Re: vcond implementation in altivec
David Edelsohn <[EMAIL PROTECTED]> wrote on 02.03.2007 19:10:58: > > Devang Patel writes: > > >> Is there a reason why op0 is V4SF > Devang> It is destination so, yes this is wrong. > > >> and op1 is V4SI (and not V8HI)? > > Devang> condition should be v4si, but it is not op1. So this is also > not correct. > > >> And also, why not use if_then_else instead of unspec (in all vcond's)? > > Devang> I did not try that path. May be I did not know about it at that time. > > >Patches welcome. > > David > I am working on the patch and will submit it soon. Sa
Re: Interoperability of Fortran array and C vector?
I noticed that in fortran/convert.c the convert() function calls convert_to_vector() if the target type is VECTOR_TYPE. When is that function triggered in Fortran frontend? Since Fortran language doesn't support vector type, why does it convert something to a vector expression? Thanks! Sa
delete LIBCALL note after split
Hi, I'm working on GCC 4.1.1 on spu target and get following problem: After splitting an insn with a note REG_LIBCALL, the insn is replaced by some other insns, which don't attach REG_LIBCALL note any more, and the original one is then deleted. While the insn REG_RETVAL still points to the LIBCALL insn, which doesn't exist after the split. After reload it tries to delete the LIBCALL insn referenced by RETVAL, but it has been deleted already. The question is how to avoid deleting the LIBCALL note twice? Is it possible to have RETVAL note point to the new insn split from LIBCALL note? Any idea to solve this problem would be appreciated... Thanks! Sa
C++ frontend can not handle vector pointer constant parameter
Hi all,
I have detected a bug in C++ frontend.
When compiling a function with parameter of a pointer to a vector constant
type, the compiler calls a recursive function and is not able to get out.
Concretely, in gcc/cp/mangle.c file, in function write_type:
if (write_CV_qualifiers_for_type (type) > 0)
/* If TYPE was CV-qualified, we just wrote the qualifiers; now
mangle the unqualified type. The recursive call is needed here
since both the qualified and unqualified types are substitution
candidates. */
write_type (TYPE_MAIN_VARIANT (type));
But TYPE_MAIN_VARIANT (type) has been set as type itself in gcc/tree.c
function make_node_stat:
case tcc_type:
...
TYPE_MAIN_VARIANT (t) = t;
Therefor the write_type function runs into a dead recursion.
I have a test case to show the incorrect behavior on Intel:
void bar(
int __attribute__((vector_size(16))) * a,
int __attribute__((vector_size(16))) * const b);
int instance(void)
{
int __attribute__((vector_size(16))) a[1], b[1];
bar(a, b);
}
The error also occurs on other machines supporting vector type. I suppose
that some flags for the qualifier are not properly set.
Does someone knowing frontend well could provide any help?
Thanks!
Sa
Re: C++ frontend can not handle vector pointer constant parameter
"Andrew Pinski" <[EMAIL PROTECTED]> wrote on 02.08.2007 19:36:30: > This used to work. Can you file a bug at the very least? > > Thanks, > Andrew Pinski I have opened a bug against it. http://gcc.gnu.org/bugzilla/show_bug.cgi?id=32970 Sa
Re: delete LIBCALL note after split
Ian Lance Taylor <[EMAIL PROTECTED]> wrote on 03.08.2007 17:24:45: > Thanks. The general idea looks right, but the implementation is > incorrect. > Thank you! I have made the correction: Index: gcc/emit-rtl.c === --- gcc.orig/emit-rtl.c +++ gcc/emit-rtl.c @@ -3121,7 +3121,7 @@ try_split (rtx pat, rtx trial, int last) rtx before = PREV_INSN (trial); rtx after = NEXT_INSN (trial); int has_barrier = 0; - rtx tem; + rtx tem, note_retval; rtx note, seq; int probability; rtx insn_last, insn; @@ -3257,6 +3257,17 @@ try_split (rtx pat, rtx trial, int last) break; #endif + case REG_LIBCALL: + /* Relink the insns with REG_LIBCALL note and with REG_RETVAL note + after split. */ + REG_NOTES (insn_last) + = gen_rtx_EXPR_LIST (REG_LIBCALL, +XEXP (note, 0), +REG_NOTES (insn_last)); + note_retval = find_reg_note (XEXP (note, 0), REG_RETVAL, NULL); + XEXP (note_retval, 0) = insn_last; + break; + default: break; } Sa
Re: delete LIBCALL note after split
Ian Lance Taylor <[EMAIL PROTECTED]> wrote on 30.07.2007 18:59:28: > If the compiler splits an insn with a REG_LIBCALL note, it needs to > remove the corresponding REG_RETVAL note, or it needs to relink the > insns. This looks like a compiler bug, in that try_split doesn't > handle REG_LIBCALL notes at all. It's quite unusual to be able to > split a libcall insn, so it's not surprising that this has not been > noticed previously. > > Ian Thanks for replying, Ian! I have added a piece of code in try_split to handle the links. Not sure whether this is O.K. for the other platforms. Index: gcc/emit-rtl.c === --- gcc.orig/emit-rtl.c +++ gcc/emit-rtl.c @@ -3121,7 +3121,7 @@ try_split (rtx pat, rtx trial, int last) rtx before = PREV_INSN (trial); rtx after = NEXT_INSN (trial); int has_barrier = 0; - rtx tem; + rtx tem, note_libcall, note_retval; rtx note, seq; int probability; rtx insn_last, insn; @@ -3257,6 +3257,17 @@ try_split (rtx pat, rtx trial, int last) break; #endif + case REG_LIBCALL: + /* Relink the insns with REG_LIBCALL note and with REG_RETVAL note + after split. */ + note_libcall += find_reg_note (trial, REG_LIBCALL, NULL); + REG_NOTES (insn_last) = note_libcall; + note_retval += find_reg_note (XEXP (note_libcall, 0), REG_RETVAL, NULL); + XEXP (note_retval, 0) = insn_last; + break; + default: break; } Thanks, Sa
Re: C++ frontend can not handle vector pointer constant parameter
"Andrew Pinski" <[EMAIL PROTECTED]> wrote on 02.08.2007 19:36:30: > On 8/2/07, Sa Liu <[EMAIL PROTECTED]> wrote: > > Hi all, > > > > I have detected a bug in C++ frontend. > > This used to work. Can you file a bug at the very least? > Hi Andrew, Do you still remember on which version it used to work? Thanks, Sa
