On Sat, Jan 23, 2016 at 12:28 PM, Tom de Vries wrote:
> That was my original patch, and Richard commented: 'I think avoiding a NULL
> access_fns is ok but it should be done unconditionally, not only for the
> DECL_P case'. In order words, he asked me to do the exact opposite of the
> change you no
---
gcc/testsuite/gcc.dg/graphite/pr69292.c | 19 +++
1 file changed, 19 insertions(+)
create mode 100644 gcc/testsuite/gcc.dg/graphite/pr69292.c
diff --git a/gcc/testsuite/gcc.dg/graphite/pr69292.c
b/gcc/testsuite/gcc.dg/graphite/pr69292.c
new file mode 100644
index 000..b9
The patch disables all fuse-*.c tests when configuring gcc with isl 0.14 or
earlier.
ChangeLog:
* Makefile.in: Regenerate.
* Makefile.tpl: Export ISLVER.
* configure: Regenerate.
* config/isl.m4: Detect isl-0.15.
gcc/
* Makefile.in: Set ISLVER in site.ex
Keep unchanged the implementation for isl-0.14.
* graphite-poly.c (apply_poly_transforms): Simplify.
(print_isl_set): Use more readable format: ISL_YAML_STYLE_BLOCK.
(print_isl_map): Same.
(print_isl_union_map): Same.
(print_isl_schedule): New.
(debu
Tom de Vries wrote:
> diff --git a/gcc/tree-data-ref.c b/gcc/tree-data-ref.c
> index a40f40d..4c29fc2 100644
> --- a/gcc/tree-data-ref.c
> +++ b/gcc/tree-data-ref.c
> @@ -1510,8 +1510,9 @@ initialize_data_dependence_relation (struct
> data_reference *a,
>if (operand_equal_p (DR_REF (a), DR_REF
2016-01-26 Abderrazek Zaafrani
Sebastian Pop
* graphite-isl-ast-to-gimple.c (ternary_op_to_tree): Handle
isl_ast_op_cond and isl_ast_op_select.
(gcc_expression_from_isl_expr_op): Same.
* gcc.dg/graphite/isl-ast-op-select.c: New.
---
gcc/graphite
* graphite-isl-ast-to-gimple.c (get_rename_from_scev): Assert instead of
setting codegen_error to fail codegen.
---
gcc/graphite-isl-ast-to-gimple.c | 26 ++
1 file changed, 6 insertions(+), 20 deletions(-)
diff --git a/gcc/graphite-isl-ast-to-gimple.c b/gc
* graphite-isl-ast-to-gimple.c (class translate_isl_ast_to_gimple):
Remove comments from class declarations: they are already in the code
close by the defs.
---
gcc/graphite-isl-ast-to-gimple.c | 437 ---
1 file changed, 83 insertions(+),
* graphite-isl-ast-to-gimple.c (binary_op_to_tree): Call
codegen_error_p.
(ternary_op_to_tree): Same.
(unary_op_to_tree): Same.
(nary_op_to_tree): Same.
(gcc_expression_from_isl_expr_op): Same.
(gcc_expression_from_isl_expression): Same.
(gra
2016-01-28 Abderrazek Zaafrani
* graphite-optimize-isl.c (optimize_isl): Print a different debug
message when isl does not return a valid schedule.
---
gcc/graphite-optimize-isl.c | 11 +--
1 file changed, 9 insertions(+), 2 deletions(-)
diff --git a/gcc/graphite-optim
* config/isl.m4: Add comments about isl-0.16.
* configure: Regenerate.
gcc/
* doc/install.texi: Document that isl-0.16 is supported.
---
config/isl.m4| 6 +++---
configure| 12 ++--
gcc/doc/install.texi | 2 +-
3 files changed, 10 insertions(+
This fixes bootstrap of GCC with BOOT_CFLAGS="-g -O2 -fgraphite-identity
-floop-nest-optimize -floop-block -floop-interchange -floop-strip-mine".
It passes regstrap on amd64-linux.
Ok to commit to trunk?
Thanks.
