Re: [4.9] PR 62146

2014-09-04 Thread Easwaran Raman
for the test to go? - ChangeLog 2014-09-04 Easwaran Raman era...@google.com PR rtl-optimization/62146 * ifcvt.c (dead_or_predicable): Make removal of REG_EQUAL note of hoisted instruction unconditional. 2014-09-04 Easwaran Raman era...@google.com PR rtl

Re: [4.9] PR 62146

2014-09-02 Thread Easwaran Raman
On Fri, Aug 29, 2014 at 1:06 PM, Jeff Law l...@redhat.com wrote: On 08/25/14 16:42, Easwaran Raman wrote: This patch deletes REG_EQUAL note when a src register is replaced by a constant in an assignment. This is to prevent spurious equivalences between the constant and the expression

Re: [4.9] PR 62146

2014-08-27 Thread Easwaran Raman
On Mon, Aug 25, 2014 at 3:42 PM, Easwaran Raman era...@google.com wrote: This patch deletes REG_EQUAL note when a src register is replaced by a constant in an assignment. This is to prevent spurious equivalences between the constant and the expression in the REG_EQUAL note. In the bug reported

[4.9] PR 62146

2014-08-25 Thread Easwaran Raman
? This patch looks applicable to trunk as well, but I don't have a test case to reproduce the issue in trunk. ChangeLog: 2014-08-25 Easwaran Raman era...@google.com PR rtl-optimization/62146 * cprop.c (try_replace_reg): Remove REG_EQUAL note when a constant

Re: Fix for PR c++/59031

2013-11-15 Thread Easwaran Raman
Ping. On Mon, Nov 11, 2013 at 9:46 AM, Easwaran Raman era...@google.com wrote: Ping. On Thu, Nov 7, 2013 at 11:14 AM, Easwaran Raman era...@google.com wrote: Before r193504, if a method can not be overridden, LOOKUP_NONVIRTUAL is set and the call is direct. The changes at r193504 (to fix

Re: Fix for PR c++/59031

2013-11-11 Thread Easwaran Raman
Ping. On Thu, Nov 7, 2013 at 11:14 AM, Easwaran Raman era...@google.com wrote: Before r193504, if a method can not be overridden, LOOKUP_NONVIRTUAL is set and the call is direct. The changes at r193504 (to fix PR c++/11750) caused a regression to this behavior. This patch attempts to fix

Fix for PR c++/59031

2013-11-07 Thread Easwaran Raman
and is this ok for trunk? Thanks, Easwaran 2013-11-07 Easwaran Raman era...@google.com PR c++/59031 * call.c (build_new_method_call_1): Comnpare function context with BASELINK_BINFO type rather than instance type before marking the call with LOOKUP_NONVIRTUAL. testsuite/ChangeLog: 2013-11-07

Re: Add a param to decide stack slot sharing at -O0

2013-10-09 Thread Easwaran Raman
On Wed, Oct 9, 2013 at 4:11 AM, Richard Biener richard.guent...@gmail.com wrote: On Tue, Oct 8, 2013 at 11:04 PM, Easwaran Raman era...@google.com wrote: In cfgexpand.c, variables in non-overlapping lexical scopes are assigned same stack locations at -O1 and above. At -O0, this is attempted

Add a param to decide stack slot sharing at -O0

2013-10-08 Thread Easwaran Raman
the stack. We are ok with a slight increase in compilation time to get smaller stack frames even at -O0 and this patch would allow us do that easily. Bootstraps on x86_64/linux. Is this ok for trunk? Thanks, Easwaran 2013-10-08 Easwaran Raman era...@google.com * params.def

Don't drop attributes on template member functions

2013-10-02 Thread Easwaran Raman
tests pass with the patch on x86_64/linux. Is this ok for trunk? 2013-10-02 Easwaran Raman era...@google.com PR c++/33911 * parser.c (cp_parser_init_declarator): Do not drop attributes of template member functions. 2013-10-02 Easwaran Raman era...@google.com

[Google] Adjust comdat-sharing-probability param for -Os

2013-09-27 Thread Easwaran Raman
This patch increases comdat-sharing-probability to 80 under -Os. This reduces the amount of inlining and helps internal benchmarks. Unfortunately, this causes slight regression on spec 2006. Ok for google branches if all tests pass? - Easwaran comdat_sharing.patch Description: Binary data

Re: [Google] Adjust comdat-sharing-probability param for -Os

2013-09-27 Thread Easwaran Raman
at 2:57 PM, Easwaran Raman era...@google.com wrote: This patch increases comdat-sharing-probability to 80 under -Os. This reduces the amount of inlining and helps internal benchmarks. Unfortunately, this causes slight regression on spec 2006. Ok for google branches if all tests pass? - Easwaran

Re: Fix PR middle-end/57393

2013-09-23 Thread Easwaran Raman
Ping. On Mon, Sep 16, 2013 at 4:42 PM, Easwaran Raman era...@google.com wrote: There are two separate root causes for the problem reported in PR 57393. This patch attempts to fix both. First is due to newly created stmts that have the default UID of 0 which are compared with statements

