Commit Graph

58113 Commits

Author SHA1 Message Date
kato
32d33222f8 Merged from sys/i386/conf/GENERIC revision 1.305 (moved the fxp driver
into the miibus section.)
2001-03-17 14:04:24 +00:00
peter
b2463eb0ef s/over 3500 ports/over 4500 ports/
MFC candidate. :-)  The RELENG_4 version says 3000 still.
2001-03-17 12:37:28 +00:00
phk
f029c2324e Unrevert something which shouldn't have been reverted in the previous revert.
(Yes, I'm going to bed now to get some much needed sleep...)
2001-03-17 11:43:36 +00:00
brian
4b8b237f29 Fix some further style nits
Pointed out by: bde
2001-03-17 10:09:52 +00:00
phk
f3af16b3ba Add a new entrypoint to the hashes in libmd:
char *
  FooFileChunk(const char *filename, char *buf, off_t offset, off_t length)
Which only hashes part of a file.
Implement FooFile() in terms of this function.

Submitted by:	roam
2001-03-17 10:00:50 +00:00
peter
156ac8e9c7 Use a generic implementation of the Fowler/Noll/Vo hash (FNV hash).
Make the name cache hash as well as the nfsnode hash use it.

As a special tweak, create an unsigned version of register_t.  This allows
us to use a special tweak for the 64 bit versions that significantly
speeds up the i386 version (ie: int64 XOR int64 is slower than int64
XOR int32).

The code layout is a little strange for the string function, but I was
able to get between 5 to 10% improvement over the original version I
started with. The layout affects gcc code generation choices and this way
was fastest on x86 and alpha.

Note that 'CPUTYPE=p3' etc makes a fair difference to this.  It is
around 45% faster with -march=pentiumpro on a p6 cpu.
2001-03-17 09:31:06 +00:00
phk
43893d1859 Backout my previous commit, it appearantly doesn't work for asynchronous
messages.
2001-03-17 08:59:31 +00:00
markm
02f3701a3d Unbreak "make installworld" 2001-03-17 07:38:57 +00:00
peter
3a6e7a5b69 Dramatically improve the **lame** nfs_hash(). This is based on the
Fowler / Noll / Vo Hash (http://www.isthe.com/chongo/tech/comp/fnv/).

This improves hash coverage a *massive* amount.  We were seeing one
set of machines that were using 0.84% of their 131072 entry nfsnode
hash buckets with maximum chain lengths of up to ~500 entries.  The
machine was spending nearly 100% of its time in 'system'.
A test with this has pushed the coverage from a few perCent up to 91%
utilization with a max chain length of 11.

Submitted by:  David Filo
2001-03-17 05:43:01 +00:00
scottl
037a4bfc84 Fix the building of the aac driver as a module. Add a comment about enabling
debug information for the driver.
2001-03-17 04:42:33 +00:00
nyan
b7a7320ec6 Always call resource_int_value function for getting portsize and msize.
It was not set resource size (portsize/msize) if resource address was set.

This is MFC candidate.
2001-03-17 04:23:12 +00:00
jlemon
3b82041343 Really fix NWAY negotiation for the 82553 PHY. Locking down the
media interface selection should not imply disabling NWAY negotiaton
as well.

Problem pointed out by: peter
2001-03-17 02:50:20 +00:00
peter
55c62eab20 Stop cron from shouting in the syslog. Use setproctitle() instead, since
smashing argv[0] doesn't have the intended effect of changing the ps(1)
output these days.

PR:		25850
2001-03-17 00:21:54 +00:00
gibbs
e5522118c5 This is an MFC candidate.
Add the AAC_DEBUG option to enable debugging in the aac driver.

Correct a race condition in the interrupt handler where the
controller may queue a fib to a response queue after the driver
has serviced the queue but before the interrupt is cleared.
This could leave a completed fib stranded in the response queue
unless another I/O completed and generated another interrupt.

Reviewed by:	msmith
2001-03-17 00:12:15 +00:00
gibbs
c98f5f11b3 Add the AAC_DEBUG option to enable debugging in the aac driver.
Reviewed by:	msmith
2001-03-17 00:09:47 +00:00
murray
e9f9238b26 Clean up the grammar a bit from the last commit. 2001-03-16 22:55:55 +00:00
jlemon
194c9e1235 When doing a recv(.. MSG_WAITALL) for a message which is larger than
the socket buffer size, the receive is done in sections.  After completing
a read, call pru_rcvd on the underlying protocol before blocking again.

This allows the the protocol to take appropriate action, such as
sending a TCP window update to the peer, if the window happened to
close because the socket buffer was filled.  If the protocol is not
notified, a TCP transfer may stall until the remote end sends a window
probe.
2001-03-16 22:37:06 +00:00
ken
2b4d5a50a9 Fix a few things in the aic(4) driver:
- enable 10MHz (fast SCSI) operation on boards that support it.  (only
   aic6360 boards with fast SCSI enabled can do it)

 - bounds check sync periods and offsets passed in from the transport layer

 - tell the user which resource allocation failed (for the ISA probe) if we
   weren't able to allocate an IRQ, DRQ or I/O port.
2001-03-16 22:20:19 +00:00
dillon
da83455a02 Fix type-o
Submitted by: okazaki
2001-03-16 22:18:26 +00:00
ken
7b3c474f9f Fix inquiry length detection for the ses(4) driver. It was using the
inq_len member of the ccb_getdev structure, but we've never filled that
value in..

So we now get the length from the inquiry data returned by the drive.
(Since we will fetch as much inquiry data as the drive claims to support.)

