;
- cgraph_process_module_scope_statics ();
- /* Now perform link to allow cross module inlining. */
- cgraph_do_link ();
- varpool_do_link ();
- cgraph_unify_type_alias_sets ();
-
return TODO_rebuild_cgraph_edges;
}
On Wed, Sep 18, 2013 at 5:16 PM, Xinliang David Li davi...@google.com wrote:
On Wed, Sep 18, 2013 at 4
Ok.
David
On Wed, Sep 18, 2013 at 10:21 AM, Dehao Chen de...@google.com wrote:
This patch disables SLP for AutoFDO.
Bootstrapped and passed unittests.
OK for google-4_8?
Thanks,
Dehao
Index: gcc/opts.c
===
--- gcc/opts.c
On Tue, Sep 17, 2013 at 1:20 AM, Richard Biener
richard.guent...@gmail.com wrote:
On Mon, Sep 16, 2013 at 10:24 PM, Xinliang David Li davi...@google.com
wrote:
On Mon, Sep 16, 2013 at 3:13 AM, Richard Biener
richard.guent...@gmail.com wrote:
On Fri, Sep 13, 2013 at 5:16 PM, Xinliang David Li
On Wed, Sep 18, 2013 at 4:51 PM, Dehao Chen de...@google.com wrote:
This patch fixup the call graph edge targets during AutoFDO pass, so
that when rebuilding call graph edges, it can find the correct callee.
Bootstrapped and passed regression test. Benchmark tests on-going.
Ok for google-4_8
On Tue, Sep 17, 2013 at 1:20 AM, Richard Biener
richard.guent...@gmail.com wrote:
On Mon, Sep 16, 2013 at 10:24 PM, Xinliang David Li davi...@google.com
wrote:
On Mon, Sep 16, 2013 at 3:13 AM, Richard Biener
richard.guent...@gmail.com wrote:
On Fri, Sep 13, 2013 at 5:16 PM, Xinliang David Li
On Tue, Sep 17, 2013 at 8:45 AM, Jakub Jelinek ja...@redhat.com wrote:
On Tue, Sep 17, 2013 at 08:37:57AM -0700, Xinliang David Li wrote:
char a[1];
void foo(int n)
{
int* b = (int*)(a+n);
int i = 0;
for (; i 1000; ++i)
b[i] = 1;
}
int main(int argn, char
ok.
David
On Tue, Sep 17, 2013 at 4:53 PM, Dehao Chen de...@google.com wrote:
This patch makes AutoFDO honor system paths stored in the profile.
Bootstrapped and passed regression tests.
OK for google-4_8 branch?
Thanks,
Dehao
Index: gcc/auto-profile.c
I noticed there are a couple of dump_printf_loc instances in
coverage.c not ended with newline. They should be fixed.
David
On Tue, Sep 10, 2013 at 6:32 AM, Teresa Johnson tejohn...@google.com wrote:
On Mon, Sep 9, 2013 at 9:55 PM, Xinliang David Li davi...@google.com wrote:
looks fine to me
16, 2013 at 11:06 AM, Teresa Johnson tejohn...@google.com wrote:
On Mon, Sep 16, 2013 at 10:57 AM, Xinliang David Li davi...@google.com
wrote:
I noticed there are a couple of dump_printf_loc instances in
coverage.c not ended with newline. They should be fixed.
I committed this change
I incorporated all the comments and committed the change (also fixed a
test failure with --help=optimizers).
thanks,
David
On Mon, Sep 16, 2013 at 3:07 AM, Richard Biener
richard.guent...@gmail.com wrote:
On Fri, Sep 13, 2013 at 6:56 PM, Xinliang David Li davi...@google.com wrote:
Updated
On Mon, Sep 16, 2013 at 3:13 AM, Richard Biener
richard.guent...@gmail.com wrote:
On Fri, Sep 13, 2013 at 5:16 PM, Xinliang David Li davi...@google.com wrote:
On Fri, Sep 13, 2013 at 1:30 AM, Richard Biener
richard.guent...@gmail.com wrote:
On Thu, Sep 12, 2013 at 10:31 PM, Xinliang David Li
Ok -- abandon the patch.
