On 5 August 2014 21:59, Richard Biener richard.guent...@gmail.com wrote:
On Mon, Aug 4, 2014 at 11:09 AM, Zhenqiang Chen zhenqiang.c...@arm.com
wrote:
-Original Message-
From: Bin.Cheng [mailto:amker.ch...@gmail.com]
Sent: Monday, August 04, 2014 4:41 PM
To: Zhenqiang Chen
Cc:
Jakub Jelinek wrote:
On Sat, Aug 02, 2014 at 12:09:24AM +0300, Janne Blomqvist wrote:
--- libgfortran/runtime/memory.c.jj 2014-06-18 08:50:33.0 +0200
+++ libgfortran/runtime/memory.c2014-08-01 14:41:08.385856116 +0200
@@ -56,7 +56,9 @@ xmallocarray (size_t nmemb, size_t
* PING * – of the patch with the obvious change mentioned by Alessandro
(i.e. using if(is_lock_type))?
Tobias
On 1 August 2014 21:57, Alessandro Fanfarillo wrote:
Hello,
I was implementing lock/unlock on the library side when I found a
possible problem in the patch:
if (is_lock_type ==
Hi,
On 08/06/2014 06:41 AM, Braden Obrzut wrote:
I can confirm that this is caused by a change to pt.c that happened, I
think, the day before my last patch.
This can be fixed by first checking that the template is a function
template at that line in pt.c. Since variable templates can't be
Thanks for looking at this.
Bootstrapped and reg-tested on x86_64-unknown-linux-gnu.
2014-08-06 0:18 GMT+04:00 Jeff Law l...@redhat.com:
On 08/04/14 00:08, Alexander Ivchenko wrote:
Hi,
libcilkrts is compiled with -nostdlib, that means we have to
explicitly specify the pthread library we
On Tue, 5 Aug 2014, Jeff Law wrote:
On 08/05/14 08:36, Marek Polacek wrote:
On Mon, Aug 04, 2014 at 02:04:36PM +0200, Richard Biener wrote:
Looks like .fre can optimize q - (q - 1) into 1:
bb 2:
q.0_3 = (long int) MEM[(void *)i + 4B];
_5 = (long int) i;
- _6 =
On Tue, Aug 05, 2014 at 02:14:21PM -0600, Jeff Law wrote:
My concern is the code we're removing discusses the need to simplify when
these expressions are in static initializers. What's going to ensure that
we're still simplifying instances which appear in static initializers? I
don't see
On Wed, Aug 06, 2014 at 10:22:19AM +0200, Marek Polacek wrote:
On Tue, Aug 05, 2014 at 02:14:21PM -0600, Jeff Law wrote:
My concern is the code we're removing discusses the need to simplify when
these expressions are in static initializers. What's going to ensure that
we're still
On 05/08/14 12:18, Kyrill Tkachov wrote:
Hi all,
This is a cleanup to replace usages of GET_CODE (x) == CONST_INT with
CONST_INT_P (x) and GET_CODE (x) == REG with REG_P (x). No functional
changes.
Tested on aarch64-none-elf and bootstrapped on aarch64-linux.
Ok for trunk?
Thanks,
On Wed, Aug 06, 2014 at 10:26:29AM +0200, Jakub Jelinek wrote:
Well, if you are going to implement it in fwprop AND fold-const, then the
natural place to fix that would be in *.pd on the match-and-simplify branch.
True. So I guess I'll have to put this one on hold for a while...
Marek
On Wed, 6 Aug 2014, Marek Polacek wrote:
On Tue, Aug 05, 2014 at 02:14:21PM -0600, Jeff Law wrote:
My concern is the code we're removing discusses the need to simplify when
these expressions are in static initializers. What's going to ensure that
we're still simplifying instances which
On Wed, 6 Aug 2014, Marek Polacek wrote:
On Wed, Aug 06, 2014 at 10:26:29AM +0200, Jakub Jelinek wrote:
Well, if you are going to implement it in fwprop AND fold-const, then the
natural place to fix that would be in *.pd on the match-and-simplify branch.
