public inbox for java-patches@gcc.gnu.org
 help / color / mirror / Atom feed
* Re: [JAVA PATCH] Enable more array bounds check elimination
@ 2016-02-22 23:02 Roger Sayle
  2016-02-23  9:56 ` Andrew Haley
  0 siblings, 1 reply; 12+ messages in thread
From: Roger Sayle @ 2016-02-22 23:02 UTC (permalink / raw)
  To: Andrew Haley; +Cc: java-patches


Hi Andrew,

On 02/22/2016, Andrew Haley wrote:
> I take it that this only really works with new arrays?

Alas yes, this patch only addresses the case that the array allocation is
visible (in the same method after inlining) as the array bounds check,
which fortunately is fairly frequent, especially for initializers.

I'd assumed that the middle-end optimizers were already doing a reasonable
job for the harder cases, but inspired by your question above, I've done some
more investigating and noticed other areas where improvements are possible.

For example

	int len = arr.length;
	for (int i=0; i<arr.length; i++)
	  arr[i] = arr[i]+1;

eliminates all bounds checks when arr is double[], but disappointingly
not when arr is int[].  I suspect that the middle-end optimizers assume
the worst, perhaps that arr.data[-1] potentially aliases arr.length, or
similar.  Obviously, java doesn't allow negative array indices, but I
suspect the gcj front-end's gimple doesn't manage to convey this.

I'm personally impressed that GCC recognizes that although

	for (int i=0; i<3; i++)
	  ... arr[i] ...

typically needs to perform index checking on every iteration,
that the reversed loop

	for (int i=2; i>=0; i--)
	  ... arr[i] ...

only needs to perform array bounds checking on the first iteration,
i.e. optimizes away the later checks [in unrolled loops].  This might
seem obvious, but there's a lot of analysis required to recover this
from java's bytecodes.


Please point me towards any relevant postings (of yours) on the subject of
gcj bounds check elimination, as I'd love to catch up on current thinking.

Cheers,

Roger
--

^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [JAVA PATCH] Enable more array bounds check elimination
  2016-02-22 23:02 [JAVA PATCH] Enable more array bounds check elimination Roger Sayle
@ 2016-02-23  9:56 ` Andrew Haley
  2016-02-23 17:45   ` Andrew Hughes
  0 siblings, 1 reply; 12+ messages in thread
From: Andrew Haley @ 2016-02-23  9:56 UTC (permalink / raw)
  To: Roger Sayle; +Cc: java-patches

On 22/02/16 23:02, Roger Sayle wrote:
> Please point me towards any relevant postings (of yours) on the subject of
> gcj bounds check elimination, as I'd love to catch up on current thinking.

I'm not sure that there are any, really.  The discussions I can
remember were all done in person, and I didn't get positive
feedback about the idea of adding to the type system in GCC's
middle end.

Incidentally, we have been talking about EOL for GCJ for some years
now.  GCC 6 will very likely be the last GCJ.

Andrew.

^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [JAVA PATCH] Enable more array bounds check elimination
  2016-02-23  9:56 ` Andrew Haley
@ 2016-02-23 17:45   ` Andrew Hughes
  2016-02-24  1:11     ` Roger Sayle
  0 siblings, 1 reply; 12+ messages in thread
From: Andrew Hughes @ 2016-02-23 17:45 UTC (permalink / raw)
  To: Andrew Haley; +Cc: Roger Sayle, java-patches

----- Original Message -----
> On 22/02/16 23:02, Roger Sayle wrote:
> > Please point me towards any relevant postings (of yours) on the subject of
> > gcj bounds check elimination, as I'd love to catch up on current thinking.
> 
> I'm not sure that there are any, really.  The discussions I can
> remember were all done in person, and I didn't get positive
> feedback about the idea of adding to the type system in GCC's
> middle end.
> 
> Incidentally, we have been talking about EOL for GCJ for some years
> now.  GCC 6 will very likely be the last GCJ.

Yes, I believe we agreed to regard it as deprecated in 6 and remove it
during the lifetime of 7 [0]. As such, this patch probably needs to
go to the 6 branch too if it's ever going to see the light of day.

> 
> Andrew.
> 
> 

[0] https://gcc.gnu.org/ml/java-patches/2015-q3/msg00041.html
-- 
Andrew :)

Senior Free Java Software Engineer
Red Hat, Inc. (http://www.redhat.com)

PGP Key: ed25519/35964222 (hkp://keys.gnupg.net)
Fingerprint = 5132 579D D154 0ED2 3E04  C5A0 CFDA 0F9B 3596 4222


^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [JAVA PATCH] Enable more array bounds check elimination
  2016-02-23 17:45   ` Andrew Hughes
@ 2016-02-24  1:11     ` Roger Sayle
  2016-02-24  3:23       ` Andrew Hughes
  2016-02-24  9:53       ` Andrew Haley
  0 siblings, 2 replies; 12+ messages in thread
From: Roger Sayle @ 2016-02-24  1:11 UTC (permalink / raw)
  To: Andrew Hughes; +Cc: Andrew Haley, java-patches


Hi Andrew,

[Spooky, Graham Birtwistle was also my PhD examiner, and Rob Pooley a
lecturer in the department at the time].

On 23 Feb 2016, at 18:45, Andrew Hughes <gnu.andrew@redhat.com> wrote:
> Yes, I believe we agreed to regard it as deprecated in 6 and remove it
> during the lifetime of 7 [0]. As such, this patch probably needs to
> go to the 6 branch too if it's ever going to see the light of day.
> 
> [0] https://gcc.gnu.org/ml/java-patches/2015-q3/msg00041.html

Thanks for this very useful pointer.

