| Age | Commit message (Collapse) | Author |
|---|
|
abbreviation for --keep-locals. from binutils-current.
millert@ ok.
|
|
|
|
sense this way and works around an apparent gcc optimizer bug on
macppc related to inline functions. OK drahn@
|
|
|
|
|
|
o Prevent close() from clobbering errno.
Patch from Peter Werner; closes PR 2995
|
|
time anymore.
|
|
|
|
|
|
|
|
|
|
<hshoexer at yerbouti dot franken dot de> via pr/3014.
mickey@ and weingart@ ok.
|
|
|
|
|
|
is already in the src dir. miod@ OK
|
|
with optimization.
|
|
now build libraries with propolice enabled. Without this, existing
binaries (such as ports/packages) that link with any system library
other than libc will fail with an undefined symbol of "___guard"
(__guard on ELF).
Pointed out by markus@ and discussed with deraadt@
|
|
|
|
fixes the "NEEDED crtend.o" problem that appeared in varios ports and
had to be workarounded. from thorpej@NetBSD, thanks to drahn@ for
pointing it out; ok drahn espie
|
|
attack protection scheme, into gcc.
This protection is enabled by default. It can be turned off by using the
-fno-stack-protector flag.
Code by Hiroaki Etoh (etoh at jp dot ibm dot com); work on openbsd-specific
integration by fgsch@, deraadt@ and myself; tests by fgsch@, naddy@ and
myself; beer drinking by myself.
Please note that system upgrades with this new code will require a new
libc and ld.so to be build and installed before the propolice-enabled
compiler can be installed.
|
|
|
|
allow, e.g., linking against static debug qt (or debug static mozilla for
that matter).
okay miod@
|
|
--
A CRLF injection vulnerability has been reported for Lynx that
may allow an attacker to include extra HTTP headers when viewing
web pages.
If Lynx is called from the command line, carriage return and line
feed (CRLF) characters may be included in the specified URL.
These characters are not escaped when the input is used to construct
a HTTP request.
URL: http://www.flora.org/lynx-dev/html/month082002/msg00211.html
henning@ fgs@ pjanzen@ pvalchev@ ok
|
|
|
|
|
|
problems with files > 2GB.
-print bytes in/out to stderr in verbose mode
ok millert@ fgsch@
|
|
instead so we don't need to use the (emulated) setre[ug]id() in libc.
|
|
I don't think it is needed anymore and it breaks MM_Unix's prefixify test.
|
|
=head[1-4] correctly.
|
|
macros.h on vax (noticed by hugh).
|
|
However, we define that to the empty string in OpenBSD since we
want to format and install perl's man pages ourselves. As a result,
man pages were not being generated from .pm files in 3rd party modules.
To fix this, go back to the older behavior of using $inst_man{1,3}dir.
Also, while I'm in here, improve the regexp for deciding what
consists of pod, based on ExtUtils-MakeMaker-6.05.
|
|
setre[ug]id().
|
|
|
|
|
|
|
|
submitted to perlbug@
ok millert@
|
|
|
|
|
|
|
|
do the lndir once (like apache, for example),
depend on ${.OBJDIR}/config.over .
make separate rules for the toke.*o, since
gcc blows on it on some platforms (hppa for now).
would be just nice if smth like this worked, but it does not:
.if make(toke.o)
CFLAGS+=-O0
.endif
millert@ ok
|
|
segment can share the same page in the file. Since we can't mmap it with
the same permissions, the data segment is loaded offset by a constant so
that it ends up in a different page.
On sparc that's not really enough. To avoid cache aliases we have to make
the offset big enough to make sure that we don't get incoherent aliases
that would have to be mapped uncached. This offset has been 64k this
far and noone noticed until some change actually made us fault in those
pages at the same time and incoherent aliases reduced make build time on
some hypersparcs (only hypersparcs have big enough caches to notice this)
from 14 hours to 23 hours.
Bump that offset (MAXPAGESIZE) to 1MB.
deraadt@ ok
|
|
TEST COMPILES?!
|
|
unnecessary now.
|
|
|
|
|
|
search for shared objects instead of using a LD_RUN_PATH hack. We
use -R here since it works with both the old and new (binutils) ld
flavors. The LD_RUN_PATH hack was using the wrong directory anyway
and so didn't even work. Problem pointed out by H. Steuer.
|
|
|
|
* normalized the signatures of the functions passed to each_file()
and each_full_file(). In most cases that meant adding a void *dummy.
In one case it changed an int to a void*, but the arg was only checked
for zero/not-zero so the change was not significant.
* removes unused code and structures.
* makes sure printf args match their format.
* got rid of some of the gcc -Wall warnings
OK deraadt@
|
|
|
|
From Solar Designer
|
