Re: [PATCH] constructor: Elide expand_constructor when can move by pieces is true

["H.J. Lu" <hjl.tools@gmail.com>]

On Fri, May 21, 2021 at 12:30 AM Bernd Edlinger
<bernd.edlinger@hotmail.de> wrote:
>
>
>
> On 5/21/21 8:57 AM, Richard Biener wrote:
> > On Thu, May 20, 2021 at 4:04 PM H.J. Lu <hjl.tools@gmail.com> wrote:
> >>
> >> On Thu, May 20, 2021 at 12:51 AM Richard Biener
> >> <richard.guenther@gmail.com> wrote:
> >>>
> >>> On Wed, May 19, 2021 at 3:22 PM H.J. Lu <hjl.tools@gmail.com> wrote:
> >>>>
> >>>> On Wed, May 19, 2021 at 2:33 AM Richard Biener
> >>>> <richard.guenther@gmail.com> wrote:
> >>>>>
> >>>>> On Tue, May 18, 2021 at 9:16 PM H.J. Lu <hjl.tools@gmail.com> wrote:
> >>>>>>
> >>>>>> When expanding a constant constructor, don't call expand_constructor if
> >>>>>> it is more efficient to load the data from the memory via move by pieces.
> >>>>>>
> >>>>>> gcc/
> >>>>>>
> >>>>>>         PR middle-end/90773
> >>>>>>         * expr.c (expand_expr_real_1): Don't call expand_constructor if
> >>>>>>         it is more efficient to load the data from the memory.
> >>>>>>
> >>>>>> gcc/testsuite/
> >>>>>>
> >>>>>>         PR middle-end/90773
> >>>>>>         * gcc.target/i386/pr90773-24.c: New test.
> >>>>>>         * gcc.target/i386/pr90773-25.c: Likewise.
> >>>>>> ---
> >>>>>>  gcc/expr.c                                 | 10 ++++++++++
> >>>>>>  gcc/testsuite/gcc.target/i386/pr90773-24.c | 22 ++++++++++++++++++++++
> >>>>>>  gcc/testsuite/gcc.target/i386/pr90773-25.c | 20 ++++++++++++++++++++
> >>>>>>  3 files changed, 52 insertions(+)
> >>>>>>  create mode 100644 gcc/testsuite/gcc.target/i386/pr90773-24.c
> >>>>>>  create mode 100644 gcc/testsuite/gcc.target/i386/pr90773-25.c
> >>>>>>
> >>>>>> diff --git a/gcc/expr.c b/gcc/expr.c
> >>>>>> index d09ee42e262..80e01ea1cbe 100644
> >>>>>> --- a/gcc/expr.c
> >>>>>> +++ b/gcc/expr.c
> >>>>>> @@ -10886,6 +10886,16 @@ expand_expr_real_1 (tree exp, rtx target, machine_mode tmode,
> >>>>>>                 unsigned HOST_WIDE_INT ix;
> >>>>>>                 tree field, value;
> >>>>>>
> >>>>>> +               /* Check if it is more efficient to load the data from
> >>>>>> +                  the memory directly.  FIXME: How many stores do we
> >>>>>> +                  need here if not moved by pieces?  */
> >>>>>> +               unsigned HOST_WIDE_INT bytes
> >>>>>> +                 = tree_to_uhwi (TYPE_SIZE_UNIT (type));
> >>>>>
> >>>>> that's prone to fail - it could be a VLA.
> >>>>
> >>>> What do you mean by fail?  Is it ICE or missed optimization?
> >>>> Do you have a testcase?
> >>>>
> >>>>>
> >>>>>> +               if ((bytes / UNITS_PER_WORD) > 2
> >>>>>> +                   && MOVE_MAX_PIECES > UNITS_PER_WORD
> >>>>>> +                   && can_move_by_pieces (bytes, TYPE_ALIGN (type)))
> >>>>>> +                 goto normal_inner_ref;
> >>>>>> +
> >>>>>
> >>>>> It looks like you're concerned about aggregate copies but this also handles
> >>>>> non-aggregates (which on GIMPLE might already be optimized of course).
