similar to: LP64 fix

Christian Weisgerber <naddy@mips.inka.de> wrote: > #0 0x0000000040d18810 in FLAC__lpc_compute_residual_from_qlp_coefficients_wide > (data=0x49e4c014, data_len=110, qlp_coeff=0x7f7ffffece70, order=1, > lp_quantization=14, residual=0x4fced000) at lpc.c:745 > 745 residual[i] = > data[i] - (FLAC__int32)((qlp_coeff[0] *

Running the basic (--disable-thorough-tests) test suite, I get these failures round-trip test (rt-1-24-111.raw) encode... Segmentation fault (core dumped) ERROR FAIL: ./test_flac.sh fsd24-01 (--channels=1 --bps=24 -0 -l 16 --lax -m -e -p): encode...ERROR during encode of fsd24-01 FAIL: ./test_streams.sh on alpha and amd64. By contrast, i386 is fine. (All OpenBSD/4.2.) Could be a generic LP64

Did you really experience problems with the code? Since I only care about the last bits (because of the & "(size-1)"), I thought it should work even if sizeof(int) != sizeof(void*). If it doesn't work with int, I suspect the solution with the long could also fail on some archs. Jean-Marc Le mer 11/06/2003 à 11:56, Christian Weisgerber a écrit : > Casting a pointer to

Christian Weisgerber <naddy@mips.inka.de> wrote: > | # nasm build rule: > | %.lo: %.s Even with gmake, this really works only by accident. Automake generates a competing suffix rule, and gmake apparently gives the rule above a higher priority than the suffix rule (or that just happens to be the evaluation order). Matt, since you seem to understand automake, can you come up with

Matt Zimmerman <mdz@debian.org> wrote: > Aha. In FLAC__fixed_restore_signal, the index variable 'i' is declared > unsigned, then used like so: [...] With the following trivial patch applied, > everything works on Alpha, at least with my test sample. Against 0.9 or CVS? While this clearly fixes a bug, 0.9 still dies for me. #0 0x1600692a8 in

That sounds like the application param is still used as part of the process though? Would allowing the user to specify the application work in the Solokey case? What is stored in the private keyfile? The documentation says no private key is stored there. So is it just information used to reseed the public/private key? Thanks, Kevin ________________________________________ From: openssh-unix-dev

Currently, if IPv6-enabled rsync is run as --daemon, it will perform a wildcard bind(2) on an AF_INET6 socket and expect that IPv4 traffic will be forwarded to the v6 socket (IPv4 mapped address, RFC2553). This has never worked on OpenBSD which disallows IPv4 mapped addresses for security reasons. On FreeBSD 5.0-CURRENT, where the behavior is subject to configuration, the default was switched to

Problems in FLAC 0.9: - On alpha, flac immediately dumps core for both encoding and decoding (FreeBSD/alpha). - The distribution Makefile.in files haven't been generated with "automake --include-deps". The resulting Makefiles aren't fully portable; in particular they break with BSD make. In the future, care should be taken to use "--include-deps". - What is

There are problems in the configure option handling code for http streaming that was added to the XMMS plugin for 1.1.2. In configure.c, flac_cfg.stream.proxy_host and other pointers to strings are initialized with pointers to an empty string "". Subsequent code in FLAC_XMMS__init() and flac_configurewin_ok() passes these pointers to g_free(). Since they don't point to memory that

This affects the XMMS plugin. configure.in has this test: AC_CHECK_TYPES(socklen_t, [], []) And src/plugin_xmms/http.c is the only consumer: #ifndef HAVE_SOCKLEN_T typedef unsigned int socklen_t; #endif Together this looks bogus to me. The configure check looks for socklen_t in the default headers. If it isn't found there, socklen_t will be typedef'ed. However, at least on

I'm currently trying to figure out why RC3 oggenc with iconv support will - work, but enter replacement characters into tags instead of converting to UTF-8; - dump core; or - catch an assert(); depending on the locale settings. (On FreeBSD, with Chuguev iconv.) I think I'm getting a handle on the assert(). It's this one: assertion "(!k && !ibl) || (k == (size_t)(-1)

libogg 1.1's included regression tests ("make check") fail at least on alpha: ---------------> ./test_bitwise Small preclipped packing (LSb): ok. Null bit call (LSb): ok. Large preclipped packing (LSb): ok. 32 bit preclipped packing (LSb): ok. Small unclipped packing (LSb): ok. Large unclipped packing (LSb): ok. Single bit unclipped packing (LSb): ok. Testing read past end

Here's the promised Sun audio system plug-in. I have tested this on OpenBSD. It should work without changes on NetBSD, too. Jeremy, could you please verify this? After a few minor tweaks it now also compiles on Solaris 2.7. Alas, I can't verify whether it actually plays anything there. I guess this could also with little effort be made to work on SunOS4. Attached: - Diff to

Building 1.3.2 on OpenBSD/i386, I get this error: cpu.c: In function 'ia32_cpu_info': cpu.c:128: error: 'FLAC__CPUINFO_IA32_CPUID_SSE3' undeclared (first use in this function) cpu.c:128: error: (Each undeclared identifier is reported only once cpu.c:128: error: for each function it appears in.) cpu.c:129: error: 'FLAC__CPUINFO_IA32_CPUID_SSSE3' undeclared (first use in

I uploaded the source tarball and binary releases for win32 and redhat 7. Solaris package should be ready tomorrow. See the news item for all the improvements from 1.0.3: http://flac.sourceforge.net/news.html#20020924 Developers, see the new Doxygen-based API docs: http://flac.sourceforge.net/api/index.html Thanks again to all the contributors and testers. Josh

_WIN32 is always defined by the compiler automatically when targeting that platform, while WIN32 only is defined automatically in some configurations, and e.g. in MSVC only ever is defined in project files (if at all). Some other checks in the codebase already check for both WIN32 and _WIN32; those are left untouched. --- include/speex/speex.h | 2 +- libspeex/stack_alloc.h | 2 +-

The PowerPC-related changes in FLAC 1.3.3 have caused some portability problems. libFLAC/cpu.c assumes that the <sys/auxv.h> header and the getauxval() function are universally available on PowerPC platforms. They are not. On FreeBSD/powerpc, <sys/auxv.h> is available, but getauxval() is not. Equivalent functionality is provided by elf_aux_info(). On OpenBSD/powerpc, neither is

On Sat, May 19, 2001 at 12:19:39PM +0000, Christian Weisgerber wrote: > Christian Weisgerber <naddy@mips.inka.de> wrote: > > > Problems in FLAC 0.9: > > Also: > > ===> Building for flac-0.9 > cd . && aclocal > aclocal: configure.in: 7: macro `AM_PROG_LIBTOOL' not found in library > aclocal: configure.in: 22: macro `AM_PATH_XMMS' not

I bumped the version number of vorbis-tools and am preparing it for an eventual release in a few days. A changelog of what's new may be found at [1]. Meanwhile, what's in SVN is a "Release Candidate" to find out if anything was broken since version 1.1.1. Report any (or lack of) problems you may find, so we'll be able to put out a new version of vorbis-tools. -Ivo [1]

configure uses invalid shell syntax. (Ah, it's a bash extension.) Here's a portable fix. --- configure.ac.orig Thu Dec 5 00:26:45 2013 +++ configure.ac Sun Dec 15 17:32:25 2013 @@ -333,9 +333,9 @@ AS_IF([test x"${enable_asm}" = x"yes"],[ AM_CONDITIONAL([CPU_ARM], [test "$cpu_arm" = "yes"]) AM_CONDITIONAL([OPUS_ARM_INLINE_ASM], - [test