Skip to content
Projects
Groups
Snippets
Help
This project
Loading...
Sign in / Register
Toggle navigation
R
riscv-gcc-1
Overview
Overview
Details
Activity
Cycle Analytics
Repository
Repository
Files
Commits
Branches
Tags
Contributors
Graph
Compare
Charts
Issues
0
Issues
0
List
Board
Labels
Milestones
Merge Requests
0
Merge Requests
0
CI / CD
CI / CD
Pipelines
Jobs
Schedules
Charts
Wiki
Wiki
Snippets
Snippets
Members
Members
Collapse sidebar
Close sidebar
Activity
Graph
Charts
Create a new issue
Jobs
Commits
Issue Boards
Open sidebar
lvzhengyang
riscv-gcc-1
Commits
07095a8b
Commit
07095a8b
authored
Sep 11, 1999
by
Zack Weinberg
Browse files
Options
Browse Files
Download
Email Patches
Plain Diff
Remove trailing whitespace from lines
From-SVN: r29283
parent
a611912f
Hide whitespace changes
Inline
Side-by-side
Showing
1 changed file
with
16 additions
and
16 deletions
+16
-16
gcc/genrecog.c
+16
-16
No files found.
gcc/genrecog.c
View file @
07095a8b
...
...
@@ -176,7 +176,7 @@ static struct pred_table
static
struct
decision_head
make_insn_sequence
PROTO
((
rtx
,
enum
routine_type
));
static
struct
decision
*
add_to_sequence
PROTO
((
rtx
,
struct
decision_head
*
,
const
char
*
,
const
char
*
,
enum
routine_type
,
int
));
static
int
not_both_true
PROTO
((
struct
decision
*
,
struct
decision
*
,
int
));
...
...
@@ -243,9 +243,9 @@ make_insn_sequence (insn, type)
last_real_name
=
name
;
last_real_code
=
next_insn_code
;
}
insn_name_ptr
[
next_insn_code
]
=
name
;
}
}
if
(
type
==
PEEPHOLE2
)
{
...
...
@@ -554,7 +554,7 @@ add_to_sequence (pattern, last, position, insn_type, top)
fatal
(
"mode mismatch in SET"
);
}
newpos
[
depth
]
=
'0'
;
new
=
add_to_sequence
(
SET_DEST
(
pattern
),
&
new
->
success
,
newpos
,
new
=
add_to_sequence
(
SET_DEST
(
pattern
),
&
new
->
success
,
newpos
,
insn_type
,
0
);
this
->
success
.
first
->
enforce_mode
=
1
;
newpos
[
depth
]
=
'1'
;
...
...
@@ -572,7 +572,7 @@ add_to_sequence (pattern, last, position, insn_type, top)
case
ZERO_EXTEND
:
case
STRICT_LOW_PART
:
newpos
[
depth
]
=
'0'
;
new
=
add_to_sequence
(
XEXP
(
pattern
,
0
),
&
new
->
success
,
newpos
,
new
=
add_to_sequence
(
XEXP
(
pattern
,
0
),
&
new
->
success
,
newpos
,
insn_type
,
0
);
this
->
success
.
first
->
enforce_mode
=
1
;
return
new
;
...
...
@@ -608,7 +608,7 @@ add_to_sequence (pattern, last, position, insn_type, top)
break
;
/* ... fall through ... */
case
COMPARE
:
/* Enforce the mode on the first operand to avoid ambiguous insns. */
newpos
[
depth
]
=
'0'
;
...
...
@@ -619,7 +619,7 @@ add_to_sequence (pattern, last, position, insn_type, top)
new
=
add_to_sequence
(
XEXP
(
pattern
,
1
),
&
new
->
success
,
newpos
,
insn_type
,
0
);
return
new
;
default
:
break
;
}
...
...
@@ -870,7 +870,7 @@ merge_trees (oldh, addh)
/* The semantics of pattern matching state that the tests are done in
the order given in the MD file so that if an insn matches two
patterns, the first one will be used. However, in practice, most,
if not all, patterns are unambiguous so that their order is
if not all, patterns are unambiguous so that their order is
independent. In that case, we can merge identical tests and
group all similar modes and codes together.
...
...
@@ -881,7 +881,7 @@ merge_trees (oldh, addh)
the same code and mode and the last one that tests just the same mode.
If we have no match, place NEW after the closest match we found. */
for
(
old
=
oldh
.
last
;
old
;
old
=
old
->
prev
)
{
int
our_merit
;
...
...
@@ -1183,7 +1183,7 @@ static const char *indents[]
We generate two nested switch statements, the outer statement for
testing modes, and the inner switch for testing RTX codes. It is
not worth optimizing cases when only a small number of modes or
not worth optimizing cases when only a small number of modes or
codes is tested, since the compiler can do that when compiling the
resulting function. We do check for when every test is the same mode
or code. */
...
...
@@ -1225,7 +1225,7 @@ write_tree_1 (tree, prevpos, afterward, type)
done. The exception is a branch from a node that does not test a
mode or code to one that does. In such cases, we set the `retest_mode'
or `retest_code' flags. That will ensure that we start a new switch
at that position and put the label before the switch.
at that position and put the label before the switch.
The branches in the latter case must set the position to that of the
target node. */
...
...
@@ -1566,7 +1566,7 @@ write_tree_1 (tree, prevpos, afterward, type)
inner_indent
-=
2
;
printf
(
"%s}
\n
"
,
indents
[
inner_indent
]);
}
else
else
{
if
(
p
->
num_clobbers_to_add
)
{
...
...
@@ -1756,7 +1756,7 @@ write_tree (tree, prevpos, afterward, initial, type)
actions are necessary to move to NEWPOS. If we fail to move to the
new state, branch to node AFTERWARD if non-zero, otherwise return.
INDENT says how many blanks to place at the front of lines.
INDENT says how many blanks to place at the front of lines.
Failure to move to the new state can only occur if we are trying to
match multiple insns and we try to step past the end of the
...
...
@@ -1802,12 +1802,12 @@ change_state (oldpos, newpos, indent, afterward)
/* We can only fail if we're moving down the tree. */
if
(
old_has_insn
>=
0
&&
oldpos
[
old_has_insn
]
>=
newpos
[
depth
])
{
printf
(
"%s_last_insn = recog_next_insn (insn, %d);
\n
"
,
printf
(
"%s_last_insn = recog_next_insn (insn, %d);
\n
"
,
indents
[
indent
],
newpos
[
depth
]
-
'A'
);
}
else
{
printf
(
"%stem = recog_next_insn (insn, %d);
\n
"
,
printf
(
"%stem = recog_next_insn (insn, %d);
\n
"
,
indents
[
indent
],
newpos
[
depth
]
-
'A'
);
printf
(
"%sif (tem == NULL_RTX)
\n
"
,
indents
[
indent
]);
...
...
@@ -1931,7 +1931,7 @@ from the machine description file `md'. */\n\n");
else
if
(
GET_CODE
(
desc
)
==
DEFINE_PEEPHOLE2
)
peephole2_tree
=
merge_trees
(
peephole2_tree
,
make_insn_sequence
(
desc
,
PEEPHOLE2
));
if
(
GET_CODE
(
desc
)
==
DEFINE_PEEPHOLE
||
GET_CODE
(
desc
)
==
DEFINE_EXPAND
)
next_insn_code
++
;
...
...
Write
Preview
Markdown
is supported
0%
Try again
or
attach a new file
Attach a file
Cancel
You are about to add
0
people
to the discussion. Proceed with caution.
Finish editing this message first!
Cancel
Please
register
or
sign in
to comment