OpenE2K/gcc
2
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

tree-ssa-threadupdate.c, [...]: Fix comment typos.

Browse Source 
* tree-ssa-threadupdate.c, tree-vectorizer.c: Fix comment
	typos.
	* config/arm/arm.c: Follow spelling conventions.

From-SVN: r90963
This commit is contained in:
Kazu Hirata 2004-11-20 12:48:15 +00:00 committed by Kazu Hirata
parent 2075b05db6
commit e7a531ae61
4 changed files with 13 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
gcc/ChangeLog
View File

@ -1,3 +1,9 @@
2004-11-20 Kazu Hirata <kazu@cs.umass.edu>
* tree-ssa-threadupdate.c, tree-vectorizer.c: Fix comment
typos.
* config/arm/arm.c: Follow spelling conventions.
2004-11-20 Richard Earnshaw <rearnsha@arm.com>
* arm.c (arm_override_options): Use arm_ld_sched rather than testing

2
gcc/config/arm/arm.c
View File

@ -7258,7 +7258,7 @@ push_minipool_fix (rtx insn, HOST_WIDE_INT address, rtx *loc,
minipool_fix_tail = fix;
}
/* Return the cost of synthesising the const_double VAL inline.
/* Return the cost of synthesizing the const_double VAL inline.
Returns the number of insns needed, or 99 if we don't know how to
do it. */
int

4
gcc/tree-ssa-threadupdate.c
View File

@ -106,7 +106,7 @@ struct el
/* We need to efficiently record the unique thread destinations of this
block and specific information associated with those destinations. We
may have many incoming edges threaded to the same outgoing edge. This
can be naturaly implemented with a hash table. */
can be naturally implemented with a hash table. */
struct redirection_data
{
@ -467,7 +467,7 @@ thread_block (basic_block bb)
bool all = true;
/* To avoid scanning a linear array for the element we need we instead
use a hash table. For normal code there should be no noticable
use a hash table. For normal code there should be no noticeable
difference. However, if we have a block with a large number of
incoming and outgoing edges such linear searches can get expensive. */
redirection_data = htab_create (EDGE_COUNT (bb->succs),

8
gcc/tree-vectorizer.c
View File

@ -574,7 +574,7 @@ slpeel_update_phis_for_duplicate_loop (struct loop *orig_loop,
- IS_NEW_LOOP: TRUE if LOOP is a new loop (a duplicated copy of another
"original" loop). FALSE if LOOP is an original loop (not a newly
created copy). The SSA_NAME_AUX fields of the defs in the origianl
created copy). The SSA_NAME_AUX fields of the defs in the original
loop are the corresponding new ssa-names used in the new duplicated
loop copy. IS_NEW_LOOP indicates which of the two args of the phi
nodes in UPDATE_BB takes the original ssa-name, and which takes the
@ -929,14 +929,14 @@ slpeel_verify_cfg_after_peeling (struct loop *first_loop,
case, first-loop is the newly created loop, and second-loop is LOOP.
- NITERS: the number of iterations that LOOP iterates.
- FIRST_NITERS: the number of iterations that the first-loop should iterate.
- UPDATE_FIRST_LOOP_COUNT: specified whether this function is responssible
- UPDATE_FIRST_LOOP_COUNT: specified whether this function is responsible
for updating the loop bound of the first-loop to FIRST_NITERS. If it
is false, the caller of this function may want to take care of this
(this can be usefull is we don't want new stmts added to first-loop).
(this can be useful if we don't want new stmts added to first-loop).
Output:
The function returns a pointer to the new loop-copy, or NULL if it failed
to perform the trabsformation.
to perform the transformation.
The function generates two if-then-else guards: one before the first loop,
and the other before the second loop: