Bug#146305: celestia: Arithmetic exception on alpha
tags 146305 + moreinfo thanks I cannot reproduce this with 1.3.2-3.1+b1. Can anybody still reproduce it? -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#193494: freeciv-server: Crashes on first client connect on alpha
tags 193494 + moreinfo I cannot reproduce this as of 2.0.7-2. Can anybody? -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#353532: FTBFS (alpha): va_list abuse
Package: poldi Version: 0.3.0-1 Severity: important Justification: fails to build from source poldi fails to build on Alpha: [...] if alpha-linux-gnu-gcc -DHAVE_CONFIG_H -I. -I. -I../.. -I../../intl -fPIC -Wall -g -O2 -Wall -MT logging.o -MD -MP -MF ".deps/logging.Tpo" -c -o logging.o logging.c; \ then mv -f ".deps/logging.Tpo" ".deps/logging.Po"; else rm -f ".deps/logging.Tpo"; exit 1; fi logging.c: In function 'set_file_fd': logging.c:298: warning: assignment from incompatible pointer type logging.c: In function 'do_logv': logging.c:606: error: incompatible type for argument 3 of 'vsyslog' make[4]: *** [logging.o] Error 1 make[4]: Leaving directory `/tmp/poldi-0.3.0/src/jnlib' see full log at http://buildd.debian.org/fetch.php?pkg=poldi&arch=alpha&ver=0.3.0-2&stamp=1140319236&file=log This is because of: vsyslog (syslog_priorities[JNLIB_LOG_INFO], fmt_prefix, NULL); NULL is not a va_list. I'm unsure what this is supposed to do, probably you can just use syslog instead. Falk -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.16-rc4-dirty Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#238687: Submit Debian release and subarchs
Hi, I'd like to lobby again for the inclusion of my patch at http://bugs.debian.org/238687 again, which reports Alpha subarchitectures and the Debian version. The motivation is to decide whether to drop ev4 or ev56 support. The objections were: * Subarchitecture information should come from another package. Well, there is no such package right now, and it seems silly to start one containing 10 lines of code. Moreover, I would prefer, as a matter of principle, to only report information we are sure is needed for some concrete decision. * /etc/debian_version is unreliable. However, I don't think the percentage of users actively setting their /etc/debian_version to something bogus is that high, so it would be still useful. It would be nice to get some more opinions on this... -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#348723: gcc-3.3 [powerpc] ICE in simplify_subeg, at simplify-rtx.c:2368
tags 348723 + moreinfo thanks Hi, gcc-3.3 isn't supported upstream anymore. Can you please test with gcc 3.4? -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#353563: Do not build on alpha
Hi, this would be caused by trying to generate a shared object from sources not compiled with -fPIC. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#353563: Do not build on alpha
Ola Lundqvist <[EMAIL PROTECTED]> writes: > On Mon, Feb 20, 2006 at 08:09:51AM +0100, Falk Hueffner wrote: >> this would be caused by trying to generate a shared object from >> sources not compiled with -fPIC. > > Ok, thanks. Do you have a patch for that? > > I thought that I have made it so, but obviously I have missed > something. Sorry, I don't have a patch. But it shouldn't be that difficult to hack the Makefiles or something. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#353922: Bug #353922: kdepim FTBFS on alpha
tags 353922 + patch thanks Hi, this should fix it (haven't tested though). --- kdepim-3.5.1/kresources/groupwise/soap/stdsoap2.h~ 2005-09-10 10:23:46.0 +0200 +++ kdepim-3.5.1/kresources/groupwise/soap/stdsoap2.h 2006-02-22 23:41:23.0 +0100 @@ -170,7 +170,7 @@ # endif #endif -#ifdef __alpha +#ifdef __osf__ # ifndef TRU64 # define TRU64 # endif -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#238687: Submit Debian release and subarchs
Bill Allombert <[EMAIL PROTECTED]> writes: > On Sun, Feb 19, 2006 at 01:38:35PM +0100, Falk Hueffner wrote: >> I'd like to lobby again for the inclusion of my patch at >> http://bugs.debian.org/238687 again, which reports Alpha >> subarchitectures and the Debian version. The motivation is to decide >> whether to drop ev4 or ev56 support. The objections were: >> >> * Subarchitecture information should come from another package. Well, >> there is no such package right now, and it seems silly to start one >> containing 10 lines of code. Moreover, I would prefer, as a matter >> of principle, to only report information we are sure is needed for >> some concrete decision. > > At least you can see what kernels are in use: That does not help in answering the above question for Alpha, since there are no subarchitecture specific kernels. > More importantly at this point there are still no reliable way to > find out the sub-architecture. I don't understand your point. I've posted a patch. Does it seem unreliable? > Collecting data is one thing, processing them is another. I would > not mind too much reporting the sub-architecture if there was a > Debian official way to get it. Maybe. But anyway, such a way does not exist (and "subarchitecture" is way to fuzzy for this to work reliably and in a useful way anyway, IMHO.) So, meanwhile, what about my patch? >> * /etc/debian_version is unreliable. However, I don't think the >> percentage of users actively setting their /etc/debian_version to >> something bogus is that high, so it would be still useful. > > Actually you can easily know the numbers of submissions from each > distributions using the version information: > woody (unknown): 449 > sarge (1.28): 2986 > testing/unstable (1.31/1.32): 6333 > outdated snapshot of testing/unstable (others): 641 > > I think it is better than what the debian_version would give you. Hmm, maybe it is, at least if we release a new version of popularity-contest after each stable release. So I wouldn't insist on that part of the patch. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#348072: qsort: wrong claim about strcmp being suitable as "compar" argument
Package: manpages-dev Version: 2.17-1 Severity: normal File: /usr/share/man/man3/qsort.3.gz The man page claims: NOTE Library routines suitable for use as the compar argument include str- cmp(), alphasort(), and versionsort(). But strcmp is not suitable (see http://www.lysator.liu.se/c/c-faq/c-12.html#12-2). Falk -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.13.2 Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15) Versions of packages manpages-dev depends on: ii manpages 2.17-1 Manual pages about using a GNU/Lin manpages-dev recommends no packages. -- no debconf information -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#348072: qsort: wrong claim about strcmp being suitable as "compar" argument
"Michael Kerrisk" <[EMAIL PROTECTED]> writes: > I have fixed this in the upstream 2.21 release by including a small > example that demonstrates how strcmp() should be used (like in the > page you refer to). Thanks for your quick reply. Just another minor suggestion, the example code uses: qsort(months, nr_of_months, sizeof(struct mi), compmi); [...] res = bsearch(&key, months, nr_of_months, sizeof(struct mi), compmi); I think it would be better to use qsort(months, nr_of_months, sizeof *months, compmi); [...] res = bsearch(&key, months, nr_of_months, sizeof *months, compmi); since this is more robust against changes and easier to verify for a human reader. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#348117: g++-4.0: g++ upgrade broke ncurses c++ demo
Thomas Dickey <[EMAIL PROTECTED]> writes: > char buf[5]; [...] > The buf variable contains "Key01" So 6 bytes. Does this still happen if you don't overrun buf? Also, can you send a complete test case? -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#186882: Crashes immediately on Alpha
Petter Reinholdtsen <[EMAIL PROTECTED]> writes: > Can you provide a gdb backtrace from the crash, preferably from a > binary with debug information to get line numbers and function > argument information? > > This does not happen on my i386, so I am not sure how to reproduce > it. :) Hmm, I thought the project is dead anyway or superceded by something else. Is this really worth bothering? Also, it FTBFS currently due to xlibs-dev dependency. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#186882: Crashes immediately on Alpha
Petter Reinholdtsen <[EMAIL PROTECTED]> writes: > [Falk Hueffner] >> Hmm, I thought the project is dead anyway or superceded by something >> else. Is this really worth bothering? > > Well, I like playing it, so I try to keep the package in Debian > working. :) The problem is that "long" is used for 32-bit types, which is fixed in 04_64bit_archs_fix.dpatch apparently, but that is disabled, and it doesn't seem to apply. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#348117: g++-4.0: g++ upgrade broke ncurses c++ demo
reassign 348117 ncurses severity 348117 normal retitle 348117 Buffer overflow in example code c++/demo.cc thanks Thomas Dickey <[EMAIL PROTECTED]> writes: > On Sun, 15 Jan 2006, Falk Hueffner wrote: > >> Thomas Dickey <[EMAIL PROTECTED]> writes: >> >>> char buf[5]; >> >> [...] >> >>> The buf variable contains "Key01" >> >> So 6 bytes. Does this still happen if you don't overrun buf? Also, >> can you send a complete test case? > > It doesn't happen if I don't overrun. (Do you still need a complete > test case?) No, I'll just reassign it to ncurses then. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#348351: Leaks potentially sensitive information
Package: gri Version: 2.12.9-2 Severity: normal gri includes by default in the .eps output the user name, hostname, creation date, and the contents of the creation script, including comments. All of these are potentially things I wouldn't want to let everybody know when distributing the .eps. Please don't enable this by default. Falk -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.13.2 Locale: LANG=C, LC_CTYPE=C (charmap=ANSI_X3.4-1968) Versions of packages gri depends on: ii libc6.1 2.3.5-11 GNU C Library: Shared libraries an ii libgcc1 1:4.0.2-6 GCC support library ii libstdc++51:3.3.6-12 The GNU Standard C++ Library v3 ii netcdfg3 3.5.0-7.1 An interface for scientific data a gri recommends no packages. -- no debconf information -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#351575: Crashes immediately on Alpha
Package: ickle Version: 0.3.2-6 Severity: important When I start ickle with my existing config, or after I remove my .ickle and re-enter my uid, ickle crashes with this message: [EMAIL PROTECTED]:~% ickle gtkspell: Timed out waiting for spell command. gtkspell: Timed out waiting for spell command. gtkspell: Timed out waiting for spell command. *** glibc detected *** free(): invalid pointer: 0x0001202d45f0 *** zsh: abort (core dumped) ickle I'm pretty sure it used to work a month or two ago. Falk -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.13.2 Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15) Versions of packages ickle depends on: ii ickle-common 0.3.2-6Common elements between the ickle ii libc6.1 2.3.5-12.1 GNU C Library: Shared libraries an ii libgcc1 1:4.0.2-8 GCC support library ii libglib1.21.2.10-10.1The GLib library of C routines ii libgtk1.2 1.2.10-18 The GIMP Toolkit set of widgets fo ii libgtkmm1.2-0c2a 1.2.10-8 C++ wrappers for GTK+ 1.2 (shared ii libicq2000c2 0.3.2-5An ICQ2000/2001 C++ library ii libsigc++0c2 1.0.4-9.1 Type-safe Signal Framework for C++ ii libstdc++64.0.2-8The GNU Standard C++ Library v3 ii libx11-6 6.9.0.dfsg.1-4 X Window System protocol client li ii libxext6 6.9.0.dfsg.1-4 X Window System miscellaneous exte ii libxi66.9.0.dfsg.1-4 X Window System Input extension li ickle recommends no packages. -- no debconf information -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#351586: FTBFS (alpha): #error "Must be compiled with option -mieee"
Package: mednafen Version: 0.4.7-1 Severity: serious Justification: no longer builds from source mednafen 0.4.7-1 does not build on Alpha: [...] source='trio/trionan.c' object='trio/trionan.o' libtool=no \ depfile='trio/.deps/trionan.Po' tmpdepfile='trio/.deps/trionan.TPo' \ depmode=gcc3 /bin/sh ../depcomp \ alpha-linux-gnu-gcc -DLOCALEDIR=\"/usr/share/locale\" -DHAVE_CONFIG_H -I.. -I../include -I../intl -Ifidlib -pthread -DFIDLIB_LINUX -Wall -Winline -fomit-frame-pointer -ffast-math -finline-limit=6000 --param large-function-growth=800 --param inline-unit-growth=175 -I/usr/include/SDL -D_REENTRANT -pthread -DFIDLIB_LINUX -Wall -Winline -fomit-frame-pointer -ffast-math -finline-limit=6000 --param large-function-growth=800 --param inline-unit-growth=175 -Wall -O2 -Wl,-z,defs -I/usr/include/SDL -D_REENTRANT -c -o trio/trionan.o `test -f 'trio/trionan.c' || echo './'`trio/trionan.c trio/trionan.c:89:5: error: #error "Must be compiled with option -mieee" make[3]: *** [trio/trionan.o] Error 1 make[3]: Leaving directory `/tmp/buildd/mednafen-0.4.7/src' [...] -mieee is on by default with Debian's Alpha gcc, but disabled by -ffast-math. If this code really needs precise IEEE semantics (and not only IEEE representation of finite values or something), it needs to avoid -ffast-math (on any architecture). Falk -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.13.2 Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#352042: Bug #352042: openssh-server: incompatible pointer types break gssapi auth on alpha, possibly others
Hi, some comments: * ints are 32 bit on Alpha, like on any Linux target * casting an unsigned int* to int* should have absolutely no effect on the generated code. So this might be a gcc bug. Can you show a diff of the assembly? -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#352042: Bug #352042: openssh-server: incompatible pointer types break gssapi auth on alpha, possibly others
Hi,
the actual problem is that gss_buffer_desc is defined like this in
/usr/include/gssapi/gssapi.h from libkrb5-dev:
typedef struct gss_buffer_desc_struct {
size_t length;
void *value;
} gss_buffer_desc, *gss_buffer_t;
size_t is 64 bit. So you need something like (untested):
--- monitor.c~ 2006-02-10 10:19:16.0 +0100
+++ monitor.c 2006-02-10 10:32:27.0 +0100
@@ -1946,8 +1946,10 @@
{
gss_buffer_desc data, hash;
OM_uint32 major, minor;
+ u_int length;
- data.value = buffer_get_string(m, &data.length);
+ data.value = buffer_get_string(m, &length);
+ &data.length = length;
if (data.length != 20)
fatal("%s: data length incorrect: %d", __func__, data.length);
and the same elsewhere.
