with BAD144 tables if the driver cannot grok it.

Reviewed by:    bde
1999-09-20 17:04:39 +00:00
Poul-Henning Kamp
624aff9227 if_edreg.h moved... 1999-09-20 16:59:55 +00:00
Matthew Dillon
d909b5635d Fix bug in brelse() regarding redirtying buffers on B_ERROR. brelse()
improperly ignored the B_INVAL flag when acting on the B_ERROR.
    If both B_INVAL and B_ERROR are set the buffer is typically out of the
    underlying device's block range and must be destroyed.  If only B_ERROR
    is set (for a write), a write error occured and operation remains as it
    was before:  the buffer must be redirtied to avoid corrupting the
    filesystem state.

Reviewed by:	David Greenman <dg@root.com>
Submitted by:	Tor.Egge@fast.no
1999-09-20 16:19:24 +00:00
Mitsuru IWASAKI
43834dfd27 Return immediately from apm_suspend() when APM BIOS wasn't initialized.
Pressing Alt-Pause key will cause machine to reboot with apm disabled.

PR:		i386/13817
Submitted by:	yokota
1999-09-20 15:29:23 +00:00
Mark Murray
e9a56ad5ca Big code cleanup. (Inspired by Brandon Gillespie). Also move as
much as possible away from secure/ to make extending easier.
1999-09-20 12:45:49 +00:00
Mark Murray
e1e54354b5 Make this completely dependant on the exportable libcrypt, to avoid
duplication of effort. Also a large cleanup of the code, inspired
by Brandon Gillespie.
1999-09-20 12:40:06 +00:00
Doug Rabson
5ed9724025 This file was never used and appears to be a leftover from the NetBSD
bootstrap days.
1999-09-20 09:46:19 +00:00
Alexey Zelkin
c6d6e7726f Correct spelling : ascii -> ASCII
PR:		docs/13702
Submitted by:	Stephen J. Roznowski <sjr@home.com>
Reviewed by:	mpp
1999-09-20 09:15:23 +00:00
Ruslan Ermilov
08e13cdfa7 Yet another align.
It will be required if sizeof(struct sockaddr_in) changes in the future.

Not objected to by:	wollman
1999-09-20 09:10:46 +00:00
Bill Paul
fac1f39b19 Grrr. Okay, changing the devnames was a bad idea. Put them back the way
they were.
1999-09-20 08:47:11 +00:00
Sheldon Hearn
2e9d38764f Remove the wait.1 manpage. There is no userland wait(1) utility, and the
manpage does not describe the builtin wait command provided by either
of the standard shells. There is already an entry for wait.1 in the
builtins.1 MLINKS list.
1999-09-20 08:45:43 +00:00
Bill Paul
b95a9362a0 Fix the strings in the driver_t structs so that they match the new names
in the DRIVER_MODULES() declarations. *sigh*
1999-09-20 08:14:39 +00:00
Mike Smith
f3722ef609 If we don't appear to have a module loaded supporting the interface
we're about to operate on, try to load one.  Don't complain if the
load fails, and always press on regardless (there may not be a module
suitable or required).

With the renaming of the PCI ethernet driver modules and the addition
of appropriate miibus dependancies on those modules that need it, it is
now no longer necessary to compile many ethernet drivers into the kernel;
they will be loaded on demand the first time they are ifconfig'ed.

Inspiration from: mount
Reviewed by:	obrien
1999-09-20 07:58:08 +00:00
Poul-Henning Kamp
24514292fa On PIIX4 based SMP systems use the PMTMR register for timecounting.
It is about 2.5 microseconds or roughly 3 times faster to use this
"PIIX" timecounter than the "i8254" timecounter.  Resolution is
also 3 times better.

The code cheats and don't register the PCI device, because other pieces
of code want to use it too.

Originally spotted by:  msmith
1999-09-20 07:52:39 +00:00
David E. O'Brien
abad681b03 Goofed and didn't change the second DRIVER_MODULE() linking these with
the miibus.