True. So I guess I'll have to put
On Wed, Aug 06, 2014 at 10:28:13AM +0200, Richard Biener wrote:
On Wed, 6 Aug 2014, Marek Polacek wrote:
On Wed, Aug 06, 2014 at 10:26:29AM +0200, Jakub Jelinek wrote:
Well, if you are going to implement it in fwprop AND fold-const, then the
natural place to fix that would be in *.pd on
On 05/08/14 22:29, Jeff Law wrote:
On 08/03/14 08:02, Richard Sandiford wrote:
This also fixes what I think is a bug: find_memory used to stop at the
first MEM it found. If that MEM was nonvolatile and nonconstant, we'd
return MEMREF_NORMAL even if there was another volatile MEM.
gcc/
On 06/08/14 06:54, Hale Wang wrote:
Refer to: https://gcc.gnu.org/ml/gcc-patches/2014-06/msg01429.html.
Sorry for an extra whitespace.
-Original Message-
From: gcc-patches-ow...@gcc.gnu.org [mailto:gcc-patches-
ow...@gcc.gnu.org] On Behalf Of Hale Wang
Sent: 2014年8月6日 13:50
To:
On Wed, Aug 6, 2014 at 3:28 AM, tsaund...@mozilla.com wrote:
From: Trevor Saunders tsaund...@mozilla.com
hi,
just what it says on the tin.
bootstrapped + regtested on x86_64-unknown-linux-gnu, also bootstrapped on
i686-unknown-linux-gnu, ran config-list.mk, ok? gcc/
Ok.
Time to remove
On Wed, Aug 6, 2014 at 3:28 AM, tsaund...@mozilla.com wrote:
From: Trevor Saunders tsaund...@mozilla.com
hi,
just what it says on the tin.
bootstrapped + regtested on x86_64-unknown-linux-gnu, also bootstrapped on
i686-unknown-linux-gnu, ran config-list.mk, ok? gcc/
Ok.
Thanks,
Hi!
I've cleaned up the testcase some more, tested on 4.8/4.9/trunk that
it fails without the sched-deps.c fix too (both -m32 and -m64) and
works with the fix. Committed to all branches.
2014-08-06 Jakub Jelinek ja...@redhat.com
PR rtl-optimization/61801
*
The following fixes the remaining ICEs I see when testing all
languages (but ada and go).
The tree-cfg.c hunk highlights one change in the behavior
of fold_stmt, namely that it now follows SSA edges by default.
Maybe that's undesired? On a related note, fold_stmt_inplace
preserves the actual
On 5 August 2014 21:29, Siva Chandra wrote:
Hi Jonathan,
Thanks a lot for taking a look. The patch in question, and the GDB
support, do not yet work with Python3. If that is a necessary
requirement, I can make the changes and send a new version of the
patch.
Some GNU/Linux distros already
Forgot the patch~
On Wed, Aug 6, 2014 at 10:32 AM, Bin.Cheng amker.ch...@gmail.com wrote:
On Fri, Jul 25, 2014 at 8:35 PM, Richard Biener
richard.guent...@gmail.com wrote:
On Thu, Jul 17, 2014 at 11:08 AM, Bin Cheng bin.ch...@arm.com wrote:
Hi,
As quoted from the function
On Wed, 6 Aug 2014, Richard Biener wrote:
The following fixes the remaining ICEs I see when testing all
languages (but ada and go).
The tree-cfg.c hunk highlights one change in the behavior
of fold_stmt, namely that it now follows SSA edges by default.
Maybe that's undesired? On a
On Wed, 6 Aug 2014, David Edelsohn wrote:
D'oh, there's even a predicate procedure in our test framework already to
cover it. Thanks for straightening me out, an updated patch follows.