Following the thread above, there seems to be an unclear lack of distinction
between different aspects and roles of GCJ, that I hope you can clear up.

I completely agree that maintaining libjava/classpath has been a pain, tracking a
continual moving target of a huge API, and now obsolete thanks to OpenJDK
and IcedTea.  But what I'm interested in is the Free Software Foundation's
ahead-of-time compiler for transforming Java bytecodes/class files to binary
executables, i.e. jc1.

My understanding is that the JVM bytecode instruction set has been 
relatively stable over the ages, and that "gcj" compiles the contents of
most modern class files without problems, it's only the run-time library
support that hasn't kept pace with the times.

Is there any reason why gcj 7.x couldn't/doesn't use OpenJDK as its runtime library?

When a better open source front-end came along (ecj), gcj switched to using
that to reduce the overhead of tracking syntax changes to the Java language.
Now that a better run-time library exists, reducing the overhead of tracking
library API changes, it seems odd not to switch to it, but to instead end-of-life
the ahead of time compiler, specifically the translation to gimple front-end.

To me obsoleting GNU classpath, which is holding gcj back to Java 1.5 [for
example no java.io.IOException(String,Throwable) constructor] is independent
of the potential performance benefits of a gcj compiled tomcat, eclipse or
LibreOffice.

Perhaps, I'm missing something or under-estimating the complexity of porting
the bits of OpenJDK that aren't written in Java, but I'd imagine this is significantly
simpler than attempting to update the current libjava to 1.8 and beyond.

Again apologies for being late to the discussion,  Perhaps the thread you cited
didn't capture the entirety of the discussion, and the distinct benefits/overheads
between gcj's runtime and the front-end itself were covered elsewhere.

Thanks in advance,

Roger
--

^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [JAVA PATCH] Enable more array bounds check elimination
  2016-02-24  1:11     ` Roger Sayle
@ 2016-02-24  3:23       ` Andrew Hughes
  2016-02-24  9:53       ` Andrew Haley
  1 sibling, 0 replies; 12+ messages in thread
From: Andrew Hughes @ 2016-02-24  3:23 UTC (permalink / raw)
  To: Roger Sayle; +Cc: Andrew Haley, java-patches

----- Original Message -----
> 
> Hi Andrew,
> 
> [Spooky, Graham Birtwistle was also my PhD examiner, and Rob Pooley a
> lecturer in the department at the time].
> 

Heh, have you been Googling me? :-)

> On 23 Feb 2016, at 18:45, Andrew Hughes <gnu.andrew@redhat.com> wrote:
> > Yes, I believe we agreed to regard it as deprecated in 6 and remove it
> > during the lifetime of 7 [0]. As such, this patch probably needs to
> > go to the 6 branch too if it's ever going to see the light of day.
> > 
> > [0] https://gcc.gnu.org/ml/java-patches/2015-q3/msg00041.html
> 
> Thanks for this very useful pointer.
> 
> Following the thread above, there seems to be an unclear lack of distinction
> between different aspects and roles of GCJ, that I hope you can clear up.
> 
> I completely agree that maintaining libjava/classpath has been a pain,
> tracking a
> continual moving target of a huge API, and now obsolete thanks to OpenJDK
> and IcedTea.  But what I'm interested in is the Free Software Foundation's
> ahead-of-time compiler for transforming Java bytecodes/class files to binary
> executables, i.e. jc1.
> 
> My understanding is that the JVM bytecode instruction set has been
> relatively stable over the ages, and that "gcj" compiles the contents of
> most modern class files without problems, it's only the run-time library
> support that hasn't kept pace with the times.

I wouldn't be too sure on this. Bytecode changes were introduced in 7 & 8,
notably invokedynamic. I'm not sure if gcj can handle this. Andrew Haley
would know more, as I believe he's been keeping it on life support these
last few years.

> 
> Is there any reason why gcj 7.x couldn't/doesn't use OpenJDK as its runtime
> library?

I don't think there's any particular technical reason, though one would have
to investigate further to know for sure. The idea was mused when OpenJDK first
appeared but no-one has doing the porting work in the last (nearly) nine years.

From working on supporting CACAO+OpenJDK and JamVM+OpenJDK (also former Classpath
JVMs), it's not just a one-stop porting effort. Oracle regularly make changes
to what is still an undocumented interface between the JVM and the OpenJDK
class library, and we've had to do our best to mirror these changes in
CACAO and JamVM e.g. [0].

> 
> When a better open source front-end came along (ecj), gcj switched to using
> that to reduce the overhead of tracking syntax changes to the Java language.
> Now that a better run-time library exists, reducing the overhead of tracking
> library API changes, it seems odd not to switch to it, but to instead
> end-of-life
> the ahead of time compiler, specifically the translation to gimple front-end.

The difference is that then gcj had an active community of people working on it
and a strong need to do that in order to support 1.5. It wasn't just a case
of ecj being 'better'. The alternative would have been to implement support
for the 1.5 features, like generics and annotations, in the old gcj frontend.
Given there was already a FOSS implementation of this work, using ecj was
the quicker solution.

> 
> To me obsoleting GNU classpath, which is holding gcj back to Java 1.5 [for
> example no java.io.IOException(String,Throwable) constructor] is independent
> of the potential performance benefits of a gcj compiled tomcat, eclipse or
> LibreOffice.