Fix PR middle-end/57393

2013-09-16 Thread Easwaran Raman
reassociation to decouple them from the SSA variables involved in reassociation. This bootstraps in x86_64 and I am running the tests. Ok for trunk? Thanks, Easwaran 2013-09-16 Easwaran Raman era...@google.com PR middle-end/57393 * tree-ssa-reassoc.c (get_stmt_uid_with_default

Re: Fix PR middle-end/57370

2013-09-06 Thread Easwaran Raman
On Fri, Sep 6, 2013 at 12:05 AM, Richard Biener richard.guent...@gmail.com wrote: On Thu, Sep 5, 2013 at 9:23 PM, Easwaran Raman era...@google.com wrote: On Tue, Aug 27, 2013 at 3:35 AM, Richard Biener richard.guent...@gmail.com wrote: On Thu, Aug 1, 2013 at 1:34 AM, Easwaran Raman era

Re: Fix PR middle-end/57370

2013-09-05 Thread Easwaran Raman
On Tue, Aug 27, 2013 at 3:35 AM, Richard Biener richard.guent...@gmail.com wrote: On Thu, Aug 1, 2013 at 1:34 AM, Easwaran Raman era...@google.com wrote: I have a new patch that supersedes this. The new patch also fixes PR tree-optimization/57393 and PR tree-optimization/58011. Bootstraps

Re: Fix PR middle-end/57370

2013-09-04 Thread Easwaran Raman
Submitted the patch (r202262) without the insert_stmt_after hunk. Also fixed nits pointed out by Jakub. - Easwaran On Mon, Sep 2, 2013 at 2:31 AM, Richard Biener richard.guent...@gmail.com wrote: On Fri, Aug 30, 2013 at 6:13 PM, Easwaran Raman era...@google.com wrote: On Fri, Aug 30, 2013 at 1

Re: Fix PR middle-end/57370

2013-08-30 Thread Easwaran Raman
On Fri, Aug 30, 2013 at 1:25 AM, Richard Biener richard.guent...@gmail.com wrote: On Fri, Aug 30, 2013 at 2:30 AM, Easwaran Raman era...@google.com wrote: Richard, Do you want me to commit everything but the insert_stmt_after hunk now? Yes. There are more similar failures reported

Re: Fix PR middle-end/57370

2013-08-30 Thread Easwaran Raman
On Fri, Aug 30, 2013 at 9:26 AM, Jakub Jelinek ja...@redhat.com wrote: On Fri, Aug 30, 2013 at 09:13:34AM -0700, Easwaran Raman wrote: There are more similar failures reported in http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57393 and I have attached the updated patch there. Shall I send

Re: [Google] Refine hot caller heuristic

2013-08-29 Thread Easwaran Raman
the check accordingly. I have attached the new patch. - Easwaran David On Tue, Aug 20, 2013 at 12:26 PM, Easwaran Raman era...@google.com wrote: The current hot caller heuristic simply promotes edges whose caller is hot. This patch does the following: * Turn it off for applications with large

Re: [Google] Refine hot caller heuristic

2013-08-29 Thread Easwaran Raman
+return true; +} +} return false; } On Tue, Aug 20, 2013 at 12:26 PM, Easwaran Raman era...@google.com wrote: The current hot caller heuristic simply promotes edges whose caller is hot. This patch does the following: * Turn it off for applications with large footprint since

Re: Fix PR middle-end/57370

2013-08-29 Thread Easwaran Raman
it is used (find_insert_point and appears_later_in_bb) instead of where the stmt is created? I think that would be less brittle. - Easwaran On Tue, Aug 27, 2013 at 3:35 AM, Richard Biener richard.guent...@gmail.com wrote: On Thu, Aug 1, 2013 at 1:34 AM, Easwaran Raman era...@google.com wrote: I

[Google] Refine hot caller heuristic

2013-08-20 Thread Easwaran Raman
The current hot caller heuristic simply promotes edges whose caller is hot. This patch does the following: * Turn it off for applications with large footprint since the size increase hurts them * Be more selective by considering arguments to callee when the heuristic is enabled. This performs

Re: Fix PR middle-end/57370

2013-08-07 Thread Easwaran Raman
Ping. On Wed, Jul 31, 2013 at 4:34 PM, Easwaran Raman era...@google.com wrote: I have a new patch that supersedes this. The new patch also fixes PR tree-optimization/57393 and PR tree-optimization/58011. Bootstraps and no test regression on x86_64/linux. Ok for trunk? 2013-07-31 Easwaran

Re: Fix PR middle-end/57370

2013-07-31 Thread Easwaran Raman
I have a new patch that supersedes this. The new patch also fixes PR tree-optimization/57393 and PR tree-optimization/58011. Bootstraps and no test regression on x86_64/linux. Ok for trunk? 2013-07-31 Easwaran Raman era...@google.com PR middle-end/57370 * tree-ssa-reassoc.c