2015-07-15 Aditya Kumar
Sebastian Pop
PR middle
gcc/ChangeLog:
2015-07-16 Aditya Kumar
Sebastian Pop
* common.opt (floop-fuse): New.
* doc/invoke.texi (floop-fuse): Documented.
* graphite-optimize-isl.c (optimize_isl): Use
ISL_SCHEDULE_FUSE_MAX when using flag_loop_fuse.
* graphite
Tom de Vries wrote:
> >>>graphite dependence analysis is too slow to be enabled unconditionally.
> >>>(read: hours in some simple cases - see bugzilla)
> >>
> >>Haha, "cool"! ;-)
> >>
> >>Maybe it is still reasonable to use graphite to analyze the code inside
> >>OpenACC kernels regions -- maybe s
Tom de Vries wrote:
> So I wondered, why not always use the graphite dependency analysis
> in parloops. (Of course you could use -floop-parallelize-all, but
> that also changes the heuristic). So I wrote a patch for parloops to
> use graphite dependency analysis by default (so without
> -floop-para
Regstrapped on amd64-linux. Ok for trunk?
Thanks,
Sebastian
2015-07-20 Sebastian Pop
* Makefile.in: Remove use of TREEBROWSER.
* config.in: Regenerated.
* configure: Regenerated.
* configure.ac: Remove definition of TREEBROWSER
Tom de Vries wrote:
> Fix reduction safety checks
>
> * graphite-sese-to-poly.c (is_reduction_operation_p): Limit
> flag_associative_math to SCALAR_FLOAT_TYPE_P. Honour
> TYPE_OVERFLOW_TRAPS and TYPE_OVERFLOW_WRAPS for INTEGRAL_TYPE_P.
> Only allow wrapping fixed-point oth
omebody with access to sourceware.org upload a tar.bz2 of the required
version of isl from http://isl.gforge.inria.fr/isl-0.15.tar.bz2?
Also, once that is done, I will commit the following patch updating the
documentation.
Thanks,
Sebastian
>From 189565d5f0cdcb54aac309316316456b88957ef6 Mon
Richard Biener wrote:
> On Wed, Jul 22, 2015 at 6:00 PM, Tom de Vries wrote:
> > Currently bootstrapping and reg-testing on x86_64.
> >
> > OK for trunk?
> >
> > OK 5 and 4.9 release branches?
>
> Ok if Sebastian is fine with it.
Ok to backport as well.
Thanks Tom for the patches.
Jeff Law wrote:
> On 07/21/2015 02:32 PM, Sebastian Pop wrote:
> >
> >Could somebody with access to sourceware.org upload a tar.bz2 of the required
> >version of isl from http://isl.gforge.inria.fr/isl-0.15.tar.bz2?
> >
> >Also, once that is done, I will commit
When trying to analyze why Graphite does not handle a loop nest, it is easy to
look in the dumps of -fdump-tree-graphite-all to guess what has to be changed to
catch the loop. This patch makes the dumps a bit more verbose and useful.
---
gcc/graphite-scop-detection.c | 72
On Sun, Jul 26, 2015 at 4:21 PM, Tom de Vries wrote:
> I wrote an equivalent test-case in C:
> ...
> $ cat src/gcc/testsuite/gcc.dg/autopar/outer-7.c
> /* { dg-do compile } */
> /* { dg-options "-O2 -ftree-parallelize-loops=2 -fdump-tree-parloops-details
> -fdump-tree-optimized" } */
>
> void abor
Aditya Kumar wrote:
> More than 3 params consumes too much memory while bootstrapping gcc
> with graphite enabled.
Ok. I will commit the patch.
Thanks for fixing bootstrap with graphite enabled.
Sebastian
Sebastian Pop wrote:
> Aditya Kumar wrote:
> > More than 3 params consumes too much memory while bootstrapping gcc
> > with graphite enabled.
>
> Ok. I will commit the patch.
> Thanks for fixing bootstrap with graphite enabled.
We will increase the max when we will use
The patch makes the detection of scop parameters in parameter_index_in_region a
bit more conservative by discarding scalar variables defined in function of data
references defined in the scop.
