Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
* [PATCH][PowerPC] Fix powerpc tests
@ 2020-02-03 20:59 Rogerio Alves
  2020-02-04  3:39 ` Simon Marchi
  0 siblings, 1 reply; 2+ messages in thread
From: Rogerio Alves @ 2020-02-03 20:59 UTC (permalink / raw)
  To: gdb-patches, uweigand, pedromfc

This patch fixes test failures power8 and power9 caused by changes on
opcodes:

The dissasembler does not emit whitespace for instructions
anymore (c2b1c2754526acff8aae2fe8f5a56c2dd11d0b7f)

The dissasembler generates extended mnemonics for some instructions
instead (aae9718e4d4e8d01dcee22684e82b000203d3e52)

The ldmx instruction was removed. This instruction was never
implemented (6fbc939cfdbdf02f205c20925583738b0f835e62)

gdb/testsuite/ChangeLog:
2020-02-03  Rogerio A. Cardoso  <rcardoso@linux.ibm.com>

	* gdb.arch/powerpc-power8.exp: Delete trailing whitespace of
	tbegin., tend. instructions. Replace bctar-, bctar+, bctarl-,
	bctarl+ extended mnemonics when avaliable by bgttar, bnstarl,
	blttar, bnetarl.
	* gdb.arch/powerpc-power8.s: Fix comments. Fix instructions
	binary for blttar, bnetarl.
	* gdb.arch/powerpc-power9.exp: Delete trailing whitespace of
	wait instruction. Delete ldmx test.
	* gdb.arch/powerpc-power9.s: Delete ldmx instruction.

---
  gdb/testsuite/gdb.arch/powerpc-power8.exp | 18 +++++++++---------
  gdb/testsuite/gdb.arch/powerpc-power8.s   |  8 ++++----
  gdb/testsuite/gdb.arch/powerpc-power9.exp |  6 ++----
  gdb/testsuite/gdb.arch/powerpc-power9.s   |  2 --
  4 files changed, 15 insertions(+), 19 deletions(-)