Actually, Classpath has had that since 2014:

    Add missing IOException constructors.
    
    2014-05-07  Andrew John Hughes  <gnu_andrew@member.fsf.org>
    
        * java/io/IOException.java:
        (IOException(String,Throwable)): Add missing constructor.
        (IOException(Throwable)): Likewise.


and a number of other changes beyond 1.5. Classpath is not holding back gcj
and has technically seen more development than gcj has recently. It will continue
to be used to bootstrap IcedTea.

gcj doesn't use Classpath directly, but rather instead maintains a fork which
has to be occasionally merged with upstream. This hasn't been done in over
three years: http://git.savannah.gnu.org/cgit/classpath.git/log/?ofs=50

> 
> Perhaps, I'm missing something or under-estimating the complexity of porting
> the bits of OpenJDK that aren't written in Java, but I'd imagine this is
> significantly
> simpler than attempting to update the current libjava to 1.8 and beyond.
> 

It is but it still needs someone to do it. This move also doesn't mean that
can't happen in the future. The code will still be available. It's not
going to disappear completely. But keeping it as part of GCC means it has
to be maintained and kept in sync with the rest of GCC, as well as being
built and tested. This all adds overhead for GCC developers who aren't
really interested in gcj. gcj also adds a significant size footprint
as it bundles pre-compiled class files for the entire class library.

FWIW, from my side, an OpenJDK-using gcj would defeat its usefulness for me,
as I use it to bootstrap OpenJDK/IcedTea. If gcj started requiring OpenJDK
to run, it would no longer help in this situation; effectively it would
introduce a circular dependency.

> Again apologies for being late to the discussion,  Perhaps the thread you
> cited
> didn't capture the entirety of the discussion, and the distinct
> benefits/overheads
> between gcj's runtime and the front-end itself were covered elsewhere.
> 
> Thanks in advance,
> 
> Roger
> --
> 
> 

I hope that helps clarify the issues involved,
-- 
Andrew :)

Senior Free Java Software Engineer
Red Hat, Inc. (http://www.redhat.com)

PGP Key: ed25519/35964222 (hkp://keys.gnupg.net)
Fingerprint = 5132 579D D154 0ED2 3E04  C5A0 CFDA 0F9B 3596 4222


^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [JAVA PATCH] Enable more array bounds check elimination
  2016-02-24  1:11     ` Roger Sayle
  2016-02-24  3:23       ` Andrew Hughes
@ 2016-02-24  9:53       ` Andrew Haley
  1 sibling, 0 replies; 12+ messages in thread
From: Andrew Haley @ 2016-02-24  9:53 UTC (permalink / raw)
  To: Roger Sayle, Andrew Hughes; +Cc: java-patches

On 24/02/16 01:11, Roger Sayle wrote:
> 
> Following the thread above, there seems to be an unclear lack of
> distinction between different aspects and roles of GCJ, that I hope
> you can clear up.
> 
> I completely agree that maintaining libjava/classpath has been a
> pain, tracking a continual moving target of a huge API, and now
> obsolete thanks to OpenJDK and IcedTea.  But what I'm interested in
> is the Free Software Foundation's ahead-of-time compiler for
> transforming Java bytecodes/class files to binary executables,
> i.e. jc1.
> 
> My understanding is that the JVM bytecode instruction set has been
> relatively stable over the ages, and that "gcj" compiles the
> contents of most modern class files without problems, it's only the
> run-time library support that hasn't kept pace with the times.
> 
> Is there any reason why gcj 7.x couldn't/doesn't use OpenJDK as its
> runtime library?

Yes.  It's important to realize that, while bytecodes are added very
infrequently, new entry points to the virtual machine are added
frequently.  Some of these are simply optimizations, but many are
needed for correct operation.  GCJ has never supported any of the
entry points required by the OpenJDK class library.

> When a better open source front-end came along (ecj), gcj switched
> to using that to reduce the overhead of tracking syntax changes to
> the Java language.  Now that a better run-time library exists,
> reducing the overhead of tracking library API changes, it seems odd
> not to switch to it, but to instead end-of-life the ahead of time
> compiler, specifically the translation to gimple front-end.

It's a lot of work, and the work gets greater with every day that
passes.  A couple of years ago I guesstimated the work that would be
involved, and I reckoned it would be at least six months, and that's
six months for me.  Nobody could do it any faster.  Now it's much
longer, more than a year.  I would love to reactivate GCJ with a
modern class library, but I can't figure out any way to justify the
time.

The other point is that it would be unsatisfying.  OpenJDK's
optimizers do many optimizations which can't be done with any
ahead-of-time compiler, so GCJ would always fall behind.

Andrew.

^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [JAVA PATCH] Enable more array bounds check elimination
  2016-07-14 17:12   ` Andrew Hughes
@ 2016-07-14 17:43     ` Andrew Haley
  0 siblings, 0 replies; 12+ messages in thread
From: Andrew Haley @ 2016-07-14 17:43 UTC (permalink / raw)
  To: java-patches

On 14/07/16 18:12, Andrew Hughes wrote:
> snip...
> 
>> At a very high level, you should be aware of a general belief that GCJ's
>> life is limited.  There's been various calls to deprecate it.  So
>> spending a lot of time optimizing GCJ's output may not be the best use
>> of your skills :-)
> 
> Unless things have changed since it was last discussed, the plan was
> for it to be deprecated in GCC 6 and removed in 7. If that's the case,
> these changes may sadly not see a release.
> 
> It does seem that no-one has bitten the bullet of removing it yet though.

