This is the mail archive of the binutils@sources.redhat.com 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]

[patch] fix mips testsuite mips16 test bogon.


the mips testsuite mips16 test output checker assumes assumes that
.p2align while mips16 code is being compiled will pad with zero.  In
fact, it pads with 0x6500 (the mips16 nop).

the patch below fixes the testcase to expect those nops.

(it's not clear to me that it's "right" to interpret all-0 words as
nops when disassembling mips16 code, but that bug was there before...
I also noticed that, in general, while mips-elf seems to make check
mostly-OK, mipsel-elf fails several additional test cases, this being
one of them.  Any suggestions on which targets i should be checking?)


to apply in src/gas/testsuite:

2000-10-07  Chris Demetriou  <cgd@sibyte.com>

	* gas/mips/mips16.d: Expect mips16 nops to be used for padding
	when mips16 code generation is enabled.

Index: gas/mips/mips16.d
===================================================================
RCS file: /cvs/src/src/gas/testsuite/gas/mips/mips16.d,v
retrieving revision 1.1.1.1
diff -c -r1.1.1.1 mips16.d
*** mips16.d	1999/05/03 07:28:51	1.1.1.1
--- mips16.d	2000/10/08 01:15:42
***************
*** 676,683 ****
   85e:	ef49      	exit	\$s0
   860:	efa9      	exit	\$s0-\$s1,\$ra
   862:	ef29      	exit	\$ra
!  864:	0000      	addiu	\$s0,\$sp,0
! 	...
  
  0+000868 <bar>:
  	...
--- 676,683 ----
   85e:	ef49      	exit	\$s0
   860:	efa9      	exit	\$s0-\$s1,\$ra
   862:	ef29      	exit	\$ra
!  864:	6500      	nop
!  866:	6500      	nop
  
  0+000868 <bar>:
  	...

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