* Re: Disallow 3-operand cmp[l][i] for ppc64
[not found] <20160929060240.GD4877@bubble.grove.modra.org>
@ 2016-09-29 21:21 ` Peter Bergner
2016-09-30 1:17 ` Alan Modra
0 siblings, 1 reply; 2+ messages in thread
From: Peter Bergner @ 2016-09-29 21:21 UTC (permalink / raw)
To: Alan Modra; +Cc: binutils, gdb-patches
On 9/29/16 1:02 AM, Alan Modra wrote:
> cmp[l][o] get an optional L field only when generating 32-bit code.
> dcbf, tlbie and tlbiel keep their optional L field, ditto for R field
> of tbegin. cmprb, tsr., wlcr[all] and mtsle all change to a
> compulsory L field.
FYI, I just committed the following as obvious to clean up the gdb
testsuite fallout from this change.
Peter
* gdb.arch/powerpc-power.exp <cmprb>: Update tests to account for
the compulsory L operand changes.
* gdb.arch/powerpc-power.s: Likewise.
diff --git a/gdb/testsuite/gdb.arch/powerpc-power.exp b/gdb/testsuite/gdb.arch/powerpc-power.exp
index 3bdfc15..8594ad5 100644
--- a/gdb/testsuite/gdb.arch/powerpc-power.exp
+++ b/gdb/testsuite/gdb.arch/powerpc-power.exp
@@ -1117,8 +1117,8 @@ func_check +4168 "xvxsigdp vs59,vs60"
func_check +4172 "xvxsigsp vs60,vs61"
func_check +4176 "cmpeqb cr0,r6,r7"
func_check +4180 "cmpeqb cr7,r6,r7"
-func_check +4184 "cmprb cr0,r8,r9"
-func_check +4188 "cmprb cr7,r8,r9"
+func_check +4184 "cmprb cr0,0,r8,r9"
+func_check +4188 "cmprb cr7,0,r8,r9"
func_check +4192 "cmprb cr0,1,r8,r9"
func_check +4196 "cmprb cr7,1,r8,r9"
func_check +4200 "setb r15,cr0"
diff --git a/gdb/testsuite/gdb.arch/powerpc-power.s b/gdb/testsuite/gdb.arch/powerpc-power.s
index 507b61e..6fe2520 100644
--- a/gdb/testsuite/gdb.arch/powerpc-power.s
+++ b/gdb/testsuite/gdb.arch/powerpc-power.s
@@ -1078,8 +1078,8 @@ func:
.long 0xf389ef6f /* <+4172>: xvxsigsp vs60,vs61 */
.long 0x7c0639c0 /* <+4176>: cmpeqb cr0,r6,r7 */
.long 0x7f8639c0 /* <+4180>: cmpeqb cr7,r6,r7 */
- .long 0x7c084980 /* <+4184>: cmprb cr0,r8,r9 */
- .long 0x7f884980 /* <+4188>: cmprb cr7,r8,r9 */
+ .long 0x7c084980 /* <+4184>: cmprb cr0,0,r8,r9 */
+ .long 0x7f884980 /* <+4188>: cmprb cr7,0,r8,r9 */
.long 0x7c284980 /* <+4192>: cmprb cr0,1,r8,r9 */
.long 0x7fa84980 /* <+4196>: cmprb cr7,1,r8,r9 */
.long 0x7de00100 /* <+4200>: setb r15,cr0 */
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: Disallow 3-operand cmp[l][i] for ppc64
2016-09-29 21:21 ` Disallow 3-operand cmp[l][i] for ppc64 Peter Bergner
@ 2016-09-30 1:17 ` Alan Modra
0 siblings, 0 replies; 2+ messages in thread
From: Alan Modra @ 2016-09-30 1:17 UTC (permalink / raw)
To: Peter Bergner; +Cc: binutils, gdb-patches
On Thu, Sep 29, 2016 at 03:34:40PM -0500, Peter Bergner wrote:
> On 9/29/16 1:02 AM, Alan Modra wrote:
> > cmp[l][o] get an optional L field only when generating 32-bit code.
> > dcbf, tlbie and tlbiel keep their optional L field, ditto for R field
> > of tbegin. cmprb, tsr., wlcr[all] and mtsle all change to a
> > compulsory L field.
>
> FYI, I just committed the following as obvious to clean up the gdb
> testsuite fallout from this change.
Thanks!
--
Alan Modra
Australia Development Lab, IBM
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2016-09-29 21:21 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
[not found] <20160929060240.GD4877@bubble.grove.modra.org>
2016-09-29 21:21 ` Disallow 3-operand cmp[l][i] for ppc64 Peter Bergner
2016-09-30 1:17 ` Alan Modra
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).