> >>>>
> >>>> Here I check if we copy more than 2 words and we can move more than
> >>>> a word in a single instruction.
> >>>>
> >>>>> Also you say "if it's cheaper" but I see no cost considerations.  How do
> >>>>> we generally handle immed const vs. load from constant pool costs?
> >>>>
> >>>> This trades 2 (update to 8) stores with one load plus one store.  Is there
> >>>> a way to check which one is faster?
> >>>
> >>> I'm not sure - it depends on whether the target can do stores from immediates
> >>> at all or what restrictions apply, what the immediate value actually is
> >>> (zero or all-ones should be way cheaper than sth arbitrary) and how the
> >>> pressure on the load unit is.  can_move_by_pieces (bytes, TYPE_ALIGN (type))
> >>> also does not guarantee it will actually move pieces larger than UNITS_PER_WORD,
> >>> that might depend on alignment.  There's by_pieces_ninsns that might provide
> >>> some hint here.
> >>>
> >>> I'm sure it works well for x86.
> >>>
> >>> I wonder if the existing code is in the appropriate place and we
> >>> shouldn't instead
> >>> handle this somewhere upthread where we ask to copy 'exp' into some other
> >>> memory location.  For your testcase that's expand_assignment but I can
> >>> imagine passing array[0] by value to a function resulting in similar copying.
> >>> Testing that shows we get
> >>>
> >>>         pushq   array+56(%rip)
> >>>         .cfi_def_cfa_offset 24
> >>>         pushq   array+48(%rip)
> >>>         .cfi_def_cfa_offset 32
> >>>         pushq   array+40(%rip)
> >>>         .cfi_def_cfa_offset 40
> >>>         pushq   array+32(%rip)
> >>>         .cfi_def_cfa_offset 48
> >>>         pushq   array+24(%rip)
> >>>         .cfi_def_cfa_offset 56
> >>>         pushq   array+16(%rip)
> >>>         .cfi_def_cfa_offset 64
> >>>         pushq   array+8(%rip)
> >>>         .cfi_def_cfa_offset 72
> >>>         pushq   array(%rip)
> >>>         .cfi_def_cfa_offset 80
> >>>         call    bar
> >>>
> >>> for that.  We do have the by-pieces infrastructure to generally do this kind of
> >>> copying but in both of these cases we do not seem to use it.  I also wonder
> >>> if the by-pieces infrastructure can pick up constant initializers automagically
> >>> (we could native_encode the initializer part and feed the by-pieces
> >>> infrastructure with an array of bytes).  There for example might be easy to
> >>> immediate-store byte parts and difficult ones where we could decide on a
> >>> case-by-case basis whether to load+store or immediate-store them.
> >>
> >> I opened:
> >>
> >> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100704
> >>
> >>> For example if I change your testcase to have the array[] initializer
> >>> all-zero we currently emit
> >>>
> >>>         pxor    %xmm0, %xmm0
> >>>         movups  %xmm0, (%rdi)
> >>>         movups  %xmm0, 16(%rdi)
> >>>         movups  %xmm0, 32(%rdi)
> >>>         movups  %xmm0, 48(%rdi)
> >>>         ret
> >>>
> >>> will your patch cause us to emit 4 loads?  OTHO if I do
> >>>
> >>> const struct S array[] = {
> >>>   { 0, 0, 0, 7241, 124764, 48, 16, 33, 10, 96, 2, 0, 0, 4 }
> >>> };
> >>>
> >>> we get
> >>>
> >>>         movq    $0, (%rdi)
> >>>         movl    $0, 8(%rdi)
> >>>         movl    $0, 12(%rdi)
> >>>         movl    $7241, 16(%rdi)
> >>> ...
> >>>
> >>> ideally we'd have sth like
> >>>
> >>>     pxor %xmm0, %xmm0
> >>>     movups  %xmm0, (%rdi)
> >>>     movaps array+16(%rip), %xmm0
> >>>     movups %xmm0, 16(%rdi)
> >>> ...
