Sebastian Andrzej Siewior sebastian@breakpoint.cc
Sun Jun 6 15:22:00 GMT 2010

• Previous message (by thread): [PATCH 7/8] add sync and lwsync to the e500 testuite
• Next message (by thread): [PATCH 6/8] add eieio handling on e500 into testsuite
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

---

sync is both a simple and an extended mnemnonic. We have
sync => sync 0
lwsync => sync 1
ptesync => sync 2

Book-E only defines sync 0 with its symple mnemnonic msync. The other two are
Book-S only.
Without -many the simple mnemnonic is treated as an invalid opcode. I see very
little benefit by throwing patches at the software in order to replace sync
with msync which turns out to be the very same opcode.
The extended form like sync 0 is still invalid for BOOKE in order to avoid
getting opcodes with the L field != 0.

Signed-off-by: Sebastian Andrzej Siewior <sebastian@breakpoint.cc>
---
 opcodes/ppc-opc.c |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/opcodes/ppc-opc.c b/opcodes/ppc-opc.c
index 0e882e7..1b3e344 100644
--- a/opcodes/ppc-opc.c
+++ b/opcodes/ppc-opc.c
@@ -4469,6 +4469,7 @@ const struct powerpc_opcode powerpc_opcodes[] = {
 {"ptesync",	XSYNC(31,598,2), 0xffffffff, PPC64,	PPCNONE,	{0}},
 {"sync",	X(31,598),	XSYNC_MASK,  PPCCOM,	BOOKE|PPC476,	{LS}},
 {"msync",	X(31,598),	0xffffffff, BOOKE|PPCA2|PPC476, PPCNONE, {0}},
+{"sync",	X(31,598),	0xffffffff, BOOKE|PPC476, PPCNONE, {0}},
 {"dcs",		X(31,598),	0xffffffff,  PWRCOM,	PPCNONE,	{0}},
 
 {"lfdx",	X(31,599),	X_MASK,      COM,	PPCNONE,	{FRT, RA0, RB}},
-- 
1.7.1

---

• Previous message (by thread): [PATCH 7/8] add sync and lwsync to the e500 testuite
• Next message (by thread): [PATCH 6/8] add eieio handling on e500 into testsuite
• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]

---

More information about the Binutils mailing list