| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
| |
It should only be declared in <signal.h> if at all. Modload used
to depend on getting it from <unistd.h> because it only included
<sys/signal.h>.
|
|
|
|
| |
posix standard on the topic.
|
| |
|
| |
|
|
|
|
|
|
|
|
| |
This will make a number of things easier in the future, as well as (finally!)
avoiding the Id-smashing problem which has plagued developers for so long.
Boy, I'm glad we're not using sup anymore. This update would have been
insane otherwise.
|
| |
|
|
|
|
| |
Submitted by: Sandro Sigala <sandro@cat.locao.net> as part of PR# 2134
|
|
|
|
|
|
| |
Abort if a file is still present, and make output file mode 0600.
Reviewed by: bde
|
| |
|
|
|
|
| |
can be loaded (e.g. ./mymod.o)
|
|
|
|
|
|
|
|
|
|
|
|
| |
1: generate the outfile in /tmp if it's not specified explicitly.
2: if the outfile was implicitly placed in /tmp, automatically remove it.
This means that you can type: modload /lkm/ipfw_mod.o and it'll work, it
wont try and write to /lkm, and it wont leave the (normally) useless
symbol file.
This should not interfere with things like ibcs2 and atapi, which
export some symbols from one LKM to the other by leaving the symbol file.
|
|
|
|
|
|
|
|
|
| |
of the recent changes to bsd.kmod.mk, which enforces that the entry
point be <modname>_mod, unless overridden.
This means that when modloading "/lkm/foo_mod.o", it'll use "foo_mod"
as the default entry point (instead of xxxinit) unless you specify
another via the -e option.
|
| |
|
| |
|
|
|
|
| |
`ld' by calling execl() directly rather than using system().
|
| |
|
|
|
|
|
| |
-q just shut up, will you?!
-u unlink output file after successful load
|
|
|
|
|
|
|
|
|
| |
modload doesn't honor it's -p argument.
It also will destroy the input file when you don't specify an output
symbol file with -o.
Submitted by: John Kohl
|
| |
|
| |
|
|
utilities w/improvements from the NetBSD group.
|