cgd
e1078396e5
copy down from libkern:
...
>use p->hashfraction when doing non-time-critical calculations, rather than
>using HASHFRACTION directly. in time-critical calculations, if HASHFRACTION
>is a power of two, check that p->hashfraction == HASHFRACTION and if so do
>the calculation with the compiled-in value so that the compiler can optimize
>out (potentially) expensive divisions. if p->hashfraction != HASHFRACTION,
>actually do the division. This has the result that on machines with slow
>division, the division can be optimized out of the common case, but that
>if HASHFRACTION changes from the compiled-in value (for whatever reason),
>profiling will still work. Changes suggested by Chris Torek.
1996-06-12 04:16:53 +00:00
cgd
352d972c94
locc() is unused. Remove it from the machine-independent kernel interface.
1996-06-10 15:33:33 +00:00
pk
433594b99d
missing semicolon
1996-06-03 22:11:23 +00:00
ragge
28061842a2
Fix include file order to confirm to KNF in master tree also.
1996-06-02 13:28:21 +00:00
cgd
efd9d2dc7f
Copy from libkern:
...
Add some casts so that we're not bitten by type promotions on 64-bit
systems. Not a great concern since the only 64-bit system currently
supported by NetBSD (Alpha) uses assembly versions of these routines.
1996-05-29 23:39:30 +00:00
mrg
663808dfc5
use ' not " when passing variables to a sub-make. pr#1199/pr#1296
1996-05-20 17:23:25 +00:00
leo
c9a8578d2b
Fix gcc -Wall warnings about printf arguments.
1996-05-14 10:28:18 +00:00
mycroft
540f9550c0
Make `make depend' work for libraries.
1996-05-11 16:11:30 +00:00
mycroft
e901286aad
When sending an ARP request, use the interface address for the route, rather
...
than the first address assigned. This gives slightly different behaviour in
the presence of aliases. From Bill Fenner, via Pete Bentley.
1996-05-11 12:59:55 +00:00
scottr
ba554befe5
Make sure we clean up libkern.o and libkern.po, too. Closes PRs 1951
...
and 2387.
1996-05-10 21:27:32 +00:00
mark
d4fcd14ee3
No longer needed. The netbsd config for gcc/arm has been fixed to generate
...
correct profiling code.
1996-04-26 20:07:43 +00:00
mark
39b959c810
Added copyright message.
1996-04-26 20:05:28 +00:00
christos
0ca6a87952
Remove memset.c; it is not currently used, and .c files should belong in
...
the MI portion of libkern. When/if we move to use mem* routines in the kernel
we can add it to the MI portion/write a memset.s.
1996-04-23 23:05:22 +00:00
cgd
4a134a0780
don't include user-land headers in libkern files when compiling for
...
the kernel.
1996-04-18 02:19:59 +00:00
cgd
1813cc71d0
don't include user-land headers in libkern files when compiling for
...
the kernel. memset.c does not belong here. it either belongs in
the MI libkern directory, or does not belong in libkern at all.
1996-04-18 02:19:07 +00:00
cgd
55afd9912f
don't get rid of the -D_KERNEL in KERNCPPFLAGS
1996-04-18 02:14:48 +00:00
cgd
a5dc646be0
use assembly version of strcat (copied down from libc)
1996-04-18 01:53:04 +00:00
cgd
eccf43e127
files copied from libc
1996-04-17 22:46:34 +00:00
cgd
e70b1375c9
add strncasecmp.c to ports' libkern Makefile.inc's. When adding a new
...
source file to libkern, it should be added to _ALL_ ports makefiles if
it is machine-independent and used by machine-independent code (in
this case adosfs). (There are several arguments that say that
strncasecmp should not be in libkern at all, and i think they're
good ones, but as long as ADOSFS needs it to link, it has to be here
for all architectures.)
1996-04-13 01:17:27 +00:00
christos
fb901eb966
- Add missing prototypes
...
- Bring prototypes into scope
- Correct prototype for skpc
1996-03-14 18:52:03 +00:00
scottr
e494b8793f
Back out previous change.
1996-03-11 05:46:51 +00:00
scottr
c31eae0510
Back out previous change.