Re: Fix PR middle-end/57370

2013-07-12 Thread Easwaran Raman
Ping. On Thu, Jun 27, 2013 at 10:15 AM, Easwaran Raman era...@google.com wrote: A newly generated statement in build_and_add_sum function of tree-ssa-reassoc.c has to be assigned the UID of its adjacent statement. In one instance, it was assigned the wrong uid (of an earlier phi statement

Fix PR middle-end/57370

2013-06-27 Thread Easwaran Raman
regressions on x86_64/linux. Ok for trunk? Thanks, Easwaran 2013-06-27 Easwaran Raman era...@google.com PR middle-end/57370 * tree-ssa-reassoc.c (build_and_add_sum): Do not use the UID of a phi node for a non-phi gimple statement. testsuite/ChangeLog: 2013-06-27 Easwaran

[google gcc-4_8] Change size accounting during inlining in lipo mode

2013-06-20 Thread Easwaran Raman
In lipo mode, this patch updates the overall unit size only when the eventual function to which the callee is inlined is in primary module. This is to avoid the situation where the module growth budget is used up by inlines into auxiliary module functions that never get inlined into some primary

Re: PR tree-optimization/57337

2013-05-28 Thread Easwaran Raman
On Tue, May 28, 2013 at 4:11 AM, Richard Biener richard.guent...@gmail.com wrote: On Mon, May 27, 2013 at 10:20 AM, Richard Biener richard.guent...@gmail.com wrote: On Sun, May 26, 2013 at 5:53 AM, Easwaran Raman era...@google.com wrote: On Sat, May 25, 2013 at 4:46 AM, Richard Biener

Fix PR 57442

2013-05-28 Thread Easwaran Raman
outside the loop. Ok for trunk? Thanks, Easwaran 2013-05-28 Easwaran Raman era...@google.com PR tree-optimization/57442 * tree-ssa-reassoc.c (appears_later_in_bb): Return correct value when control exits the main loop. 2013-05-28 Easwaran Raman era...@google.com

Re: PR tree-optimization/57337

2013-05-25 Thread Easwaran Raman
On Sat, May 25, 2013 at 4:46 AM, Richard Biener richard.guent...@gmail.com wrote: Easwaran Raman era...@google.com wrote: In that case, if my insert_stmt immediately follows dep_stmt and both have the same UID, not_dominated_by would return true and I will end up updating insert_stmt to dep_stmt

Re: PR tree-optimization/57337

2013-05-24 Thread Easwaran Raman
23, 2013 at 7:26 PM, Easwaran Raman era...@google.com wrote: This addresses the case where UID alone is not sufficient to figure out which statement appears earlier in a BB. Bootstraps and no test regressions in x86_64 on linux. Ok for trunk? Why not simply conservatively use gimple_uid

PR tree-optimization/57337

2013-05-23 Thread Easwaran Raman
This addresses the case where UID alone is not sufficient to figure out which statement appears earlier in a BB. Bootstraps and no test regressions in x86_64 on linux. Ok for trunk? Thanks, Easwaran 2013-05-23 Easwaran Raman era...@google.com PR tree-optimization/57337 * tree-ssa-reassoc.c

Re: Fix PR tree-optimization/57322

2013-05-20 Thread Easwaran Raman
, Easwaran On Mon, May 20, 2013 at 4:43 AM, Steven Bosscher stevenb@gmail.com wrote: On Mon, May 20, 2013 at 1:41 AM, Easwaran Raman era...@google.com wrote: PR tree-optimization/57322 * (build_and_add_sum): If a BB is empty, set the UID of the statement added to the BB

Fix PR tree-optimization/57322

2013-05-19 Thread Easwaran Raman
--- 2013-05-19 Easwaran Raman era...@google.com PR tree-optimization/57322 * (build_and_add_sum): If a BB is empty, set the UID of the statement added to the BB to be 1. Index: gcc/tree-ssa-reassoc.c

Re: Minimize downward code motion during reassociation

2013-05-17 Thread Easwaran Raman
On Thu, Apr 25, 2013 at 4:42 AM, Richard Biener richard.guent...@gmail.com wrote: On Fri, Dec 7, 2012 at 9:01 PM, Easwaran Raman era...@google.com wrote: It seems I need to reset the debug uses of a statement before moving the statement itself. The attached patch starts from the leaf to root

Print column information in dump_loc

2013-05-14 Thread Easwaran Raman
This patch dumps the column number as part of dump_loc making the output similar to inform(). This allows these messages to be pattern matched by dg_message. Bootstraps with this change. Ok for trunk? - Easwaran - 2013-05-14 Easwaran Raman era...@google.com * dumpfile.c (dump_loc

Re: Minimize downward code motion during reassociation

2013-04-24 Thread Easwaran Raman
I want to resend this patch for consideration. I applied the patch to trunk and confirmed that it bootstraps and doesn't cause test regressions. Is this ok for trunk? Thanks, Easwaran On Fri, Dec 7, 2012 at 12:01 PM, Easwaran Raman era...@google.com wrote: It seems I need to reset the debug

Re: Minimize downward code motion during reassociation

2012-12-07 Thread Easwaran Raman
. Ok if there are no test failures? Thanks, Easwaran 2012-12-07 Easwaran Raman era...@google.com * tree-ssa-reassoc.c(find_insert_point): New function. (insert_stmt_after): Likewise. (get_def_stmt): Likewise. (ensure_ops_are_available): Likewise. (rewrite_expr_tree): Do not move statements