So this is OpenBSD's flagship security product? Frankly, I'm baffled.
This is a rookie mistake, and it was even obvious from the compiler
warnings.
--
Falk
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#342126: gcc-4.0: gcc generates bad assembly code for simple code
Vagner Pedrotti <[EMAIL PROTECTED]> writes:
> When I compile the following simple code, gcc fails with:
> /tmp/ccySlj9q.s: Assembler messages:
> /tmp/ccySlj9q.s:56: Error: missing or invalid immediate expression `'
> taken as 0
> /tmp/ccySlj9q.s:56: Error: suffix or operands invalid for `call'
This boils down to:
void f() {
$();
}
Apparently i386 doesn't like symbols starting with $. There should
probably be a better error message, but I wouldn't give it very high
priority... Or maybe there is some way to "escape" the symbol name?
--
Falk
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#342148: FTBFS (alpha): initializer element is not constant
Package: xoscope
Version: 1.12-2
Severity: important
Justification: fails to build from source
xoscope fails to build on Alpha:
[...]
cc -DHAVE_CONFIG_H -I. -I. -I. -I/usr/include/gtk-1.2 -I/usr/include/glib-1.2
-I/usr/lib/glib/include-g -Wall -O2 -c gr_gtk.c
gr_gtk.c: In function 'setbits':
gr_gtk.c:524: warning: cast from pointer to integer of different size
gr_gtk.c: At top level:
gr_gtk.c:670: error: initializer element is not constant
gr_gtk.c:670: error: (near initialization for 'menu_items[2].callback_action')
gr_gtk.c:671: error: initializer element is not constant
gr_gtk.c:671: error: (near initialization for 'menu_items[3].callback_action')
gr_gtk.c:676: error: initializer element is not constant
[...]
The lines look like this:
{"/File/Quit", NULL, hit_key, (int)"\e", NULL},
Casting is not allowed in an initializer, and moreover casting a
pointer to int will only yield garbage on a 64-bit machine. It only
happens to "work" on i386 due to a gcc quirk.
-- System Information:
Debian Release: testing/unstable
APT prefers unstable
APT policy: (500, 'unstable')
Architecture: alpha
Shell: /bin/sh linked to /bin/bash
Kernel: Linux 2.6.13.2
Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15)
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#342484: Error: unknown pseudo-op: `.weakref'
Mathieu Malaterre <[EMAIL PROTECTED]> writes: > Package: gcc-snapshot > Version: 20051124-1 > Severity: important > > /tmp/ccFywWsc.s:2650: Error: unknown pseudo-op: `.weakref' Newer binutils is needed. 2.16.1cvs20051109-1 should do. Can you please try that? -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#342484: Compilation is working
Mathieu Malaterre <[EMAIL PROTECTED]> writes: > Thanks Falk, you can close the bug. Updating binutils from unstable > did the trick Well, it shouldn't really be closed, but gcc-snapshot should depend on that version. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#339061: bug in binutils
On Wed, Dec 07, 2005 at 08:54:03PM -0500, Mike Furr wrote: > This is actually a bug in binutils (338148). I'll leave this bug > here just so it doesn't get reported again. AFAICT, 338148 is about ocaml not setting .prologue correctly, and not about a binutils bug. Can you give more details? -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#338148: Bug#339061: bug in binutils
Mike Furr <[EMAIL PROTECTED]> writes: > After some more investigation described in 338148, we found out that it > was actually OCaml which was at fault. A new version of OCaml was > uploaded yesterday which has a work-around for this bug and appears to > produce working binaries on alpha. Thanks, that clears it up (I had forgotten felix uses OCaml). Is anybody working on a proper patch for OCaml? If not, I could give it a try. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#338148: [alpha] regression: ocamlopt.opt segfaults
Hi, AFAICS, the caml_apply2 posted to http://caml.inria.fr/mantis/view.php?id=3912 does not indicate an actual problem. ".prologue 0" means the function will not rely on pv ($27) being set correctly, and it will not clobber gp ($29). Clobbering pv is OK. So the problem is probably elsewhere. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#338148: [alpha] regression: ocamlopt.opt segfaults
Mike Furr <[EMAIL PROTECTED]> writes: > Falk Hueffner wrote: >> AFAICS, the caml_apply2 posted to >> http://caml.inria.fr/mantis/view.php?id=3912 does not indicate an >> actual problem. ".prologue 0" means the function will not rely on pv >> ($27) being set correctly, and it will not clobber gp ($29). >> Clobbering pv is OK. So the problem is probably elsewhere. > > My understanding of alpha assembly and calling conventions is very > minimal at best, so this is entirely possible. I was using[1] as a > reference which says that "[.prologue] 0 indicates that $27 is not > used". Since it is "used" in caml_apply2, it looked like an error > to me (note the $29 was a typo in my report, see the notes at the > bottom). Also if I change the assembly by hand from '.prologue 0' > to 1 (or 2), the segfault does not occur (at least in that > function). Hmm. I was looking at the Assembly Language Programmer's Guide for Digital UNIX, which says: .prologue flag Marks the end of the prologue section of a procedure. A flag of 0 indicates that the procedure does not use $gp; the caller does not need to set up $pv prior to calling the procedure or restore $gp on return from the procedure. A flag of 1 indicates that the procedure does use $gp; the caller must set up $pv prior to calling the procedure and restore $gp on return from the procedure. If flag is not specified, the behavior is as if a value of 1 was specified. My guess at what actually happens is that whatever caml_apply2 calls clobbers gp, and the linker elides the reconstruction of gp on return. I haven't looked, but this is probably a indirect tail call at the intermediate language level. Setting .prologue to 1 there is probably the right thing to do. Maybe like this: --- alpha/emit.mlp~ 2005-10-14 18:41:06.0 +0200 +++ alpha/emit.mlp 2005-12-10 19:38:46.0 +0100 @@ -59,7 +59,7 @@ | Lop(Iconst_symbol s) -> true (* turned into ldq ($gp) *) | Lop(Icall_ind) -> false (* does ldgp if needed afterwards *) | Lop(Icall_imm s) -> true (* does lda $27, *) -| Lop(Itailcall_ind) -> false +| Lop(Itailcall_ind) -> true | Lop(Itailcall_imm s) -> if s = f.fun_name then label_needs_gp tailrec_entry_point else true | Lop(Iextcall(_, _)) -> true (* does lda $27, *) -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#338148: [alpha] regression: ocamlopt.opt segfaults
Falk Hueffner <[EMAIL PROTECTED]> writes:
> Maybe like this:
Actually, what I really want is:
--- ocaml-3.09.0.orig/asmcomp/alpha/emit.mlp2005-10-14 18:41:06.0 +0
200
+++ ocaml-3.09.0/asmcomp/alpha/emit.mlp 2005-12-10 20:50:11.0 +0100
@@ -740,7 +740,7 @@
` .fmask 0x0, 0\n`
end;
`.frame $sp, {emit_int n}, $26\n`;
- `.prologue {emit_int(if needs_gp then 1 else 0)}\n`;
+ `.prologue {emit_int(if !contains_calls || needs_gp then 1 else 0)}\n`;
tailrec_entry_point := new_label();
`{emit_label !tailrec_entry_point}:\n`;
emit_all true fundecl.fun_body;
which gives .prologue 1 for caml_apply2. However, ocaml still
segfaults when bootstrapping without --no-relax...
--
Falk
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#335684: rrdtool graph FPE on alpha
tags 335684 + moreinfo thanks Hi, this is probably fixed by gcc now defaulting to -mieee again. Can anybody please test it, or tell me how to reproduce it? -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#342939: FTBFS: Hardcoded "-march=pentium"
Package: toolame Version: 02l-1 Severity: serious Justification: no longer builds from source toolame fails to build on almost all architectures: make[1]: Entering directory `/tmp/toolame-02l' gcc -O3 -DNDEBUG -DINLINE=inline -march=pentium -fomit-frame-pointer -Wall -DNEWENCODE -DNEWATAN -c common.c -o common.o cc1: error: invalid option 'arch=pentium' make[1]: *** [common.o] Error 1 make[1]: Leaving directory `/tmp/toolame-02l' This option also mustn't be present on i386, since it breaks the program on 486 class machines, which we still support. Falk -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.13.2 Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15) Versions of packages toolame depends on: ii libc6.1 2.3.5-8.1 GNU C Library: Shared libraries an toolame recommends no packages. -- no debconf information -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#343511: g++-3.4 -O1: wrong code generation
Bill Allombert <[EMAIL PROTECTED]> writes:
> Package: g++-3.4
> Version: 3.4.5-1
> Severity: normal
>
> Hello Debian GCC maintainers,
>
> g++-3.4 incorrectly compiles the attached file at -O1 and up by causing
> an infinite loop.
This reduces to:
__attribute__ ((pure)) inline long f(void) {
return 0;
}
void __attribute__ ((noinline)) g(long *e)
{
e[f()] += 1;
}
int main(void)
{
long e[163];
g(e);
return 0;
}
which looks to be the same problem as this:
http://gcc.gnu.org/PR17972
so unfortunately, it looks like this won't be fixed upstream. It looks
pretty nasty, though, maybe we should ask again.
--
Falk
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#334339: Is bug #334339 a gcc issue?