1996-03-11 05:31:59 +00:00
thorpej
e142a2195f
Provide a memset() for the i386 (assembly version), and move memset.c
...
out into the libkern proper in case other ports want to use it.
i386 kernels with COMPAT_LINUX should compile now.
1996-03-11 03:19:37 +00:00
gwr
056810ed99
Make this code independent of <sys/nfs> now that we use very little
...
from there anyway, so we can ignore changes in the NFS code...
1996-02-26 23:05:14 +00:00
scottr
6a0803ca9a
Make these build again in the wake of the v3 changes
1996-02-26 21:50:07 +00:00
scottr
c24557134f
Needed for NFS booting; we aren't likely to use v3 here any time soon.
1996-02-26 21:49:34 +00:00
christos
1cb0b13391
Add declaration of strncasecmp.
1996-02-13 23:48:26 +00:00
christos
14d9cd33af
netinet prototypes
1996-02-13 23:40:59 +00:00
gwr
b3ba75d64e
use vprintf where appropriate
1996-02-09 16:10:59 +00:00
gwr
7ce27395a6
Need vprintf() for things like panic().
1996-02-08 20:19:36 +00:00
christos
b07937f401
Make it really work like strncasecmp, instead of 0/1 return codes.
1996-02-06 21:35:31 +00:00
christos
70552d3392
Add strncasecmp.c; extracted from ADOSFS adlookup.c.
1996-02-06 21:22:00 +00:00
mycroft
f96fc7f12e
Simplify, and add support for passing CPPFLAGS.
1996-02-02 20:34:38 +00:00
mark
f56b429c95
Initial commit of the NetBSD/arm32 port.
1996-01-31 23:44:43 +00:00
thorpej
43349cd7ef
Add a memcpy() function, just a cheezy wrapper around bcopy(). Yes,
...
it's really true...we didn't have one before, even though the Makefile
thought we did. (Yes, I checked the Attic...)
1996-01-30 19:45:34 +00:00
gwr
38c7f668a2
Clean-up open/close logic of "net" device. Fixes PR 1726
1996-01-29 23:54:14 +00:00
leo
b7b97847ce
Some files containing more than one function are split. This reduces the
...
size of the resulting loader. The read_inode() function now returns the
result of the strategy function instead of success in all cases (ufs.c).
Fixes pr#1817
1996-01-13 22:25:32 +00:00
pk
284c0cd0b1
Make `time difference' a signed quantity explicitly.
1995-12-13 23:38:10 +00:00
gwr
63a836a44a
Check return value from sendrecv so that ARP timeout does not cause
...
garbage to be entered into the ARP table. Also fix ARP_DEBUG code.
1995-12-08 04:12:53 +00:00
gwr
63e145ebd7
Correct error which caused truncation of port number on i386 (PR#1823)
1995-12-08 04:10:28 +00:00
jtc
0cd793449e
merge in changes from 1.1 release branch
1995-11-30 00:56:23 +00:00
thorpej
87093145ce
Use bzero() instead of memset(..., 0, ...).
1995-11-29 06:12:08 +00:00
cgd
dfad729a16
make netinet work on systems where pointers and longs are 64 bits
...
(like the alpha). Biggest problem: IP headers were overlayed with
structure which included pointers, and which therefore didn't overlay
properly on 64-bit machines. Solution: instead of threading pointers
through IP header overlays, add a "queue element" structure to do
the threading, and point it at the ip headers.
1995-11-21 01:07:34 +00:00
cgd
40a4907f5e
oops; i forgot these u_int -> size_t conversions.
1995-10-20 01:35:25 +00:00
cgd
6b49a2ceed
copy from libkern
1995-10-20 01:16:02 +00:00
cgd
35742ffe0a
use NetBSD RCS id format. use bzero.S, not bzerro.c
1995-10-20 01:15:37 +00:00
cgd
47667c7df1
set, don't add to, CFLAGS from KERNCFLAGS. Do assignment of CC and
...
CFLAGS earlier, so that the arch-specific Makefile can add to them.
1995-10-20 01:10:06 +00:00
cgd
51f39e4840
fix prototypes here, to match u_int->size_t changes in ufs.c
1995-10-20 00:54:17 +00:00
cgd
d7a207fd49
update from libc
1995-10-20 00:53:28 +00:00
cgd
6c46625cf7
cast pointer to long, not int
1995-10-20 00:47:47 +00:00