It's on my list.  The first thing to do before removing the sources
is to change the top-level autoconf and automake scripts not to resurse
into the gcj scripts, but my automake-fu is rather rusty.

Andrew.



^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [JAVA PATCH] Enable more array bounds check elimination
  2016-07-13 20:07 ` Jeff Law
@ 2016-07-14 17:12   ` Andrew Hughes
  2016-07-14 17:43     ` Andrew Haley
  0 siblings, 1 reply; 12+ messages in thread
From: Andrew Hughes @ 2016-07-14 17:12 UTC (permalink / raw)
  To: Jeff Law; +Cc: roger, java-patches, gcc-patches

snip...

> 
> At a very high level, you should be aware of a general belief that GCJ's
> life is limited.  There's been various calls to deprecate it.  So
> spending a lot of time optimizing GCJ's output may not be the best use
> of your skills :-)
> 

Unless things have changed since it was last discussed, the plan was
for it to be deprecated in GCC 6 and removed in 7. If that's the case,
these changes may sadly not see a release.

It does seem that no-one has bitten the bullet of removing it yet though.
-- 
Andrew :)

Senior Free Java Software Engineer
Red Hat, Inc. (http://www.redhat.com)

PGP Key: ed25519/35964222 (hkp://keys.gnupg.net)
Fingerprint = 5132 579D D154 0ED2 3E04  C5A0 CFDA 0F9B 3596 4222


^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [JAVA PATCH] Enable more array bounds check elimination
  2016-02-22 18:10 roger
  2016-02-22 18:13 ` Andrew Haley
  2016-02-22 18:18 ` Jeff Law
@ 2016-07-13 20:07 ` Jeff Law
  2016-07-14 17:12   ` Andrew Hughes
  2 siblings, 1 reply; 12+ messages in thread
From: Jeff Law @ 2016-07-13 20:07 UTC (permalink / raw)
  To: roger, java-patches, gcc-patches

On 02/22/2016 11:10 AM, roger@nextmovesoftware.com wrote:
>
> It has been a while since my last contribution.  The following patch allows GCC's optimizers
> to more aggressively eliminate and optimize java array bounds checks.  The results are
> quite impressive, for example producing a 26% performance improvement on the sieve.java
> benchmark given at http://keithlea.com/javabench/ on x86_64-pc-linux-gnu, reducing the
> runtime for 1 million iterations from 31.5 seconds on trunk, to 25.0s with this patch, in fact
> eliminating all array bounds checks.  This is close to the 22.3s of an equivalent C/C++
> implementation, and significantly closes the gap to Java HotSpot(TM) JIT at 23.0 seconds.
>
> The approach is to provide sufficient information in the gimple generated by the gcj front-end
> to allow the optimizers to do their thing.  For array allocations of constant length, I propose
> generating an additional (cheap) write to the array length field returned from _Jv_NewPrimArray,
> which is then sufficient to allow this value to propagate throughout the optimizers.
>
> This is probably best explained by a simple example.  Consider the array initializer below:
Thanks.  The example helped a lot.

At a very high level, you should be aware of a general belief that GCJ's 
life is limited.  There's been various calls to deprecate it.  So 
spending a lot of time optimizing GCJ's output may not be the best use 
of your skills :-)

>
> With the patch below, we now generate the much more informative .004t.gimple for this:
>
>     D.926 = _Jv_NewPrimArray (&_Jv_intClass, 3);
>     D.926->length = 3;
Essentially you're just storing back into the result the length that 
we'd passed to the allocator.  Cute.  Good to see that all the work 
we've done to propagate the RHS of that kind of statement into uses has 
paid off.

Presumably there's no reasonable way this could fail (like you're 
getting objects from a readonly part of memory), or the result gets used 
in some other thread which changes its size prior to the re-storing of 
the initial size?


>
>
> Achieving this result required two minor tweaks.   The first is to allow the array length constant
> to reach the newarray call, by allowing constants to remain on the quickstack.  This allows the
> call to _Jv_NewPrimArray to have a constant integer argument instead of the opaque #slot#0#0.
> Then in the code that constructs the call to _Jv_NewPrimArray we wrap it in a COMPOUND_EXPR
> allowing us to insert the superfluous, but helpful, write to the length field.
>
> Whilst working on this improvement I also noticed that the array bounds checks we were
> initially generating could also be improved.  Currently, an array bound check in 004t.gimple
> looks like:
>
>     D.925 = MEM[(struct int[] *)_ref_1_4.6].length;
>     D.926 = (unsigned int) D.925;
>     if (_slot_2_5.9 >= D.926) goto <D.927>; else goto <D.921>;
>     <D.927>:
>     _Jv_ThrowBadArrayIndex (_slot_2_5.8);
>     if (0 != 0) goto <D.928>; else goto <D.921>;
>     <D.928>:
>     iftmp.7 = 1;
>     goto <D.922>;
>     <D.921>:
>     iftmp.7 = 0;
>     <D.922>:
>
> Notice the unnecessary "0 != 0" and the dead assignments to iftmp.7 (which is unused).
FWIW, we're generally moving away from optimization in the language 
front-ends -- in particular folding, which you introduce in this patch 
on the array index.  Given the trajectory of GCJ I'm not going to worry 
about it though.


>
> With the patch below, we now not only avoid this conditional but also use __builtin_expect
> to inform the compiler that throwing an BadArrayIndex exception is typically unlikely.  i.e.
Sounds like a good thing as well.

