This is the mail archive of the
binutils@sourceware.cygnus.com
mailing list for the binutils project.
Re: XCOFF64 gas patch
- To: cpopetz at cygnus dot com
- Subject: Re: XCOFF64 gas patch
- From: Ian Lance Taylor <ian at zembu dot com>
- Date: 10 Apr 2000 11:02:25 -0700
- CC: binutils at sourceware dot cygnus dot com
- References: <20000410114121.A9393@cpopetz.com>
Date: Mon, 10 Apr 2000 11:41:21 -0500
From: Clinton Popetz <cpopetz@cygnus.com>
*** obj-coff.c 2000/02/22 07:50:13 1.22
--- obj-coff.c 2000/04/10 16:40:54
*************** static void
*** 430,434 ****
add_lineno (frag, offset, num)
fragS *frag;
! int offset;
int num;
{
--- 430,434 ----
add_lineno (frag, offset, num)
fragS *frag;
! bfd_vma offset;
int num;
{
This is basically right, but it should be addressT, not bfd_vma, since
add_lineno() is only called with the result of frag_now_fix(), and
frag_now_fix() returns addressT, not bfd_vma.
+ int
+ ppc_subseg_align()
Should have a space before the left paren.
+ extern char*
+ target_format()
Here too. Also, I think this should be named something like
ppc_target_format.
*************** ppc_xcoff_cons (log_size)
*** 3151,3154 ****
--- 3185,3195 ----
static void
+ ppc_machine(dummy)
+ int dummy;
+ {
+ discard_rest_of_line();
+ }
More spaces before parens.
*************** md_apply_fix3 (fixp, valuep, seg)
*** 4803,4807 ****
&& operand->bits == 16
&& operand->shift == 0
! && operand->insert == NULL
&& fixp->fx_addsy != NULL
&& ppc_is_toc_sym (fixp->fx_addsy))
--- 4844,4848 ----
&& operand->bits == 16
&& operand->shift == 0
! /* && operand->insert == NULL*/
&& fixp->fx_addsy != NULL
&& ppc_is_toc_sym (fixp->fx_addsy))
This is not a good sort of change, because it is not maintainable.
Either the test for insert == NULL is simply incorrect, and we should
simply remove it, or there is some more subtle issue here, and we need
a comment explaining what is going on.
Ian