Christian Hammers <[EMAIL PROTECTED]> writes: > sql_analyse.cc: In member function 'virtual void field_longlong::add()': > sql_analyse.cc:506: internal compiler error: in invert_exp_1, at jump.c:1719 Please provide the preprocessed source (.ii), as obtained by adding -save-temps to the compilation flags. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#340660: FTBFS (alpha): Assertion `sizeof(IEEE_64) == ((2)*sizeof(obj))' failed
Package: rscheme Version: 0.7.3.4.b1-2 Severity: serious Justification: no longer builds from source rscheme fails to build on Alpha: [...] gcc -g -O2 -g -Wall -Wno-unused '-DINSTALL_DIR="/usr/lib/rscheme"' -I. -I- -I/tmp/rscheme-0.7.3.4.b1/debian/tmproot/include shell.o -o rs -L/tmp/rscheme-0.7.3.4.b1/debian/tmproot/lib -lrs -rdynamic -lm -ldl -lgmp make[2]: Leaving directory `/tmp/rscheme-0.7.3.4.b1/src/rshell' rshell/rs -image tmp/system.bas -c.repl system.img rs: interim.c:54: make_float: Assertion `sizeof(IEEE_64) == ((2)*sizeof(obj))' failed. RScheme (v0.7.3.4-b1u, 2005-09-16) Copyright (C) 1995-2005 Donovan Kolbly <[EMAIL PROTECTED]> RScheme comes with ABSOLUTELY NO WARRANTY. Doing first-time initialization... make[1]: *** [system.img] Aborted (core dumped) make[1]: Leaving directory `/tmp/rscheme-0.7.3.4.b1/src' make: *** [build-stamp] Error 2 Full build log at http://buildd.debian.org/fetch.php?&pkg=rscheme&ver=0.7.3.4.b1-2&arch=alpha&stamp=1132058047&file=log&as=raw -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.13.2 Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#340664: FTBFS (alpha): arg_max_32bit_linux_bug fails
Package: findutils Version: 4.2.26-1 Severity: serious Justification: no longer builds from source findutils does not build on Alpha: [...] Running /src/xargs/findutils-4.2.26/xargs/testsuite/xargs.posix/arg_max_32bit_linux_bug.exp ... FAIL: arg_max_32bit_linux_bug, wrong stdout output: xargs.out /src/xargs/findutils-4.2.26/xargs/testsuite/xargs.posix/arg_max_32bit_linux_bug.xo differ: char 32764, line 1 child process exited abnormally Running /src/xargs/findutils-4.2.26/xargs/testsuite/xargs.posix/arg_max_64bit_linux_bug.exp ... FAIL: arg_max_64bit_linux_bug, wrong stdout output: xargs.out /src/xargs/findutils-4.2.26/xargs/testsuite/xargs.posix/arg_max_64bit_linux_bug.xo differ: char 32764, line 1 child process exited abnormally [...] === xargs Summary === # of expected passes90 # of unexpected failures2 /src/xargs/findutils-4.2.26/xargs/testsuite/../xargs version 4.2.26 make[4]: *** [check-DEJAGNU] Error 1 make[4]: Leaving directory `/src/xargs/findutils-4.2.26/xargs/testsuite' For arg_max_32bit_linux_bug, the testsuite excepts 1 line of 32766 y's, and 1 line of 1 y. The freshly built xarg produces 2 lines of 16382 y's each, and 1 line of 3 y's. So the number of y's is correct, only their partitioning isn't as expected. The same happens on ia64. Full log at http://buildd.debian.org/fetch.php?&pkg=findutils&ver=4.2.26-1&arch=alpha&stamp=1132483101&file=log&as=raw -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.13.2 Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15) Versions of packages findutils depends on: ii libc6.1 2.3.5-8GNU C Library: Shared libraries an findutils recommends no packages. -- no debconf information -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#340851: FTBFS (alpha): autogen segfaults on license.test
Package: autogen Version: 1:5.7.2-1 Severity: serious Justification: no longer builds from source autogen does not build on Alpha: [...] FAIL: license.test [...] === 1 of 34 tests failed Please report to [EMAIL PROTECTED] === Full log at http://buildd.debian.org/fetch.php?&pkg=autogen&ver=1%3A5.7.3-1&arch=alpha&stamp=1131841737&file=log&as=raw The reason is that autogen segfaults on this example. juist:/tmp/autogen-5.7.3/agen5/test/FAILURES# ../../autogen -b license --no-def -T license.tpl Segmentation fault (core dumped) The backtrace is not very helpful, unfortunately: juist:/tmp/autogen-5.7.3/agen5/test/FAILURES# gdb ../../.libs/lt-autogen GNU gdb 6.3.90_20051119-debian Copyright 2004 Free Software Foundation, Inc. GDB is free software, covered by the GNU General Public License, and you are welcome to change it and/or distribute copies of it under certain conditions. Type "show copying" to see the conditions. There is absolutely no warranty for GDB. Type "show warranty" for details. This GDB was configured as "alpha-linux-gnu"...Using host libthread_db library "/lib/libthread_db.so.1". (gdb) r -b license --no-def -T license.tpl Starting program: /tmp/autogen-5.7.3/agen5/.libs/lt-autogen -b license --no-def -T license.tpl [Thread debugging using libthread_db enabled] [New Thread 16384 (LWP 27322)] Program received signal SIGSEGV, Segmentation fault. [Switching to Thread 16384 (LWP 27322)] 0x02008a68 in _dl_rtld_di_serinfo () from /lib/ld-linux.so.2 (gdb) bt #0 0x02008a68 in _dl_rtld_di_serinfo () from /lib/ld-linux.so.2 #1 0x02008efc in _dl_rtld_di_serinfo () from /lib/ld-linux.so.2 #2 0x0200c540 in _dl_rtld_di_serinfo () from /lib/ld-linux.so.2 #3 0x0200c058 in _dl_rtld_di_serinfo () from /lib/ld-linux.so.2 #4 0x00012001a9fc in ag_scm_license () #5 0x0001200930b0 in ?? () warning: Hit heuristic-fence-post without finding enclosing function for address 0x1200930b0 This warning occurs if you are debugging a function without any symbols (for example, in a stripped executable). In that case, you may wish to increase the size of the search with the `set heuristic-fence-post' command. Otherwise, you told GDB there was a function where there isn't one, or (more likely) you have encountered a bug in GDB. #6 0x0001200930b0 in ?? () warning: Hit heuristic-fence-post without finding enclosing function for address 0x1200930b0 Previous frame identical to this frame (corrupt stack?) (gdb) It also fails with autogen 1:5.7.2-1 from sid: [EMAIL PROTECTED]:/var/cache/pbuilder/build/5817/tmp/autogen-5.7.3/agen5/test/FAILURES# autogen -b license --no-def -T license.tpl AutoGen aborting on signal 10 (Bus error) in state LOAD_TPL processing template NULL file name on line 0 for function pseudo-macro (-1) zsh: abort (core dumped) autogen -b license --no-def -T license.tpl Interestingly, the package builds OK outside of pbuilder. The problem seems to be somewhat volatile. -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.13.2 Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15) Versions of packages autogen depends on: ii guile-1.6-libs1.6.7-1.1 Main Guile libraries ii libc6.1 2.3.5-8GNU C Library: Shared libraries an ii libguile-ltdl-1 1.6.7-1.1 Guile's patched version of libtool ii libopts25 1:5.7.2-1 automated option processing librar ii libopts25-dev 1:5.7.2-1 automated option processing librar ii libqthreads-121.6.7-1.1 QuickThreads library for Guile ii libxml2 2.6.22-2 GNOME XML library ii zlib1g1:1.2.3-8 compression library - runtime autogen recommends no packages. -- no debconf information -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#340853: FTBFS (alpha): __remq is undefined
Package: gclcvs Version: 2.7.0-50.1 Severity: serious Justification: no longer builds from source gclcvs does not build on Alpha: [...] GCL (GNU Common Lisp) 2.7.0 CLtL1Oct 15 2005 08:35:10 Source License: LGPL(gcl,gmp), GPL(unexec,bfd) Binary License: GPL due to GPL'ed components: (READLINE BFD UNEXEC) Modifications of this banner must retain notice of a compatible license Dedicated to the memory of W. Schelter Use (help) to get some basic information on how to use GCL. > Loading ../tkl.o __remq is undefined Error: Cannot get relocated section contents Fast links are on: do (si::use-fast-links nil) for debugging Error signalled by EVAL. [...] Full log at http://buildd.debian.org/fetch.php?&pkg=gclcvs&ver=2.7.0-50.1&arch=alpha&stamp=1129366266&file=log&as=raw This was already discussed at http://lists.debian.org/debian-alpha/2005/09/msg00027.html , but without a resolution. Since apparently this isn't reproducible on escher, I can offer an account on my machine if anybody wants to debug this. -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.13.2 Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#340865: FTBFS (alpha): asm-specifier conflicts with asm clobber list
Package: pistachio
Severity: important
Justification: fails to build from source
pistachio fails to build on Alpha:
[...]
cd src/generic/ && gcc-3.4 -nostdinc
-I/tmp/buildd/pistachio-0.4+20050706/build-kernel/include
-I/tmp/buildd/pistachio-0.4+20050706/kernel/include
-I/usr/lib/gcc/alpha-linux-gnu/3.4.5/include -Ualpha -Ua21164 -Umiata -Uv4
-D__ARCH__=alpha -D__CPU__=a21164 -D__PLATFORM__=miata -D__API__=v4 -imacros
/tmp/buildd/pistachio-0.4+20050706/build-kernel/config/config.h -imacros
/tmp/buildd/pistachio-0.4+20050706/kernel/include/macros.h -include
/tmp/buildd/pistachio-0.4+20050706/kernel/include/config.h -include
/tmp/buildd/pistachio-0.4+20050706/kernel/include/types.h -fno-rtti
-fno-builtin -fomit-frame-pointer -fno-exceptions -Wall -Wno-format
-Wconversion -O2 -mno-fp-regs -freg-struct-return -mcpu=ev5-c
/tmp/buildd/pistachio-0.4+20050706/kernel/src/generic/kmemory.cc
/tmp/buildd/pistachio-0.4+20050706/kernel/include/arch/alpha/palcalls.h: In
member function `void* kmem_t::alloc_aligned(word_t, word_t, word_t)':
/tmp/buildd/pistachio-0.4+20050706/kernel/include/arch/alpha/palcalls.h:108:
error: asm-specifier for variable `a0' conflicts with asm clobber list
/tmp/buildd/pistachio-0.4+20050706/kernel/include/arch/alpha/palcalls.h:108:
error: asm-specifier for variable `a1' conflicts with asm clobber list
make[2]: *** [src/generic/kmemory.o] Error 1
make[2]: Leaving directory `/tmp/buildd/pistachio-0.4+20050706/build-kernel'
make[1]: *** [all] Error 2
[...]
Full log at
http://buildd.debian.org/fetch.php?&pkg=pistachio&ver=0.4%2B20050706-2&arch=alpha&stamp=1125115586&file=log&as=raw
With newer gcc, a local "register asm" variable mustn't be clobbered.
Instead, it should get an output constraint ("=r").
-- System Information:
Debian Release: testing/unstable
APT prefers unstable
APT policy: (500, 'unstable')
Architecture: alpha
Shell: /bin/sh linked to /bin/bash
Kernel: Linux 2.6.13.2
Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15)
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#341345: gcc-4.0: Please warn on impossible && conditionals
Justin Pryzby <[EMAIL PROTECTED]> writes: > Please consider also warning on things like: > > if (0==c && 1==c) How would this be different from -Wunreachable-code? -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#346003: FTBFS (alpha): -mieee kludge in debian/rules broken
Package: gsl Severity: serious Justification: no longer builds from source gsl 1.7-3 does not build on Alpha: [...] mv libtool-2 libtool chmod 755 libtool touch configure-stamp dh_testdir /usr/bin/make CFLAGS="-g -O2 -D_REENTRANT" " -mieee" make[1]: Entering directory `/build/buildd/gsl-1.7' make[1]: *** No rule to make target ` -mieee'. Stop. make[1]: Leaving directory `/build/buildd/gsl-1.7' make: *** [build-stamp] Error 2 (http://buildd.debian.org/fetch.php?pkg=gsl&arch=alpha&ver=1.7-3&stamp=1136325772&file=log) You need this: --- debian/rules~ 2006-01-04 20:50:55.0 +0100 +++ debian/rules2006-01-04 20:52:32.0 +0100 @@ -20,13 +20,13 @@ endif # edd 12 Mar 2002 m68k barks when using -O2 -compilerflags = "-g -O2 -D_REENTRANT" +compilerflags = -g -O2 -D_REENTRANT ifeq ($(arch),m68k) -compilerflags = "-g -O0 -D_REENTRANT" +compilerflags = -g -O0 -D_REENTRANT endif # edd 29 Sep 2005 alpha needs -mieee with gcc 4.0 ifeq ($(arch),alpha) -compilerflags += " -mieee" +compilerflags += -mieee endif #export DH_VERBOSE=1 -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.13.2 Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#346011: FTBFS (alpha): #error architecture not supported
Package: widelands Version: build9half-1 Severity: important Justification: fails to build from source widelands build9half-1 fails to build on Alpha: [...] src/machdep.h:44:2: error: #error architecture not supported (http://buildd.debian.org/fetch.php?pkg=widelands&arch=alpha&ver=build9half-1&stamp=1136359129&file=log) Please don't attempt to "fix" this by adding more cases in src/machdep.h. Endianness can be checked with SDL_BYTEORDER == SDL_LIL_ENDIAN (or POSIX macros if you don't want to rely on SDL). P_ALIGNMENT should be true everywhere except i386 (even there it is not a clear win for all CPUs). It being unset for __x86_64__ is bogus, for example. So hardcode it for __i386__, or just let it be. Falk -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.13.2 Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#346003: FTBFS (alpha): -mieee kludge in debian/rules broken
Sorry, you also need $(MAKE) CFLAGS="$(compilerflags)" -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#346011: Acknowledgement (FTBFS (alpha): #error architecture not supported)
Falk Hueffner <[EMAIL PROTECTED]> writes: > So hardcode it for __i386__ actually, strike that. The memcpy should generate optimal code for this, and it does for i386, so just use it everywhere, and file a bug against gcc if it doesn't give optimal code. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#346011: Acknowledgement (FTBFS (alpha): #error architecture not supported)
Martin Quinson <[EMAIL PROTECTED]> writes:
> Could you propose a clean machdep.h file?
OK.
