This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [patch, m68k] A couple of fixes for ColdFire Linux

Andreas Schwab wrote:
Maxim Kuvyrkov <maxim@codesourcery.com> writes:

--- gcc/config/m68k/lb1sf68.asm
+++ gcc/config/m68k/lb1sf68.asm
@@ -163,6 +163,8 @@ see the files COPYING3 and COPYING.RUNTI
 #if defined (__mcoldfire__) && !defined (__mcfisab__) && !defined (__mcfisac__)
 	lea	\addr-.-8,a0
 	jsr	pc@(a0)
+#elif defined (__mcfisab__) || defined (__mcfisac__)
+	bsr.l	\addr
 #else
 	bsr	\addr
 #endif

I think this should use jbsr for all.

jbsr is just an alias for bsr.l, they assemble to same binary code. Assembling jbsr for an ISA_A[+] cpu produces "Warning: Can't use long branches on this architecture". So I don't see any advantages in using jbsr over bsr.l. Or am I missing something?

--
Maxim

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]