Bug#958091: wtdbg2: Baseline violation on amd64/i386 and FTBFS everywhere else

2020-04-18 Thread Michael Crusoe 
This looks like something I can fix for all architectures and still respect
the baselines with SIMDe, yes.

On Sat, Apr 18, 2020 at 6:26 PM Andreas Tille  wrote:

> Hi Adrian,
>
> recently Michael Crusoe injected simde which should solve the SSE issue
> in a more elegant way.  Michael, would you comment on this whether we
> should apply this patch or rather use simde here?
>
> Kind regards
>
>   Andreas.
>
> On Sat, Apr 18, 2020 at 02:54:12PM +0300, Adrian Bunk wrote:
> > On Sat, Apr 18, 2020 at 02:44:54PM +0300, Adrian Bunk wrote:
> > > Source: wtdbg2
> > > Version: 2.5-1
> > > Severity: serious
> > > Tags: ftbfs patch
> > >
> > > https://buildd.debian.org/status/package.php?p=wtdbg2
> > >
> > > ...
> > > gcc: error: unrecognized command line option ‘-mpopcnt’
> > > gcc: error: unrecognized command line option ‘-msse4.2’
> > > make[1]: *** [Makefile:27: kbm2] Error 1
> > >
> > >
> > > Fix attached.
> >
> > Additionally the package should become
> >   Architecture: any-amd64
> > and the i386 package removed.
> >
> > It uses SSE (but not SSE4.2) unconditionally, so cannot be built
> > on i386 without baseline violation and in any case not anywhere else.
> >
> > cu
> > Adrian
> >
> > ___
> > Debian-med-packaging mailing list
> > debian-med-packag...@alioth-lists.debian.net
> >
> https://alioth-lists.debian.net/cgi-bin/mailman/listinfo/debian-med-packaging
>
> --
> http://fam-tille.de
>


-- 
Michael R. Crusoe

Bug#958091: wtdbg2: Baseline violation on amd64/i386 and FTBFS everywhere else

2020-04-18 Thread Andreas Tille 
Hi Adrian,

recently Michael Crusoe injected simde which should solve the SSE issue
in a more elegant way.  Michael, would you comment on this whether we
should apply this patch or rather use simde here?

Kind regards

  Andreas.

On Sat, Apr 18, 2020 at 02:54:12PM +0300, Adrian Bunk wrote:
> On Sat, Apr 18, 2020 at 02:44:54PM +0300, Adrian Bunk wrote:
> > Source: wtdbg2
> > Version: 2.5-1
> > Severity: serious
> > Tags: ftbfs patch
> > 
> > https://buildd.debian.org/status/package.php?p=wtdbg2
> > 
> > ...
> > gcc: error: unrecognized command line option ‘-mpopcnt’
> > gcc: error: unrecognized command line option ‘-msse4.2’
> > make[1]: *** [Makefile:27: kbm2] Error 1
> > 
> > 
> > Fix attached.
> 
> Additionally the package should become
>   Architecture: any-amd64
> and the i386 package removed.
> 
> It uses SSE (but not SSE4.2) unconditionally, so cannot be built
> on i386 without baseline violation and in any case not anywhere else.
> 
> cu
> Adrian
> 
> ___
> Debian-med-packaging mailing list
> debian-med-packag...@alioth-lists.debian.net
> https://alioth-lists.debian.net/cgi-bin/mailman/listinfo/debian-med-packaging

-- 
http://fam-tille.de

Bug#958091: wtdbg2: Baseline violation on amd64/i386 and FTBFS everywhere else

2020-04-18 Thread Adrian Bunk 
On Sat, Apr 18, 2020 at 02:44:54PM +0300, Adrian Bunk wrote:
> Source: wtdbg2
> Version: 2.5-1
> Severity: serious
> Tags: ftbfs patch
> 
> https://buildd.debian.org/status/package.php?p=wtdbg2
> 
> ...
> gcc: error: unrecognized command line option ‘-mpopcnt’
> gcc: error: unrecognized command line option ‘-msse4.2’
> make[1]: *** [Makefile:27: kbm2] Error 1
> 
> 
> Fix attached.

Additionally the package should become
  Architecture: any-amd64
and the i386 package removed.

It uses SSE (but not SSE4.2) unconditionally, so cannot be built
on i386 without baseline violation and in any case not anywhere else.

cu
Adrian

Bug#958091: wtdbg2: Baseline violation on amd64/i386 and FTBFS everywhere else

2020-04-18 Thread Adrian Bunk 
Source: wtdbg2
Version: 2.5-1
Severity: serious
Tags: ftbfs patch

https://buildd.debian.org/status/package.php?p=wtdbg2

...
gcc: error: unrecognized command line option ‘-mpopcnt’
gcc: error: unrecognized command line option ‘-msse4.2’
make[1]: *** [Makefile:27: kbm2] Error 1


Fix attached.
Description: Don't build with -mpopcnt -msse4.2
 This is a baseline violation on amd64 and i386,
 and causes FTBFS on all other release architectures.
Author: Adrian Bunk 

--- wtdbg2-2.5.orig/Makefile
+++ wtdbg2-2.5/Makefile
@@ -11,9 +11,9 @@ TIMESTAMP=$(shell date)
 endif
 
 ifeq (1, ${DEBUG})
-CFLAGS+=-g3 -W -Wall -Wno-unused-but-set-variable -O0 -DDEBUG=1 
-DVERSION="$(VERSION)" -DRELEASE="$(RELEASE)" -D_FILE_OFFSET_BITS=64 
-D_GNU_SOURCE -mpopcnt -msse4.2
+CFLAGS+=-g3 -W -Wall -Wno-unused-but-set-variable -O0 -DDEBUG=1 
-DVERSION="$(VERSION)" -DRELEASE="$(RELEASE)" -D_FILE_OFFSET_BITS=64 
-D_GNU_SOURCE
 else
-CFLAGS+=-g3 -W -Wall -Wno-unused-but-set-variable -O4 -DVERSION="$(VERSION)" 
-DRELEASE="$(RELEASE)" -D_FILE_OFFSET_BITS=64 -D_GNU_SOURCE -mpopcnt -msse4.2
+CFLAGS+=-g3 -W -Wall -Wno-unused-but-set-variable -O4 -DVERSION="$(VERSION)" 
-DRELEASE="$(RELEASE)" -D_FILE_OFFSET_BITS=64 -D_GNU_SOURCE
 endif
 
 GLIBS=-lm -lrt -lpthread -lz $(LDFLAGS)