diff --git a/gdb/testsuite/gdb.arch/powerpc-power8.exp 
b/gdb/testsuite/gdb.arch/powerpc-power8.exp
index 109a5cbb64..d0da3e4f09 100644
--- a/gdb/testsuite/gdb.arch/powerpc-power8.exp
+++ b/gdb/testsuite/gdb.arch/powerpc-power8.exp
@@ -66,10 +66,10 @@ func_check "tabortwc. 7,r8,r16"
  func_check "tabortdc. 20,r11,r10"
  func_check "tabortwci. 17,r10,-13"
  func_check "tabortdci. 29,r3,-5"
-func_check "tbegin. "
+func_check "tbegin."
  func_check "tcheck  cr7"
-func_check "tend.   "
-func_check "tend.   "
+func_check "tend."
+func_check "tend."
  func_check "tendall."
  func_check "tendall."
  func_check "treclaim. r24"
@@ -82,12 +82,12 @@ func_check "ori     r2,r2,0"
  func_check "nop"
  func_check "ori     r2,r2,0"
  func_check "rfebb   0"
-func_check "rfebb   "
-func_check "rfebb   "
-func_check "bctar-  12,4*cr5+gt"
-func_check "bctarl- 4,4*cr1+so"
-func_check "bctar+  12,4*cr3+lt"
-func_check "bctarl+ 4,eq"
+func_check "rfebb"
+func_check "rfebb"
+func_check "bgttar  cr5"
+func_check "bnstarl cr1"
+func_check "blttar+ cr3"
+func_check "bnetarl+"
  func_check "bctar   4,4*cr2+lt,1"
  func_check "bctarl  4,4*cr1+so,2"
  func_check "waitasec"
diff --git a/gdb/testsuite/gdb.arch/powerpc-power8.s 
b/gdb/testsuite/gdb.arch/powerpc-power8.s
index 3e466a06e8..0cc3f84b3d 100644
--- a/gdb/testsuite/gdb.arch/powerpc-power8.s
+++ b/gdb/testsuite/gdb.arch/powerpc-power8.s
@@ -41,10 +41,10 @@ func:
  	.long  0x4c000124    /* rfebb   0                  */
  	.long  0x4c000924    /* rfebb                      */
  	.long  0x4c000924    /* rfebb                      */
-	.long  0x4d950460    /* bctar-  12,4*cr5+gt        */
-	.long  0x4c870461    /* bctarl- 4,4*cr1+so         */
-	.long  0x4dac0460    /* bctar+  12,4*cr3+lt        */
-	.long  0x4ca20461    /* bctarl+ 4,eq               */
+	.long  0x4d950460    /* bgttar  cr5                */
+	.long  0x4c870461    /* bnstarl cr1                */
+	.long  0x4dec0460    /* blttar+ cr3                */
+	.long  0x4ce20461    /* bnetarl+                   */
  	.long  0x4c880c60    /* bctar   4,4*cr2+lt,1       */
  	.long  0x4c871461    /* bctarl  4,4*cr1+so,2       */
  	.long  0x7c00003c    /* waitasec                   */
diff --git a/gdb/testsuite/gdb.arch/powerpc-power9.exp 
b/gdb/testsuite/gdb.arch/powerpc-power9.exp
index d3ab9af15e..c92b2008d9 100644
--- a/gdb/testsuite/gdb.arch/powerpc-power9.exp
+++ b/gdb/testsuite/gdb.arch/powerpc-power9.exp
@@ -409,11 +409,9 @@ func_check "stwat   r23,0,0"
  func_check "stwat   r23,r13,28"
  func_check "urfid"
  func_check "rmieg   r30"
-func_check "ldmx    r10,0,r15"
-func_check "ldmx    r10,r3,r15"
  func_check "stop"
-func_check "wait    "
-func_check "wait    "
+func_check "wait"
+func_check "wait"
  func_check "darn    r3,0"
  func_check "darn    r3,1"
  func_check "darn    r3,2"
diff --git a/gdb/testsuite/gdb.arch/powerpc-power9.s 
b/gdb/testsuite/gdb.arch/powerpc-power9.s
index e47809caae..b3dc6ea83a 100644
--- a/gdb/testsuite/gdb.arch/powerpc-power9.s
+++ b/gdb/testsuite/gdb.arch/powerpc-power9.s
@@ -366,8 +366,6 @@ func:
  	.long  0x7eede58c    /* stwat   r23,r13,28         */
  	.long  0x4c000264    /* urfid                      */
  	.long  0x7c00f6e4    /* rmieg   r30                */
-	.long  0x7d407a6a    /* ldmx    r10,0,r15          */
-	.long  0x7d437a6a    /* ldmx    r10,r3,r15         */
  	.long  0x4c0002e4    /* stop                       */
  	.long  0x7c00003c    /* wait                       */
  	.long  0x7c00003c    /* wait                       */
-- 
2.17.1


^ permalink raw reply	[flat|nested] 2+ messages in thread

* Re: [PATCH][PowerPC] Fix powerpc tests
  2020-02-03 20:59 [PATCH][PowerPC] Fix powerpc tests Rogerio Alves
@ 2020-02-04  3:39 ` Simon Marchi
  0 siblings, 0 replies; 2+ messages in thread
From: Simon Marchi @ 2020-02-04  3:39 UTC (permalink / raw)
  To: Rogerio Alves, gdb-patches, uweigand, pedromfc

On 2020-02-03 3:55 p.m., Rogerio Alves wrote:
> This patch fixes test failures power8 and power9 caused by changes on
> opcodes:
> 
> The dissasembler does not emit whitespace for instructions
> anymore (c2b1c2754526acff8aae2fe8f5a56c2dd11d0b7f)
> 
> The dissasembler generates extended mnemonics for some instructions
> instead (aae9718e4d4e8d01dcee22684e82b000203d3e52)
> 
> The ldmx instruction was removed. This instruction was never
> implemented (6fbc939cfdbdf02f205c20925583738b0f835e62)

Hi Rogerio,

Thanks for the patch.  Note that applying the patch with an older git version (1.8.3.1,
as found on gcc135 of the compile farm) failed.  When applying it locally with a more
recent git (2.25.0), it worked but said:

  warning: Patch sent with format=flowed

That probably means you sent the patch using your email client.  Please consider using
git-send-email when sending patches by email in the future to avoid formatting issues.

I've tested the patch on that machine (a Power9) and confirmed that the tests did fail
and now pass with this patch applied, so I've pushed it.

Thanks!

Simon


^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2020-02-04  3:39 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-02-03 20:59 [PATCH][PowerPC] Fix powerpc tests Rogerio Alves
2020-02-04  3:39 ` Simon Marchi

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox