This is the mail archive of the binutils@sourceware.org mailing list for the binutils 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: Rebased: [Patch mach-o/gas] provide a base test-suite entry for mach-o.


Nick, Alan,

is it ok for you ?

Tristan.

On Dec 13, 2011, at 12:50 PM, Iain Sandoe wrote:

> Rebased after the changes to objdump -P
> 
> (doesn't affect this patch much - but I've elected to roll in a couple of changes that kept needing repeating in dependent sub-trees).
> 
> On 12 Dec 2011, at 08:22, Tristan Gingold wrote:
> 
>> 
>> On Dec 11, 2011, at 3:13 PM, Iain Sandoe wrote:
>> 
>>> This provides the basic entries for testing the mach-o additions to gas.
>>> a couple of trivial tests are included - but the main motivation is to provide a structure to add tests to for other pending patches.
>>> cheers
>> 
>> Makes sense to me.  I suppose cpu specific tests will go to the already existing cpu subdirs.
>> 
>> I don't think I can approve this however.
> 
> Who should I ask?
> thanks
> Iain
> 
>>> gas/testsuite:
>>> 
>>> 	* gas/mach-o: New.
>>> 	* gas/mach-o/mach-o.exp: New.
>>> 	* gas/mach-o/warn-1.s: New.
>>> 	* gas/mach-o/lcomm-1.s: New.
>>> 	* gas/mach-o/lcomm-1.d: New.
> 	* gas/mach-o/empty.s: New.
> 
> gas/testsuite/gas/mach-o/empty.s    |    1 +
> gas/testsuite/gas/mach-o/lcomm-1.d  |   12 ++++++++++++
> gas/testsuite/gas/mach-o/lcomm-1.s  |   11 +++++++++++
> gas/testsuite/gas/mach-o/mach-o.exp |   20 ++++++++++++++++++++
> gas/testsuite/gas/mach-o/warn-1.s   |    5 +++++
> 5 files changed, 49 insertions(+), 0 deletions(-)
> 
> diff --git a/gas/testsuite/gas/mach-o/empty.s b/gas/testsuite/gas/mach-o/empty.s
> new file mode 100644
> index 0000000..b2a4ba5
> --- /dev/null
> +++ b/gas/testsuite/gas/mach-o/empty.s
> @@ -0,0 +1 @@
> +# nothing here
> diff --git a/gas/testsuite/gas/mach-o/lcomm-1.d b/gas/testsuite/gas/mach-o/lcomm-1.d
> new file mode 100644
> index 0000000..a9e29bc
> --- /dev/null
> +++ b/gas/testsuite/gas/mach-o/lcomm-1.d
> @@ -0,0 +1,12 @@
> +# nm: -B
> +#
> +# match for 32 and 64 bits.
> +(00000000)?00008000 b align_max
> +(00000000)?00000010 b another_align
> +(00000000)?00000000 b column0
> +(00000000)?00000002 b column1
> +(00000000)?00000004 b local3_noalign
> +(00000000)?00000008 b local4_aligned_2
> +(00000000)?00000020 b nospaces
> +(00000000)?00000040 b trailingspace
> +#
> diff --git a/gas/testsuite/gas/mach-o/lcomm-1.s b/gas/testsuite/gas/mach-o/lcomm-1.s
> new file mode 100644
> index 0000000..2e44f37
> --- /dev/null
> +++ b/gas/testsuite/gas/mach-o/lcomm-1.s
> @@ -0,0 +1,11 @@
> +
> +.lcomm column0,1,1
> + .lcomm column1, 2,1
> +	.lcomm local3_noalign, 3,1
> +
> +	.lcomm local4_aligned_2, 4, 2
> +	.lcomm another_align, 5, 4
> +	.lcomm nospaces,6,5
> +	.lcomm trailingspace,7,6	 	
> +
> +	.lcomm align_max, 8, 15
> diff --git a/gas/testsuite/gas/mach-o/mach-o.exp b/gas/testsuite/gas/mach-o/mach-o.exp
> new file mode 100644
> index 0000000..854cb0c
> --- /dev/null
> +++ b/gas/testsuite/gas/mach-o/mach-o.exp
> @@ -0,0 +1,20 @@
> +#
> +# mach-o tests
> +#
> +
> +# We're testing bits in obj-mach-o -- don't run on anything else.
> +
> +if {[istarget "*-*-darwin*"]} then {
> +
> +load_lib gas-dg.exp
> +
> +run_dump_tests [lsort [glob -nocomplain $srcdir/$subdir/*.d]]
> +
> +dg-init
> +
> +dg-runtest [lsort [glob -nocomplain $srcdir/$subdir/err-*.s $srcdir/$subdir/warn-*.s]] "" ""
> +
> +dg-finish
> +
> +}
> +# mach-o
> diff --git a/gas/testsuite/gas/mach-o/warn-1.s b/gas/testsuite/gas/mach-o/warn-1.s
> new file mode 100644
> index 0000000..5ed9369
> --- /dev/null
> +++ b/gas/testsuite/gas/mach-o/warn-1.s
> @@ -0,0 +1,5 @@
> +# { dg-do assemble }
> +
> +	.lcomm align_too_big,9,100
> +
> +# { dg-warning "Warning: Alignment (100) too large: 15 assumed." "" { xfail *-*-darwin* } 3 }
> 
> 
> 


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