>
>     D.930 = MEM[(struct int[] *)_ref_1_4.4].length;
>     D.931 = D.930 <= 1;
>     D.932 = __builtin_expect (D.931, 0);
>     if (D.932 != 0) goto <D.933>; else goto <D.934>;
>     <D.933>:
>     _Jv_ThrowBadArrayIndex (0);
>     <D.934>:
>
>
> The following patch has been tested on x86_64-pc-linux-gnu with a full make bootstrap
> and make check, with no new failures/regressions.
>
> Please let me know what you think (for stage 1 once it reopens)?
>
> Roger
> --
> Roger Sayle, Ph.D.
> CEO and founder
> NextMove Software Limited
> Registered in England No. 07588305
> Registered Office: Innovation Centre (Unit 23), Cambridge Science Park, Cambridge CB4 0EY
>
> 2016-02-21  Roger Sayle  <roger@nextmovesoftware.com>
>
> 	* expr.c (push_value): Only call flush_quick_stack for non-constant
> 	arguments.
> 	(build_java_throw_out_of_bounds_exception): No longer wrap calls
> 	to _Jv_ThowBadArrayIndex in a COMPOUND_EXPR as no longer needed.
> 	(java_check_reference): Annotate COND_EXPR with __builtin_expect
> 	to indicate that calling _Jv_ThrowNullPointerException is unlikely.
> 	(build_java_arrayaccess): Construct an unlikely COND_EXPR instead
> 	of a TRUTH_ANDIF_EXPR in a COMPOUND_EXPR.  Only generate array
> 	index MULT_EXPR when size_exp is not unity.
> 	(build_array_length_annotation): When optimizing, generate a write
> 	to the allocated array's length field to expose constant lengths
> 	to GCC's optimizers.
> 	(build_newarray): Call new build_array_length_annotation.
> 	(build_anewarray): Likewise.
Looks generally OK.  There's a whitespace nit in the call to build3 in 
build_java_arrayaccess (missing space between the function name and open 
paren).

I think this is OK for trunk after fixing the whitespace nit.

jeff


^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [JAVA PATCH] Enable more array bounds check elimination
  2016-02-22 18:10 roger
  2016-02-22 18:13 ` Andrew Haley
@ 2016-02-22 18:18 ` Jeff Law
  2016-07-13 20:07 ` Jeff Law
  2 siblings, 0 replies; 12+ messages in thread
From: Jeff Law @ 2016-02-22 18:18 UTC (permalink / raw)
  To: roger, java-patches, gcc-patches

On 02/22/2016 11:10 AM, roger@nextmovesoftware.com wrote:
>
> It has been a while since my last contribution.

It has been.  I ran into some of your eyesopen.com colleagues a couple 
years back at SC12 or SC13.  It'd obviously be great to have you 
contributing regularly again.

Note that we're rapidly approaching the gcc-6 release; since this isn't 
a regression bugfix, it'll be tabled until gcc-7 is open for development.

Jeff

^ permalink raw reply	[flat|nested] 12+ messages in thread

* Re: [JAVA PATCH] Enable more array bounds check elimination
  2016-02-22 18:10 roger
@ 2016-02-22 18:13 ` Andrew Haley
  2016-02-22 18:18 ` Jeff Law
  2016-07-13 20:07 ` Jeff Law
  2 siblings, 0 replies; 12+ messages in thread
From: Andrew Haley @ 2016-02-22 18:13 UTC (permalink / raw)
  To: java-patches

On 02/22/2016 06:10 PM, roger@nextmovesoftware.com wrote:
> Please let me know what you think (for stage 1 once it reopens)?

It's an interesting approach.  I have a few times discussed with
optimization people the idea of marking some fields as "effectively
const" so that we can do the bounds check elimination.  However, these
discussions came to naught.

I take it that this only really works with new arrays?

Thanks,

Andrew.

^ permalink raw reply	[flat|nested] 12+ messages in thread

* [JAVA PATCH] Enable more array bounds check elimination
@ 2016-02-22 18:10 roger
  2016-02-22 18:13 ` Andrew Haley
                   ` (2 more replies)
  0 siblings, 3 replies; 12+ messages in thread
From: roger @ 2016-02-22 18:10 UTC (permalink / raw)
  To: java-patches, gcc-patches

[-- Attachment #1: Type: text/plain, Size: 6877 bytes --]


It has been a while since my last contribution.  The following patch allows GCC's optimizers
to more aggressively eliminate and optimize java array bounds checks.  The results are
quite impressive, for example producing a 26% performance improvement on the sieve.java
benchmark given at http://keithlea.com/javabench/ on x86_64-pc-linux-gnu, reducing the
runtime for 1 million iterations from 31.5 seconds on trunk, to 25.0s with this patch, in fact
eliminating all array bounds checks.  This is close to the 22.3s of an equivalent C/C++
implementation, and significantly closes the gap to Java HotSpot(TM) JIT at 23.0 seconds.

The approach is to provide sufficient information in the gimple generated by the gcj front-end
to allow the optimizers to do their thing.  For array allocations of constant length, I propose
generating an additional (cheap) write to the array length field returned from _Jv_NewPrimArray,
which is then sufficient to allow this value to propagate throughout the optimizers.