This scores:
UNSUPPORTED: gcc.dg/pr44194-1.c
in my testing, like the previous version. OK to
Hi,
The issue was firstly observed on NDK gcc since TLS is not supported
in Android bionic. I also see the same failure on gcc configured for
linux with –disable-tls, libgomp make check log:
FAIL: libgomp.c/affinity-1.c execution test
FAIL: libgomp.c/icv-2.c execution test
FAIL:
Fixed with the following, committed. Incidentially this is the
only constant folding pattern that also applies to floats - otherwise
the use of the integer_* predicates prevent that.
Richard.
2014-08-06 Richard Biener rguent...@suse.de
* match-constant-folding.pd (minus @0 @0):
Here's a patch for the more conservative option.
- Braden Obrzut
2014-08-06 Braden Obrzut ad...@maniacsvault.net
* pt.c (check_explicit_specialization): Ensure tmpl is a function
template before checking if it is inline for COMDAT.
diff --git a/gcc/cp/pt.c b/gcc/cp/pt.c
index
The following fixes PR61320 - we were not properly treating
explicitely misaligned loads as misaligned.
Tested by various people on their STRICT_ALIGN targets, applied
to trunk and branch.
Richard.
2014-08-06 Richard Biener rguent...@suse.de
PR tree-optimization/61320
*
The only thing that I don't like about that, is that the user would
still have stdio.h fixed if gcc is built with sysroots older than r10.
But I guess it is not that critical :)
We still have to remove fix for compiler.h:
diff --git a/fixincludes/ChangeLog b/fixincludes/ChangeLog
index
On Wed, Aug 06, 2014 at 11:03:02AM +0100, Maciej W. Rozycki wrote:
On Wed, 6 Aug 2014, David Edelsohn wrote:
D'oh, there's even a predicate procedure in our test framework already to
cover it. Thanks for straightening me out, an updated patch follows.
This scores:
UNSUPPORTED:
On 08/03/14 17:44, Mircea Namolaru wrote:
2014-08-03 Mircea Namolarumircea.namol...@inria.fr
Replacement of isl-int by isl_val
* graphite-clast-to-gimple.c: include isl/val.h, isl/val_gmp.h
(compute_bounds_for_param): use isl_val instead of isl_int
On Tue, Aug 5, 2014 at 4:21 PM, Jakub Jelinek ja...@redhat.com wrote:
On Tue, Aug 05, 2014 at 04:17:41PM +0200, Richard Biener wrote:
what's the semantic of setting SRP_SIGNED_AND_UNSIGNED
on the subreg? That is, for the created (subreg:lhs_mode
(reg:PROMOTE_MODE of ssa N))?
Rewrite the constraint-checking code so that it doesn't instantiate
concept definitions. Instead of doing a simple constexpr evaluation on
the associated constraints, this now iterates over the decomposed
assumptions to determine satisfaction.
2014-08-06 Andrew Sutton andrew.n.sut...@gmail.com
On Wed, Aug 6, 2014 at 8:34 AM, Zhenqiang Chen
zhenqiang.c...@linaro.org wrote:
On 5 August 2014 21:59, Richard Biener richard.guent...@gmail.com wrote:
On Mon, Aug 4, 2014 at 11:09 AM, Zhenqiang Chen zhenqiang.c...@arm.com
wrote:
-Original Message-
From: Bin.Cheng
On 06/08/14 22:09, Richard Biener wrote:
On Tue, Aug 5, 2014 at 4:21 PM, Jakub Jelinek ja...@redhat.com wrote:
On Tue, Aug 05, 2014 at 04:17:41PM +0200, Richard Biener wrote:
what's the semantic of setting SRP_SIGNED_AND_UNSIGNED
on the subreg? That is, for the created (subreg:lhs_mode
On Wed, Aug 6, 2014 at 3:21 PM, Kugan kugan.vivekanandara...@linaro.org wrote:
On 06/08/14 22:09, Richard Biener wrote:
On Tue, Aug 5, 2014 at 4:21 PM, Jakub Jelinek ja...@redhat.com wrote:
On Tue, Aug 05, 2014 at 04:17:41PM +0200, Richard Biener wrote:
what's the semantic of setting
On Wed, Aug 6, 2014 at 2:47 AM, Jonathan Wakely jwakely@gmail.com wrote:
Some GNU/Linux distros already build GDB using Python3, so they will
be unable to use these xmethods.
