public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
From: "spop at gcc dot gnu dot org" <gcc-bugzilla@gcc.gnu.org>
To: gcc-bugs@gcc.gnu.org
Subject: [Bug middle-end/38446] [graphite] The def for a var exists inside one of the scops bb's but an appropriate phi is not created to allow the phi to reach the use of that def ouside the scop.
Date: Thu, 11 Dec 2008 19:38:00 -0000	[thread overview]
Message-ID: <20081211193707.5320.qmail@sourceware.org> (raw)
In-Reply-To: <bug-38446-12919@http.gcc.gnu.org/bugzilla/>



------- Comment #6 from spop at gcc dot gnu dot org  2008-12-11 19:37 -------
Subject: Bug 38446

Author: spop
Date: Thu Dec 11 19:35:41 2008
New Revision: 142688

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=142688
Log:
2008-12-11  Harsha Jagasia  <harsha.jagasia@amd.com>

        PR tree-optimization/38446
        * graphite.c (register_bb_in_sese): New.
        (bb_in_sese_p): Check if bb belongs to sese region by explicitly
        looking at the bbs in the region.
        * graphite.h (sese): Add region_basic_blocks pointer set to
        structure and initialize at the time of defining new scop.

2008-12-11  Tobias Grosser  <grosser@fim.uni-passau.de>

        * graphite.c (new_graphite_bb): Initialize GBB_STATIC_SCHEDULE.
        (find_params_in_bb): Do not free data refs.
        (free_graphite_bb): Add FIXME on disabled free_data_refs.

2008-12-11  Sebastian Pop  <sebastian.pop@amd.com>

        * graphite.c (struct ivtype_map_elt): New.
        (debug_ivtype_elt, debug_ivtype_map_1, debug_ivtype_map,
        new_ivtype_map_elt, ivtype_map_elt_info, eq_ivtype_map_elts,
        gcc_type_for_cloog_iv): New.
        (loop_iv_stack_patch_for_consts): Use the type of the induction
        variable from the original loop, except for the automatically
        generated loops, i.e., in the case of a strip-mined loop, in
        which case there is no original loop: in that case just use
        integer_type_node.
        (new_graphite_bb): Initialize GBB_CLOOG_IV_TYPES.
        (free_graphite_bb): Free GBB_CLOOG_IV_TYPES.
        (clast_name_to_gcc): Accept params to be NULL.
        (clast_to_gcc_expression): Take an extra parameter for the type.
        Convert to that type all the expressions built by this function.
        (gcc_type_for_clast_expr, gcc_type_for_clast_eq): New.
        (graphite_translate_clast_equation): Compute the type of the
        clast_equation before translating its LHS and RHS.
        (clast_get_body_of_loop, gcc_type_for_iv_of_clast_loop): New.
        (graphite_create_new_loop): Compute the type of the induction
        variable before translating the lower and upper bounds and before
        creating the induction variable.
        (rename_variables_from_edge, rename_phis_end_scop): New.
        (copy_bb_and_scalar_dependences): Call rename_phis_end_scop.
        (sese_add_exit_phis_edge): Do not use integer_zero_node.
        (find_cloog_iv_in_expr, compute_cloog_iv_types_1,
        compute_cloog_iv_types): New.
        (gloog): Call compute_cloog_iv_types before starting the
        translation of the clast.

        * graphite.h (struct graphite_bb): New field cloog_iv_types.
        (GBB_CLOOG_IV_TYPES): New.
        (debug_ivtype_map): Declared.
        (oldiv_for_loop): New.

2008-12-10  Tobias Grosser  <grosser@fim.uni-passau.de>

        PR middle-end/38459
        * graphite.c (new_scop): Initialize SCOP_ADD_PARAMS.
        (param_index): Assert if parameter is not know after parameter
        detection.
        (find_params_in_bb): Detect params directly in GBB_CONDITIONS.
        (find_scop_parameters): Mark, that we have finished parameter
        detection.
        (graphite_transform_loops): Move condition detection before parameter
        detection.
        * graphite.h (struct scop): Add SCOP_ADD_PARAMS.

2008-12-11  Harsha Jagasia  <harsha.jagasia@amd.com>

        PR tree-optimization/38446
        * gcc.dg/graphite/pr38446.c: New.

2008-12-11  Sebastian Pop  <sebastian.pop@amd.com>

        * gcc.dg/graphite/scop-16.c: Test only scop specific info.
        * gcc.dg/graphite/scop-17.c: Same.
        * gcc.dg/graphite/block-5.c: New.
        * gcc.dg/graphite/block-6.c: New.
        * gcc.dg/graphite/pr37485.c: Clean dump file after.
        * gcc.dg/graphite/pr37684.c: Same.
        * gcc.dg/graphite/block-2.c: Same.

2008-12-10  Tobias Grosser  <grosser@fim.uni-passau.de>

        PR middle-end/38459
        * gfortran.dg/graphite/pr38459.f90: New.


Added:
    trunk/gcc/testsuite/gcc.dg/graphite/block-2.c
    trunk/gcc/testsuite/gcc.dg/graphite/block-5.c
    trunk/gcc/testsuite/gcc.dg/graphite/block-6.c
    trunk/gcc/testsuite/gcc.dg/graphite/pr38446.c
    trunk/gcc/testsuite/gfortran.dg/graphite/pr38459.f90
Modified:
    trunk/gcc/ChangeLog
    trunk/gcc/graphite.c
    trunk/gcc/graphite.h
    trunk/gcc/testsuite/ChangeLog
    trunk/gcc/testsuite/gcc.dg/graphite/pr37485.c
    trunk/gcc/testsuite/gcc.dg/graphite/pr37684.c
    trunk/gcc/testsuite/gcc.dg/graphite/scop-15.c
    trunk/gcc/testsuite/gcc.dg/graphite/scop-16.c
    trunk/gcc/testsuite/gcc.dg/graphite/scop-17.c


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38446


  parent reply	other threads:[~2008-12-11 19:38 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-12-08 17:26 [Bug c/38446] New: " hjagasia at gcc dot gnu dot org
2008-12-08 17:27 ` [Bug c/38446] " hjagasia at gcc dot gnu dot org
2008-12-10 21:10 ` [Bug middle-end/38446] " hjagasia at gcc dot gnu dot org
2008-12-10 22:35 ` hjagasia at gcc dot gnu dot org
2008-12-10 22:37   ` Sebastian Pop
2008-12-10 22:39 ` sebpop at gmail dot com
2008-12-11 17:29 ` hjagasia at gcc dot gnu dot org
2008-12-11 19:38 ` spop at gcc dot gnu dot org [this message]
2008-12-11 19:42 ` spop at gcc dot gnu dot org

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20081211193707.5320.qmail@sourceware.org \
    --to=gcc-bugzilla@gcc.gnu.org \
    --cc=gcc-bugs@gcc.gnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).