Re: [Qemu-devel] [PATCH 00/52] 680x0 instructions emulation

[Andreas Schwab <schwab@linux-m68k.org>]

Andreas Schwab <schwab@linux-m68k.org> writes:

> Here is a testcase:
>
> when compiling tlink.c from the gcc sources with gcc-6 r234449 (which
> was bootstrapped in aranym):
>
> $ g++ -fno-PIE -c   -fomit-frame-pointer -O2 -DIN_GCC     -fno-exceptions -fno-rtti -fasynchronous-unwind-tables -W -Wall -Wno-narrowing -Wwrite-strings -Wcast-qual -Wmissing-format-attribute -Woverloaded-virtual -pedantic -Wno-long-long -Wno-variadic-macros -Wno-overlength-strings -Werror   -DHAVE_CONFIG_H -I. -I. -I../../gcc -I../../gcc/. -I../../gcc/../include -I../../gcc/../libcpp/include  -I../../gcc/../libdecnumber -I../../gcc/../libdecnumber/dpd -I../libdecnumber -I../../gcc/../libbacktrace   -o tlink.o ../../gcc/tlink.c
>
> I get this difference for the generated assembler:

I have accidentially used gcc-5 for the test.  With gcc-6 as above this
is the only difference:

--- aranym/tlink.s	2016-05-06 16:45:31.924100718 +0200
+++ qemu/tlink.s	2016-05-06 16:44:54.568052423 +0200
@@ -2471,7 +2471,8 @@ _Z8do_tlinkPPcS0_:
 	move.l 4(%a5),%d7
 	cmp.l %d2,%d7
 	jne .L329
-	btst #0,3(%a5)
+	move.b 3(%a5),%d0
+	btst #0,%d0
 	jeq .L384
 	move.l -14(%fp),%a1
 	clr.l 4(%a1)

Andreas.

-- 
Andreas Schwab, schwab@linux-m68k.org
GPG Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."