2015-09-25 Aditya Kumar
Sebastian Pop
PR tree-optimization/67700
the data
dependence graph.)
Patch passed bootstrap and check on x86_64-linux with ISL-0.15.
I will commit this patch to trunk.
2015-09-29 Sebastian Pop
Aditya Kumar
* graphite-sese-to-poly.c (gsi_for_phi_node): Remove.
(nb_data_writes_in_bb): Remove
This patch fixes PR66754 by reverting an earlier unintended change.
We now generate a much simpler AST for interchange-1.c:
ISL AST generated by ISL:
{
for (int c1 = 0; c1 <= 1334; c1 += 1) {
S_7(c1);
for (int c3 = 0; c3 <= 1334; c3 += 1)
S_4(c1, c3);
S_5(c1);
}
for (int c1
Tobias Grosser wrote:
> On 09/29/2015 06:26 PM, Sebastian Pop wrote:
> >This code is not used anymore after we removed the previous loop optimizer
> >(not
> >based on the ISL scheduler.) We will add back the detection of commutative
> >reductions after we improve th
We now check that all data references in the current loop and inner loops
contained within loop are valid in an outer region before declaring that the
outer loop is a valid scop.
2015-09-30 Sebastian Pop
Aditya Kumar
PR tree-optimization/67754
* graphite-scop
-linux when compiled with
-m32. Patch passed bootstrap with BOOT_CFLAGS="-g -O2 -fgraphite-identity
-floop-nest-optimize" and check on x86_64-linux using ISL-0.15.
2015-09-28 Sebastian Pop
Aditya Kumar
PR tree-optimization/67754
* graphite-scop-d
PR tree-optimization/66980
* graphite-scop-detection.c (stmt_has_simple_data_refs_p): Return false
when data reference analysis has failed.
---
gcc/graphite-scop-detection.c| 7 +++
gcc/testsuite/gcc.dg/graphite/scop-pr66980.c | 10 ++
2 files c
---
gcc/params.def | 2 +-
gcc/testsuite/gcc.dg/graphite/scop-sor.c | 3 +--
2 files changed, 2 insertions(+), 3 deletions(-)
diff --git a/gcc/params.def b/gcc/params.def
index 3f91992..da2c6a3 100644
--- a/gcc/params.def
+++ b/gcc/params.def
@@ -835,7 +835,7 @@ DEFPARAM
From: hiraditya
---
gcc/graphite-scop-detection.c | 1943 +
gcc/graphite-sese-to-poly.c | 184 +---
2 files changed, 1207 insertions(+), 920 deletions(-)
diff --git a/gcc/graphite-scop-detection.c b/gcc/graphite-scop-detection.c
index 77fe25f..1704265
---
gcc/graphite-poly.c | 68 ++-
gcc/graphite-poly.h | 11 +-
gcc/graphite-scop-detection.c| 217 ++---
gcc/graphite-sese-to-poly.c | 324 ---
gcc/graphite-sese-to-poly.h
---
gcc/graphite-dependences.c | 36
gcc/graphite-poly.h| 3 ---
2 files changed, 39 deletions(-)
diff --git a/gcc/graphite-dependences.c b/gcc/graphite-dependences.c
index e39394a..37535d5 100644
--- a/gcc/graphite-dependences.c
+++ b/gcc/graphite-de
---
gcc/graphite-dependences.c | 255 +
gcc/graphite-poly.h| 3 -
2 files changed, 1 insertion(+), 257 deletions(-)
diff --git a/gcc/graphite-dependences.c b/gcc/graphite-dependences.c
index e39394a..2c4f92c 100644
--- a/gcc/graphite-dependenc
On Mon, Oct 5, 2015 at 6:02 PM, Joseph Myers wrote:
> Is this Samsung Austin R&D Center? If so, yes, there's an assignment
Yes.
2015-10-05 Aditya Kumar
Sebastian Pop
* graphite-dependences.c (scop_get_transformed_schedule):
Remove.
(no_violations): Remove.
