[Bug target/58622] New: With -fomit-frame-pointer, A64 does not generate post-decrement stores

[Bug target/58622] New: With -fomit-frame-pointer, A64 does not generate post-decrement stores

[b.grayson at samsung dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58622

            Bug ID: 58622
           Summary: With -fomit-frame-pointer, A64 does not generate
                    post-decrement stores
           Product: gcc
           Version: 4.9.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: target
          Assignee: unassigned at gcc dot gnu.org
          Reporter: b.grayson at samsung dot com
            Target: AArch64
             Build: 4.9.0 20130602

In A64, if one compiles a simple program under -O3, one gets code like this:

int bar(int i);
int foo() { return bar(5)+4; }

A64 -O3 assembly:

foo:
        stp     x29, x30, [sp, -16]!
        add     x29, sp, 0
        mov     w0, 5
        bl      bar
        add     w0, w0, 4
        ldp     x29, x30, [sp], 16
        ret

Note the use of update-form loads and stores for the SP.

But if one uses -O3 -fomit-frame-pointer, the following is obtained:

foo:
        sub     sp, sp, #16
        mov     w0, 5
        str     x30, [sp]
        bl      bar
        add     w0, w0, 4
        ldr     x30, [sp]
        add     sp, sp, 16
        ret

The sub and str could be merged into str x30, [sp, #-16]!, and the ldr/add
could be merged into ldr x30, [sp], #16 (if I have my assembly correct), as
they were in the with-frame-pointer case.  On some ARM implementations, the
updates are "for free", so one would get better performance with the merged
load/store instructions, not to mention better instruction-cache density.

Note that under A32, identical code (using update/post-decrement stores) is
generated regardless of omit-frame-pointer settings.

[Bug target/58622] With -fomit-frame-pointer, A64 does not generate post-decrement stores

[rearnsha at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58622

Richard Earnshaw <rearnsha at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2014-01-28
     Ever confirmed|0                           |1

--- Comment #1 from Richard Earnshaw <rearnsha at gcc dot gnu.org> ---
Confirmed.  Prologue and epilogue code is constructed in the backend late on. 
Any optimizations have to be done during the code generation.  The particular
path with -fomit-frame-pointer is not on by default and clearly hasn't been
tuned yet.

[Bug target/58622] With -fomit-frame-pointer, A64 does not generate post-decrement stores

[kuganv at linaro dot org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58622

Kugan <kuganv at linaro dot org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |kuganv at linaro dot org

--- Comment #2 from Kugan <kuganv at linaro dot org> ---
Created attachment 32342
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=32342&action=edit
Proposed RFC patch

Proposed RFC patch - In the process rgression testing.

[Bug target/58622] With -fomit-frame-pointer, A64 does not generate post-decrement stores

[ramana at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=58622

Ramana Radhakrishnan <ramana at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |ASSIGNED
                 CC|                            |ramana at gcc dot gnu.org
           Assignee|unassigned at gcc dot gnu.org      |mshawcroft at gcc dot gnu.org

--- Comment #3 from Ramana Radhakrishnan <ramana at gcc dot gnu.org> ---
(In reply to Kugan from comment #2)
> Created attachment 32342 [details]
> Proposed RFC patch
> 
> Proposed RFC patch - In the process rgression testing.

Kugan,

This also needs to handle ILP32 and probably needs a bit more TLC than just
this. We have something ongoing that can help address this, so assigning to
Marcus as he's been looking into some of this.

However this is stage1 material and cannot make it in for 4.9.

Ramana


(In reply to Kugan from comment #2)
> Created attachment 32342 [details]
> Proposed RFC patch
> 
> Proposed RFC patch - In the process rgression testing.

[Bug target/58622] With -fomit-frame-pointer, A64 does not generate post-decrement stores

[mshawcroft at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=58622

mshawcroft at gcc dot gnu.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |4.10.0

[Bug target/58622] With -fomit-frame-pointer, A64 does not generate post-decrement stores

[mshawcroft at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=58622

mshawcroft at gcc dot gnu.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|---                         |FIXED

--- Comment #4 from mshawcroft at gcc dot gnu.org ---
Issue resolved with Jiong's prologue / epilogue patch series committed to
trunk:

The relevant optimization commits are r212976, r212999, r213000.