This is probably best explained by a simple example.  Consider the array initializer below:

    private static int mk1[] = { 71,85,95 };

which gets compiled into the java byte code sequence below:

       0: iconst_3
       1: newarray       int
       3: dup
       4: iconst_0
       5: bipush        71
       7: iastore
       8: dup
       9: iconst_1
      10: bipush        85
      12: iastore
      13: dup
      14: iconst_2
      15: bipush        95
      17: iastore
      18: putstatic     #3                  // Field mk1:[I
      21: return

Currently, the .004t.gimple generated by gcj for the array allocation is the cryptic:

    #slot#0#0 = 3;
    #ref#0#2 = _Jv_NewPrimArray (&_Jv_intClass, #slot#0#0);
    #ref#1#4 = #ref#0#2;
    _ref_1_4.6 = #ref#1#4;

which unfortunately doesn't provide many clues for the middle-end, so we end up
generating the following .210t.optimized:

  void * _3 = _Jv_NewPrimArray (&_Jv_intClass, 3);
  int _4 = MEM[(struct int[] *)_3].length;
  unsigned int _5 = (unsigned int) _4;
  if (_4 == 0)
    goto <bb 3>;
  else
    goto <bb 4>;
  <bb 3>:
  _Jv_ThrowBadArrayIndex (0);
  <bb 4>:
  MEM[(int *)_3 + 12B] = 71;
  if (_5 == 1)
    goto <bb 5>;
  else
    goto <bb 6>;
  <bb 5>:
  _Jv_ThrowBadArrayIndex (1);
  <bb 6>:
  MEM[(int *)_3 + 16B] = 85;
  if (_5 == 2)
    goto <bb 7>;
  else
    goto <bb 8>;
  <bb 7>:
  _Jv_ThrowBadArrayIndex (2);
  <bb 8>:
  MEM[(int *)_3 + 20B] = 95;
  mk1 = _3;
  return;

which obviously contains three run-time array bounds checks.  These same checks appear
in the x86_64 assembly language:

	subq	$8, %rsp
	xorl	%eax, %eax
	movl	$3, %esi
	movl	$_Jv_intClass, %edi
	call	_Jv_NewPrimArray
	movl	8(%rax), %edx
	testl	%edx, %edx
	je	.L13
	cmpl	$1, %edx
	movl	$71, 12(%rax)
	je	.L14
	cmpl	$2, %edx
	movl	$85, 16(%rax)
	je	.L15
	movl	$95, 20(%rax)
	movq	%rax, _ZN9TestArray3mk1E(%rip)
	addq	$8, %rsp
	ret
.L13:	xorl	%edi, %edi
	xorl	%eax, %eax
	call	_Jv_ThrowBadArrayIndex
.L15:	movl	$2, %edi
	xorl	%eax, %eax
	call	_Jv_ThrowBadArrayIndex
.L14:	movl	$1, %edi
	xorl	%eax, %eax
	call	_Jv_ThrowBadArrayIndex


With the patch below, we now generate the much more informative .004t.gimple for this:

    D.926 = _Jv_NewPrimArray (&_Jv_intClass, 3);
    D.926->length = 3;

This additional write to the array length field of the newly allocated array enables much
more simplification.  The resulting .210t.optimized for our array initialization now becomes:

  struct int[3] * _3;
  _3 = _Jv_NewPrimArray (&_Jv_intClass, 3);
  MEM[(int *)_3 + 8B] = { 3, 71, 85, 95 };
  mk1 = _3;
  return;

And the x86_64 assembly code is also much prettier:

	subq	$8, %rsp
        movl    $3, %esi
        movl    $_Jv_intClass, %edi
        xorl    %eax, %eax
        call    _Jv_NewPrimArray
        movdqa  .LC0(%rip), %xmm0
        movq    %rax, _ZN9TestArray3mk1E(%rip)
        movups  %xmm0, 8(%rax)
	addq	$8, %rsp
	ret
.LC0:	.long 3
	.long 71
	.long 85
	.long 95


Achieving this result required two minor tweaks.   The first is to allow the array length constant
to reach the newarray call, by allowing constants to remain on the quickstack.  This allows the
call to _Jv_NewPrimArray to have a constant integer argument instead of the opaque #slot#0#0.
Then in the code that constructs the call to _Jv_NewPrimArray we wrap it in a COMPOUND_EXPR
allowing us to insert the superfluous, but helpful, write to the length field.

Whilst working on this improvement I also noticed that the array bounds checks we were
initially generating could also be improved.  Currently, an array bound check in 004t.gimple
looks like:

    D.925 = MEM[(struct int[] *)_ref_1_4.6].length;
    D.926 = (unsigned int) D.925;
    if (_slot_2_5.9 >= D.926) goto <D.927>; else goto <D.921>;
    <D.927>:
    _Jv_ThrowBadArrayIndex (_slot_2_5.8);
    if (0 != 0) goto <D.928>; else goto <D.921>;
    <D.928>:
    iftmp.7 = 1;
    goto <D.922>;
    <D.921>:
    iftmp.7 = 0;
    <D.922>:

Notice the unnecessary "0 != 0" and the dead assignments to iftmp.7 (which is unused).

With the patch below, we now not only avoid this conditional but also use __builtin_expect
to inform the compiler that throwing an BadArrayIndex exception is typically unlikely.  i.e.

    D.930 = MEM[(struct int[] *)_ref_1_4.4].length;
    D.931 = D.930 <= 1;
    D.932 = __builtin_expect (D.931, 0);
    if (D.932 != 0) goto <D.933>; else goto <D.934>;
    <D.933>:
    _Jv_ThrowBadArrayIndex (0);
    <D.934>:


The following patch has been tested on x86_64-pc-linux-gnu with a full make bootstrap
and make check, with no new failures/regressions.