Noticed by:	wpaul
1999-09-20 07:50:10 +00:00
Brian Somers
7e795ebe38 NetBSD has moved ``extern int errno;'' to signal.h :-/
Submitted by:	Kazuyoshi Kato <kazk@yyy.or.jp>
1999-09-20 07:36:46 +00:00
Brian Somers
e83358b444 NetBSD has changed the el_init() prototype
Submitted by:	Kazuyoshi Kato <kazk@yyy.or.jp>
1999-09-20 07:22:22 +00:00
Brian Somers
a77cab77c5 NetBSD requires net/if.h for the IFF_* defines (as well as OpenBSD) ! 1999-09-20 07:18:50 +00:00
David E. O'Brien
bd8a15ce8a Change the name we register with DRIVER_MODULE() to include the leading
"if_".

Reviewed by:	msmith, wpaul
1999-09-20 06:50:52 +00:00
Mark Murray
33f891d293 Common Error libraries are needed here. 1999-09-20 06:23:16 +00:00
Mark Murray
1099209e99 Prepare for K5. 1999-09-20 06:21:51 +00:00
David E. O'Brien
40ec3caac8 Change the name of the installed KLM to contain a leading "if_".
Except for miibus which is now installed as miibus.ko.

Discussed by:	msmith, peter, and wpaul
1999-09-20 06:11:48 +00:00
Peter Wemm
814e16091b Make if_ed work again on pci, isa, isapnp. The hack to make it work on
PCCARD is pretty revolting but should buy us time while the pccard driver
angle is sorted out.  A commit for the MCA ed attachment will follow
shortly.
1999-09-20 05:48:16 +00:00
Bill Paul
1b4226d3e4 Add an alternate transmit strategy for 3c90xB adapters based on the transmit
strategy used in the 3Com Linux driver. The new strategy is to use transmit
descriptor polling -- that is, the NIC polls the descriptors to see when
new packets are available for transmission. The advantage to the new scheme
is that no register accesses are needed in the transmit routine. The old
scheme requires several register accesses to stall the TX engine, update the
TX DMA list pointer register, then unstall the TX engine. Hopefully the new
scheme will provide improved transmit performance with less CPU overhead.

This only affects the 3c90xB or 3c90xC cards, not the 3c90x cards. This
means the original 3c900 and 3c905 cards are unaffected. Newer cards include
the 3c900B series, the 3c905B, 3c980, 3c980B, 3c905C and 3c905C, and the
3cSOHO100-TX OfficeConnect.
1999-09-20 00:24:11 +00:00
Bill Paul
977f4ed148 RealTek driver module wasn't being built; add it to the Makefile. 1999-09-19 23:19:59 +00:00
Bill Paul
ca2190f252 Make some small performance tweaks to the tl driver. This should hopefully
close PR , however I'm waiting on user feedback before declaring the
PR officially closed. Among other things, this improves UDP transmit
performance, and tx underruns are now detected and the TX start threshold
adjusted accordingly.
1999-09-19 22:39:24 +00:00
Mark Murray
38efe8d18c Change distribution to krb4 in preparation for K5. 1999-09-19 22:32:09 +00:00
Mark Murray
df3819c1b4 Prepare for K5. 1999-09-19 22:30:39 +00:00
Mark Murray
f7f09ffb6c Change edistribution to krb4 in preaparation for K5 1999-09-19 22:26:02 +00:00
Mark Murray
105b74f110 Fix for new KerberosFix for new Kerberos44 1999-09-19 22:14:33 +00:00
Mark Murray
69746547bd Fix for new Kerberos4. 1999-09-19 22:11:15 +00:00
Mark Murray
6c9134c067 Fix for new Kerberos4. Make a fist cut at PAM-ising while I'm here. 1999-09-19 22:05:32 +00:00
Bill Paul
ae3b8c19b1 Convert the VIA Rhine driver to miibus. 1999-09-19 22:03:31 +00:00