--
Falk
/*
* Copyright (C) 2002, 2003 by the Widelands Development Team
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version 2
* of the License, or (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
*
*/
#ifndef MACHDEP_H
#define MACHDEP_H
#include
#include "types.h"
#if SDL_BYTEORDER == SDL_LIL_ENDIAN
#define Little16(x) (x)
#define Little32(x) (x)
#define LittleFloat(x) (x)
#define Big16(x) Swap16((x))
#define Big32(x) Swap32((x))
#define BigFloat(x) SwapFloat((x))
#else
#define Little16(x) Swap16((x))
#define Little32(x) Swap32((x))
#define LittleFloat(x) SwapFloat((x))
#define Big16(x) (x)
#define Big32(x) (x)
#define BigFloat(x) (x)
#endif
inline short Swap16(short x) {
short s;
((uchar *)&s)[0] = ((uchar *)&x)[1];
((uchar *)&s)[1] = ((uchar *)&x)[0];
return s;
}
inline int Swap32(int x) {
int s;
((uchar *)&s)[0] = ((uchar *)&x)[3];
((uchar *)&s)[1] = ((uchar *)&x)[2];
((uchar *)&s)[2] = ((uchar *)&x)[1];
((uchar *)&s)[3] = ((uchar *)&x)[0];
return s;
}
inline float SwapFloat(float x)
{
float s;
((uchar *)&s)[0] = ((uchar *)&x)[3];
((uchar *)&s)[1] = ((uchar *)&x)[2];
((uchar *)&s)[2] = ((uchar *)&x)[1];
((uchar *)&s)[3] = ((uchar *)&x)[0];
return s;
}
inline char Deref8(const void* ptr)
{
return *reinterpret_cast(ptr);
}
inline short Deref16(const void* ptr)
{
short r;
memcpy(&r, ptr, sizeof(r));
return r;
}
inline int Deref32(const void* ptr)
{
int r;
memcpy(&r, ptr, sizeof(r));
return r;
}
inline float DerefFloat(const void* ptr)
{
float r;
memcpy(&r, ptr, sizeof(r));
return r;
}
#endif
Bug#346025: FTBFS (alpha): Bad build dependency on libc6-dev
Package: csync2 Version: 1.29-1 Severity: important Justification: fails to build from source csync2 fails to build on Alpha: [...] dpkg-checkbuilddeps: Unmet build dependencies: libc6-dev (>= 2.3.5) dpkg-buildpackage: Build dependencies/conflicts unsatisfied; aborting. (http://buildd.debian.org/fetch.php?pkg=csync2&arch=alpha&ver=1.29-2&stamp=1136398406&file=log) There libc package is called "libc6.1" on Alpha for historical reasons. You do not need a build dependency on libc, since it is build essential, so you can just drop the libc6-dev depencency. -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.13.2 Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#346026: FTBFS (alpha): Bad build dependency on libc6-dev
Package: cvsnt Version: 2.5.03.2151-1 Severity: important Justification: fails to build from source cvsnt fails to build on Alpha: [...] dpkg-checkbuilddeps: Unmet build dependencies: libc6-dev (>= 2.3.5) dpkg-buildpackage: Build dependencies/conflicts unsatisfied; aborting. (http://buildd.debian.org/fetch.php?pkg=cvsnt&arch=alpha&ver=2.5.03.2151-1&stamp=1136387869&file=log) There libc package is called "libc6.1" on Alpha for historical reasons. You do not need a build dependency on libc, since it is build essential, so you can just drop the libc6-dev depencency. -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.13.2 Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#334766: binutils: ld infinite loops on alpha
Hi, this makes firefox FTBFS. Linking takes about 200 minutes on an 800MHz machine, but the buildd kills it after 150 minutes without output. Maybe the time limit should be raised as a workaround? Interestingly, I couldn't reproduce this with CVS gcc and CVS mozilla. So maybe it will go away all by itself. Unfortunately, I don't really have much time to investigate it... -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#334766: binutils: ld infinite loops on alpha
Steve Langasek <[EMAIL PROTECTED]> writes: > On Thu, Jan 12, 2006 at 01:48:57PM +0100, Falk Hueffner wrote: > >> this makes firefox FTBFS. Linking takes about 200 minutes on an 800MHz >> machine, but the buildd kills it after 150 minutes without output. > >> Maybe the time limit should be raised as a workaround? > >> Interestingly, I couldn't reproduce this with CVS gcc and CVS >> mozilla. So maybe it will go away all by itself. Unfortunately, I >> don't really have much time to investigate it... > > Hmm, so it went away with gcc changes only, no binutils changes required? Yes. And a different mozilla version (maybe they changed the linking scheme?) Maybe I was also using different compiler flags... Sorry for not being more helpful at the moment... -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#347822: FTBFS: libprelude/list.h: No such file or directory
Package: prelude-nids Version: 0.8.6-1 Severity: serious Justification: no longer builds from source prelude-nids fails to build on all architectures: [...] gcc -I../ -I./include/ -Ilibpcap/ -DDETECT_PLUGIN_DIR=\"/usr/lib/prelude-nids/detects/\" -DPROTOCOL_PLUGIN_DIR=\"/usr/lib/prelude-nids/protocols/\" -DCONFIG_DIR=\"/etc/prelude-nids\" -DLOCALSTATEDIR=\"/usr/var\" -I. -I. -I.. -g -O2 -Wall -Wstrict-prototypes -Wmissing-prototypes -Wmissing-declarations -Wbad-function-cast -Wcast-qual -Wcast-align -Wnested-externs -Wunused -c `test -f 'detect-plugins.c' || echo './'`detect-plugins.c detect-plugins.c:35:30: error: libprelude/list.h: No such file or directory detect-plugins.c:37:38: error: libprelude/plugin-common.h: No such file or directory detect-plugins.c:38:42: error: libprelude/plugin-common-prv.h: No such file or directory [...] see e.g. http://buildd.debian.org/fetch.php?pkg=prelude-nids&arch=alpha&ver=0.8.6-1%2Bb1&stamp=1136620477&file=log -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.13.2 Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#347842: mkfontdir segfaults on 64-bit architectures
Package: xutils Version: 6.9.0.dfsg.1-2 Severity: normal File: /usr/X11R6/bin/mkfontdir mkfontdir segfaults in the presence of a certain file, which I have attached. This makes jnethack FTBFS. Since jnethack fails only on ia64 and alpha, this seems to be a 64-bit problem. [EMAIL PROTECTED]:/tmp/a/jnethack% ls -l total 8 -rw-r--r-- 1 falk falk 2200 Jan 13 00:00 dungeon [EMAIL PROTECTED]:/tmp/a/jnethack% ls -la total 8 drwxr-xr-x 2 falk falk 60 Jan 13 00:01 ./ drwxr-xr-x 12 falk falk 260 Jan 13 00:00 ../ -rw-r--r-- 1 falk falk 2200 Jan 13 00:00 dungeon [EMAIL PROTECTED]:/tmp/a/jnethack% md5sum dungeon 7ba98ba4593f442983740d496651b90a dungeon [EMAIL PROTECTED]:/tmp/a/jnethack% mkfontdir . zsh: segmentation fault (core dumped) mkfontdir . [EMAIL PROTECTED]:/tmp/a/jnethack% -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.13.2 Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15) Versions of packages xutils depends on: ii cpp 4:4.0.2-2 The GNU C preprocessor (cpp) ii libc6.1 2.3.5-11 GNU C Library: Shared libraries an ii libfreetype6 2.1.10-1 FreeType 2 font engine, shared lib ii libfs66.9.0.dfsg.1-2 X Font Server library ii x11-common6.9.0.dfsg.1-3 X Window System (X.Org) infrastruc ii zlib1g1:1.2.3-9 compression library - runtime xutils recommends no packages. -- no debconf information dungeon Description: Binary data
Bug#350782: FTBFS (alpha): conflicting types for 'strlen'
Package: asclassic Version: 1.1b-28.1 Severity: serious Justification: no longer builds from source asclassic fails to build on Alpha: [...] make[2]: Entering directory `/tmp/buildd/asclassic-1.1b/afterstep' gcc -O2 -g -Wall -I/usr/X11R6/include-Dlinux -D__alpha__ -D_POSIX_C_SOURCE=199309L -D_POSIX_SOURCE -D_XOPEN_SOURCE -D_BSD_SOURCE -D_SVID_SOURCE -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -DFUNCPROTO=15 -DNARROWPROTO-DDEBIAN -c -o afterstep.o afterstep.c In file included from misc.h:364, from afterstep.c:56: alpha_header.h:12: error: conflicting types for 'strlen' afterstep.c:155: warning: return type of 'main' is not 'int' afterstep.c: In function 'Reborder': afterstep.c:848: warning: unused variable 'tmp' make[2]: *** [afterstep.o] Error 1 Probably the same problem as #337870. Falk -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.13.2 Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#350793: htons() has wrong type at -O
Package: libc6.1-dev
Version: 2.3.5-12
Severity: normal
File: /usr/include/arpa/inet.h
[EMAIL PROTECTED]:/tmp% cat test.c
#include
int f() { return ntohs(1) > 1u; }
[EMAIL PROTECTED]:/tmp% gcc -c -W test.c
[EMAIL PROTECTED]:/tmp% gcc -c -W -O test.c
test.c: In function 'f':
test.c:2: warning: comparison between signed and unsigned
This is because htons is inlined to:
int f() { return (__extension__ ({ unsigned short int __bsx = (1); __bsx)
>> 8) & 0xff) | (((__bsx) & 0xff) << 8)); })) > 1u; }
and because of C's promotion rules, the ({}) expression has type
signed int. However, htons is supposed to return uint16_t.
This makes nepenthes FTBFS (it has -Werror). I tried to imagine a
situation where it might lead to wrong code, but couldn't think of
one, short of taking sizeof htons().
Falk
-- System Information:
Debian Release: testing/unstable
APT prefers unstable
APT policy: (500, 'unstable')
Architecture: alpha
Shell: /bin/sh linked to /bin/bash
Kernel: Linux 2.6.13.2
Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15)
Versions of packages libc6.1-dev depends on:
ii libc6.12.3.5-12 GNU C Library: Shared libraries an
ii linux-kernel-headers 2.6.13+0rc3-2 Linux Kernel Headers for developme
Versions of packages libc6.1-dev recommends:
ii gcc [c-compiler] 4:4.0.2-2 The GNU C compiler
ii gcc-3.3 [c-compiler] 1:3.3.6-12 The GNU C compiler
ii gcc-4.0 [c-compiler] 4.0.2-8The GNU C compiler
-- no debconf information
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#337340: sentinel attribute should support non-NULL sentinels
Josh Triplett <[EMAIL PROTECTED]> writes: > The sentinel attribute tells GCC to ensure that the argument list of a > variadic function always ends with a final NULL argument. It should be > possible to use this attribute for a variadic function where the > sentinel is some constant other than NULL, as long as the value of the > constant can be determined at compile time and supplied as a parameter > to the sentinel attribute. Does this really occur frequently enough to be worth the trouble? Can you name a few libraries that do this? -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#337617: Dollar in msgid scores 9.4 points
Package: spamassassin Version: 3.1.0a-1 Severity: normal File: /usr/bin/spamassassin Have a look at the attached mail. It is extremely hammy, except it contains $ in its Message-ID. This triggers all three of 3.5 RATWARE_OUTLOOK_NONAME Bulk email fingerprint (Outlook no name) found 2.7 RATWARE_MS_HASHBulk email fingerprint (msgid ms hash) found 3.2 MSGID_DOLLARS Message-Id has pattern used in spam No single evidence should score 9.4 points. -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.13.2 Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15) Versions of packages spamassassin depends on: ii libdigest-sha1-perl 2.10-1 NIST SHA-1 message digest algorith ii libhtml-parser-perl 3.46-1 A collection of modules that parse ii perl 5.8.7-7Larry Wall's Practical Extraction Versions of packages spamassassin recommends: ii libnet-dns-perl 0.53-2 Perform DNS queries from a Perl sc ii perl [libmime-base64-perl]5.8.7-7Larry Wall's Practical Extraction ii spamc 3.1.0a-1 Client for SpamAssassin spam filte -- debconf information: spamassassin/upgrade/2.40: spamassassin/upgrade/2.40w: spamassassin/upgrade/cancel: Continue spamassassin/upgrade/2.42m: No spamassassin/upgrade/2.42u: No testmail.gz Description: Binary data
Bug#336463: g++-4.0: FPE ICE on mipsen also affects fltk1.1
Hi, Very likely, this is the same as PR 24514, which has a patch here: http://gcc.gnu.org/ml/gcc-patches/2005-11/msg00103.html -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#300930: 300930 not fixed