Reviewed by:	mjacob
Reported by:	Andrzej Tobola <san@iem.pw.edu.pl>
2001-03-16 22:16:54 +00:00
gad
3fecedd0b7 Fix a number of warnings that come up when compiling with BDECFLAGS. 2001-03-16 21:54:12 +00:00
gallatin
aba1a5cf5f prevent gdb from panic'ing an alpha with "mutex Giant not owned"
reviewed by: jhb
2001-03-16 21:44:21 +00:00
msmith
16abde54ad Fix a typo which would cause containers between 1GB and 2GB to have the wrong
geometry reported.

Submitted by:	mark tinguely <tinguely@web.cs.ndsu.NoDak.edu>
2001-03-16 21:43:32 +00:00
cg
80a5db8b67 fix a typo preventing the second dma channel being released
use isa_dma_release when releasing dma channels
2001-03-16 21:04:14 +00:00
cg
87ce9dff8c don't leak memory allocated for feeders at module unload
kill the fake channel when unregistering
2001-03-16 20:58:44 +00:00
phk
7254466c42 <sys/queue.h> makeover. 2001-03-16 20:00:53 +00:00
phk
4225b0140e Fix a style(9) nit. 2001-03-16 19:36:23 +00:00
phk
0f1180106c Add -ledit to ngctl.
Somebody submitted this long time ago, and it has been sitting in my
tree for months because I thought archie would pick it up.

Submitted by:	(sorry, lost track)
2001-03-16 19:35:07 +00:00
jlemon
0e5439e79f Bump MAX_GLOBENTRIES up to 16384, so it is a power of two. Add
some comments explaining that this is an arbitrary limit.

Requested by:  jkh
2001-03-16 19:05:20 +00:00
ume
cdc44b350e nuke IPSEC_SRCSEL which does not do the right thing.
adjust state->ro if the tunnel endpoint is offlink.
KAME PR 233.

PR:		kern/21079
2001-03-16 17:52:48 +00:00
rwatson
ce67b919f8 o The revenge of the mdoc(7) police:
- These pages abused Ar macro (they should have used Fa).
   - NULL and other numeric constants should be marked with Dv.
   - VOP_* in the ERRORS section for the EOPNOTSUPP entry should be marked
     with Fn.

Submitted by:	ru
2001-03-16 17:42:38 +00:00
rwatson
b796adf323 o The mdoc(7) police strike with avengance:
- These pages abused Ar macro (they should have used Fa).
   - NULL constant should be marked with Dv.
   - VOP_* in the ERRORS section for the EOPNOTSUPP entry should be marked
     with Fn.

Submitted by:	ru
2001-03-16 15:02:30 +00:00
jlemon
108dc7ffa2 Enable some undocumented bits in the DP83840 PHY, which is needed
when using it with the Intel fxp driver.

Tested by: peter, Ollivier Robert <roberto@eurocontrol.fr>,
	   Peter Schultz <pete@jocose.org>
2001-03-16 14:17:02 +00:00
ru
20a25d49d2 mdoc(7) police: simplify bogus ``.Ql Fl'' constructs. 2001-03-16 14:11:41 +00:00
ru
b96f425cc7 mdoc(7) police: -mdoc still has a nine-arguments limitation. 2001-03-16 13:58:34 +00:00
ru
4e2b569f18 mdoc(7) police: empty lines outside displays cause warnings in -mdocNG. 2001-03-16 13:50:50 +00:00
ru
f2bd8ad705 mdoc(7) police: apply formatting fixes to rev 1.14. 2001-03-16 13:38:49 +00:00
alfred
456151f2ac s/0x7f000001/INADDR_LOOPBACK 2001-03-16 13:10:26 +00:00
sos
9714e01a36 Sigh, try to get this in sync with an offical src tree...
I hate it when this happens...
2001-03-16 13:08:19 +00:00
ru
7be937b474 It is the ``nostrictjoliet'' option equivalent to -b.
Reviewed by:	bp
2001-03-16 12:55:54 +00:00
ache
c6d439193e Add et_EE 2001-03-16 12:33:50 +00:00
ache
50fbc40590 Make empty unused am/pm 2001-03-16 12:29:09 +00:00
ache
e50633d5e3 Add et_EE
PR:		25495
Submitted by:	Vallo Kallaste <vallo@matti.ee>
2001-03-16 12:27:20 +00:00
ache
4487169d3f Add et_EE
PR:		25495
Submitted by:	Vallo Kallaste <vallo@matti.ee>
2001-03-16 12:21:11 +00:00
ache
1a947a46aa Add et_EE
PR:		25495
Submitted by:	Vallo Kallaste <vallo@matti.ee>
2001-03-16 12:11:30 +00:00
ru
df6d18791c mdoc(7) police: ``It'' macro does not take argument in -enum lists.
(In -mdocNG, this only causes warning.  In current implementation,
it is fatal.)

Pointy hat to:	markm (for not checking stderr)
2001-03-16 12:06:26 +00:00
ache
d063440cda Add et_EE.ISO_8859-15 2001-03-16 12:02:36 +00:00
sos
3cfbe461cf Remove the now defunct ATA_ENABLE* options
Spotted by: phk
2001-03-16 11:52:25 +00:00
kuriyama
1c016fe073 Merge from kame (1.175 -> 1.176):
cope with freebsd4 bridge code.
2001-03-16 10:58:16 +00:00
ru
df326be943 Apply the README's requirement a) to the previous commit. 2001-03-16 08:43:07 +00:00