Re: Minimize downward code motion during reassociation

2012-11-05 Thread Easwaran Raman
statements would be very helpful. Thanks, Easwaran On Sun, Nov 4, 2012 at 9:10 AM, Richard Biener richard.guent...@gmail.com wrote: On Wed, Oct 31, 2012 at 8:16 PM, Easwaran Raman era...@google.com wrote: On Wed, Oct 31, 2012 at 4:36 AM, Richard Biener richard.guent...@gmail.com wrote: On Wed

Re: GCC 4.8.0 Status Report (2012-10-29), Stage 1 to end soon

2012-11-05 Thread Easwaran Raman
I'd like to get a small patch to tree reassociation ( http://gcc.gnu.org/ml/gcc-patches/2012-10/msg01761.html ) in. Thanks, Easwaran On Mon, Oct 29, 2012 at 10:56 AM, Jakub Jelinek ja...@redhat.com wrote: Status == I'd like to close the stage 1 phase of GCC 4.8 development on Monday,

Re: Minimize downward code motion during reassociation

2012-11-02 Thread Easwaran Raman
Ping. - Easwaran On Wed, Oct 31, 2012 at 12:16 PM, Easwaran Raman era...@google.com wrote: On Wed, Oct 31, 2012 at 4:36 AM, Richard Biener richard.guent...@gmail.com wrote: On Wed, Oct 24, 2012 at 4:02 AM, Easwaran Raman era...@google.com wrote: On Tue, Oct 23, 2012 at 2:52 AM, Richard

Re: Fix bugs introduced by switch-case profile propagation

2012-10-31 Thread Easwaran Raman
On Fri, Oct 26, 2012 at 8:05 AM, Jan Hubicka hubi...@ucw.cz wrote: Hi, On Tue, Oct 23, 2012 at 3:03 AM, Jan Hubicka hubi...@ucw.cz wrote: Ping. On Wed, Oct 17, 2012 at 1:48 PM, Easwaran Raman era...@google.com wrote: Hi, This patch fixes bugs introduced by my previous patch

Re: Minimize downward code motion during reassociation

2012-10-31 Thread Easwaran Raman
On Wed, Oct 31, 2012 at 4:36 AM, Richard Biener richard.guent...@gmail.com wrote: On Wed, Oct 24, 2012 at 4:02 AM, Easwaran Raman era...@google.com wrote: On Tue, Oct 23, 2012 at 2:52 AM, Richard Biener richard.guent...@gmail.com wrote: On Mon, Oct 22, 2012 at 8:31 PM, Easwaran Raman era

Re: Fix bugs introduced by switch-case profile propagation

2012-10-25 Thread Easwaran Raman
Hi, On Tue, Oct 23, 2012 at 3:03 AM, Jan Hubicka hubi...@ucw.cz wrote: Ping. On Wed, Oct 17, 2012 at 1:48 PM, Easwaran Raman era...@google.com wrote: Hi, This patch fixes bugs introduced by my previous patch to propagate profiles during switch expansion. Bootstrap and profiledbootstrap

Re: Fix bugs introduced by switch-case profile propagation

2012-10-23 Thread Easwaran Raman
On Tue, Oct 23, 2012 at 3:03 AM, Jan Hubicka hubi...@ucw.cz wrote: Ping. On Wed, Oct 17, 2012 at 1:48 PM, Easwaran Raman era...@google.com wrote: Hi, This patch fixes bugs introduced by my previous patch to propagate profiles during switch expansion. Bootstrap and profiledbootstrap

Re: Minimize downward code motion during reassociation

2012-10-23 Thread Easwaran Raman
On Tue, Oct 23, 2012 at 2:52 AM, Richard Biener richard.guent...@gmail.com wrote: On Mon, Oct 22, 2012 at 8:31 PM, Easwaran Raman era...@google.com wrote: On Mon, Oct 22, 2012 at 12:59 AM, Richard Biener richard.guent...@gmail.com wrote: On Fri, Oct 19, 2012 at 12:36 AM, Easwaran Raman era

Re: Minimize downward code motion during reassociation

2012-10-22 Thread Easwaran Raman
On Mon, Oct 22, 2012 at 12:59 AM, Richard Biener richard.guent...@gmail.com wrote: On Fri, Oct 19, 2012 at 12:36 AM, Easwaran Raman era...@google.com wrote: Hi, During expression reassociation, statements are conservatively moved downwards to ensure that dependences are correctly satisfied

Re: Fix bugs introduced by switch-case profile propagation

2012-10-22 Thread Easwaran Raman
Ping. On Wed, Oct 17, 2012 at 1:48 PM, Easwaran Raman era...@google.com wrote: Hi, This patch fixes bugs introduced by my previous patch to propagate profiles during switch expansion. Bootstrap and profiledbootstrap successful on x86_64. Confirmed that it fixes the crashes reported in PR

