bug#9568: Bug#642291: FTBFS on powerpc: misc/seq-long-double test fails,
From:
Michael Stone
Subject:
bug#9568: Bug#642291: FTBFS on powerpc: misc/seq-long-double test fails, should be skipped
Date:
Wed, 21 Sep 2011 07:30:47 -0400
User-agent:
Mutt/1.5.20 (2009-06-14)
On Wed, Sep 21, 2011 at 10:18:47AM +0200, Michel Dänzer wrote:
The test fails because it doesn't pass $CFLAGS to the compiler. The attached
patch fixes this, so the test is skipped as expected.
The point wasn't to skip the test, it was to fix the bug. :-)
Unfortunately, the CFLAGS change doesn't do that after all. If anything,
it's worse:
./seq 9223372036854775807 9223372036854775808 | head
9223372036854775808
9223372036854775808
9223372036854775808
9223372036854775808
9223372036854775808
9223372036854775808
9223372036854775808
9223372036854775808
9223372036854775808
9223372036854775808
Before the CFLAGS change (add -mlong-double-64) it just did this:
whereas with 8.5 it did this:
seq 9223372036854775807 9223372036854775808
9223372036854775807
9223372036854775808
(running on pescetti.d.o/powerpc)
Mike Stone