(subtract_commutative_associative_deps): Remove.
(compute_deps): Do not call
2015-10-06 Aditya Kumar
Sebastian Pop
* graphite-poly.c (free_data_refs_aux): Remove.
(free_gimple_poly_bb): Do not call free_data_refs_aux.
* graphite-poly.h (struct base_alias_pair): Remove.
* graphite-sese-to
2015-10-06 Aditya Kumar
Sebastian Pop
* graphite-poly.c (new_scop): Initialize drs.
* graphite-poly.h (struct dr_info): New.
(struct scop): Add drs.
* graphite-sese-to-poly.c (pdr_add_alias_set): Use dr_info
2015-10-06 Aditya Kumar
Sebastian Pop
* graphite-poly.c (new_poly_dr): Remove dr_base_object_set.
Do not set PDR_BASE_OBJECT_SET.
* graphite-poly.h (poly_dr): Same.
(PDR_BASE_OBJECT_SET): Remove
2015-10-06 Aditya Kumar
Sebastian Pop
* graphite-isl-ast-to-gimple.c (translate_isl_ast_to_gimple):
Use
an sese_info_p.
(copy_def): Same.
(copy_internal_parameters): Same
On Wed, Oct 7, 2015 at 3:11 AM, Richard Biener
wrote:
> On Tue, Oct 6, 2015 at 10:45 PM, Sebastian Pop wrote:
>> 2015-10-06 Aditya Kumar
>> Sebastian Pop
>>
>> * graphite-poly.c (free_data_refs_aux): Remove.
>>
---
gcc/graphite-scop-detection.c | 3 ---
gcc/sese.c| 1 -
gcc/sese.h| 8 +---
3 files changed, 1 insertion(+), 11 deletions(-)
diff --git a/gcc/graphite-scop-detection.c b/gcc/graphite-scop-detection.c
index 7e5039c..43e03a6 100644
--- a/gcc/graphite-
2015-10-07 Aditya Kumar
Sebastian Pop
* graphite-scop-detection.c (parameter_index_in_region): Remove
use of SESE_ADD_PARAMS.
(find_scop_parameters): Same.
* sese.c (new_sese_info): Same.
* sese.h
2015-10-06 Aditya Kumar
Sebastian Pop
* graphite-dependences.c (scop_get_dependences): Do not use
SCOP_BBS.
* graphite-isl-ast-to-gimple.c (get_max_schedule_dimensions):
Same.
(generate_isl_schedule): Same
will use that vector in a patch that removes the out-of-ssa
translation of scalar dependences: we will iterate through the basic blocks of a
region to record scalar dependences crossing bbs or going out of the region.
The patch passes bootstrap and regtest on x86_64-linux.
Sebastian Pop wrote:
&
Hi Abe,
could you please avoid double negations, and
please use early returns rather than huge right indentations:
+ if (! not_a_scratchpad_candidate)
+ {
+if (MEM_SIZE_KNOWN_P (orig_x))
+{
+ const size_t size_of_MEM = MEM_SIZE (orig_x);
+
+ if (size_of_MEM <= SCRATCHPAD_MAX_S
the
+ THEN block and the set of 'a', if any. */
This comment duplicates the same content as the comment before the loop.
Please remove.
On Thu, Oct 8, 2015 at 8:08 AM, Sebastian Pop wrote:
> Hi Abe,
>
> could you please avoid double negations, and
> please use early return
2015-10-07 Aditya Kumar
Sebastian Pop
* graphite-dependences.c (scop_get_dependences): Add dump of the
data dependence graph.
* graphite-poly.c (print_isl_union_map): New.
(debug_isl_union_map): New
2015-10-07 Aditya Kumar
Sebastian Pop
* graphite-poly.c (print_iteration_domain): Remove verbosity.
Remove OpenScop formatting.
(print_iteration_domains): Same.
(debug_iteration_domain): Same.
(debug_iteration_domains): Same
We used to insert a trivial integer_onep condition around loops we know have
more than an iteration. This patch avoids generating those conditions.