Re: Minimize downward code motion during reassociation

2012-10-19 Thread Easwaran Raman
On Fri, Oct 19, 2012 at 5:13 PM, Xinliang David Li davi...@google.com wrote: On Thu, Oct 18, 2012 at 3:36 PM, Easwaran Raman era...@google.com wrote: Hi, During expression reassociation, statements are conservatively moved downwards to ensure that dependences are correctly satisfied after

Minimize downward code motion during reassociation

2012-10-18 Thread Easwaran Raman
on x86_64/linux. OK for trunk? Thanks, Easwaran 2012-10-18 Easwaran Raman era...@google.com * tree-ssa-reassoc.c(assign_uids): New function. (assign_uids_in_relevant_bbs): Likewise. (ensure_ops_are_available): Likewise. (rewrite_expr_tree): Do not move statements beyond what is necessary. Remove

Fix bugs introduced by switch-case profile propagation

2012-10-17 Thread Easwaran Raman
Hi, This patch fixes bugs introduced by my previous patch to propagate profiles during switch expansion. Bootstrap and profiledbootstrap successful on x86_64. Confirmed that it fixes the crashes reported in PR middle-end/54957. OK for trunk? - Easwaran 2012-10-17 Easwaran Raman era

Re: Propagate profile counts during switch expansion

2012-10-15 Thread Easwaran Raman
On Sun, Oct 14, 2012 at 8:09 AM, Jan Hubicka hubi...@ucw.cz wrote: Hi, Index: optabs.c === --- optabs.c(revision 191879) +++ optabs.c(working copy) @@ -4249,7 +4249,7 @@ prepare_operand (enum insn_code icode, rtx x, int

Re: Propagate profile counts during switch expansion

2012-10-13 Thread Easwaran Raman
Ping. On Mon, Oct 8, 2012 at 5:46 PM, Easwaran Raman era...@google.com wrote: I have attached a revised patch. The updated ChangeLog is given below and I have responded to your comments inline: 2012-10-08 Easwaran Raman era...@google.com * optabs.c (emit_cmp_and_jump_insn_1): Add a new

Re: Move statements upwards after reassociation

2012-10-12 Thread Easwaran Raman
On Fri, Oct 12, 2012 at 1:45 AM, Richard Biener richard.guent...@gmail.com wrote: On Fri, Oct 12, 2012 at 3:09 AM, Easwaran Raman era...@google.com wrote: Thanks for the comments. As David wrote, the intent of the patch is not to do a general purpose scheduling, but to compensate

[google] Move delete with size to its own file (issue6655052)

2012-10-11 Thread Easwaran Raman
regressions on a x86_64 machine running linux. Ok for google/4_7 and google/main branches? Google ref b/6982747 2012-10-11 Easwaran Raman era...@google.com * libsupc++/Makefile.am: Add del_opsz.cc to sources. * libsupc++/Makefile.in: Regenerated. * libsupc++/del_opsz.cc: New file

Re: Move statements upwards after reassociation

2012-10-11 Thread Easwaran Raman
, 2012 at 3:52 AM, Easwaran Raman era...@google.com wrote: +/* Move STMT up within its BB until it can not be moved any further. */ + +static void move_stmt_upwards (gimple stmt) +{ + gimple_stmt_iterator gsi, gsistmt; + tree rhs1, rhs2; + gimple rhs1def = NULL, rhs2def = NULL; + rhs1

Move statements upwards after reassociation

2012-10-10 Thread Easwaran Raman
in reassociation and pushes them upwards in the BB as far as possible without violating dependences. Bootstraps and no tests regressions on a x86_64 machine running linux. Ok for trunk? - Easwaran --- 2012-10-10 Easwaran Raman era...@google.com * tree-ssa-reassoc.c (move_stmt_upwards

Re: Propagate profile counts during switch expansion

2012-10-04 Thread Easwaran Raman
Hi Honza, I am addressing some of the questions you raise here. Will send an updated patch later. On Thu, Oct 4, 2012 at 6:19 AM, Jan Hubicka hubi...@ucw.cz wrote: @@ -560,7 +577,6 @@ compute_outgoing_frequencies (basic_block b) return; } } - if (single_succ_p (b))

Propagate profile counts during switch expansion

2012-10-02 Thread Easwaran Raman
corresponding to the cases are (nearly the) same as at the gimple level. Bootstrapped and profile-bootstrapped on an x86_64/linux machine. OK for trunk? - Easwaran -- 2012-10-02 Easwaran Raman era...@google.com * cfgbuild.c (gen_probabilities_from_existing_counts): New function

Change the ordering of cdce pass

2012-06-14 Thread Easwaran Raman
pow, which can then be shrink-wrapped by cdce. So it seems reasonable to do this reordering. Bootstraps on x86_64 on linux with no test regression. OK for trunk? - Easwaran -- 2012-06-14 Easwaran Raman era...@google.com * gcc/passes.c (init_optimization_passes

Re: Change the ordering of cdce pass

2012-06-14 Thread Easwaran Raman
ChangeLog entry has a gcc/ prefix that shouldn't be there. Here is the revised entry: 2012-06-14 Easwaran Raman era...@google.com * passes.c (init_optimization_passes): Remove pass_call_cdce from its current position and insert after pass_dce. On Thu, Jun 14, 2012 at 6:38

