PATCH: Add a testcase for PR 876

View: New views
3 Messages — Rating Filter:   Alert me  

PATCH: Add a testcase for PR 876

by H.J. Lu-10 :: Rate this Message:

Reply to Author | View Threaded | Show Only this Message

I am checking in this patch to add a testcase for PR 876.

H.J.
Index: ChangeLog
===================================================================
--- ChangeLog (revision 7236)
+++ ChangeLog (working copy)
@@ -1,3 +1,11 @@
+2009-11-04  H.J. Lu  <hongjiu.lu@...>
+
+ PR gas/876
+ * gas/i386/i386.exp: Run space1.
+
+ * gas/i386/space1.l: New.
+ * gas/i386/space1.s: Likewise.
+
 2009-11-03  Paul Brook  <paul@...>
 
  * gas/arm/vfp-neon-syntax.d: Update expected results.
Index: gas/i386/i386.exp
===================================================================
--- gas/i386/i386.exp (revision 7236)
+++ gas/i386/i386.exp (working copy)
@@ -17,6 +17,9 @@ proc gas_32_check { } {
     return [regexp "targets:.*i386" [lindex $status 1]];
 }
 
+# Common tests
+run_list_test "space1" "-al"
+
 if [expr ([istarget "i*86-*-*"] ||  [istarget "x86_64-*-*"]) && [gas_32_check]]  then {
 
     global ASFLAGS
Index: gas/i386/space1.l
===================================================================
--- gas/i386/space1.l (revision 0)
+++ gas/i386/space1.l (revision 0)
@@ -0,0 +1,15 @@
+.*: Assembler messages:
+.*:2: Error: .space specifies non-absolute value
+.*:3: Error: .space specifies non-absolute value
+.*:4: Error: .space specifies non-absolute value
+.*:5: Error: .space specifies non-absolute value
+.*:6: Error: .space specifies non-absolute value
+GAS LISTING .*
+
+
+[ ]*1[ ]+\.text
+[ ]*2[ ]+\.space foo
+[ ]*3[ ]+\.space foo\*4
+[ ]*4[ ]+\.space foo\+4
+[ ]*5[ ]+\.space foo-4
+[ ]*6[ ]+\.space foo << 4
Index: gas/i386/space1.s
===================================================================
--- gas/i386/space1.s (revision 0)
+++ gas/i386/space1.s (revision 0)
@@ -0,0 +1,6 @@
+ .text
+ .space foo
+ .space foo*4
+ .space foo+4
+ .space foo-4
+ .space foo << 4

Re: PATCH: Add a testcase for PR 876

by Hans-Peter Nilsson :: Rate this Message:

Reply to Author | View Threaded | Show Only this Message

> Date: Wed, 4 Nov 2009 10:49:34 -0800
> From: "H.J. Lu" <hongjiu.lu@...>

> Index: gas/i386/i386.exp
> ===================================================================
> --- gas/i386/i386.exp (revision 7236)
> +++ gas/i386/i386.exp (working copy)
> @@ -17,6 +17,9 @@ proc gas_32_check { } {
>      return [regexp "targets:.*i386" [lindex $status 1]];
>  }
>  
> +# Common tests
> +run_list_test "space1" "-al"
> +
>  if [expr ([istarget "i*86-*-*"] ||  [istarget "x86_64-*-*"]) && [gas_32_check]]  then {

This is wrong, you've put the test before the target-gate-test,
so it's attempted on all targets, not just the x86 and
relatives.

My autotester complains about the failure for cris-axis-elf and
cris-axis-linux-gnu:

Running /tmp/hpautotest-binutils/bsrc/src/gas/testsuite/gas/i386/i386.exp ...
FAIL: i386 space1

gas.log:
regexp_diff match failure
regexp "^GAS LISTING .*$"
line   "GAS for CRIS /tmp/hpautotest-binutils/bsrc/src/gas/testsuite/gas/i386/space1.s                  page 1"
FAIL: i386 space1

(That's due to a target-specific assumption of the default
LISTING_HEADER.)

If you really mean for it to run everywhere, it shouldn't sit in
testsuite/gas/i386 anyways.

brgds, H-P

Re: PATCH: Add a testcase for PR 876

by H.J. Lu-30 :: Rate this Message:

Reply to Author | View Threaded | Show Only this Message

On Wed, Nov 4, 2009 at 6:37 PM, Hans-Peter Nilsson
<hans-peter.nilsson@...> wrote:

>> Date: Wed, 4 Nov 2009 10:49:34 -0800
>> From: "H.J. Lu" <hongjiu.lu@...>
>
>> Index: gas/i386/i386.exp
>> ===================================================================
>> --- gas/i386/i386.exp (revision 7236)
>> +++ gas/i386/i386.exp (working copy)
>> @@ -17,6 +17,9 @@ proc gas_32_check { } {
>>      return [regexp "targets:.*i386" [lindex $status 1]];
>>  }
>>
>> +# Common tests
>> +run_list_test "space1" "-al"
>> +
>>  if [expr ([istarget "i*86-*-*"] ||  [istarget "x86_64-*-*"]) && [gas_32_check]]  then {
>
> This is wrong, you've put the test before the target-gate-test,
> so it's attempted on all targets, not just the x86 and
> relatives.
>

I will fix it tomorrow.


--
H.J.