However, I think it can be committed now and fixed later.
For the libstdc++ side, it is a very small fix (using
The following avoids recursing for popping SCCs.
LTO bootstrapped and tested on x86_64-unknown-linux-gnu, applied.
Richard.
2014-08-06 Richard Biener rguent...@suse.de
PR lto/62034
* lto-streamer-in.c (lto_input_tree_1): Assert we do not read
SCCs here.
Wasn't done properly.
Committed.
Richard.
2014-08-06 Richard Biener rguent...@suse.de
* tree-ssa-forwprop.c (pass_forwprop::execute): Properly
clean EH info on folded stmts.
Index: gcc/tree-ssa-forwprop.c
===
Hi,
On Wed, Jul 30, 2014 at 06:56:05PM +0200, Martin Jambor wrote:
Hi,
IPA-CP can wreck havoc to transactional memory support as described in
the summary of the PR in bugzilla. It seems the cause is that IPA-CP
clones of nodes created by trans-mem do not have their tm_clone flag
set. For
Hi,
On Wed, Aug 6, 2014 at 4:51 AM, Alexander Ivchenko aivch...@gmail.com wrote:
We still have to remove fix for compiler.h:
Correct. Thank you.
Bruce, I think I formally have to ask for your approval again :)
I don't think so. You've selected one of the changes we wrote about,
so With
On 07/30/2014 04:56 PM, Marat Zakirov wrote:
On 07/23/2014 05:33 PM, Marat Zakirov wrote:
Hi all!
This is a friendly reminder message.
On 07/17/2014 03:22 PM, Marat Zakirov wrote:
On 07/16/2014 01:32 PM, Kyrill Tkachov wrote:
On 16/07/14 10:22, Marat Zakirov wrote:
Christophe,
Please
As discussed in the other thread, I can't just remove folding from the
C FE and implement it on GIMPLE level, because that regressed some of
those not-really-kosher static initializers. Instead, fold-const.c
has to be taught how to fold PTR0 - (PTR1 p+ A). (Now it sounds so
obvious.) I added
This is OK thanks.
Ramana
This avoids shadowing the outer ops[] array with inner ones
during code generation. This makes the results of inner
expressions disappear ... oops.
Applied.
Richard.
2014-08-06 Richard Biener rguent...@suse.de
* genmatch.c (gen_transform): Add depth parameter everywhere.
We fold stmts from non-SSA so when we simplify a single stmt
into multiple ones (like strcat (x, foo) - memcpy (x + strlen (foo),
)) then gimple_build fails because it unconditionally builds
SSA names.
Fixed.
Richard.
2014-08-06 Richard Biener rguent...@suse.de
* gimple-fold.c
$subject, applied.
Richard.
2014-08-06 Richard Biener rguent...@suse.de
* gimple-fold.h (gimple_simplify): Add two-parameter builtin
function overload.
* gimple-match-head.c (gimple_simplify): Implement it.
Index: gcc/gimple-fold.h
On Wed, 6 Aug 2014, Richard Biener wrote:
$subject, applied.
Err, too fast. Fixed.
Richard.
2014-08-06 Richard Biener rguent...@suse.de
* gimple-match-head.c (gimple_simplify): Fix implementation.
Index: gcc/gimple-match-head.c
On 06/08/14 15:14, Ramana Radhakrishnan wrote:
This is OK thanks.
Ramana
Hmm, minor nit.