Re: Propagate profile counts after switch case expansion (issue5896043)

2012-04-16 Thread Easwaran Raman
Ping. On Mon, Apr 9, 2012 at 2:33 PM, Easwaran Raman era...@google.com wrote: On Sun, Mar 25, 2012 at 9:40 PM, Easwaran Raman era...@google.com wrote: On Sun, Mar 25, 2012 at 12:22 PM, Jan Hubicka hubi...@ucw.cz wrote: This patch propagates execution count of thee case labels of a switch-case

Re: Propagate profile counts after switch case expansion (issue5896043)

2012-04-09 Thread Easwaran Raman
On Sun, Mar 25, 2012 at 9:40 PM, Easwaran Raman era...@google.com wrote: On Sun, Mar 25, 2012 at 12:22 PM, Jan Hubicka hubi...@ucw.cz wrote: This patch propagates execution count of thee case labels of a switch-case statement after its expansion. Bootstraps and all tests pass. OK for trunk

Re: [google]Add support for sampled profile collection (issue4438083)

2012-03-30 Thread Easwaran Raman
I want to revive this patch for mainline and have some questions on Honza's comments. On Fri, Apr 29, 2011 at 1:48 PM, Jan Hubicka hubi...@ucw.cz wrote: A known limitation is that value profiling is not yet sampled, but it does not seem to cause problems. For the performance alone, we

Re: Propagate profile counts after switch case expansion (issue5896043)

2012-03-25 Thread Easwaran Raman
On Sun, Mar 25, 2012 at 12:22 PM, Jan Hubicka hubi...@ucw.cz wrote: This patch propagates execution count of thee case labels of a switch-case statement after its expansion. Bootstraps and all tests pass. OK for trunk? Hi, while this is resonable thing to do, I belive it would make most

Propagate profile counts after switch case expansion (issue5896043)

2012-03-23 Thread Easwaran Raman
This patch propagates execution count of thee case labels of a switch-case statement after its expansion. Bootstraps and all tests pass. OK for trunk? 2012-03-23 Easwaran Raman era...@google.com * cfgbuild.c (non_zero_profile_counts): New function

Re: Propagate profile counts after switch case expansion (issue5896043)

2012-03-23 Thread Easwaran Raman
equally which can cause poor optimization of hot code. This patch ensures that the counts collected during profile collection are correctly propagated allowing hot code to be better optimized by RTL optimizations. Patch tested on x86_64. - Easwaran On Fri, Mar 23, 2012 at 10:43 AM, Easwaran Raman era

Re: Propagate profile counts after switch case expansion (issue5896043)

2012-03-23 Thread Easwaran Raman
On Fri, Mar 23, 2012 at 3:29 PM, Andi Kleen a...@firstfloor.org wrote: Easwaran Raman era...@google.com writes: Some more background on this patch: Right now, while the execution counts of different case labels of a switch statement are obtained during profile collection

[google] Use delete with size parameter in STL deallocate (issue5794070)

2012-03-12 Thread Easwaran Raman
that compiling a C++ program that uses std::vector with -fsized-delete invokes the two parameter version of operator delete. OK for google/main and google/4_6 branches? c-family/ChangeLog.google-main: 2012-03-12 Easwaran Raman era...@google.com * c-cppbuiltin.c (c_cpp_builtins): Define

Re: [google] Propagate profile information to RTL level during switch expansion

2012-02-01 Thread Easwaran Raman
On Tue, Jan 31, 2012 at 10:16 PM, Xinliang David Li davi...@google.com wrote: Ok for google branch with minor changes below. thanks, David +#define case_probability(x, y) ((y) ? ((x) * REG_BR_PROB_BASE  / (y))  : -1) + Using upper case for macro? From

[google] Propagate profile information to RTL level during switch expansion

2012-01-31 Thread Easwaran Raman
This patch propagates profile information to RTL level when expanding switch statements using jump table or a binary tree of branches. Ok for google/gcc-4_6 branch? I would like the patch to be considered for trunk when stage 1 opens again. -Easwaran 2012-01-31 Easwaran Raman era

Re: [google] Add support for delete operator that takes the size of the object as a parameter

2011-12-17 Thread Easwaran Raman
and no test regression. OK for google/main and google/gcc-4_6 branches? - 011-12-17 Easwaran Raman era...@google.com * common.opt (fsized-delete): New option. cp/ChangeLog.google-main: 2011-12-17 Easwaran Raman era...@google.com * decl.c (cxx_init_decl_processing

Re: [google] Add support for delete operator that takes the size of the object as a parameter

2011-12-13 Thread Easwaran Raman
On Mon, Dec 12, 2011 at 1:52 PM, Easwaran Raman era...@google.com wrote: On Mon, Dec 12, 2011 at 12:41 PM, Paolo Carlini paolo.carl...@oracle.com wrote: On 12/12/2011 09:37 PM, Easwaran Raman wrote: Thanks for the comments Paolo. I have attached the new patch. I have bumped the version