# not fixed in 0.9.3-1.1, see http://buildd.debian.org/fetch.php?&pkg=sim&ver=0.9.3-1.1&arch=alpha&stamp=1128326800&file=log&as=raw tags 300930 - fixed thanks -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#337803: FTBFS (alpha): incompatible type for argument 3 of 'vsprintf'
Package: robotour Version: 3.1.1-3 Severity: important Justification: fails to build from source robotour fails to build on Alpha: [...] if gcc -DPACKAGE_NAME=\"\" -DPACKAGE_TARNAME=\"\" -DPACKAGE_VERSION=\"\" -DPACKAGE_STRING=\"\" -DPACKAGE_BUGREPORT=\"\" -DPACKAGE=\"lrt\" -DVERSION=\"0.0.0\" -DSTDC_HEADERS=1 -DHAVE_SYS_TYPES_H=1 -DHAVE_SYS_STAT_H=1 -DHAVE_STDLIB_H=1 -DHAVE_STRING_H=1 -DHAVE_MEMORY_H=1 -DHAVE_STRINGS_H=1 -DHAVE_INTTYPES_H=1 -DHAVE_STDINT_H=1 -DHAVE_UNISTD_H=1 -I. -I. -g -O2 -D__UNIX__ -MT svprintf.o -MD -MP -MF ".deps/svprintf.Tpo" \ -c -o svprintf.o `test -f 'svprintf.c' || echo './'`svprintf.c; \ then mv -f ".deps/svprintf.Tpo" ".deps/svprintf.Po"; \ else rm -f ".deps/svprintf.Tpo"; exit 1; \ fi svprintf.c: In function 'Svprintf': svprintf.c:42: error: incompatible type for argument 3 of 'vsprintf' svprintf.c:44: warning: incompatible implicit declaration of built-in function 'strlen' make[3]: *** [svprintf.o] Error 1 make[3]: Leaving directory `/tmp/robotour-3.1.1/libRT/simpsock' make[2]: *** [all-recursive] Error 1 Full log at http://buildd.debian.org/fetch.php?&pkg=robotour&ver=3.1.1-3&arch=alpha&stamp=1126039289&file=log&as=raw -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.13.2 Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#326252: rekall: FTBFS (amd64): cast from 'ELShared*' to 'int' loses precision
found 326252 2.2.6-2 reopen 326252 thanks Hi, this is still present in 2.2.6-2, see http://buildd.debian.org/fetch.php?&pkg=rekall&ver=2.2.6-2&arch=alpha&stamp=1131207039&file=log&as=raw -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#337807: FTBFS (alpha): conflicting types for 'int64_t'
Package: qemu Version: 0.7.2-1 Severity: important Justification: fails to build from source qemu fails to build on Alpha: [...] gcc-3.4 -Wall -O2 -g -fno-reorder-blocks -fno-optimize-sibling-calls -I. -I/tmp/qemu-0.7.2/./target-i386 -I/tmp/qemu-0.7.2/. -I/tmp/qemu-0.7.2/./linux-user -I/tmp/qemu-0.7.2/./linux-user/i386 -D_GNU_SOURCE -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -I/tmp/qemu-0.7.2/./fpu -I/tmp/qemu-0.7.2/./slirp -c -o op.o /tmp/qemu-0.7.2/./target-i386/op.c In file included from /usr/include/inttypes.h:28, from /tmp/qemu-0.7.2/./cpu-defs.h:25, from /tmp/qemu-0.7.2/./target-i386/exec.h:30, from /tmp/qemu-0.7.2/./target-i386/op.c:22: /usr/include/stdint.h:41: error: conflicting types for 'int64_t' /tmp/qemu-0.7.2/./dyngen-exec.h:44: error: previous declaration of 'int64_t' was here /usr/include/stdint.h:56: error: conflicting types for 'uint64_t' /tmp/qemu-0.7.2/./dyngen-exec.h:35: error: previous declaration of 'uint64_t' was here In file included from /tmp/qemu-0.7.2/./target-i386/op.c:735: /tmp/qemu-0.7.2/./target-i386/ops_template.h: In function `op_jb_subb': /tmp/qemu-0.7.2/./target-i386/ops_template.h:278: warning: implicit declaration of function `GOTO_LABEL_PARAM' /tmp/qemu-0.7.2/./target-i386/op.c: In function `op_goto_tb0': /tmp/qemu-0.7.2/./target-i386/op.c:1300: warning: cast to pointer from integer of different size /tmp/qemu-0.7.2/./target-i386/op.c:1300: warning: cast to pointer from integer of different size /tmp/qemu-0.7.2/./target-i386/op.c: In function `op_goto_tb1': /tmp/qemu-0.7.2/./target-i386/op.c:1305: warning: cast to pointer from integer of different size /tmp/qemu-0.7.2/./target-i386/op.c:1305: warning: cast to pointer from integer of different size make[2]: *** [op.o] Error 1 make[2]: Leaving directory `/tmp/qemu-0.7.2/i386-user' Full log at http://buildd.debian.org/fetch.php?&pkg=qemu&ver=0.7.2-1&arch=alpha&stamp=1130732292&file=log&as=raw -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.13.2 Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#337857: FTBFS (alpha): conflicting types for 'memcpy'
Package: libobjc-lf2 Version: 2.95.3r115-1 Severity: important Justification: fails to build from source libobjc-lf2 fails to build on Alpha: [...] gcc sarray.c -c \ -MMD -MP -pipe -DSTDC_HEADERS=1 -DHAVE_STDLIB_H -DOBJC_THREAD_BACKEND='"thr-backends/single.h"' -DOBJC_WITHOUT_THREADING -DOBJC_WITH_GC=0 -DDEBUG_OBJC_GC=0 -DLIB_FOUNDATION_LIBRARY=1 -DGNU_RUNTIME=1 -D_REENTRANT -fPIC -DGSWARN -DGSDIAGNOSE -O0 -O2 -Wall -Iconfig/alpha/linux-gnu -Iconfig/alpha/generic -I.. -I. -I. -I/home/falk/GNUstep/Library/Headers -I/usr/local/lib/opengroupware.org/Local/Library/Headers -I/usr/local/lib/opengroupware.org/Network/Library/Headers -I/usr/lib/opengroupware.org/System/Library/Headers -I/home/falk/GNUstep/Library/Headers/libFoundation -I/usr/local/lib/opengroupware.org/Local/Library/Headers/libFoundation -I/usr/local/lib/opengroupware.org/Network/Library/Headers/libFoundation -I/usr/lib/opengroupware.org/System/Library/Headers/libFoundation -I/home/falk/GNUstep/Library/Headers/libFoundation/./GNU -I/usr/local/lib/opengroupware.org/Local/Library/Headers/libFoundation/./GNU -I/usr/local/lib/opengroupware.org/Network/Library/Headers/libFoundation/./GNU -I/usr/lib/opengroupware.org/System/Library/Headers/libFoundation/./GNU \ -o shared_obj/sarray.o sarray.c:44: error: conflicting types for 'memcpy' make[2]: *** [shared_obj/sarray.o] Error 1 make[1]: *** [libobjc.all.library.variables] Error 2 make[1]: Leaving directory `/tmp/libobjc-lf2-2.95.3r115' Full log at http://buildd.debian.org/fetch.php?&pkg=libobjc-lf2&ver=2.95.3r115-1&arch=alpha&stamp=1125170315&file=log&as=raw -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.13.2 Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#337863: FTBFS (alpha): storage size of 'ht_data' isn't known
Package: gsoap Version: 2.7.6c-1 Severity: important Justification: fails to build from source gsoap fails to build on Alpha: [...] source='stdsoap2.c' object='libgsoap_a-stdsoap2.o' libtool=no \ depfile='.deps/libgsoap_a-stdsoap2.Po' tmpdepfile='.deps/libgsoap_a-stdsoap2.TPo' \ depmode=gcc3 /bin/sh ../depcomp \ alpha-linux-gnu-gcc -DHAVE_CONFIG_H -I. -I. -I.. -DLINUX -Wall -g -O2 -c -o libgsoap_a-stdsoap2.o `test -f 'stdsoap2.c' || echo './'`stdsoap2.c stdsoap2.c: In function 'tcp_gethost': stdsoap2.c:3048: error: storage size of 'ht_data' isn't known stdsoap2.c:3048: warning: unused variable 'ht_data' make[3]: *** [libgsoap_a-stdsoap2.o] Error 1 make[3]: Leaving directory `/tmp/gsoap-2.7.6c/soapcpp2' Full log at http://buildd.debian.org/fetch.php?&pkg=gsoap&ver=2.7.6c-1&arch=alpha&stamp=1127262240&file=log&as=raw -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.13.2 Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#337867: FTBFS (alpha): Bad inline assembly
Package: helix-player Version: 1.0.6-2 Severity: important Justification: fails to build from source helix-player fails to build on Alpha: [...] make[2]: Entering directory `/tmp/helix-player-1.0.6/common/system' g++-3.4 --permissive -pipe -W -O2 -I../../common/runtime/pub -I/usr/X11R6/include -Ipub/platform/unix -I../include -I../container/pub -I../dbgtool/pub -I../util/pub -I../fileio/pub -I../runtime/pub -I./pub -I. -include rel/common_system_ribodefs.h -fPIC -DPIC -o rel/obj/platform/unix/unix_dll_common.o -c platform/unix/unix_dll_common.cpp In file included from ../include/hxcom.h:808, from ../util/pub/hxstrutl.h:122, from platform/unix/unix_dll_common.cpp:56: ../include/atomicbase.h:857:19: c_asm.h: No such file or directory In file included from ../include/hxcom.h:808, from ../util/pub/hxstrutl.h:122, from platform/unix/unix_dll_common.cpp:56: ../include/atomicbase.h: In function `INT32 HXAtomicIncRetINT32(INT32*)': ../include/atomicbase.h:863: error: expected primary-expression before "asm" ../include/atomicbase.h:863: error: expected `;' before "asm" [...] make[2]: *** [rel/obj/platform/unix/unix_dll_common.o] Error 1 make[2]: Leaving directory `/tmp/helix-player-1.0.6/common/system' It tries to use DEC C inline assembly. It should not be too hard to change those to gcc inline assembly. Full log at: http://buildd.debian.org/fetch.php?&pkg=helix-player&ver=1.0.6-2&arch=alpha&stamp=1130079217&file=log&as=raw -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.13.2 Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#337868: FTBFS (alpha): operator '&&' has no right operand
Package: hs-plugins Version: 0.9.10-1 Severity: important Justification: fails to build from source hs-plugins does not build on Alpha: [...] ghc -package-name altdata -O -Wall -Werror -fno-warn-missing-signatures -threaded -c AltData/Dynamic.hs -o AltData/Dynamic.o -ohi AltData/Dynamic.hi ld -x -r -o HSaltdata.o AltData/Dynamic.o AltData/Typeable.o :1:35: error: operator '&&' has no right operand /bin/sh: /plugins.conf.inplace: Permission denied /bin/sh: src/altdata/altdata.conf.inplace.in: No such file or directory make[3]: *** [/plugins.conf.inplace] Error 1 make[3]: Leaving directory `/tmp/hs-plugins-0.9.10/src/altdata' [...] Full log at http://buildd.debian.org/fetch.php?&pkg=hs-plugins&ver=0.9.10-1&arch=alpha&stamp=1130280498&file=log&as=raw -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.13.2 Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#337870: FTBFS (alpha): conflicting types for 'strlen'
Package: fvwm1 Version: 1.24r-47 Severity: serious Justification: no longer builds from source fvwm1 fails to build on Alpha: [...] gcc -g -O2 -fno-strict-aliasing -I/usr/X11R6/include-Dlinux -D__alpha__ -D_POSIX_C_SOURCE=199309L -D_POSIX_SOURCE -D_XOPEN_SOURCE -D_BSD_SOURCE -D_SVID_SOURCE-DFUNCPROTO=15 -DNARROWPROTO -c -o fvwm.o fvwm.c In file included from misc.h:338, from fvwm.c:47: alpha_header.h:12: error: conflicting types for 'strlen' fvwm.c: In function 'main': fvwm.c:143: warning: return type of 'main' is not 'int' make[2]: *** [fvwm.o] Error 1 make[2]: Leaving directory `/tmp/fvwm1-1.24r/fvwm' [...] Full log at http://buildd.debian.org/fetch.php?&pkg=fvwm1&ver=1.24r-47&arch=alpha&stamp=1126601338&file=log&as=raw -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.13.2 Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#337871: FTBFS (alpha): invalid lvalue in assignment
Package: crash Version: 4.0-2.2-1 Severity: serious Justification: no longer builds from source crash fails to build on Alpha: [...] gcc -DHAVE_CONFIG_H -I. -I. -I. -D_GNU_SOURCE -DTRAD_CORE -I. -I. -I./../include -I./../intl -I../intl -W -Wall -Wstrict-prototypes -Wmissing-prototypes -g -O2 -c coff-alpha.c -o coff-alpha.o coff-alpha.c: In function 'alpha_adjust_reloc_in': coff-alpha.c:671: warning: comparison is always true due to limited range of data type coff-alpha.c: In function 'alpha_relocate_section': coff-alpha.c:1458: error: invalid lvalue in assignment make[6]: *** [coff-alpha.lo] Error 1 make[5]: *** [all-recursive] Error 1 make[4]: *** [all-recursive-am] Error 2 make[3]: *** [all-bfd] Error 2 make[2]: *** [gdb_merge] Error 2 make[1]: *** [all] Error 2 make[1]: Leaving directory `/tmp/crash-4.0-2.2' [...] Full log at http://buildd.debian.org/fetch.php?&pkg=crash&ver=4.0-2.2-1&arch=alpha&stamp=1124884999&file=log&as=raw -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.13.2 Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#337872: FTBFS (alpha): invalid va_list* usage
Package: drivel
Version: 2.0.2-4
Severity: serious
Justification: no longer builds from source
drivel fails to build on Alpha:
[...]
if alpha-linux-gnu-gcc -DHAVE_CONFIG_H -I. -I. -I..