* graphite-isl-ast-to-gimple.c (graphite_create_new_loop_guard): Do not
call
create_empty_if_region_on_edge when cond_expr is true.
* graphite-scop-detection.c (build_scops): Do not handle scops with more
than PARAM_GRAPHITE_MAX_ARRAYS_PER_SCOP arrays.
* params.def (PARAM_GRAPHITE_MAX_ARRAYS_PER_SCOP): New.
---
gcc/graphite-scop-detection.c | 11 +++
gcc/params.def| 7 +++
2
When a scalar is defined in function of an array reference in the current scop,
it does variate. Graphite cannot represent the condition in scop-11.c, as a[*]
variates in the current region, and it is not an affine condition:
for (j = 0; j <= 20; j++)
a[j] = b + i;
if (a[12] == 23)
b
From: Aditya Kumar
- fix printing of ISL stmt and parameter names
- move dot_scop* functions outside of anonymous namespace.
* graphite-isl-ast-to-gimple.c: Include tree-cfg.h.
(translate_isl_ast_node_user): Add more dumps: call print_loops_bb.
* graphite-scop-detection.c (dot_all_scops_1)
This map was used in the transition to the new scop detection: with the new scop
detection, we do not need this map anymore.
* graphite-isl-ast-to-gimple.c (gcc_expression_from_isl_ast_expr_id):
Remove use of parameter_rename_map.
(copy_def): Remove.
(copy_internal_para
* graphite-optimize-isl.c (optimize_isl): Call
isl_options_set_schedule_maximize_band_depth.
* gcc.dg/graphite/fuse-1.c: New.
* gcc.dg/graphite/fuse-2.c: New.
* gcc.dg/graphite/interchange-13.c: Remove bogus check.
---
gcc/graphite-optimize-isl.c
* graphite-scop-detection.c (loop_is_valid_scop): Call
optimize_loop_nest_for_speed_p.
---
gcc/graphite-scop-detection.c | 7 +++
1 file changed, 7 insertions(+)
diff --git a/gcc/graphite-scop-detection.c b/gcc/graphite-scop-detection.c
index ae8497d..f22e9cc 100644
---
On Thu, Nov 5, 2015 at 10:12 PM, Aditya Kumar wrote:
> Irreducible regions are not going to be optimized by ISL
> so discard them early.
>
> gcc/ChangeLog:
>
> 2015-11-06 Aditya Kumar
>
> * graphite-scop-detection.c (scop_detection::merge_sese):
> (scop_detection::can_represent_
On Fri, Nov 6, 2015 at 2:56 AM, Bernd Schmidt wrote:
> Formatting problem, here and in a few other places. I didn't fully read the
> patch this time around.
>
> I'm probably not reviewing further patches because I don't see this
> progressing to a state where it's acceptable. Others may do so, but
On Fri, Nov 6, 2015 at 6:32 AM, Bernd Schmidt wrote:
> On 11/06/2015 03:10 PM, Sebastian Pop wrote:
>>
>> On Fri, Nov 6, 2015 at 2:56 AM, Bernd Schmidt wrote:
>>>
>>> Formatting problem, here and in a few other places. I didn't fully read
>>> the
&
From: Abderrazek Zaafrani
* graphite-optimize-isl.c (optimize_isl): Call isl_union_map_is_equal.
* graphite-poly.c (new_scop): Initialize original_schedule.
(free_scop): Free original_schedule.
* graphite-poly.h (struct scop): Add field original_schedule.
* graphite-sese-to-po
* graphite-sese-to-poly.c (build_pbb_scattering_polyhedrons): Remove.
(build_pbb_minimal_scattering_polyhedrons): New.
(build_scop_scattering): Remove.
(build_scop_minimal_scattering): New.
(build_scop_scattering): Call
build_pbb_minimal_scatt
The patch builds the original schedule based on the now optimized scattering
dimension instead of building one based on the loop index only.