Please let me know what you think (for stage 1 once it reopens)?

Roger
--
Roger Sayle, Ph.D.
CEO and founder
NextMove Software Limited
Registered in England No. 07588305
Registered Office: Innovation Centre (Unit 23), Cambridge Science Park, Cambridge CB4 0EY

2016-02-21  Roger Sayle  <roger@nextmovesoftware.com>

	* expr.c (push_value): Only call flush_quick_stack for non-constant
	arguments.
	(build_java_throw_out_of_bounds_exception): No longer wrap calls
	to _Jv_ThowBadArrayIndex in a COMPOUND_EXPR as no longer needed.
	(java_check_reference): Annotate COND_EXPR with __builtin_expect
	to indicate that calling _Jv_ThrowNullPointerException is unlikely.
	(build_java_arrayaccess): Construct an unlikely COND_EXPR instead
	of a TRUTH_ANDIF_EXPR in a COMPOUND_EXPR.  Only generate array
	index MULT_EXPR when size_exp is not unity.
	(build_array_length_annotation): When optimizing, generate a write
	to the allocated array's length field to expose constant lengths
	to GCC's optimizers.
	(build_newarray): Call new build_array_length_annotation.
	(build_anewarray): Likewise.


[-- Attachment #2: patch7a.txt --]
[-- Type: text/plain, Size: 7574 bytes --]

Index: expr.c
===================================================================
--- expr.c	(revision 233190)
+++ expr.c	(working copy)
@@ -37,6 +37,7 @@
 #include "jcf.h"
 #include "parse.h"
 #include "tree-iterator.h"
+#include "tree-eh.h"
 
 static void flush_quick_stack (void);
 static void push_value (tree);
@@ -54,6 +55,7 @@
 static void expand_java_pushc (int, tree);
 static void expand_java_return (tree);
 static void expand_load_internal (int, tree, int);
+static void expand_store_internal (tree, int, int);
 static void expand_java_NEW (tree);
 static void expand_java_INSTANCEOF (tree);
 static void expand_java_CHECKCAST (tree);
@@ -273,10 +275,12 @@
   /* If the value has a side effect, then we need to evaluate it
      whether or not the result is used.  If the value ends up on the
      quick stack and is then popped, this won't happen -- so we flush
-     the quick stack.  It is safest to simply always flush, though,
-     since TREE_SIDE_EFFECTS doesn't capture COMPONENT_REF, and for
-     the latter we may need to strip conversions.  */
-  flush_quick_stack ();
+     the quick stack.  It is safest to always flush non-constant
+     operands.  */
+  if (! TREE_CONSTANT (value)
+      || TREE_SIDE_EFFECTS (value)
+      || tree_could_trap_p (value))
+    flush_quick_stack ();
 }
 
 /* Pop a type from the type stack.
@@ -778,19 +782,13 @@
 {
   tree node;
 
-  /* We need to build a COMPOUND_EXPR because _Jv_ThrowBadArrayIndex()
-     has void return type.  We cannot just set the type of the CALL_EXPR below
-     to int_type_node because we would lose it during gimplification.  */
+  /* _Jv_ThrowBadArrayIndex() has void return type.  */
   gcc_assert (VOID_TYPE_P (TREE_TYPE (TREE_TYPE (soft_badarrayindex_node))));
   node = build_call_nary (void_type_node,
-			       build_address_of (soft_badarrayindex_node),
-			       1, index);
+			  build_address_of (soft_badarrayindex_node),
+			  1, index);
   TREE_SIDE_EFFECTS (node) = 1;
-
-  node = build2 (COMPOUND_EXPR, int_type_node, node, integer_zero_node);
-  TREE_SIDE_EFFECTS (node) = 1;	/* Allows expansion within ANDIF */
-
-  return (node);
+  return node;
 }
 
 /* Return the length of an array. Doesn't perform any checking on the nature
@@ -833,10 +831,12 @@
 {
   if (!flag_syntax_only && check)
     {
+      tree test;
       expr = save_expr (expr);
-      expr = build3 (COND_EXPR, TREE_TYPE (expr),
-		     build2 (EQ_EXPR, boolean_type_node,
-			     expr, null_pointer_node),
+      test = build2 (EQ_EXPR, boolean_type_node, expr, null_pointer_node);
+      test = build_call_expr (builtin_decl_implicit (BUILT_IN_EXPECT), 2,
+			      test, boolean_false_node);
+      expr = build3 (COND_EXPR, TREE_TYPE (expr), test,
 		     build_call_nary (void_type_node, 
 				      build_address_of (soft_nullpointer_node),
 				      0),
@@ -865,7 +865,7 @@
 tree
 build_java_arrayaccess (tree array, tree type, tree index)
 {
-  tree node, throw_expr = NULL_TREE;
+  tree node;
   tree data_field;
   tree ref;
   tree array_type = TREE_TYPE (TREE_TYPE (array));
@@ -882,9 +882,9 @@
     {
       /* Generate:
        * (unsigned jint) INDEX >= (unsigned jint) LEN
-       *    && throw ArrayIndexOutOfBoundsException.
+       *    ? throw ArrayIndexOutOfBoundsException : INDEX.
        * Note this is equivalent to and more efficient than:
-       * INDEX < 0 || INDEX >= LEN && throw ... */
+       * INDEX < 0 || INDEX >= LEN ? throw ... : INDEX. */
       tree test;
       tree len = convert (unsigned_int_type_node,
 			  build_java_array_length_access (array));
@@ -893,19 +893,14 @@
 			  len);
       if (! integer_zerop (test))
 	{
-	  throw_expr
-	    = build2 (TRUTH_ANDIF_EXPR, int_type_node, test,
-		      build_java_throw_out_of_bounds_exception (index));
-	  /* allows expansion within COMPOUND */
-	  TREE_SIDE_EFFECTS( throw_expr ) = 1;
+	  test = build_call_expr (builtin_decl_implicit (BUILT_IN_EXPECT), 2,
+				  test, boolean_false_node);
+          index = build3(COND_EXPR, int_type_node, test,
+			 build_java_throw_out_of_bounds_exception (index),
+                         index);
 	}
     }
 