-DGNOMELOCALEDIR=\""/usr/share/locale"\"
-DDRIVEL_GLADE_DIR=\"/usr/share/drivel/glade\"
-DDRIVEL_GLADE_FILE=\"drivel.glade\"
-DDRIVEL_LANGUAGES_DIR=\"/usr/share/drivel/languages\"
-DDATADIR=\""/usr/share"\" -DPREFIX=\""/usr"\" -DLIBDIR=\""/usr/lib"\"
-DG_LOG_DOMAIN=\"Drivel\" -DSYSCONFDIR=\""/usr/etc"\"
-DPIXMAPDIR=\""/usr/share/pixmaps"\" -I.. -I../src/libegg -Wall
-Wmissing-prototypes -I/usr/include -DXTHREADS -DORBIT2=1 -pthread
-I/usr/include/glib-2.0 -I/usr/lib/glib-2.0/include -I/usr/include/gtk-2.0
-I/usr/lib/gtk-2.0/include -I/usr/X11R6/include -I/usr/include/atk-1.0
-I/usr/include/pango-1.0 -I/usr/include/freetype2 -I/usr/include/gconf/2
-I/usr/include/orbit-2.0 -I/usr/include/gnome-vfs-2.0
-I/usr/lib/gnome-vfs-2.0/include -I/usr/include/bonobo-activation-2.0
-I/usr/include/libbonobo-2.0 -I/usr/include/libgnome-2.0
-I/usr/include/libgnomeui-2.0 -I/usr/include/libgnomecanvas-2.0
-I/usr/include/libart-2.0 -I/usr/include/libbonoboui-2.0 -I/usr/include/libxml2
-I/usr/include/libglade-2.0 -I/usr/include/gtksourceview-1.0
-I/usr/include/libgnomeprint-2.2 -I/usr/include/gtkspell-2.0-std=c89 -g -O2
-MT xmlrpc.o -MD -MP -MF ".deps/xmlrpc.Tpo" \
-c -o xmlrpc.o `test -f 'xmlrpc.c' || echo './'`xmlrpc.c; \
then mv -f ".deps/xmlrpc.Tpo" ".deps/xmlrpc.Po"; \
else rm -f ".deps/xmlrpc.Tpo"; exit 1; \
fi
xmlrpc.c: In function 'parse_parameter_types':
xmlrpc.c:44: error: used struct type value where scalar is required
xmlrpc.c:52: warning: pointer targets in passing argument 3 of 'xmlNewChild'
differ in signedness
[...]
make[4]: *** [xmlrpc.o] Error 1
make[4]: Leaving directory `/tmp/drivel-2.0.2/src'
[...]
The problem is:
static void
parse_parameter_types (va_list *args, xmlNodePtr node)
{
xmlNodePtr xml_param, xml_value;
gchar *value;
gint type, value_int;
gdouble value_double;
GTimeVal value_time;
g_return_if_fail (node);
g_return_if_fail (*args); // <- here
Dereferencing a va_list* makes no sense at all, I have no idea what it
is supposed to do, probably it can simply be removed.
Full log at:
http://buildd.debian.org/fetch.php?&pkg=drivel&ver=2.0.2-4&arch=alpha&stamp=1129027124&file=log&as=raw
-- System Information:
Debian Release: testing/unstable
APT prefers unstable
APT policy: (500, 'unstable')
Architecture: alpha
Shell: /bin/sh linked to /bin/bash
Kernel: Linux 2.6.13.2
Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15)
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#337867: FTBFS (alpha): Bad inline assembly
tags 337867 + patch
thanks
Hi,
here's a patch, but I haven't tested it yet.
--- common/include/atomicbase.h~2004-12-22 21:00:41.0 +0100
+++ common/include/atomicbase.h 2005-11-07 01:12:23.0 +0100
@@ -852,7 +852,7 @@
* know whether you need to retry or not.
*
*/
-#elif defined (__alpha)
+#elif defined (__alpha) && defined (__DECC)
#include
@@ -971,6 +971,101 @@
#define HXAtomicSubUINT32(p,n) HXAtomicSubRetUINT32((p),(n))
+#elif defined (__alpha) && defined (__GNUC__)
+
+/* Add n and return new value */
+inline INT32
+HXAtomicAddRetINT32(INT32* pNum, INT32 n)
+{
+INT32 tmp, ret;
+asm volatile(
+ "1: ldl_l %[ret], %[val] \n"
+ " addl%[ret], %[inc], %[tmp] \n"
+ " addl%[ret], %[inc], %[ret] \n"
+ " stl_c %[tmp], %[val] \n"
+ " beq %[tmp], 2f \n"
+ ".subsection 2 \n"
+ "2: br 1b \n"
+ ".previous"
+ : [ret] "=&r" (ret),
+ [tmp] "=&r" (tmp),
+ [val] "=m" (*pNum)
+ : [inc] "Ir" (n), "m" (*pNum));
+return ret;
+}
+
+/* Subtract n and return new value */
+inline INT32
+HXAtomicSubRetINT32(INT32* pNum, INT32 n)
+{
+INT32 tmp, ret;
+asm volatile(
+ "1: ldl_l %[ret], %[val] \n"
+ " subl%[ret], %[inc], %[tmp] \n"
+ " subl%[ret], %[inc], %[ret] \n"
+ " stl_c %[tmp], %[val] \n"
+ " beq %[tmp], 2f \n"
+ ".subsection 2 \n"
+ "2: br 1b \n"
+ ".previous"
+ : [ret] "=&r" (ret),
+ [tmp] "=&r" (tmp),
+ [val] "=m" (*pNum)
+ : [inc] "Ir" (n), "m" (*pNum));
+return ret;
+}
+
+/* Increment by 1 and return new value */
+inline INT32
+HXAtomicIncRetINT32(INT32* pNum)
+{
+return HXAtomicAddRetINT32(pNum, 1);
+}
+
+/* Decrement by 1 and return new value */
+inline INT32
+HXAtomicDecRetINT32(INT32* pNum)
+{
+return HXAtomicSubRetINT32(pNum, 1);
+}
+
+/* Increment by 1 and return new value */
+inline UINT32
+HXAtomicIncRetUINT32(UINT32* pNum)
+{
+return HXAtomicIncRetINT32((INT32*) pNum);
+}
+
+/* Decrement by 1 and return new value */
+inline UINT32
+HXAtomicDecRetUINT32(UINT32* pNum)
+{
+return HXAtomicDecRetINT32((INT32*) pNum);
+}
+
+/* Add n and return new value */
+inline UINT32
+HXAtomicAddRetUINT32(UINT32* pNum, UINT32 n)
+{
+return HXAtomicAddRetINT32((INT32*) pNum, n);
+}
+
+/* Subtract n and return new value */
+inline UINT32
+HXAtomicSubRetUINT32(UINT32* pNum, UINT32 n)
+{
+return HXAtomicSubRetINT32((INT32*) pNum, n);
+}
+
+#define HXAtomicIncINT32(p)HXAtomicIncRetINT32((p))
+#define HXAtomicDecINT32(p)HXAtomicDecRetINT32((p))
+#define HXAtomicAddINT32(p,n) HXAtomicAddRetINT32((p),(n))
+#define HXAtomicSubINT32(p,n) HXAtomicSubRetINT32((p),(n))
+
+#define HXAtomicIncUINT32(p) HXAtomicIncRetUINT32((p))
+#define HXAtomicDecUINT32(p) HXAtomicDecRetUINT32((p))
+#define HXAtomicAddUINT32(p,n) HXAtomicAddRetUINT32((p),(n))
+#define HXAtomicSubUINT32(p,n) HXAtomicSubRetUINT32((p),(n))
/***
* AIX / PowerPC (Native compiler)
--
Falk
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#337570: apparent gcc4 regression in xprint on powerpc
Drew Parsons <[EMAIL PROTECTED]> writes: > Current gcc is 4.0.2-1. Is the segfault in xprint -12 something you > might expect to occur, and would a rebuilt with gcc 4.0.2 be likely to > repair it? Not very likely, but it should definitely be done before trying anything else. Trying gcc-snapshot would also be nice. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#338487: Bug#338501: octave: dies with illegal instruction
Paul Brossier <[EMAIL PROTECTED]> writes: > right, as vorlon pointed me on irc, the trick is not to use -O0, but to > not use -msse, which is the flag producing buggy code. There is no CPU that has SSE but no CMOV; therefore, I am not convinced it is a bug in gcc that -msse enables cmov. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#130007: sample rate is always 44100 on alpha
tags 130007 + moreinfo thanks Hi, can you please provide the test file and describe how exactly to reproduce the problem? -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#147187: xdelta: Deltas generated on i386 fail to apply on alpha
found 147187 1.1.3-6.1 severity 147187 grave thanks This is still present in 1.1.3-6.1 and makes xdelta completely unusable on all 64-bit platforms, so setting to grave. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#303465: Cannot load modules on hppa (and likely also alpha)
found 303465 1.1-1 thanks Hi, this is still reproducible on Alpha with 1.1-1: [EMAIL PROTECTED]:~% lush LUSH Lisp Universal Shell (compiled on Aug 3 2005) Copyright (C) 2002 Leon Bottou, Yann LeCun, AT&T, NECI. Includes parts of TL3: Copyright (C) 1987-1999 Leon Bottou and Neuristique. Includes selected parts of SN3.2: Copyright (C) 1991-2001 AT&T Corp. This program is free software distributed under the terms of the GNU Public Licence (GPL) with ABSOLUTELY NO WARRANTY. Type `(helptool)' for details. +[/usr/share/lush/sys/stdenv.dump] *** First time user: Creating directory /home/falk/.lush *** First time user: Creating initial file /home/falk/.lush/lushrc.lsh [lushrc.lsh] ? (de harmonic (n) ((-double-) n) (let* ((z 0) (i 0)) ((-double-) z i) (while (< i n) (incr i) (incr z (/ i) = harmonic ? (dhc-make "junk" harmonic) *** Warning: Creating /home/falk/C *** Warning: Creating /home/falk/C/alphaev68-unknown-linux-gnu Preprocessing and parsing harmonic ... Generating C for harmonic ... gcc -DHAVE_CONFIG_H -DNO_DEBUG -Wall -O3 -pthread -I/usr/share/lush/include -c /home/falk/C/junk.c -o /home/falk/C/alphaev68-unknown-linux-gnu/junk.o *** module-load : dld/bfd error *** Overflow during relocation : "/home/falk/C/alphaev68-unknown-linux-gnu/junk.o" Debug toplevel [y/N] ? -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#324051: Bogus free space display on Alpha
"Martin v. Löwis" <[EMAIL PROTECTED]> writes:
It would be helpful to get the preprocessor output for posixmodule.i.
I'll include the relevant fragments from SF below; on the real build
environment, you should see the same (somewhat differently formatted).
With python 2.4_2.4.2-1 and libc6.1-dev 2.3.5-7 I get exactly the same
code, except that statvfs is now declared
extern int statvfs (__const char *__restrict __file,
struct statvfs *__restrict __buf)
__asm__ ("" "statvfs64")
__attribute__ ((__nothrow__))
__attribute__ ((__nonnull__ (1, 2)));
but that shouldn't really change anything...
--
Falk
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#329578: xserver-xorg: Xorg causes unaligned traps (Alpha arch.)
Hi,
I also get this, although only at one address:
dccproc(20112): unaligned trap at 00012001a428: 00011ff1c552 28 1
[EMAIL PROTECTED]:~# addr2line -e =Xorg-debug 000120e29738
/build/buildd/xorg-x11-6.8.2.dfsg.1/build-tree/xc-xserver-xorg-dbg/programs/Xserver/render/render.c:1307
which is somewhere around here:
[...]
buffer = (CARD8 *) (stuff + 1);
end = (CARD8 *) stuff + (client->req_len << 2);
nglyph = 0;
nlist = 0;
while (buffer + sizeof (xGlyphElt) < end)
{
elt = (xGlyphElt *) buffer;
buffer += sizeof (xGlyphElt);
if (elt->len == 0xff)
{
buffer += 4;
}
else
{
nlist++;
nglyph += elt->len;
space = size * elt->len;
if (space & 3)
space += 4 - (space & 3);
buffer += space;
}
}
[...]
probably elt is misaligned, and elt->len traps.