The implementation is simpler and catches more cases where the original schedule
and the transformed schedule are the same, such as the one below:
for (i =
---
gcc/testsuite/gcc.dg/graphite/fuse-1.c | 10 +++---
gcc/testsuite/gcc.dg/graphite/fuse-2.c | 4 +---
gcc/testsuite/gcc.dg/graphite/graphite.exp | 2 ++
gcc/testsuite/gcc.dg/graphite/isl-ast-gen-blocks-1.c| 3 ---
gcc/testsuite/gcc.dg/gr
On Fri, Nov 13, 2015 at 6:15 AM, Richard Biener
wrote:
> On Fri, Nov 13, 2015 at 12:43 PM, Martin Liška wrote:
>> Hello.
>>
>> Patch survives regbootstrap on x86_64-linux-gnu.
>> Ready for trunk?
Thanks Martin for the patch.
Sebastian
>
> Ok.
>
> Richard.
>
>> Thanks,
>> Martin
On Fri, Nov 13, 2015 at 3:08 PM, Tom de Vries wrote:
> On 11/11/15 23:54, Aditya Kumar wrote:
>>
>> Earlier, graphite used to translate portions of the original program after
>> scop-detection in order to represent the SCoP into polyhedral model. This
>> was
>> required because each basic block w
Martin, thanks for getting this patch out. I like the patch.
Jeff, clang-format has scripts that allow formatting only the lines
touched by a patch.
It also has a script to integrate with git:
https://llvm.org/svn/llvm-project/cfe/trunk/tools/clang-format/git-clang-format
We could use those script
Fixed in r230625.
David, please test on your systems, we were not able to reproduce the fails on
x86_64-linux: the linker does optimize away the functions that are not used.
Thanks,
Sebastian
Aditya K wrote:
> Thanks for the update. I'll fix that asap.
>
>
> -Aditya
>
>
> ---
---
gcc/graphite-isl-ast-to-gimple.c | 96 ++--
1 file changed, 62 insertions(+), 34 deletions(-)
diff --git a/gcc/graphite-isl-ast-to-gimple.c b/gcc/graphite-isl-ast-to-gimple.c
index 3e0907d..91bda33 100644
--- a/gcc/graphite-isl-ast-to-gimple.c
+++ b/gcc/gra
---
gcc/graphite-scop-detection.c | 6 +-
gcc/testsuite/gcc.dg/graphite/pr68428.c | 23 +++
2 files changed, 28 insertions(+), 1 deletion(-)
create mode 100644 gcc/testsuite/gcc.dg/graphite/pr68428.c
diff --git a/gcc/graphite-scop-detection.c b/gcc/graphite-sco
---
gcc/testsuite/gfortran.dg/graphite/pr68335.f90 | 45 ++
1 file changed, 45 insertions(+)
create mode 100644 gcc/testsuite/gfortran.dg/graphite/pr68335.f90
diff --git a/gcc/testsuite/gfortran.dg/graphite/pr68335.f90
b/gcc/testsuite/gfortran.dg/graphite/pr68335.f90
new
On Sat, Nov 21, 2015 at 4:03 PM, David Edelsohn wrote:
> Graphite relies on the ISL library and includes multiple ISL headers.
> The ISL headers refer to identifiers that are poisoned for use in GCC.