(define_insn *thumb1_movhi_insn
[(set (match_operand:HI 0 nonimmediate_operand =l,l,m,*r,*h,l)
- (match_operand:HI 1 general_operand l,m,l,*h,*r,I))]
+ (match_operand:HI 1
On 08/05/2014 10:48 AM, Paolo Carlini wrote:
+ (VOID_TYPE_P (TREE_TYPE (type1))
+ || comptypes (TYPE_MAIN_VARIANT (TREE_TYPE (type0)),
+TYPE_MAIN_VARIANT (TREE_TYPE (type1)),
+
I've tested the modified version of Graphite using the gcc test suite
and haven't found out new failed tests.
However, pr35356-2.c is still not suitable for testing. The ISL AST
generated from its source code doesn't contain MIN or MAX.
if (k = -1) {
for (int c1 = 0; c1 n; c1 += 1)
Hi Tobias,
I've attached the patch, which should eliminate CLooG library
installation dependency from GCC. The CLooG AST generator is still the
main code generator, but the isl ast generator will be chosen in case
of nonavailability of CLooG library.
However, I've found out a problem. Almost all
On 06/08/2014 17:21, Roman Gareev wrote:
I've tested the modified version of Graphite using the gcc test suite
and haven't found out new failed tests.
However, pr35356-2.c is still not suitable for testing. The ISL AST
generated from its source code doesn't contain MIN or MAX.
if (k = -1)
The following patch fixes PR61923. The details of the problem can be
found on
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=61923
The patch was tested and bootstrapped on x86/x86-64.
The patch has been committed as rev. 213674 to gcc-4.9 branch and as
rev. 213675 to the trunk.
2014-08-06
On 06/08/2014 17:21, Roman Gareev wrote:
Hi Tobias,
I've attached the patch, which should eliminate CLooG library
installation dependency from GCC. The CLooG AST generator is still the
main code generator, but the isl ast generator will be chosen in case
of nonavailability of CLooG library.
On 08/06/14 02:27, Richard Biener wrote:
which we may restrict better with checking whether the pointer
uses a partial integer mode. Not sure how PSImode - SImode
extends on RTL?
Well, at least on the mn102, I defined both a zero and sign extension
for PSI - SI. So whichever one the generic
On Wed, 6 Aug 2014, Marek Polacek wrote:
Applied, thanks.
I see
ERROR: gcc.dg/pr44194-1.c: unknown dg option: \} for }
now (x86_64-unknown-linux-gnu).
I pushed a stale version of the patch, sorry. Fixed up now.
2014-08-06 Maciej W. Rozycki ma...@codesourcery.com
*
This is the first step for moving Fortran to use the common
diagnostics machinery. This patch makes Fortran use the common
machinery for those warnings that don't have a location or a
controlling option.
Before:
manuel@gcc10:~/test1$ echo end | ./213518/build/gcc/f951
-fdiagnostics-color -Werror
Hi,
On 08/06/2014 05:19 PM, Jason Merrill wrote:
On 08/05/2014 10:48 AM, Paolo Carlini wrote:
+(VOID_TYPE_P (TREE_TYPE (type1))
+ || comptypes (TYPE_MAIN_VARIANT (TREE_TYPE (type0)),
+ TYPE_MAIN_VARIANT (TREE_TYPE (type1)),
+
gcc/
* rtl.h (JUMP_LABEL): Add a note that this isn't always a LABEL.
---
gcc/rtl.h | 6 +-
1 file changed, 5 insertions(+), 1 deletion(-)
diff --git a/gcc/rtl.h b/gcc/rtl.h
index 51cfae5..b9b069a 100644
--- a/gcc/rtl.h
+++ b/gcc/rtl.h
@@ -1194,7 +1194,11 @@ enum label_kind
/* In
This gives a slight improvement in typesafety in cfgexpand.c
gcc/
* cfgexpand.c (lab_rtx_for_bb): Convert from pointer_map_t to
pointer_maprtx.
(label_rtx_for_bb): Update for conversion of lab_rtx_for_bb to
a pointer_maprtx, eliminating casts from void* to rtx.
DF_REF_INSN looks up the insn field of the referenced df_insn_info.
This will eventually be an rtx_insn *, but for now is just an rtx.
As further scaffolding: for now, convert DF_REF_INSN to a function,
adding a checked downcast to rtx_insn *. This can eventually be
converted back to macro when
gcc/
* config/mn10300/mn10300.c (mn10300_adjust_sched_cost): Fix the
handling of PARALLEL to work on PATTERN (insn) and PATTERN (dep),
rather than just on insn, dep themselves. The latter are insns,
and thus can't be PARALLEL.
---
gcc/config/mn10300/mn10300.c | 8
This is the patch series I spoke about at Cauldron in the talk
A proposal for typesafe RTL; slides here:
http://dmalcolm.fedorapeople.org/presentations/cauldron-2014/rtl
They can also be seen at:
https://dmalcolm.fedorapeople.org/gcc/patch-backups/rtx-classes/v20/
The aim of the patch series is
This is an enabling patch, splitting existing macros in two, covering
the rvalue and lvalue uses separately.
Followup patches will replace these with functions, and gradually convert
the types from rtx to rtx_insn *, but we need to do this separately for
the lvalue vs rvalue use-cases, hence this
For now, convert DEP_PRO and DEP_CON into functions. We will eventually
change them back to macros once the relevant fields are of type
rtx_insn *.
gcc/
* sched-int.h (DEP_PRO): struct _dep's pro and con fields will
eventually be rtx_insn *, but to help with transition, for now,
gcc/
* rtl.h (entry_of_function): Strengthen return type from rtx to
rtx_insn *.
* cfgrtl.c (entry_of_function): Likewise.
---
gcc/cfgrtl.c | 2 +-
gcc/rtl.h| 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/gcc/cfgrtl.c b/gcc/cfgrtl.c
index
These should all eventually require an rtx_insn * as an argument,
but we'll save that for a later patch.
gcc/
* rtl.h (previous_insn): Strengthen return type from rtx to
rtx_insn *.
(next_insn): Likewise.
(prev_nonnote_insn): Likewise.
gcc/
* rtl.h (gen_label_rtx): Strengthen return type from rtx to
rtx_code_label *.
* emit-rtl.c (gen_label_rtx): Likewise.
---
gcc/emit-rtl.c | 7 ---
gcc/rtl.h | 2 +-
2 files changed, 5 insertions(+), 4 deletions(-)
diff --git a/gcc/emit-rtl.c b/gcc/emit-rtl.c
This commit is a placeholder for me when rebasing, to help organize the
patch kit.
/
* rtx-classes-status.txt: New file
---
rtx-classes-status.txt | 9 +
1 file changed, 9 insertions(+)
create mode 100644 rtx-classes-status.txt
diff --git a/rtx-classes-status.txt
Ultimately, the underlying fields should become rtx_insn *, but for now we
can do this with a checked cast.
Note to self:
config/m32c/m32c.c: m32c_leaf_function_p directly manipulates
x_first_insn and x_last_insn, using sequence_stack.
gcc/
* emit-rtl.h (get_insns): Strengthen return
gcc/
* asan.h (asan_emit_stack_protection): Strengthen return type from
rtx to rtx_insn *.
* asan.c (asan_emit_stack_protection): Likewise. Add local
insns to hold the return value.
---
gcc/asan.c | 7 ---
gcc/asan.h | 4 ++--
2 files changed, 6 insertions(+),
gcc/
* coretypes.h (class rtx_insn): Add forward declaration.
* rtl.h: Include is-a.h
(struct rtx_def): Add dummy desc and tag GTY options as a
workaround to ensure gengtype knows inheritance is occurring,
whilst continuing to use the pre-existing
This is further scaffolding; convert the BB_* and SET_BB_* macros
into functions. Convert the BB_* rvalue-style functions into returning
rtx_insn * rather than plain rtx.
For now, this is done by adding a checked cast, but this will eventually
become a field lookup. The lvalue form for now
gcc/
* rtl.h (for_each_rtx_in_insn): New function.
* rtlanal.c (for_each_rtx_in_insn): Likewise.