--
Falk
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#324051: Bogus free space display on Alpha
reassign 324051 libc6.1-dev retitle 324051 bad struct statvfs declaration on Alpha when _FILE_OFFSET_BITBITS is defined thanks As Sergey Tikhonov found out, this is really a bug in glibc's /usr/include/bits/statvfs.h. When _FILE_OFFSET_BITS is set, a different struct definition is taken, although there is really only one correct one on Alpha. Martin's test program shows this: [EMAIL PROTECTED]:/tmp% df / Filesystem 1K-blocks Used Available Use% Mounted on /dev/hda1 127383 88516 32290 74% / [EMAIL PROTECTED]:/tmp% gcc test.c && ./a.out / 32290 [EMAIL PROTECTED]:/tmp% gcc -D_FILE_OFFSET_BITS=64 test.c && ./a.out / 116483808061937 -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#335237: glpk: GLPK should produce a shared library (patch for libtool included)
Brady Hunsaker <[EMAIL PROTECTED]> writes: > Falk, I'd like to get your opinion here. The fundamental problem as I > see it is that the upstream author (Andrew Makhorin) sees GLPK as still > under significant development. > > I have been reading up on shared libraries, though I don't have > experience with them yet. As I understand it, creating a shared library > would require that the code be thread-safe. As I looked through the > source code, there is one obvious place where it is not thread-safe. > This shouldn't be too hard to correct, but there may be other places I > didn't notice. Yes, this should indeed be feasible. > More significantly, the API is not really stable, so I expect the soname > of a shared library would have to be incremented almost every release. > Some places I read suggest that it may not be appropriate to create a > shared library for 'stable' in this situation. (These API changes are > also the issue with the much earlier bug report, which I still need to > address.) Well, each library would need to have its own package name, leading to major hassles when upgrading. > On the other hand, I understand the frustration from other packages that > want to depend on GLPK and have a shared library. > > I am willing to create and update a shared library if we decide that > it's appropriate given that the API (and ABI) is likely to change > frequently. I don't know enough about Debian packaging culture to know > whether it makes sense to create a shared library in this situation. My > main concern is that as I understand it, we would need to create a new > GLPK package (with a new name) every time the API or ABI changes. > > I would also contact Andrew (upstream) about the possibility of > attempting to plan out an API in advance, though I don't know whether > he'd be receptive. > > What do you think? I think having a shared library would be higly desireable and not too hard, but without some commitment from upstream to keep the ABI stable it is not reasonable. It would be nice if you could achieve that... -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#338942: FTBFS: missing libz dependency
Package: hula Severity: serious Justification: no longer builds from source hula 0.1.0+svn379-2.1 fails to build on all autobuilders: [...] gcc -DHAVE_CONFIG_H -I. -I. -I../../.. -I../../../include -Wall -g -O2 -Wchar-subscripts -Wmissing-declarations -Wmissing-prototypes -Wnested-externs -Wpointer-arith -Wcast-align -Wsign-compare -DLINUX -MT msgaddr.lo -MD -MP -MF .deps/msgaddr.Tpo -c msgaddr.c -o msgaddr.o >/dev/null 2>&1 /bin/sh ../../../libtool --mode=link gcc -Wall -g -O2 -Wchar-subscripts -Wmissing-declarations -Wmissing-prototypes -Wnested-externs -Wpointer-arith -Wcast-align -Wsign-compare -DLINUX -o libhulamsgapi.la -rpath /usr/lib msgapi.lo msgdate.lo msgaddr.lo ../connio/libhulaconnio.la ../hulautil/libhulautil.la ../memmgr/libhulamemmgr.la ../mdb/libhulamdb.la ../xpl/libhulaxpl.la -ldl -lssl -lcrypto -ldl -lz gcc -shared .libs/msgapi.o .libs/msgdate.o .libs/msgaddr.o -Wl,--rpath -Wl,/build/buildd/hula-0.1.0+svn379/src/libs/connio/.libs -Wl,--rpath -Wl,/build/buildd/hula-0.1.0+svn379/src/libs/hulautil/.libs -Wl,--rpath -Wl,/build/buildd/hula-0.1.0+svn379/src/libs/memmgr/.libs -Wl,--rpath -Wl,/build/buildd/hula-0.1.0+svn379/src/libs/mdb/.libs -Wl,--rpath -Wl,/build/buildd/hula-0.1.0+svn379/src/libs/xpl/.libs ../connio/.libs/libhulaconnio.so ../hulautil/.libs/libhulautil.so ../memmgr/.libs/libhulamemmgr.so ../mdb/.libs/libhulamdb.so ../xpl/.libs/libhulaxpl.so -lssl -lcrypto -ldl -lz -Wl,-soname -Wl,libhulamsgapi.so.0 -o .libs/libhulamsgapi.so.0.0.0 /usr/bin/ld: cannot find -lz collect2: ld returned 1 exit status make[5]: *** [libhulamsgapi.la] Error 1 make[5]: Leaving directory `/build/buildd/hula-0.1.0+svn379/src/libs/msgapi' Looks like a dependency on libz-dev or similar is missing. Full log: http://buildd.debian.org/fetch.php?&pkg=hula&ver=0.1.0%2Bsvn379-2.1&arch=alpha&stamp=1131319048&file=log&as=raw -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.13.3 Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#338945: FTBFS (alpha): FAILED tests 278-280, 285
Package: libmath-gmp-perl
Version: 2.04-1
Severity: important
Justification: fails to build from source
libmath-gmp-perl fails to build on Alpha:
[...]
/usr/bin/make test
make[1]: Entering directory `/tmp/libmath-gmp-perl-2.04'
PERL_DL_NONLAZY=1 /usr/bin/perl "-MExtUtils::Command::MM" "-e" "test_harness(0,
'blib/lib', 'blib/arch')" t/*.t
t/gmppm# Test 278 got: "99" (t/gmppm.t at line 118 fail #278)
# Expected: "1410065407"
# t/gmppm.t line 118 is: print "# Tried '$try'\n" if !ok("$ans1",$ans); #
need "$ans1" due to bug
# Test 279 got: "999" (t/gmppm.t at line 118 fail #279)
# Expected: "1215752191"
# Test 280 got: "" (t/gmppm.t at line 118 fail #280)
# Expected: "3567587327"
# Test 285 got: "99" (t/gmppm.t at line 118 fail #285)
# Expected: "1410065407"
FAILED tests 278-280, 285
Failed 4/350 tests, 98.86% okay
Failed Test Stat Wstat Total Fail Failed List of Failed
---
t/gmppm.t3504 1.14% 278-280 285
Failed 1/1 test scripts, 0.00% okay. 4/350 subtests failed, 98.86% okay.
make[1]: *** [test_dynamic] Error 255
make[1]: Leaving directory `/tmp/libmath-gmp-perl-2.04'
Full log at:
http://buildd.debian.org/fetch.php?&pkg=libmath-gmp-perl&ver=2.04-1&arch=alpha&stamp=1131900079&file=log&as=raw
-- System Information:
Debian Release: testing/unstable
APT prefers unstable
APT policy: (500, 'unstable')
Architecture: alpha
Shell: /bin/sh linked to /bin/bash
Kernel: Linux 2.6.13.3
Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15)
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#339061: FTBFS (alpha): test suite fails: ERROR CODE 0x8b for test/drt1
Package: felix Severity: serious Justification: no longer builds from source felix fails to build on Alpha and IA64. This is the Alpha log: [...] RUNNING TESTS TRANSLATING FELIX TEST CODE TO C++ .. REGRESSION TESTS and TUTORIAL EXAMPLES PATH=bin:"$PATH" LD_LIBRARY_PATH=rtl:"$LD_LIBRARY_PATH" bin/flxg -Ilib test/drt1 .. ERROR CODE 0x8b TESTFILE -- ERROR! test/drt1 ** ** Terminating due to error make[1]: *** [all] Error 1 make[1]: Leaving directory `/tmp/felix-1.1.1' The IA64 log looks slightly different. Most likely this is a 64-bit problem, since Alpha and IA64 are exactly the two 64-bit architectures in Debian. Full logs at: http://buildd.debian.org/fetch.php?&pkg=felix&ver=1.1.1-1&arch=alpha&stamp=1131851005&file=log&as=raw http://buildd.debian.org/fetch.php?&pkg=felix&ver=1.1.1-1&arch=ia64&stamp=1131850705&file=log&as=raw -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.13.2 Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#335237: Forking glpk
Rafael Laboissiere <[EMAIL PROTECTED]> writes: > [Cc to the maintainer of glpk. Let me just point out I'm mostly only the uploader, the actual work is done by Brady Hunsaker, whom I've cc'd. > Falk: for the discussion context, see > http://lists.alioth.debian.org/pipermail/pkg-octave-devel/2005-November/000705.html > please.] > > * John W. Eaton <[EMAIL PROTECTED]> [2005-11-15 13:35]: > >> On 15-Nov-2005, Rafael Laboissiere wrote: >> >> | So, you are suggesting that we effectively fork the glpk package for our >> own >> | use. Your suggestion above is quite sensible, in particular because this >> | new package will only be useful for Octave. I will take a look at this >> when >> | time permits. >> >> It would be best if we could avoid the fork. How can we convince the >> maintainers of glpk that it would be best to generate shared >> libraries? What are their objections? Just that it is still under >> development? Hmm. Pretty much all software is under development, is >> it not? I've never heard anyone say that shared librarires are only >> for software that is no longer changing. The difference is that the *ABI* changes frequently. It doesn't for most projects. The X11 ABI hasn't changed for like 12 years, although quite a lot of development has happened. >> If the upstream author won't accept your patch, then wouldn't it be >> best for the maintainer of the Debian package of glpk to apply it >> rather than have two glpk packages for Debian? >> >> If neither will apply the patch, I don't see that we have much choice >> if we want to use glpk for Octave. > > Although it is a waste of disk space (duplicated upstream tarball, for > instance), having two packages in Debian is not that bad. This may be > the only feasible short-term solution for the octave2.9 problem. If/when > the upstream authors decide to generate the shared library, we can > withdraw the package. > [...] > > This package does not conflict with glpk. With both packages installed, > octave2.9 will find the header files (in glpk) at compile time and the > libglpk.so.0 shared library (in libglpk0) at link time. This will break horribly if the headers become out of sync with the library. You absolutely need to provide matching headers. > What do you think? Should we go ahead with this? I'm not really convinced. The problem is that you'll need a new package name each time you upgrade, and this will leave lots of stale library packages behind and generally be a hassle. If the goal is only to provide a solution for octave, it would probably be easier to just add a copy of glpk's source into the octave tarball and install it as glpk-octave or something. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#339873: FTBFS (alpha): hard-coded use of gcc-3.3 without build dependency
Package: linux-wlan-ng Version: 0.2.2+dfsg-6 Severity: serious Justification: no longer builds from source linux-wlan-ng fails to build on Alpha: [...] make[4]: Entering directory `/tmp/buildd/linux-wlan-ng-0.2.2+dfsg/debian/modules-build-alpha-smp/modules/linux-wlan-ng-source/src/mkmeta' gcc-3.3 -E -M -I../include -I/usr/src/linux-headers-2.6.14-1-alpha-smp/include -D__LINUX_WLAN__ ../shared/p80211types.c ../shared/p80211metamsg.c ../shared/p80211metamib.c ../shared/p80211meta.c mkmetadef.c ../shared/p80211types.c ../shared/p80211metamsg.c ../shared/p80211metamib.c ../shared/p80211meta.c mkmetastruct.c > .depend /bin/sh: gcc-3.3: command not found make[4]: *** [.depend] Error 127 make[4]: Leaving directory `/tmp/buildd/linux-wlan-ng-0.2.2+dfsg/debian/modules-build-alpha-smp/modules/linux-wlan-ng-source/src/mkmeta' make[3]: *** [all] Error 2 This is because gcc-3.3 is hardcoded in debian/rules without a build dependency. Full log at: http://buildd.debian.org/fetch.php?&pkg=linux-wlan-ng&ver=0.2.2%2Bdfsg-6&arch=alpha&stamp=1131922170&file=log&as=raw -- System Information: Debian Release: testing/unstable APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.13.2 Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15) -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#324046: Crashes when I click on any directory in the left panel
tags 324046 + patch
thanks
Jérôme COUDERC <[EMAIL PROTECTED]> writes:
> Could you try is the attached patch have also an effect for this
> problem?
No, it doesn't. This patch does fix it.