On Mon, Sep 16, 2013 at 2:04 AM, Richard Biener
richard.guent...@gmail.com wrote:
On Sat, Sep 14, 2013 at 8:10 AM, Xinliang David Li davi...@google.com wrote:
tree if conversion is an enabler pass for vectorization, so by
default, it is only turned on when
Xinliang David Li davi...@google.com
* tree-vect-slp.c (destroy_bb_vec_info): Data ref cleanup.
Index: common.opt
===
--- common.opt (revision 202229)
+++ common.opt (working copy)
@@ -1308,7 +1308,7 @@ EnumValue
Enum
Ok.
David
On Fri, Sep 13, 2013 at 7:21 AM, Teresa Johnson tejohn...@google.com wrote:
Testing passes, is the below patch ok for google/4_8?
Thanks, Teresa
On Thu, Sep 12, 2013 at 10:18 PM, Teresa Johnson tejohn...@google.com wrote:
On Thu, Sep 12, 2013 at 2:32 PM, Xinliang David Li davi
On Fri, Sep 13, 2013 at 1:30 AM, Richard Biener
richard.guent...@gmail.com wrote:
On Thu, Sep 12, 2013 at 10:31 PM, Xinliang David Li davi...@google.com
wrote:
Currently -ftree-vectorize turns on both loop and slp vectorizations,
but there is no simple way to turn on loop vectorization alone
, Xinliang David Li davi...@google.com wrote:
On Fri, Sep 13, 2013 at 1:30 AM, Richard Biener
richard.guent...@gmail.com wrote:
On Thu, Sep 12, 2013 at 10:31 PM, Xinliang David Li davi...@google.com
wrote:
Currently -ftree-vectorize turns on both loop and slp vectorizations,
but there is no simple
Ok -- then my updated patch is wrong then. The implementation in the
first version matches the requirement.
thanks,
David
On Fri, Sep 13, 2013 at 9:45 AM, Joseph S. Myers
jos...@codesourcery.com wrote:
On Fri, 13 Sep 2013, Richard Biener wrote:
@@ -1691,6 +1695,12 @@ common_handle_option
Updated patch implementing the logic that more specific option wins.
Ok for trunk?
thanks,
David
On Fri, Sep 13, 2013 at 9:48 AM, Xinliang David Li davi...@google.com wrote:
Ok -- then my updated patch is wrong then. The implementation in the
first version matches the requirement.
thanks
On Thu, Sep 12, 2013 at 1:06 PM, Teresa Johnson tejohn...@google.com wrote:
After porting r198033 from google/4_7 to google/4_8 a test case failed
with an assert when trying to take the strlen of profile_data_prefix.
In most cases this is either set from the directory specified to
;
Index: ChangeLog
===
--- ChangeLog (revision 202481)
+++ ChangeLog (working copy)
@@ -1,3 +1,24 @@
+2013-09-12 Xinliang David Li davi...@google.com
+
+ * tree-if-conv.c (main_tree_if_conversion): Check new flag.
+ * omp
...@google.com wrote:
On Thu, Sep 12, 2013 at 1:20 PM, Xinliang David Li davi...@google.com wrote:
On Thu, Sep 12, 2013 at 1:06 PM, Teresa Johnson tejohn...@google.com wrote:
After porting r198033 from google/4_7 to google/4_8 a test case failed
with an assert when trying to take the strlen
Can you add a test case to the regression suite?
When the type of arguments are unsigned short/unsigned int, GCC does
not vectorize the loop anymore -- this is worth a separate bug to
track. punpcklwd instruction can be used to do zero extension of the
short type.
David
On Wed, Sep 11, 2013 at
The following patch enables GCC to record system include paths in
module infos. This allows more precise parsing context to be
established when compiling auxiliary modules -- FE behaves slightly
differently when parsing system headers. There are also a couple of
other cleanups in the patch.
:26 PM, Joseph S. Myers jos...@codesourcery.com
wrote:
On Wed, 4 Sep 2013, Xinliang David Li wrote:
This patch submission still fails to pay attention to various of my
comments.
If you can provide inlined comments in the patch, that will be more
useful and productive.