> The source files for Graphite were organized to include the ISL
> headers first, to avoid the id
---
gcc/graphite-isl-ast-to-gimple.c| 2 ++
gcc/testsuite/gcc.dg/graphite/pr68493.c | 34 +
2 files changed, 36 insertions(+)
create mode 100644 gcc/testsuite/gcc.dg/graphite/pr68493.c
diff --git a/gcc/graphite-isl-ast-to-gimple.c b/gcc/graphite-isl-ast-t
---
gcc/graphite-isl-ast-to-gimple.c | 9 -
gcc/sese.c | 4
2 files changed, 13 deletions(-)
diff --git a/gcc/graphite-isl-ast-to-gimple.c b/gcc/graphite-isl-ast-to-gimple.c
index 557c44c..30c3a21 100644
--- a/gcc/graphite-isl-ast-to-gimple.c
+++ b/gcc/graphite-
- scop->poly_scop_p = true;
+ return true;
}
#endif /* HAVE_isl */
diff --git a/gcc/graphite-sese-to-poly.h b/gcc/graphite-sese-to-poly.h
deleted file mode 100644
index 48f0d56..0000000
--- a/gcc/graphite-sese-to-poly.h
+++ /dev/null
@@ -1,26 +0,0 @@
-/* Conversion of SESE regions to Polyhed
---
gcc/graphite-optimize-isl.c | 20 +---
gcc/graphite-poly.c | 2 -
gcc/graphite-poly.h | 3 -
gcc/graphite-sese-to-poly.c | 235 ++--
4 files changed, 58 insertions(+), 202 deletions(-)
diff --git a/gcc/graphite-optimize-isl.c b/gcc/
---
gcc/graphite-isl-ast-to-gimple.c| 98 -
gcc/testsuite/gcc.dg/graphite/pr67984.c | 17 ++
2 files changed, 77 insertions(+), 38 deletions(-)
create mode 100644 gcc/testsuite/gcc.dg/graphite/pr67984.c
diff --git a/gcc/graphite-isl-ast-to-gimple.c b/g
h"
+#include "graphite.h"
#define OPENSCOP_MAX_STRING 256
diff --git a/gcc/graphite-poly.h b/gcc/graphite-poly.h
deleted file mode 100644
index d396d3f..000
--- a/gcc/graphite-poly.h
+++ /dev/null
@@ -1,466 +0,0 @@
-/* Graphite polyhedral representation.
- Copyright (C) 2009
The patch adds close phi nodes to every outer loop exit, and to every loop
guard. For loop guards it computes an initial value that determines where we
stop inserting phi nodes. When the initial value is a constant, the initial
value is considered to be defined in the entry of the code gen region
---
gcc/graphite-isl-ast-to-gimple.c | 4 ++--
gcc/graphite-scop-detection.c | 31 ---
gcc/graphite-sese-to-poly.c | 16 +++-
gcc/testsuite/gcc.dg/graphite/pr35356-1.c | 2 +-
4 files changed, 22 insertions(+), 31 deletio
we used to generate modulo and division by zero because ISL uses big numbers
which translate to zero in modulo arithmetic. The patch also improves error
handling
and bails out early in case of wrong code gen.
---
gcc/graphite-isl-ast-to-gimple.c | 85 +++-
1 f
While enabling graphite in -O3 we found a Fortran testcase that fails
because the max of the type domain is -1. We used to add that as a constraint
to the elements accessed by the array, leading to a unfeasible constraint:
0 <= i <= -1. Having that constraint, drops the data reference as that say
---
gcc/common.opt | 2 +-
gcc/graphite-poly.c | 2 +-
gcc/graphite.c | 2 +-
gcc/toplev.c| 2 +-
4 files changed, 4 insertions(+), 4 deletions(-)
diff --git a/gcc/common.opt b/gcc/common.opt
index e1617c4..e593631 100644
--- a/gcc/common.opt
+++ b/gcc/common.opt
@@ -1379,7 +137
In case ISL did some loop peeling, like this:
S_8(0);
for (int c1 = 1; c1 <= 5; c1 += 1) {
S_8(c1);
}
S_8(6);
we should not copy loop-phi nodes in S_8(0) or in S_8(6).
* graphite-isl-ast-to-gimple.c (copy_loop_phi_nodes): Add dump.
(copy_bb_and_scalar_dependences): Do
on the testcase we used to generate code in the function entry bb_0,
and that choked the cfg verifier.
* graphite-isl-ast-to-gimple.c (copy_bb_and_scalar_dependences): Check
that insertion point is still in the region.
* gfortran.dg/graphite/id-26.f03: New.
---
gcc/graphi
>From ISL's documentation, isl_ast_op_zdiv_r is equal to zero iff the remainder
on integer division is zero. Code generate a modulo operation for that.
* graphite-isl-ast-to-gimple.c (binary_op_to_tree): Handle
isl_ast_op_zdiv_r.
(gcc_expression_from_isl_expr_op): Same.
Richard Biener wrote:
> On Wed, Dec 2, 2015 at 10:36 PM, Sebastian Paul Pop wrote:
> > Do you recommend that we add a gcc_assert that min is always lower than max?
>
> No, min can be one less than max if the array has size zero.
Maybe a typo: do you mean max can be one less than min?
If the arr
LGTM. Thanks!
Tobias Burnus wrote:
> If there are no objections, I intent to commit the following patch
> tomorrow as obvious.
>
> Tobias
>
> On Tue, Oct 27, 2015 at 11:27:41AM +0100, Tobias Burnus wrote:
> > recently, support for ISL 0.15 was added to GCC and also
> > ftp://gcc.gnu.org/pub/gcc
ks,
> Richard.
>
> > Thanks,
> > Sebastian
> >
> > -Original Message-----
> > From: Mike Stump [mailto:mikest...@comcast.net]
> > Sent: Friday, December 04, 2015 12:03 PM
> > To: Alan Lawrence
> > Cc: Sebastian Pop; seb...@gmail.com; gcc-pat
---
gcc/graphite-optimize-isl.c | 8
1 file changed, 8 insertions(+)
diff --git a/gcc/graphite-optimize-isl.c b/gcc/graphite-optimize-isl.c
index f90fcfd..50f2b3c 100644
--- a/gcc/graphite-optimize-isl.c
+++ b/gcc/graphite-optimize-isl.c
@@ -50,6 +50,7 @@ along with GCC; see the file COP
we used to add the access functions in the wrong order, Fortran style, leading
to unprofitable interchanges.
---
gcc/graphite-sese-to-poly.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/gcc/graphite-sese-to-poly.c b/gcc/graphite-sese-to-poly.c
index 887c212..480c552 100644
---
gcc/graphite-optimize-isl.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/gcc/graphite-optimize-isl.c b/gcc/graphite-optimize-isl.c
index 8727e39..f90fcfd 100644
--- a/gcc/graphite-optimize-isl.c
+++ b/gcc/graphite-optimize-isl.c
@@ -426,7 +426,7 @@ optimize_isl (scop_p s
we now check the isl version, as there are no real differences in existing files
in between isl 0.14 and isl 0.15.
---
config/isl.m4 | 41 +++--
configure | 112 --
2 files changed, 123 insertions(+), 30 deletions(-)
dif
also update ISL to isl as requested by its author Sven Verdoolaege.
---
gcc/doc/install.texi | 9 +
gcc/doc/invoke.texi | 4 ++--
2 files changed, 7 insertions(+), 6 deletions(-)
diff --git a/gcc/doc/install.texi b/gcc/doc/install.texi
index 0b71bef..b43a3ec 100644
--- a/gcc/doc/install.
On Wed, Jun 7, 2017 at 4:46 AM, Richard Biener wrote:
>
> When the order of visiting dominator children in domwalk changed
> GRAPHITE falls foul of relying on a particular order BBs are visited
> when computing the original schedule from the vector of pbbs.
>
> The following restores an order that
Steve Ellcey wrote:
> +/* This file implements an optimization where, when a variable is set
> + to a constant value and there is a path that leads from that definition
> + to a switch statement that uses that variable as its controlling
> expression
> + we duplicate the blocks on this path
Jeff Law wrote:
> I'm pretty sure jump threading *could* handle it, but after looking
> at the full testcase when it was posted, I'm not sure it's *wise* to
> handle this in jump threading.
Thanks for clearing my doubts.
Sebastian
Bin.Cheng wrote:
> >> The overflow check can be improved by using deeper inspection to prove the
> >> equality. This patch deals with that by making below two improvements:
> >> a) Handles constant cases.
> >> b) Uses affine expansion as deeper inspection to check the equality.
Looks good to
Richard Biener wrote:
>
> The following fixes missing dominator computation before fixing loops.
> Rather than doing even more such weird stuff in a pass gate function
> this puts this into a new pass scheduled before the loop passes gate.
>
Ok.
> +unsigned int
> +pass_fix_loops::execute (funct
1 - 100 of 429 matches
Mail list logo