-  /* If checking bounds, wrap the index expr with a COMPOUND_EXPR in order
-     to have the bounds check evaluated first. */
-  if (throw_expr != NULL_TREE)
-    index = build2 (COMPOUND_EXPR, int_type_node, throw_expr, index);
-
   data_field = lookup_field (&array_type, get_identifier ("data"));
 
   ref = build3 (COMPONENT_REF, TREE_TYPE (data_field),    
@@ -919,9 +914,11 @@
 
   /* Multiply the index by the size of an element to obtain a byte
      offset.  Convert the result to a pointer to the element type.  */
-  index = build2 (MULT_EXPR, sizetype, 
-		  fold_convert (sizetype, index), 
-		  size_exp);
+  index = fold_convert (sizetype, index);
+  if (! integer_onep (size_exp))
+    {
+      index = build2 (MULT_EXPR, sizetype, index, size_exp);
+    }
 
   /* Sum the byte offset and the address of the data field.  */
   node = fold_build_pointer_plus (node, index);
@@ -1026,6 +1023,34 @@
   return indexed_type;
 }
 
+/* When optimizing, wrap calls to array allocation functions taking
+   constant length arguments, in a COMPOUND_EXPR, containing an
+   explict assignment of the .length field, for GCC's optimizers.  */
+
+static tree
+build_array_length_annotation (tree call, tree length)
+{
+  if (optimize
+      && TREE_CONSTANT (length)
+      && is_array_type_p (TREE_TYPE (call)))
+    {
+      tree type, note;
+      type = TREE_TYPE (call);
+      call = save_expr(call);
+      note = build3 (COMPONENT_REF, int_type_node,
+		     build1 (INDIRECT_REF, TREE_TYPE (type), call),
+		     lookup_field (&TREE_TYPE (type),
+				   get_identifier ("length")),
+		     NULL_TREE);
+      note = build2 (MODIFY_EXPR, int_type_node, note, length);
+      TREE_SIDE_EFFECTS (note) = 1;
+      call = build2 (COMPOUND_EXPR, TREE_TYPE (call), note, call);
+      TREE_SIDE_EFFECTS (call) = 1;
+    }
+  return call;
+}
+
+
 /* newarray triggers a call to _Jv_NewPrimArray. This function should be 
    called with an integer code (the type of array to create), and the length
    of the array to create.  */
@@ -1033,7 +1058,7 @@
 tree
 build_newarray (int atype_value, tree length)
 {
-  tree type_arg;
+  tree type_arg, call;
 
   tree prim_type = decode_newarray_type (atype_value);
   tree type
@@ -1045,9 +1070,10 @@
      some work.  */
   type_arg = build_class_ref (prim_type);
 
-  return build_call_nary (promote_type (type),
+  call = build_call_nary (promote_type (type),
 			  build_address_of (soft_newarray_node),
 			  2, type_arg, length);
+  return build_array_length_annotation (call, length);
 }
 
 /* Generates anewarray from a given CLASS_TYPE. Gets from the stack the size
@@ -1061,12 +1087,14 @@
 			     tree_fits_shwi_p (length)
 			     ? tree_to_shwi (length) : -1);
 
-  return build_call_nary (promote_type (type),
-			  build_address_of (soft_anewarray_node),
-			  3,
-			  length,
-			  build_class_ref (class_type),
-			  null_pointer_node);
+  tree call = build_call_nary (promote_type (type),
+			       build_address_of (soft_anewarray_node),
+			       3,
+			       length,
+			       build_class_ref (class_type),
+			       null_pointer_node);
+
+  return build_array_length_annotation (call, length);
 }
 
 /* Return a node the evaluates 'new TYPE[LENGTH]'. */

[-- Attachment #3: Type: text/plain, Size: 2 bytes --]




^ permalink raw reply	[flat|nested] 12+ messages in thread

end of thread, other threads:[~2016-07-14 17:43 UTC | newest]

Thread overview: 12+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2016-02-22 23:02 [JAVA PATCH] Enable more array bounds check elimination Roger Sayle
2016-02-23  9:56 ` Andrew Haley
2016-02-23 17:45   ` Andrew Hughes
2016-02-24  1:11     ` Roger Sayle
2016-02-24  3:23       ` Andrew Hughes
2016-02-24  9:53       ` Andrew Haley
  -- strict thread matches above, loose matches on Subject: below --
2016-02-22 18:10 roger
2016-02-22 18:13 ` Andrew Haley
2016-02-22 18:18 ` Jeff Law
2016-07-13 20:07 ` Jeff Law
2016-07-14 17:12   ` Andrew Hughes
2016-07-14 17:43     ` Andrew Haley

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).