I have explained
looks fine to me.
In the long run, I wonder if the machinery in diagnostic messages can
be reused for opt-info dumping -- i.e., support different streams. It
has many nice features including %qD specifier for printing tree
decls.
David
On Mon, Sep 9, 2013 at 12:01 PM, Teresa Johnson
On Wed, Sep 4, 2013 at 1:53 PM, Cong Hou co...@google.com wrote:
I have made a new patch according to your comments. I found several
references saying that the precision 2p+2 is OK for the sqrt
conversion (one here:
http://www.cs.berkeley.edu/~fateman/generic/algorithms.pdf). The new
patch is
On Wed, Sep 4, 2013 at 1:59 PM, Joseph S. Myers jos...@codesourcery.com wrote:
On Wed, 4 Sep 2013, Cong Hou wrote:
I have made a new patch according to your comments. I found several
references saying that the precision 2p+2 is OK for the sqrt
conversion (one here:
From Joseph:
The
conversion is not safe for sqrt if the two types are double and long
double and long double is x86 extended, for example.
This is not reflected in the patch.
David
On Tue, Sep 3, 2013 at 2:27 PM, Joseph S. Myers jos...@codesourcery.com wrote:
On Tue, 3 Sep 2013, Cong Hou
Missing test cases?
Have you tested the optimization with SPEC2k and SPEC06? There are a
couple of benchmarks benefit greatly from devirtualization, such as
eon, povray etc. I believe astar will probably improve with this
optimization at O2 (it has hot virtual functions that are not
overridden
Except that in this form, the dump will be extremely large and not
suitable for very large applications. Besides, we might also want to
use the same machinery (dump_printf_loc etc) for dump file dumping.
The current behavior of using '-details' to turn on opt-info-all
messages for dump files are
On Fri, Aug 30, 2013 at 12:51 PM, Teresa Johnson tejohn...@google.com wrote:
On Fri, Aug 30, 2013 at 9:27 AM, Xinliang David Li davi...@google.com wrote:
Except that in this form, the dump will be extremely large and not
suitable for very large applications.
Yes. I did some measurements
On Fri, Aug 30, 2013 at 1:23 AM, Richard Biener
richard.guent...@gmail.com wrote:
On Fri, Aug 30, 2013 at 1:28 AM, Xinliang David Li davi...@google.com wrote:
I was debugging a runtime failure of SPEC06 xalancbmk built with LIPO.
Using -fdisable-pass option pinpoints the problem in slp
)
+++ ChangeLog (working copy)
@@ -1,5 +1,17 @@
2013-08-29 Xinliang David Li davi...@google.com
+ * tree-vect-data-refs.c (vect_compute_data_ref_alignment):
+ Delay base decl alignment adjustment.
+ * tree-vectorizer.c (ensure_base_alignment): New function.
+ (vectorize_loops
:
On Tue, Aug 20, 2013 at 9:35 PM, Xinliang David Li davi...@google.com wrote:
Do you need to guard the jump function access with check if
(ipa_node_params_vector.exists ())?
I believe it is not necessary since, for example, ipa_analyze_node
calls ipa_check_create_node_params that calls create
On Wed, Aug 28, 2013 at 7:09 AM, Teresa Johnson tejohn...@google.com wrote:
On Wed, Aug 28, 2013 at 4:01 AM, Richard Biener
richard.guent...@gmail.com wrote:
On Wed, Aug 7, 2013 at 7:23 AM, Teresa Johnson tejohn...@google.com wrote:
On Tue, Aug 6, 2013 at 9:29 AM, Teresa Johnson
On Wed, Aug 28, 2013 at 12:59 AM, Richard Biener rguent...@suse.de wrote:
On Tue, 27 Aug 2013, Xinliang David Li wrote:
Richard, I have some comments about the patch.
-ftree-vectorizer-verbose=numberThis switch is deprecated. Use
-fopt-info instead.
ftree-slp-vectorize
, Xinliang David Li davi...@google.com
wrote:
If this is the convention, we should probably have another patch to
fix all the existing opt-info messages.
Yes please.
Also ...
b.c:16:A::foo: note: Loop is vectorized
loop vectorized
b.c:16:A::foo: note: Loop is versioned to remove aliases
Ok.
David
On Tue, Aug 27, 2013 at 7:36 AM, Dehao Chen de...@google.com wrote:
Patch updated.
Thanks,
Dehao
On Mon, Aug 26, 2013 at 4:11 PM, Xinliang David Li davi...@google.com wrote:
Can you add missing documentation on functions like ...:get_count_info
-- documenting return value etc
+ Honza
On Tue, Aug 27, 2013 at 10:56 AM, Teresa Johnson tejohn...@google.com wrote:
Ping #3.
Thanks,
Teresa
On Mon, Aug 19, 2013 at 11:33 AM, Teresa Johnson tejohn...@google.com wrote:
Ping.
Thanks,
Teresa
On Mon, Aug 12, 2013 at 6:54 AM, Teresa Johnson tejohn...@google.com wrote:
On
Does this one look ok?
thanks,
David
On Thu, Aug 22, 2013 at 4:20 PM, Xinliang David Li davi...@google.com wrote:
Hi, In this patch, loop alignment peeling and loop versioning
transformation will be reported via -fopt-info by default. This will
help vectorizer size tuning.
It also enhances
the function name
before each block of messages, a la inform().
Thanks,
Teresa
On Tue, Aug 27, 2013 at 11:22 AM, Xinliang David Li davi...@google.com
wrote:
Does this one look ok?
thanks,
David
On Thu, Aug 22, 2013 at 4:20 PM, Xinliang David Li davi...@google.com
wrote:
Hi, In this patch
If this is the convention, we should probably have another patch to
fix all the existing opt-info messages.
thanks,
David
On Tue, Aug 27, 2013 at 1:23 PM, Mike Stump mikest...@comcast.net wrote:
On Aug 27, 2013, at 11:22 AM, Xinliang David Li davi...@google.com wrote:
Does this one look ok
Richard, I have some comments about the patch.
-ftree-vectorizer-verbose=numberThis switch is deprecated. Use
-fopt-info instead.
ftree-slp-vectorize
! Common Report Var(flag_tree_slp_vectorize) Optimization
Enable basic block vectorization (SLP) on trees
The code dealing with
Can you add missing documentation on functions like ...:get_count_info
-- documenting return value etc. Also it might be better to avoid
using 'set' as the local variable name. Change it to something more
specific.
thanks,
David
On Thu, Aug 22, 2013 at 3:56 PM, Dehao Chen de...@google.com
===
--- ChangeLog (revision 201752)
+++ ChangeLog (working copy)
@@ -1,3 +1,13 @@
+2013-08-22 Xinliang David Li davi...@google.com
+
+ * tree-vect-loop-manip.c (vect_do_peeling_for_alignment):
+ Emit alignment peeling message with default -fopt-info
Do you need to guard the jump function access with check if
(ipa_node_params_vector.exists ())?
Ideally, useful_cold_callee should be folded into the inline hints
estimation. Question about the heuristic: why filtering out
PASS_THROUGH parameter cases completely? Passing 'this' parameter in
many
ok.
David
On Fri, Aug 16, 2013 at 1:28 PM, Rong Xu x...@google.com wrote:
This patch fixed a bug in lipo varpool node linking.
C++ FE drops the initializer if it's not used in this TU. For current
varpool linking may
resolve the varpool node to the one with null initializer.
-Rong
Some suggestions for the future:
1) add summary info in the odr dump -- i.e. for each node, dump the
number of direct bases, direct subtypes, number of all bases, all
subtypes;
2) add statistics dump -- average size of a hierarchy subgraph
3) Dump the graph using top-order -- starting from
On Thu, Aug 15, 2013 at 9:46 AM, Jan Hubicka hubi...@ucw.cz wrote:
Some suggestions for the future:
1) add summary info in the odr dump -- i.e. for each node, dump the
number of direct bases, direct subtypes, number of all bases, all
subtypes;
OK, can add that.
2) add statistics dump --
yes.
thanks,
David
On Thu, Aug 15, 2013 at 9:50 AM, Wei Mi w...@google.com wrote:
Turning off X86_TUNE_USE_VECTOR_CONVERTS uses cvtss2sd instead of
unpcklps+cvtps2pd, which is better for some recent intel micro arch
such as westmere and sandybridge. So turn it off for m_GENERIC and
Ping?
David
On Mon, Aug 12, 2013 at 9:54 AM, Xinliang David Li davi...@google.com wrote:
Fixed some formatting issues and typos. There are no regressions with
the attached patch. Ok for trunk?
thanks,
David
On Wed, Aug 7, 2013 at 10:12 PM, Xinliang David Li davi...@google.com wrote:
Hi
On Wed, Aug 14, 2013 at 9:01 AM, Jan Hubicka hubi...@ucw.cz wrote:
2013-08-07 Xinliang David Li davi...@google.com
* config/i386/i386.h: Adjust macro definition.
* config/i386/i386.opt: Define two new options.
* config/i386/x86-tune.def: Add arch selector field
;
char alg_name[128];
char align[16];
Index: ChangeLog
===
--- ChangeLog (revision 201732)
+++ ChangeLog (working copy)
@@ -1,4 +1,8 @@
2013-08-14 Xinliang David Li davi...@google.com
+ * config/i386/i386.c
Ok. Will drop it in the next patch.
David
On Wed, Aug 14, 2013 at 11:23 AM, Jakub Jelinek ja...@redhat.com wrote:
On Wed, Aug 14, 2013 at 08:17:55PM +0200, Marek Polacek wrote:
On Wed, Aug 14, 2013 at 11:12:27AM -0700, Xinliang David Li wrote:
Index: ChangeLog
, align))
Index: ChangeLog
===
--- ChangeLog (revision 201735)
+++ ChangeLog (working copy)
@@ -1,6 +1,12 @@
2013-08-14 Xinliang David Li davi...@google.com
+
+ * config/i386/i386.c (ix86_option_override_internal
Fixed some formatting issues and typos. There are no regressions with
the attached patch. Ok for trunk?
thanks,
David
On Wed, Aug 7, 2013 at 10:12 PM, Xinliang David Li davi...@google.com wrote:
Hi, the attached is a follow up patch. The patch includes the following
changes:
1. new option
It might be more flexible to represent the analysis results as a
type/inheritance graph -- i.e. explicitly introduce inheritance edge
class to capture the interitence relationship (offset, etc) between
two class nodes. The 'method' should probably be augmented to include
vtable slot index info.
The attached patch defines a new parameter to decide if icall
promotion (LIPO) is using a new hotness based heuristic or not. (See
Honza's speculative call support patch for reference).
David
Index: params.def
===
--- params.def
I like the approach in general -- in the past, indirect call promotion
and function inlining heuristics are disconnected -- which can lead to
either missing promotions or useless ones. This approach solves the
problem.
On Sun, Aug 11, 2013 at 4:11 PM, Jan Hubicka hubi...@ucw.cz wrote:
Hi,
this
Is this version ok for trunk?
thanks,
David
On Thu, Aug 8, 2013 at 9:31 AM, Xinliang David Li davi...@google.com wrote:
Updated.
thanks,
David
On Thu, Aug 8, 2013 at 8:18 AM, Joseph S. Myers jos...@codesourcery.com
wrote:
On Wed, 7 Aug 2013, Xinliang David Li wrote:
Updated patch
need to do tunings
for those constants, we can revisit it.
thanks,
David
This has chance to do the right thing with optimize attribute specifying
algorithms
and with the longer term threading plan.
Honza
thanks,
David
On Thu, Aug 8, 2013 at 9:31 AM, Xinliang David Li davi
On Fri, Aug 9, 2013 at 1:24 PM, Jan Hubicka hubi...@ucw.cz wrote:
I have not looked at the details. One high level question: this form
seems to only support one indirect target case. LIPO uses TOPN
indirect target profiling (tracking multiple targets), which can be
used by LTO as well (when
On Fri, Aug 9, 2013 at 4:21 PM, Jan Hubicka hubi...@ucw.cz wrote:
On Fri, Aug 9, 2013 at 1:24 PM, Jan Hubicka hubi...@ucw.cz wrote:
I have not looked at the details. One high level question: this form
seems to only support one indirect target case. LIPO uses TOPN
indirect target profiling
Updated patch attached (fixed header, buffer overflow, and warning --
error problems).
Ok for trunk?
thanks,
David
On Wed, Aug 7, 2013 at 6:04 PM, Joseph S. Myers jos...@codesourcery.com wrote:
On Wed, 7 Aug 2013, Xinliang David Li wrote:
Why the exception? This should only be used
Updated.
thanks,
David
On Thu, Aug 8, 2013 at 8:18 AM, Joseph S. Myers jos...@codesourcery.com wrote:
On Wed, 7 Aug 2013, Xinliang David Li wrote:
Updated patch attached (fixed header, buffer overflow, and warning --
error problems).
You still have diagnostics starting with a capital
Ping?
David
On Sat, Aug 3, 2013 at 12:36 PM, Xinliang David Li davi...@google.com wrote:
Hi, GCC/i386 currently has about 73 boolean parameters/knobs (defined
in ix86_tune_features[], indexed by ix86_tune_indices) to perform
micro-arch specific performance tuning. However such settings
On Wed, Aug 7, 2013 at 8:37 AM, Jan Hubicka hubi...@ucw.cz wrote:
Ping?
David
On Sat, Aug 3, 2013 at 12:36 PM, Xinliang David Li davi...@google.com
wrote:
Hi, GCC/i386 currently has about 73 boolean parameters/knobs (defined
in ix86_tune_features[], indexed by ix86_tune_indices
Fixed the do while formatting. Ok for trunk with this version?
thanks,
David
On Tue, Aug 6, 2013 at 2:42 AM, Jan Hubicka hubi...@ucw.cz wrote:
2013-08-02 Xinliang David Li davi...@google.com
* config/i386/stringop.def: New file.
* config/i386/stringop.opt: New file
ok for google branch.
David
On Tue, Aug 6, 2013 at 3:41 PM, Dehao Chen de...@google.com wrote:
Patch updated.
http://codereview.appspot.com/12079043
Thanks,
Dehao
On Fri, Aug 2, 2013 at 11:21 AM, Xinliang David Li davi...@google.com wrote:
More to follow.
David
static void
On Wed, Aug 7, 2013 at 4:54 PM, Joseph S. Myers jos...@codesourcery.com wrote:
On Sun, 4 Aug 2013, Andi Kleen wrote:
Richard Biener richard.guent...@gmail.com writes:
The patch fails to add documentation.
That seems like a feature, it's likely not useful for the general
public. More for
On Wed, Aug 7, 2013 at 5:23 PM, Joseph S. Myers jos...@codesourcery.com wrote:
On Wed, 7 Aug 2013, Xinliang David Li wrote:
Index: config/i386/stringop.def
===
--- config/i386/stringop.def (revision 0)
+++ config/i386
On Tue, Aug 6, 2013 at 5:37 AM, Martin Jambor mjam...@suse.cz wrote:
Hi,
On Mon, Aug 05, 2013 at 10:37:00PM -0700, Teresa Johnson wrote:
This patch ports messages to the new dump framework,
It would be great this new framework was documented somewhere. I lost
track of what was agreed it
On Tue, Aug 6, 2013 at 2:42 AM, Jan Hubicka hubi...@ucw.cz wrote:
2013-08-02 Xinliang David Li davi...@google.com
* config/i386/stringop.def: New file.
* config/i386/stringop.opt: New file.
* config/i386/i386-opts.h: Include stringopt.def.
* config
, probably you could tune existing
cost-models as well? Is it possible?
the option is designed for purpose like this.
thanks,
David
Michael
On 5 August 2013 20:44, Xinliang David Li davi...@google.com wrote:
thanks. Updated patch attached.
David
On Mon, Aug 5, 2013 at 3:57 AM, Michael V
Forgot the patch.
David
On Tue, Aug 6, 2013 at 9:42 AM, Xinliang David Li davi...@google.com wrote:
Corrected two small problems reported by the style checker (The
warnings about the EnumValue for options in stringopt.opt are not
valid).
On Tue, Aug 6, 2013 at 1:46 AM, Michael Zolotukhin
at 8:57 AM, Xinliang David Li davi...@google.com
wrote:
On Tue, Aug 6, 2013 at 5:37 AM, Martin Jambor mjam...@suse.cz wrote:
Hi,
On Mon, Aug 05, 2013 at 10:37:00PM -0700, Teresa Johnson wrote:
This patch ports messages to the new dump framework,
It would be great this new framework
On Mon, Aug 5, 2013 at 8:26 AM, H.J. Lu hjl.to...@gmail.com wrote:
On Sun, Aug 4, 2013 at 9:23 AM, Xinliang David Li davi...@google.com wrote:
On Sun, Aug 4, 2013 at 4:40 AM, Richard Biener
richard.guent...@gmail.com wrote:
Xinliang David Li davi...@google.com wrote:
Hi, GCC/i386 currently has
: afair, comments should end with
dot-space-space.
Michael
On 04 Aug 20:01, Xinliang David Li wrote:
The attached is a new patch implementing the stringop inline strategy
control using two new -m options:
-mmemcpy-strategy=
-mmemset-strategy=
See changes in doc/invoke.texi
ok -- makes sense. This can be done as a follow up patch.
thanks,
David
On Mon, Aug 5, 2013 at 10:59 AM, H.J. Lu hjl.to...@gmail.com wrote:
On Mon, Aug 5, 2013 at 9:33 AM, Xinliang David Li davi...@google.com wrote:
On Mon, Aug 5, 2013 at 8:26 AM, H.J. Lu hjl.to...@gmail.com wrote:
On Sun
On Sun, Aug 4, 2013 at 4:40 AM, Richard Biener
richard.guent...@gmail.com wrote:
Xinliang David Li davi...@google.com wrote:
Hi, GCC/i386 currently has about 73 boolean parameters/knobs (defined
in ix86_tune_features[], indexed by ix86_tune_indices) to perform
micro-arch specific performance
?
thanks,
David
2013-08-02 Xinliang David Li davi...@google.com
* config/i386/stringop.def: New file.
* config/i386/stringop.opt: New file.
* config/i386/i386-opts.h: Include stringopt.def.
* config/i386/i386.opt: Include stringopt.opt.
* config/i386/i386.c
parameters to do that. Potential usage includes
per-application libcall strategy min-size tuning based on summary data
with FDO (e.g, instruction workset size).
Bootstrap and tested on x86_64/linux. Ok for trunk?
thanks,
David
2013-08-02 Xinliang David Li davi...@google.com
and names of the features.
Ok for trunk?
thanks,
David
2013-08-03 Xinliang David Li davi...@google.com
* config/i386/i386.opt: New option -mtune-ctrl=.
* config/i386/x86-tune.def: New file.
* config/i386/i386.h: include x86-tune.def.
* config/i386/i386.c
=
afdo_source_profile-get_function_instance_by_decl (
current_function_decl);
need a new line after decls.
On Thu, Aug 1, 2013 at 2:49 PM, Dehao Chen de...@google.com wrote:
On Wed, Jul 31, 2013 at 10:14 AM, Xinliang David Li davi...@google.com
wrote:
Another round of documentation and naming
strategy min-size tuning based on summary data
with FDO (e.g, instruction workset size).
Bootstrap and tested on x86_64/linux. Ok for trunk?
thanks,
David
2013-08-02 Xinliang David Li davi...@google.com
* params.def: New parameters.
* config/i386/i386.c
Sorry for the delay. The patch is ok and I have committed it to the
google branch.
thanks,
David
On Thu, Aug 1, 2013 at 4:51 PM, Dinar Temirbulatov di...@kugelworks.com wrote:
Ping?
Hi,
Here is the patch, Tested by profiledbootstrap. Ok for google gcc-4.8?
thanks, Dinar.
...@google.com wrote:
Patch updated to fix the code style and documentation.
Thanks,
Dehao
On Tue, Jul 30, 2013 at 2:24 PM, Xinliang David Li davi...@google.com wrote:
I have some preliminary comments. Mostly just related to code style
and missing documentation.
David
#define
I need to understand why this affects profile bootstrap -- is this due
to file name conflict?
The fix is wrong -- please do not remove the parameter. If it is a
problem, a better fix is to change the default parameter value to 0.
David
On Tue, Jul 30, 2013 at 11:56 AM, Teresa Johnson
mentioned, setting
the default value of the parameter to 0).
-Rong
On Tue, Jul 30, 2013 at 12:02 PM, Xinliang David Li davi...@google.com
wrote:
I need to understand why this affects profile bootstrap -- is this due
to file name conflict?
The fix is wrong -- please do not remove
On Tue, Jul 30, 2013 at 1:44 PM, Rong Xu x...@google.com wrote:
The following patch forces the command line match for -ansi option
in LIPO use build. Otherwise, it gets various undefined symbol errors.
Parsing error as you have clarified.
This is exposed in LIPO random grouping test.
Tested
;
Missing documentation for the interfaces
NameSymbolMap map_;
};
class ModuleMap {
Need documentation.
On Tue, Jul 30, 2013 at 11:03 AM, Dehao Chen de...@google.com wrote:
I just rebased the CL to head and updated the patch.
Thanks,
Dehao
On Tue, Jul 30, 2013 at 10:09 AM, Xinliang David
Ok for google branches. Many changes in coverage.c (such as
get_coverage_counts) and value-prof.c need to be in trunk too.
David
On Thu, Jul 25, 2013 at 10:02 AM, Teresa Johnson tejohn...@google.com wrote:
This patch ports the remaining -fopt-info messages that had been added
to google/gcc-4_7
indentation spaces and
On Fri, Jul 26, 2013 at 10:18 AM, Teresa Johnson tejohn...@google.com wrote:
Thanks. I'll work on a trunk patch to send next week. Teresa
On Fri, Jul 26, 2013 at 10:05 AM, Xinliang David Li davi...@google.com
wrote:
Ok for google branches. Many changes in coverage.c
, Xinliang David Li davi...@google.com wrote:
is it better to hoist the newline dump before all the branches?
David
On Fri, Jul 26, 2013 at 1:06 PM, Teresa Johnson tejohn...@google.com wrote:
After converting so many more messages to the new dump framework with my
previous change, I noticed a bug
the patch is ok.
thanks,
David
On Thu, Jul 25, 2013 at 2:53 PM, Sriraman Tallam tmsri...@google.com wrote:
On Wed, Jul 24, 2013 at 3:21 PM, Xinliang David Li davi...@google.com wrote:
Can you collect some number on ggc_memory increase with this change
in profile_gen build -- the value
Can you collect some number on ggc_memory increase with this change
in profile_gen build -- the value is recorded gcov_module_info object
in coverage.c.
thanks,
David
On Wed, Jul 24, 2013 at 3:14 PM, Sriraman Tallam tmsri...@google.com wrote:
The following test-case fails in LIPO mode during
Ok after testing.
David
On Wed, Jul 10, 2013 at 4:59 PM, Dehao Chen de...@google.com wrote:
This fix is need if you want to collect AutoFDO profile on
AutoFDO+LIPO optimized binary.
Testing on going.
OK for google branches?
Thanks,
Dehao
Index: gcc/toplev.c
You found the root cause of the bootstrap profile with vectorizer
turned on, but the patch does not look correct to me. The sorting
needs to make sure accesses with the same (base, size, offset, step,
r_w_kind) are placed together (see how the sorted array is used to
build the interleaving chain).
Is it possible to add a test case?
David
On Mon, Jul 8, 2013 at 5:55 PM, Dehao Chen de...@google.com wrote:
In lookup_stmt_eh_lp, negative return value indicates a MUST_NOT_THROW
region index. In this case, we should *not* add an EH edge during VPT.
Bootstrapped and passed regression test.
, Sharad Singhai sing...@google.com wrote:
On Mon, Jun 24, 2013 at 9:14 PM, Xinliang David Li davi...@google.com wrote:
To avoid printing twice, can you just do
opt_type = (warn_conversion ? OPT_Wconversion : OPT_Wreal_conversion);
warning_at (loc, opt_type, ...);
Thanks for the suggestion. I have
301 - 400 of 970 matches
Mail list logo
