Andy Polyakov [Mon, 27 Dec 2004 14:59:36 +0000 (14:59 +0000)]
As new major IRIX release is highly unlikely to appear [and break following],
I change from -notall to -none synonym in irix rules to improve backward
compatibility with IRIX 5.x.
PR: 987
Andy Polyakov [Wed, 1 Dec 2004 15:28:18 +0000 (15:28 +0000)]
I've introduced a bug to i386 RC4 assembler, which would emerge with
certain mix of calls to RC4 routine not covered by rc4test.c.
It's fixed now. In addition this patch inadvertently fixes minor
performance problem: in 0.9.7 context P4 was performing 12% slower
than the original implementation...
Richard Levitte [Tue, 30 Nov 2004 12:18:55 +0000 (12:18 +0000)]
Split X509_check_ca() into a small self and an internal function
check_ca(), to resolve constness issue. check_ca() is called from the
purpose checkers instead of X509_check_ca(), since the stuff done by
the latter (except for calling check_ca()) is also done by
X509_check_purpose().
Richard Levitte [Mon, 29 Nov 2004 11:28:08 +0000 (11:28 +0000)]
Make an explicit check during certificate validation to see that the
CA setting in each certificate on the chain is correct. As a side-
effect always do the following basic checks on extensions, not just
when there's an associated purpose to the check:
- if there is an unhandled critical extension (unless the user has
chosen to ignore this fault)
- if the path length has been exceeded (if one is set at all)
- that certain extensions fit the associated purpose (if one has been
given)
Andy Polyakov [Sun, 21 Nov 2004 10:36:25 +0000 (10:36 +0000)]
RC4 tune-up for Intel P4 core, both 32- and 64-bit ones. As it's
apparently impossible to compose blended code with would perform
satisfactory on all x86 and x86_64 cores, an extra RC4_CHAR
code-path is introduced and P4 core is detected at run-time. This
way we keep original performance on non-P4 implementations and
turbo-charge P4 performance by factor of 2.8x (on 32-bit core).
Add command line options -certform, -keyform and -pass to s_client and
s_server. This supports the use of alternative passphrase sources, key formats
and keys handled by an ENGINE.
Richard Levitte [Thu, 11 Nov 2004 18:18:43 +0000 (18:18 +0000)]
Some find it confusing that environment variables are set when shared
libraries aren't built or used. I can see the point, so I'm
reorganising a little for clarity.
Richard Levitte [Fri, 5 Nov 2004 09:12:10 +0000 (09:12 +0000)]
Make sure LD_PRELOAD is only set when we build shared libraries (and
therefore link with them). Add LD_PRELOAD setting code where it was
still missing.
Richard Levitte [Tue, 2 Nov 2004 01:13:04 +0000 (01:13 +0000)]
Because -rpath/-R may have been used, our settings of LD_LIBRARY_PATH
and friends may be entirely useless. In such a case, LD_PRELOAD is
the answer, at least on platforms using LD_LIBRARY_PATH. There might
be other variables to set on other platforms, please fill us in...
For now, we only do this with the tests, so they won't fail for silly
reasons like getting dynamically linked to older installed libraries
rather than the newly built ones...
Richard Levitte [Thu, 14 Oct 2004 05:48:59 +0000 (05:48 +0000)]
Because libraries on Windows lack useful version information, the zlib
guys had to change the name to differentiate with older versions when
a backward incompatibility came up. Of course, we need to adapt.
This change simply tries to load the library through the newer name
(ZLIB1) first, and if that fails, it tries the good old ZLIB.
Andy Polyakov [Tue, 28 Sep 2004 20:45:10 +0000 (20:45 +0000)]
Fix Solaris 10_x86 shared build. -Bsymbolic is required to avoid
"remaining relocations" in assembler modules. The latter seems to
be new behaviour, elder as/ld managed to resolve this relocations
as internal. It's possible to address this problem differently,
but I settle for -Bsymbolic...
PR: 546
Richard Levitte [Thu, 23 Sep 2004 22:11:39 +0000 (22:11 +0000)]
Import changed files from LPlib. The changes are logged as follows
for LPdir_unix.c in LPlib. For the other files, only the last log
entry applies.
----------------------------
revision 1.11
date: 2004/09/23 22:07:22; author: _cvs_levitte; state: Exp; lines: +20 -6
Define my own macro LP_ENTRY_SIZE to express the size of my own
buffering of directory entries, and make it depend on whichever comes
first of PATH_MAX and NAME_MAX. As a fallback, make sure it's set to
255 if neither PATH_MAX or NAME_MAX were defined. Also, if the size
given from PATH_MAX or NAME_MAX is less than 255, force LP_ENTRY_SIZE
to be 255.
It makes no harm whatsoever if LP_ENTRY_SIZE is larger than the
maximum local path name limit. It does make a lot of harm if
LP_ENTRY_SIZE is smaller. 255 seemed like a fairly acceptable default
when nothing else is available.
----------------------------
revision 1.10
date: 2004/08/26 13:36:05; author: _cvs_levitte; state: Exp; lines: +13 -13
License correction. I am not REGENTS, just a COPYRIGHT HOLDER.
----------------------------
Richard Levitte [Mon, 13 Sep 2004 09:15:06 +0000 (09:15 +0000)]
- There's no more need for the snprintf macro.
- Move the inclusion of malloc.h until after all other includes, so we
can do proper tests of system macros.
- Make sure the correct header file is included to get the builtin
"alloca" under VMS, and define a macro to map the symbol 'alloca' to
it.