---
gcc/rtl.h | 1 +
gcc/rtlanal.c | 16
2 files changed, 17 insertions(+)
diff --git a/gcc/rtl.h b/gcc/rtl.h
index 0858230..3e37ed0 100644
--- a/gcc/rtl.h
+++
gcc/
* basic-block.h (create_basic_block_structure): Strengthen third
param bb_note from rtx to rtx_note *.
* rtl.h (emit_note_before): Strengthen return type from rtx to
rtx_note *.
(emit_note_after): Likewise.
(emit_note): Likewise.
With this patch, insn and curr_insn as used from C++ fragments in .md
files are strengthened from rtx to rtx_insn *, allowing numerous
target-specific functions to have their params be similiar strengthened.
The top-level interfaces (recog, split, peephole2) continue to take
a plain rtx for insn,
More scaffolding: strengthen the return types from the various emit_
functions from rtx to rtx_insn * (or to the rtx_barrier * subclass in a
few cases).
These will ultimately have their params strengthened also, but we
postpone that until much later in the patch series. So for now there
are also
gcc/
* rtl.h (next_cc0_user): Strengthen return type from rtx to
rtx_insn *.
(prev_cc0_setter): Likewise.
* emit-rtl.c (next_cc0_user): Strengthen return type from rtx to
rtx_insn *, adding checked casts for now as necessary.
(prev_cc0_setter):
gcc/
* rtl.h (try_split): Strengthen return type from rtx to rtx_insn *.
* emit-rtl.c (try_split): Likewise, also for locals before and
after. For now, don't strengthen param trial, which requires
adding checked casts when returning it.
---
gcc/emit-rtl.c | 12
gcc/
* alias.c (init_alias_analysis): Strengthen local insn from rtx
to rtx_insn *.
---
gcc/alias.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/gcc/alias.c b/gcc/alias.c
index 0246dd7..903840c 100644
--- a/gcc/alias.c
+++ b/gcc/alias.c
@@ -2840,7 +2840,8
This commit is a placeholder for me when rebasing, to help organize the
patch kit.
/
* rtx-classes-status.txt: Update
---
rtx-classes-status.txt | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/rtx-classes-status.txt b/rtx-classes-status.txt
index 52567e7..e350eaf
gcc/
* calls.c (emit_call_1): Strengthen local call_insn from rtx to
rtx_insn *.
(internal_arg_pointer_exp_state): Likewise for field scan_start.
(internal_arg_pointer_based_exp_scan): Likewise for locals insn,
scan_start.
(load_register_parameters):
gcc/
* cfgloop.c (loop_exits_from_bb_p): Strengthen local insn from
rtx to rtx_insn *.
---
gcc/cfgloop.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/gcc/cfgloop.c b/gcc/cfgloop.c
index 73f79ef..6d1fe8d 100644
--- a/gcc/cfgloop.c
+++ b/gcc/cfgloop.c
@@
gcc/
* cfgbuild.c (make_edges): Strengthen local insn from rtx to
rtx_insn *.
(purge_dead_tablejump_edges): Likewise.
(find_bb_boundaries): Likewise for locals insn, end,
flow_transfer_insn.
---
gcc/cfgbuild.c | 14 --
1 file changed, 8
gcc/
* dce.c (worklist): Strengthen from vecrtx to vecrtx_insn *.
(deletable_insn_p): Strengthen param insn from rtx to
rtx_insn *. Add checked cast to rtx_call_insn when invoking
find_call_stack_args, since this is guarded by CALL_P (insn).
gcc/
* compare-elim.c (struct comparison_use): Strengthen field insn
from rtx to rtx_insn *.
(struct comparison): Likewise, also for field prev_clobber.
(conforming_compare): Likewise for param insn.
(arithmetic_flags_clobber_p): Likewise.
gcc/
* df-core.c (df_bb_regno_first_def_find): Strengthen local insn
from rtx to rtx_insn *.
(df_bb_regno_last_def_find): Likewise.