--
Falk
diff -Nurp easytag-1.99.9.orig/src/msgbox.c easytag-1.99.9/src/msgbox.c
--- easytag-1.99.9.orig/src/msgbox.c2005-11-19 18:50:02.0 +0100
+++ easytag-1.99.9/src/msgbox.c 2005-11-19 18:50:48.0 +0100
@@ -44,9 +44,9 @@ voidcheck_button_toggled (GtkChe
static GtkDialogClass *parent_klass = NULL;
-guint msg_box_get_type(void)
+GType msg_box_get_type(void)
{
-static guint mb_type = 0;
+static GType mb_type = 0;
if (!mb_type)
{
diff -Nurp easytag-1.99.9.orig/src/msgbox.h easytag-1.99.9/src/msgbox.h
--- easytag-1.99.9.orig/src/msgbox.h2005-11-19 18:50:02.0 +0100
+++ easytag-1.99.9/src/msgbox.h 2005-11-19 18:50:34.0 +0100
@@ -85,7 +85,7 @@ enum Message_Type
};
-guint msg_box_get_type(void);
+GType msg_box_get_type(void);
GtkWidget *msg_box_new (gchar *title, gchar *message, const gchar *icon, ...);
gint msg_box_run (MsgBox *msgbox);
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#344503: Icon build failure on arm due to gcc-4.0 regression
Jurij Smakov <[EMAIL PROTECTED]> writes: > The latest upload of icon (9.4.2-2.4) has failed to build on arm with > the current gcc-4.0 in sid (4.0.2-5). After some investigation I've > found that this is gcc-4.0 fault, and the build failures start with > the version 4.0.1-7 (4.0.1-6 builds the same source fine). Also, > 4.0.1-7 works fine if I change the optimization level from -O2 to -O0. What happens with -fno-strict-aliasing? There are zillions of warnings about aliasing problems. > Please let me know if you need more details and I'll try to dig into > it. It would be nice to have a stand-alone testcase. Can you try to produce one? -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#337871: FTBFS (alpha): invalid lvalue in assignment
Troy Heber <[EMAIL PROTECTED]> writes: > The problem is there is no "good" way to patch it at the debian > level, because the gdb included in crash is shipped as a gzip > tarball that gets unpacked at build time. It's much easier to submit > the patch upstream and get a fixed versions there, especially > because upstream is so responsive. Uhm, wouldn't it actually make *much* more sense to fix the packaging instead? Not being able to patch the package properly is a nuisance, it deterred me from debugging the package myself and is an obstacle should security updates be needed quickly. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#309677: Dumps core on Alpha
Package: asc Version: 1.16.0.0-1 Severity: grave When I start asc, I get a screen with some aircrafts for a second or so, and then: [EMAIL PROTECTED]:~% asc open /dev/sequencer: No such device zsh: abort (core dumped) asc -- System Information: Debian Release: 3.1 APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.11.3 Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15) Versions of packages asc depends on: ii aalib11.4p5-22 ascii art library ii asc-data 1.16.0.0-1 data files for the Advanced Strate ii libartsc0 1.3.2-3aRts Sound system C support librar ii libasound21.0.8-3ALSA library ii libaudio2 1.7-2 The Network Audio System (NAS). (s ii libaudiofile0 0.2.6-6Open-source version of SGI's audio ii libc6.1 2.3.2.ds1-22 GNU C Library: Shared libraries an ii libesd0 0.2.35-2 Enlightened Sound Daemon - Shared ii libexpat1 1.95.8-3 XML parsing C library - runtime li ii libfreetype6 2.1.7-2.4 FreeType 2 font engine, shared lib ii libgcc1 1:4.0-0pre9GCC support library ii libglib2.0-0 2.6.4-1The GLib library of C routines ii libjpeg62 6b-10 The Independent JPEG Group's JPEG ii libncurses5 5.4-4 Shared libraries for terminal hand ii libogg0 1.1.2-1Ogg Bitstream Library ii libparagui1.0 1.0.4-8a real crossplatform widget set ba ii libphysfs-1.0 1.0.0-4filesystem abstraction library for ii libpng12-01.2.8rel-1 PNG library - runtime ii libsdl-image1 1.2.4-1image loading library for Simple D ii libsdl-mixer1 1.2.6-1mixer library for Simple DirectMed ii libsdl1.2debi 1.2.7+1.2.8cvs20041007-4.1 Simple DirectMedia Layer ii libsigc++-1.2 1.2.5-4type-safe Signal Framework for C++ ii libsmpeg0 0.4.5+cvs20030824-1SDL MPEG Player Library - shared l ii libstdc++51:3.3.6-5 The GNU Standard C++ Library v3 ii libtiff4 3.7.2-2Tag Image File Format (TIFF) libra ii libvorbis0a 1.1.0-1The Vorbis General Audio Compressi ii libvorbisfile 1.1.0-1The Vorbis General Audio Compressi ii libx11-6 4.3.0.dfsg.1-13X Window System protocol client li ii libxext6 4.3.0.dfsg.1-13X Window System miscellaneous exte ii libxt64.3.0.dfsg.1-13X Toolkit Intrinsics ii slang11.4.9dbs-8 The S-Lang programming library - r ii xlibs 4.3.0.dfsg.1-13X Keyboard Extension (XKB) configu ii zlib1g1:1.2.2-4 compression library - runtime -- no debconf information -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#309697: Bad color rendering on cerain PDF
Package: evince Version: 0.3.0-2 Severity: normal File: /usr/bin/evince For http://www.ece.cmu.edu/%7espiral/papers/si-spiral.pdf, on page 36 colors are rendered wrong, for example text is colored red and blue. xpdf and gv get it right. Falk -- System Information: Debian Release: 3.1 APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.11.3 Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15) Versions of packages evince depends on: ii gconf2 2.8.1-6 GNOME configuration database syste ii libart-2.0-2 2.3.17-1Library of functions for 2D graphi ii libatk1.0-0 1.8.0-4 The ATK accessibility toolkit ii libaudiofile00.2.6-6 Open-source version of SGI's audio ii libbonobo2-0 2.8.1-2 Bonobo CORBA interfaces library ii libbonoboui2-0 2.8.1-2 The Bonobo UI library ii libc6.1 2.3.2.ds1-22GNU C Library: Shared libraries an ii libcairo10.4.0-1 Multi-platform 2D graphics library ii libdjvulibre13.5.14-5Runtime support for the DjVu image ii libesd0 0.2.35-2Enlightened Sound Daemon - Shared ii libfontconfig1 2.3.2-1 generic font configuration library ii libfreetype6 2.1.7-2.4 FreeType 2 font engine, shared lib ii libgcc1 1:4.0-0pre9 GCC support library ii libgconf2-4 2.8.1-6 GNOME configuration database syste ii libgcrypt11 1.2.0-11LGPL Crypto library - runtime libr ii libglade2-0 1:2.4.2-2 library to load .glade files at ru ii libglib2.0-0 2.6.4-1 The GLib library of C routines ii libgnome-keyring00.4.2-1 GNOME keyring services library ii libgnome2-0 2.8.1-2 The GNOME 2 library - runtime file ii libgnomecanvas2-02.8.0-1 A powerful object-oriented display ii libgnomeprint2.2-0 2.8.2-1 The GNOME 2.2 print architecture - ii libgnomeprintui2.2-0 2.8.2-2 GNOME 2.2 print architecture User ii libgnomeui-0 2.8.1-3 The GNOME 2 libraries (User Interf ii libgnomevfs2-0 2.8.4-4 The GNOME virtual file-system libr ii libgnutls11 1.0.16-13 GNU TLS library - runtime library ii libgpg-error01.0-1 library for common error values an ii libgtk2.0-0 2.6.4-3 The GTK+ graphical user interface ii libice6 4.3.0.dfsg.1-13 Inter-Client Exchange library ii libjpeg626b-10 The Independent JPEG Group's JPEG ii libkpathsea3 2.0.2-30path search library for teTeX (run ii liborbit21:2.12.2-1 libraries for ORBit2 - a CORBA ORB ii libpango1.0-01.8.1-1 Layout and rendering of internatio ii libpixman1 0.1.4-1 Cairo pixel manipulation library ii libpng12-0 1.2.8rel-1 PNG library - runtime ii libpoppler0 0.3.1-1 PDF rendering library ii libpoppler0-glib 0.3.1-1 PDF rendering library (GLib-based ii libpopt0 1.7-5 lib for parsing cmdline parameters ii libsm6 4.3.0.dfsg.1-13 X Window System Session Management ii libstdc++5 1:3.3.6-5 The GNU Standard C++ Library v3 ii libtasn1-2 0.2.10-4Manage ASN.1 structures (runtime) ii libx11-6 4.3.0.dfsg.1-13 X Window System protocol client li ii libxml2 2.6.16-7GNOME XML library ii libxrender1 1:0.8.3-1 X Rendering Extension client libra ii xlibs4.3.0.dfsg.1-13 X Keyboard Extension (XKB) configu ii zlib1g 1:1.2.2-4 compression library - runtime -- no debconf information -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#309802: wnpp-alert tries to write to home directory
Package: devscripts Version: 2.8.14 Severity: normal File: /usr/bin/wnpp-alert wnpp-alert tries to write to ~/.devscripts_cache and fails if that isn't writeable. Firstly, I don't really want scripts to clobber my home directory with cruft like this; secondly, this fails if the user is "nobody", as when running from a cron job. -- System Information: Debian Release: 3.1 APT prefers unstable APT policy: (500, 'unstable') Architecture: alpha Shell: /bin/sh linked to /bin/bash Kernel: Linux 2.6.11.3 Locale: LANG=C, [EMAIL PROTECTED] (charmap=ISO-8859-15) Versions of packages devscripts depends on: ii debianutils 2.13.2 Miscellaneous utilities specific t ii dpkg-dev1.10.27 Package building tools for Debian ii libc6.1 2.3.2.ds1-22 GNU C Library: Shared libraries an ii perl5.8.4-8 Larry Wall's Practical Extraction ii sed 4.1.4-2 The GNU sed stream editor -- no debconf information -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#309986: gcc-4.0 build problems on powerpc
Roger Leigh <[EMAIL PROTECTED]> writes: > I found quite a number of errors, many minor, some serious, building > 4.0.0-7, in both the gcc build and while packaging. I've put all > the detail in bug #309986. You mention warnings about "returning pointer to automatic variable" and "expr is called with too many arguments". I cannot find them in the script, can you point them out? The other warnings seem harmless. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#309986: gcc-4.0 build problems on powerpc
Roger Leigh <[EMAIL PROTECTED]> writes: > ./libtool: line 4696: /usr/bin/expr: Argument list too long > using piecewise archive linking... > > This is while creating libgcj.a, so there may be some object files > missed out of the ar command. It's happening inside libtool. The piecewise linking should take care of that. > /home/rleigh/gcc/gcc-4.0-4.0.0/src/libobjc/thr.c:117: warning: 'noreturn' > function does return > > Remove the noreturn attribute?? Rather, add one to objc_thread_exit. Not a real problem. > ../../../src/boehm-gc/mark_rts.c: In function 'GC_approx_sp': > ../../../src/boehm-gc/mark_rts.c:379: warning: function returns address of > local variable > > That's never a good idea. Not a problem either, this is just done to get the value of the stack pointer. > And there's a shlibs problem in a few packages, for example > > dh_shlibdeps -pgnat-4.0 -L libgnat-4.0 -l debian/libgnat-4.0/usr/lib > dpkg-shlibdeps: warning: could not find path for ld.so.1 > dpkg-shlibdeps: warning: could not find any packages for (ld.so.1) > dpkg-shlibdeps: warning: unable to find dependency information for shared > library ld (soname 1, path , dependency field Depends) > > $ cat /var/lib/dpkg/info/libc6.shlibs > ld 1 libc6 (>= 2.3.2.ds1-21) > > Presumably, there's a binary linked incorrectly or a problem in the > rules or the control file causing this. Maybe something's assuming > something i386-specific, like ld-linux.so.2, but I'm afraid I'm not at > all familiar with the GCC packaging. I don't know what to make of this either. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#262717: Bug #262717: gcc-2.95: checking for va_list assignment copy... configure: error: no
tags 262717 + fixed-upstream wontfix retitle 262717 [fixed in 3.0] gcc-2.95: checking for va_list assignment copy... configure: error: no thanks Hi, according to the reporter, this is fixed in 3.0, and we won't touch 2.95 anymore. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#253077: gcc-2.95: internal compiler error while compiling todays asterisk cvs head
tags 253077 +unreproducible Hi, without preprocessed source (as obtained by adding -save-temps), we cannot do anything about this. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#273442: Bug #273442: ICE building Linux 2.6.9-rc2-mm3 fs/cachefs/inode.c
tags 273442 + unreproducible thanks Hi, we cannot do anything about this without the preprocessed source, as obtained by adding -save-temps to the options. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#270620: Bug #270620: o32 ABI breakage in gcc 3.3/3.4
reassign 270620 gcc-3.4 tags 270620 + fixed-upstream retitle 270620 [fixed in 4.0] o32 ABI breakage in gcc 3.3/3.4 thanks Hi, the patch is now in gcc 4.0. I'm not sure whether we should try to backport it. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#217966: Bug #217966: lapack: Fails selftests and builds useless libs (it claims)
Hi, can the lapack maintainers say anything about the current state of this bug? Is this still a problem with 3.4 or 4.0? -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#270795: Bug #270795: monster-masher broken in fr_BE locale, Segfault at start
reassign 270795 g++-3.3 retitle 270795 [fixed in 3.4] monster-masher broken in fr_BE locale, Segfault at start tags 270795 fixed-upstream severity 270795 normal forwarded 270795 http://gcc.gnu.org/PR16011 thanks Hi, according to http://gcc.gnu.org/PR16011, this is fixed for 3.4. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#251149: Bug #251149: gcc wrapper for sparc is chronically broken
Hi, this bug has been open for quite some time as "important". Can some sparc people please comment on it? -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#283042: Bug #283042: smilutils_0.3.0-5(hppa/unstable): FTBFS: bad hppa assembly
reassign 283042 g++-3.3 tags 283042 + unreproducible thanks Hi, we cannot do anything about this without the preprocessed source, as obtained by adding -save-temps to the g++ options. -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#297112: Bug #297112: g77 compiles program that segfaults before executing the Fortran code
tags 297112 + unreproducible reassign 297112 g77-2.95 thanks Hi, we need the source to reproduce this. Also, can you please this with g77-3.3 or g77-3.4? -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#162690: Bug #162690: [parisc]gcc 3.x produces openssl test code that SEGVs with -march=2.0 and optimization >0
Hi, this is an ancient bug report about a miscompilation of openssl on hppa. Can somebody please check whether this is still an issue? -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
Bug#233891: [hppa] maxima/acl2 failures at -O1/-O2, -O0 ok
Hi, it seems that acl2 now builds on hppa. Can we close this bug? -- Falk -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