Re: [google] Add support for delete operator that takes the size of the object as a parameter

2011-12-12 Thread Easwaran Raman
Thanks for the comments Paolo. I have attached the new patch. I have bumped the version to 3.4.18 and used _ZdlPv[jmy] in gnu.ver. I have also added the symbol to baseline_symbols.txt of other targets. -Easwaran 2011-12-11 Easwaran Raman era...@google.com * common.opt

Re: [google] Add support for delete operator that takes the size of the object as a parameter

2011-12-12 Thread Easwaran Raman
On Mon, Dec 12, 2011 at 12:41 PM, Paolo Carlini paolo.carl...@oracle.com wrote: On 12/12/2011 09:37 PM, Easwaran Raman wrote: Thanks for the comments Paolo. I have attached the new patch. I have bumped the version to 3.4.18 You shouldn't: 4.7 is not out yet, thus no reason to increase

[google] Add support for delete operator that takes the size of the object as a parameter

2011-12-11 Thread Easwaran Raman
? --- 2011-12-11 Easwaran Raman era...@google.com * common.opt (fsized-delete): New option. cp/ChangeLog.google-4_6: 2011-12-11 Easwaran Raman era...@google.com * decl.c (cxx_init_decl_processing): Specify a function that takes a void* and size_t

[google] Fix bugs in sampled profile collection

2011-09-30 Thread Easwaran Raman
branches? -Easwaran 2011-09-30 Easwaran Raman era...@google.com * tree-profile.c (gcov_sample_counter_decl): Add GTY marker. (gcov_sampling_rate_decl): Likewise. (add_sampling_to_edge_counters): Do not free instrumentation_to_be_sampled

Re: [google] Linker plugin to do function reordering using callgraph edge profiles (issue5124041)

2011-09-27 Thread Easwaran Raman
+static inline hashval_t +edge_hash_function (unsigned int id1, unsigned int id2) +{ +  /* If the number of functions is less than 1000, this gives a unique value +     for every function id combination.  */ +  const int MULTIPLIER = 1000; +  return id1* MULTIPLIER + id2; Change to id1 16 | id2

Re: [google] Linker plugin to do function reordering using callgraph edge profiles (issue5124041)

2011-09-27 Thread Easwaran Raman
OK for google/gcc-4_6 and google/main branches. -Easwaran On Tue, Sep 27, 2011 at 4:07 PM, Sriraman Tallam tmsri...@google.com wrote: Made all the changes. Attaching new patch of updated files. On Tue, Sep 27, 2011 at 11:26 AM, Easwaran Raman era...@google.com wrote: +static inline

[google] Backport r175063, r175082 and r175384 from trunk to google/gcc-4_6 branch.

2011-07-24 Thread Easwaran Raman
Commited.

Re: Mark variables addressable if they are copied using libcall in RTL expander

2011-06-23 Thread Easwaran Raman
posting a consolidated patch? -- Eric Botcazou Here is the revised patch. Bootstraps and all tests pass on x86_64-unknown-linux. OK for trunk? 2011-06-23 Easwaran Raman era...@google.com PR rtl-optimization/49429 PR target/49454 * expr.c (emit_block_move_hints): Mark

Re: Mark variables addressable if they are copied using libcall in RTL expander

2011-06-23 Thread Easwaran Raman
On Thu, Jun 23, 2011 at 12:16 PM, Jakub Jelinek ja...@redhat.com wrote: On Thu, Jun 23, 2011 at 12:02:35PM -0700, Easwaran Raman wrote: +      if (y_expr) +        mark_addressable (y_expr); Please watch formatting, a tab should be used instead of 8 spaces. +      if (x_expr

Re: Mark variables addressable if they are copied using libcall in RTL expander

2011-06-22 Thread Easwaran Raman
On Wed, Jun 22, 2011 at 7:13 AM, Eric Botcazou ebotca...@adacore.com wrote: I fear this isn't enough considering pass-by-value aggregates that are callee copied. It's indeed not sufficient for arguments passed by reference but callee-copied. This is PR target/49454.  For

Re: Improve DSE in the presence of calls

2011-06-17 Thread Easwaran Raman
, Easwaran On Tue, Jun 14, 2011 at 9:34 AM, Jeff Law l...@redhat.com wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 05/10/11 13:18, Easwaran Raman wrote: I am not sure I understand the problem here.  If there is a wild read from asm, the instruction has the wild_read flag set

Re: Ping: Re: Improve DSE in the presence of calls

2011-06-07 Thread Easwaran Raman
Ping. Diego, David, Is this patch OK for google/main? -Easwaran On Thu, Jun 2, 2011 at 4:48 PM, Easwaran Raman era...@google.com wrote: Ping. On Sat, May 14, 2011 at 8:01 AM, Easwaran Raman era...@google.com wrote: http://gcc.gnu.org/ml/gcc-patches/2011-05/msg00781.html

Re: Ping: Re: Improve DSE in the presence of calls

2011-06-02 Thread Easwaran Raman
Ping. On Sat, May 14, 2011 at 8:01 AM, Easwaran Raman era...@google.com wrote: http://gcc.gnu.org/ml/gcc-patches/2011-05/msg00781.html

Re: Ping: Re: Improve DSE in the presence of calls

2011-05-20 Thread Easwaran Raman
Ping. On Sat, May 14, 2011 at 8:01 AM, Easwaran Raman era...@google.com wrote: http://gcc.gnu.org/ml/gcc-patches/2011-05/msg00781.html

Ping: Re: Improve DSE in the presence of calls

2011-05-14 Thread Easwaran Raman
http://gcc.gnu.org/ml/gcc-patches/2011-05/msg00781.html

Re: Improve DSE in the presence of calls

2011-05-10 Thread Easwaran Raman
On Tue, May 3, 2011 at 9:40 AM, Easwaran Raman era...@google.com wrote: On Mon, May 2, 2011 at 8:37 PM, Jeff Law l...@redhat.com wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 04/26/11 16:06, Easwaran Raman wrote: You're right. The patch has correctness issues. It is not possible

[google]Implement an optimization based on reuse distance profiling (issue4517049)

2011-05-09 Thread Easwaran Raman
This patch by Silvius Rus replaces calls to certain functions with a specialized version that uses non-temporal stores based on memory reuse distance profiling. Bootstraps, no test regressions and the profiling works for a small test case. Ok for google/main.? -Easwaran 2011-05-09 Silvius

[google] Backport r172837 and r172788 to google/main

2011-05-06 Thread Easwaran Raman
Backported r172788 and r172837 from trunk to google/main. 2011-05-06 Easwaran Raman era...@google.com Backport r172837: * cfgexpand.c (stack_var): Remove OFFSET... (add_stack_var): ...and its reference here... (expand_stack_vars): ...and here. (stack_var_cmp

Re: [google] Backport r172837 and r172788 to google/main

2011-05-06 Thread Easwaran Raman
Thanks. Fixed them (as well as the same issue in some earlier entries). -Easwaran On Fri, May 6, 2011 at 11:27 AM, Diego Novillo dnovi...@google.com wrote: On Fri, May 6, 2011 at 14:22, Easwaran Raman era...@google.com wrote: Backported r172788 and r172837  from trunk to google/main. Minor

Re: Improve DSE in the presence of calls

2011-05-03 Thread Easwaran Raman
On Mon, May 2, 2011 at 8:37 PM, Jeff Law l...@redhat.com wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 04/26/11 16:06, Easwaran Raman wrote: You're right. The patch has correctness issues. It is not possible to simply not call add_wild_read because it also resets

[google]Add support for sampled profile collection (issue4438083)

2011-04-28 Thread Easwaran Raman
This patch from Silvius Rus adds support for sampled edge profile collection to reduce instrumentation run overhead. Bootstraps and no test regressions. Ok for google/main? 2011-04-28 Silvius Rus silvius@gmail.com * doc/invoke.texi: Document -fprofile-generate-sampling option.

Re: Improve DSE in the presence of calls

2011-04-26 Thread Easwaran Raman
On Mon, Apr 25, 2011 at 10:03 AM, Jeff Law l...@redhat.com wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 04/22/11 14:19, Easwaran Raman wrote: Hi,  This patch improves RTL DSE by not assuming that calls read all memory locations. With this patch, calls are assumed to read any non

Re: [google]Pass --save-temps to the assembler (issue4436049)

2011-04-25 Thread Easwaran Raman
On Thu, Apr 21, 2011 at 3:20 PM, Andrew Pinski pins...@gmail.com wrote: On Thu, Apr 21, 2011 at 3:18 PM, Easwaran Raman era...@google.com wrote: When using gcc with a post-assembly tool, we use a wrapper that invokes the tool and needs to know if the tool's output (another assembly file) needs

Improve DSE in the presence of calls

2011-04-22 Thread Easwaran Raman
for trunk? Thanks, Easwaran 2011-04-22 Easwaran Raman era...@google.com PR rtl-optimization/44194 * dse.c (header files): Include tree-flow.h. (group_info): Add fields. (globals): Add a new variable kill_on_calls. (get_group_info): Initialize added fields

Re: Improve DSE in the presence of calls

2011-04-22 Thread Easwaran Raman
On Fri, Apr 22, 2011 at 1:26 PM, Jakub Jelinek ja...@redhat.com wrote: On Fri, Apr 22, 2011 at 01:19:17PM -0700, Easwaran Raman wrote: The ChangeLog entry has various issues: 2011-04-22  Easwaran Raman  era...@google.com PR rtl-optimization/44194 This should have tab before PR as well

Re: Improve stack layout heuristic.

2011-04-21 Thread Easwaran Raman
On Thu, Apr 21, 2011 at 8:43 AM, Richard Guenther richard.guent...@gmail.com wrote: On Thu, Apr 21, 2011 at 5:22 PM, Michael Matz m...@suse.de wrote: Hi, On Wed, 20 Apr 2011, Easwaran Raman wrote: But you're right - not adding that conflict doesn't actually reduce the size of bit maps

  1   2   >