* df-problems.c (df_rd_bb_local_compute): Likewise.
(df_lr_bb_local_compute): Likewise.
(df_live_bb_local_compute):
This is an example of strengthening rtx. For example, we
now have strong enough types provided by the existing scaffolding to
turn insn and insns in this:
for (insn = insns; insn; insn = NEXT_INSN (insn))
from plain rtx into rtx_insn *.
gcc/
* asan.c (asan_clear_shadow): Strengthen
gcc/
* fwprop.c (single_def_use_dom_walker::before_dom_children):
Strengthen local insn from rtx to rtx_insn *.
(use_killed_between): Likewise for param target_insn.
(all_uses_available_at): Likewise for param target_insn and
local next.
gcc/
* haifa-sched.c (bb_header): Strengthen from rtx * to rtx_insn **.
(add_delay_dependencies): Strengthen local pro from rtx to
rtx_insn *.
(recompute_todo_spec): Likewise.
(dep_cost_1): Likewise for locals insn, used.
(schedule_insn): Likewise
gcc/
* internal-fn.c (ubsan_expand_si_overflow_addsub_check):
Strengthen locals done_label, do_error from rtx to
rtx_code_label *.
(ubsan_expand_si_overflow_addsub_check): Strengthen local last
from rtx to rtx_insn *. Strengthen local sub_check from rtx to
gcc/
* loop-doloop.c (doloop_valid_p): Strengthen local insn from rtx
to rtx_insn *.
(add_test): Likewise for locals seq, jump.
(doloop_modify): Likewise for locals sequence, jump_insn.
---
gcc/loop-doloop.c | 9 +
1 file changed, 5 insertions(+), 4
gcc/
* predict.h (predict_insn_def): Strengthen param insn from rtx
to rtx_insn *.
* function.c (stack_protect_epilogue): Add checked cast to
rtx_insn for now when invoking predict_insn_def.
* predict.c (predict_insn): Strengthen param insn from rtx to
gcc/
* lra-int.h (struct lra_insn_recog_data): Strengthen field insn
from rtx to rtx_insn *.
(lra_push_insn): Likewise for 1st param.
(lra_push_insn_and_update_insn_regno_info): Likewise.
(lra_pop_insn): Likewise for return type.
gcc/
* basic-block.h (basic_block split_edge_and_insert): Strengthen
param insns from rtx to rtx_insn *.
* loop-unroll.c (struct iv_to_split): Strengthen field insn from
rtx to rtx_insn *.
(struct iv_to_split): Likewise.
(loop_exit_at_end_p):
gcc/
* resource.c (next_insn_no_annul): Strengthen local next from
rtx to rtx_insn *.
(mark_referenced_resources): Likewise for local insn.
---
gcc/resource.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/gcc/resource.c b/gcc/resource.c
index
gcc/
* regstat.c (regstat_bb_compute_ri): Strengthen local insn from
rtx to rtx_insn *.
(regstat_bb_compute_calls_crossed): Likewise.
---
gcc/regstat.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/gcc/regstat.c b/gcc/regstat.c
index
gcc/
* ree.c (struct ext_cand): Strengthen field insn from rtx to
rtx_insn *.
(combine_set_extension): Likewise for param curr_insn.
(transform_ifelse): Likewise for param def_insn.
(get_defs): Likewise for param def_insn. Strengthen param dest
from
gcc/
* regcprop.c (struct queued_debug_insn_change): Strengthen field
insn from rtx to rtx_insn *.
(replace_oldest_value_reg): Likewise for param insn.
(replace_oldest_value_addr): Likewise.
(replace_oldest_value_mem): Likewise.
gcc/
* shrink-wrap.h (requires_stack_frame_p): Strengthen param 1
insn from rtx to rtx_insn *.
(dup_block_and_redirect): Likewise for param 3 before.
* shrink-wrap.c (requires_stack_frame_p): Strengthen param insn
from rtx to rtx_insn *.
1 - 100 of 314 matches
Mail list logo