Bug#595496: FTBFS: operation on 'src32' may be undefined
Kurt Roeckx dixit:
>The libelf part probably shouldn't be a problem since that's
>mostly generic code as far as I know. But I guess the testsuite
>shows alot of warnings.
Ouch, I just see I have DEB_BUILD_OPTIONS=nocheck in the environment.
So, no comment on that.
>It normally lives somewhere in /usr/include/bits/byteswap.h,
Yeah, that one used the ({ extension though. I still suggest you
apply that patch because using ++ inside of arguments of things
that *might* be macros is generally frowned upon…
bye,
//mirabilos
--
I believe no one can invent an algorithm. One just happens to hit upon it
when God enlightens him. Or only God invents algorithms, we merely copy them.
If you don't believe in God, just consider God as Nature if you won't deny
existence. -- Coywolf Qi Hunt
--
To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#595496: FTBFS: operation on 'src32' may be undefined
On Sat, Sep 04, 2010 at 06:17:53PM +, Thorsten Glaser wrote: > Kurt Roeckx dixit: > > >Debian). As far as I know, it never even built properly on m68k. > >I've added some minimal m68k support, but the last build on m68k > >still failed to pass the regression tests. > > Works for me now. The libelf part probably shouldn't be a problem since that's mostly generic code as far as I know. But I guess the testsuite shows alot of warnings. > >I have to wonder why you get that error message in the first > >place. bswap_32 is normally implemented as a macro. So I'd > > Yes, I even fail to find the non-extension macro definition... > nevertheless, with this small patch it works for me. It normally lives somewhere in /usr/include/bits/byteswap.h, atleast on my system. But I suggest you get that fixed because other callers of that "function" might expect the same behaviour and get broken by it. Using gcc -E might help you find the location of the macro. > >I also wonder why you need elfutils to build a kernel. As far > > apt-get build-dep linux-2.6 #requires it. Didn't know that yet. Kurt -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#595496: FTBFS: operation on 'src32' may be undefined
Kurt Roeckx dixit: >Debian). As far as I know, it never even built properly on m68k. >I've added some minimal m68k support, but the last build on m68k >still failed to pass the regression tests. Works for me now. >I have to wonder why you get that error message in the first >place. bswap_32 is normally implemented as a macro. So I'd Yes, I even fail to find the non-extension macro definition… nevertheless, with this small patch it works for me. >I also wonder why you need elfutils to build a kernel. As far apt-get build-dep linux-2.6 #requires it. bye, //mirabilos -- I believe no one can invent an algorithm. One just happens to hit upon it when God enlightens him. Or only God invents algorithms, we merely copy them. If you don't believe in God, just consider God as Nature if you won't deny existence. -- Coywolf Qi Hunt -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#595496: FTBFS: operation on 'src32' may be undefined
severity 595496 important thanks On Sat, Sep 04, 2010 at 02:25:51PM +, Thorsten Glaser wrote: > Source: elfutils > Version: 0.148-1 > Severity: serious > Tags: patch > Justification: FTBFS > > I need elfutils to build the Linux kernel on m68k, which FTBFS. > The attached patch fixes this issue for me, please apply and > reupload. Since m68k is not part of Debian anymore, it's not even in unstable anymore, this obviously can't be a serious bug (for Debian). As far as I know, it never even built properly on m68k. I've added some minimal m68k support, but the last build on m68k still failed to pass the regression tests. I have to wonder why you get that error message in the first place. bswap_32 is normally implemented as a macro. So I'd have to guess that the macro you're using is not doing the right thing and first copy it to a temp variable like other implementations do. It should be doing this, since from a user point of view it should react like a normal function call. I suggest you try and fix that instead. I also wonder why you need elfutils to build a kernel. As far as I know all the tools in elfutils are also part of binutils, and you probably want to use those instead. Kurt -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
Bug#595496: FTBFS: operation on 'src32' may be undefined
Source: elfutils
Version: 0.148-1
Severity: serious
Tags: patch
Justification: FTBFS
I need elfutils to build the Linux kernel on m68k, which FTBFS.
The attached patch fixes this issue for me, please apply and
reupload.
-- System Information:
Debian Release: squeeze/sid
APT prefers etch-m68k
APT policy: (500, 'etch-m68k')
Architecture: m68k
Kernel: Linux 2.6.26-1-atari
Locale: LANG=C, LC_CTYPE=en_GB.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/mksh
diff -Nru elfutils-0.148/debian/changelog elfutils-0.148/debian/changelog
--- elfutils-0.148/debian/changelog 2010-07-03 13:18:23.0 +
+++ elfutils-0.148/debian/changelog 2010-09-04 12:40:38.0 +
@@ -1,3 +1,11 @@
+elfutils (0.148-1+m68k) unreleased; urgency=low
+
+ * Fix compilation error:
+- gnuhash_xlate.h: In function ‘elf_cvt_gnuhash’:
+ gnuhash_xlate.h:92: error: operation on ‘src32’ may be undefined
+
+ -- Thorsten Glaser Sat, 04 Sep 2010 12:39:07 +
+
elfutils (0.148-1) unstable; urgency=low
* New upstream release
diff -Nru elfutils-0.148/debian/patches/debian-changes-0.148-1+m68k elfutils-0.148/debian/patches/debian-changes-0.148-1+m68k
--- elfutils-0.148/debian/patches/debian-changes-0.148-1+m68k 1970-01-01 00:00:00.0 +
+++ elfutils-0.148/debian/patches/debian-changes-0.148-1+m68k 2010-09-04 12:42:33.0 +
@@ -0,0 +1,38 @@
+Description: Upstream changes introduced in version 0.148-1+m68k
+ This patch has been created by dpkg-source during the package build.
+ Here's the last changelog entry, hopefully it gives details on why
+ those changes were made:
+ .
+ elfutils (0.148-1+m68k) unreleased; urgency=low
+ .
+ * Fix compilation error:
+ - gnuhash_xlate.h: In function ‘elf_cvt_gnuhash’:
+ gnuhash_xlate.h:92: error: operation on ‘src32’ may be undefined
+ .
+ The person named in the Author field signed this changelog entry.
+Author: Thorsten Glaser
+
+---
+The information above should follow the Patch Tagging Guidelines, please
+checkout http://dep.debian.net/deps/dep3/ to learn about the format. Here
+are templates for supplementary fields that you might want to add:
+
+Origin: ,
+Bug:
+Bug-Debian: http://bugs.debian.org/
+Forwarded:
+Reviewed-By:
+Last-Update:
+
+--- elfutils-0.148.orig/libelf/gnuhash_xlate.h
elfutils-0.148/libelf/gnuhash_xlate.h
+@@ -89,7 +89,8 @@ elf_cvt_gnuhash (void *dest, const void
+ dest32 = (Elf32_Word *) &dest64[bitmask_words];
+ while (len >= 4)
+ {
+- *dest32++ = bswap_32 (*src32++);
++ *dest32++ = bswap_32 (*src32);
++ ++src32;
+ len -= 4;
+ }
+ }
diff -Nru elfutils-0.148/debian/patches/series elfutils-0.148/debian/patches/series
--- elfutils-0.148/debian/patches/series 2010-07-03 13:03:05.0 +
+++ elfutils-0.148/debian/patches/series 2010-09-04 12:42:33.0 +
@@ -7,3 +7,4 @@
do-autoreconf.diff
testsuite-ignore-elflint.diff
elf_additions.diff
+debian-changes-0.148-1+m68k