> >>>
> >>> thus have the zeros written as immediates and the remaining pieces
> >>> with load+stores.
> >>>
> >>> The by-pieces infrastructure eventually get's to see
> >>>
> >>> (mem/u/c:BLK (symbol_ref:DI ("array") [flags 0x2] <var_decl
> >>> 0x7ffff7ff5b40 array>) [1 array+0 S64 A256])
> >>>
> >>> where the MEM_EXPR should provide a way to access the constant initializer.
> >>>
> >>> That said I do agree the current code is a bit premature optimization
> >>> - but maybe
> >>> it should be fend off in expand_constructor which has the cheap clear_storage
> >>> first and which already does check can_move_by_pieces with some heuristics,
> >>> but that seems to be guarded by
> >>>
> >>>            || (tree_fits_uhwi_p (TYPE_SIZE_UNIT (type))
> >>>                && (! can_move_by_pieces
> >>>                    (tree_to_uhwi (TYPE_SIZE_UNIT (type)),
> >>>                     TYPE_ALIGN (type)))
> >>>                && ! mostly_zeros_p (exp))))
> >>>
> >>> which is odd (we _can_ move by pieces, but how does this apply to
> >>> TREE_CONSTANT CTORs and avoid_temp_mem?).
> >>>
> >>> That said, I wonder if we want to elide expand_constructor when the
> >>> CTOR is TREE_STATIC && TREE_CONSTANT and !mostly_zeros_p
> >>> and we can_move_by_pieces.
> >>>
> >>> So sth like
> >>>
> >>> diff --git a/gcc/expr.c b/gcc/expr.c
> >>> index 7139545d543..76b3bdf0c01 100644
> >>> --- a/gcc/expr.c
> >>> +++ b/gcc/expr.c
> >>> @@ -8504,6 +8504,12 @@ expand_constructor (tree exp, rtx target, enum
> >>> expand_modifier modifier,
> >>>                && (! can_move_by_pieces
> >>>                    (tree_to_uhwi (TYPE_SIZE_UNIT (type)),
> >>>                     TYPE_ALIGN (type)))
> >>> +              && ! mostly_zeros_p (exp))
> >>> +          || (TREE_CONSTANT (exp)
> >>> +              && tree_fits_uhwi_p (TYPE_SIZE_UNIT (type))
> >>> +              && (can_move_by_pieces
> >>> +                  (tree_to_uhwi (TYPE_SIZE_UNIT (type)),
> >>> +                   TYPE_ALIGN (type)))
> >>>                && ! mostly_zeros_p (exp))))
> >>>        || ((modifier == EXPAND_INITIALIZER || modifier == EXPAND_CONST_ADDRESS)
> >>>           && TREE_CONSTANT (exp)))
> >>>
> >>> which handles your initializer and the all-zero one optimal?
> >>>
> >>
> >> It works.  Here is the updated patch.
> >
> > So just looking at the code again I think we probably want to add
> > && avoid_temp_mem here, at least that's the case we're looking
> > at.  Not sure if we ever arrive with TREE_CONSTANT CTORs
> > and !avoid_temp_mem but if so we'd create a temporary here
> > which of course would be pointless.
> >
> > So maybe it's then clearer to split the condition out as
> >
> > diff --git a/gcc/expr.c b/gcc/expr.c
> > index 7139545d543..ee8f25f9abd 100644
> > --- a/gcc/expr.c
> > +++ b/gcc/expr.c
> > @@ -8523,6 +8523,19 @@ expand_constructor (tree exp, rtx target, enum
> > expand_modifier modifier,
> >        return constructor;
> >      }
> >
> > +  /* If the CTOR is available in static storage and not mostly
> > +     zeros and we can move it by pieces prefer to do so since
> > +     that's usually more efficient than performing a series of
> > +     stores from immediates.  */
> > +  if (avoid_temp_mem
> > +      && TREE_STATIC (exp)
> > +      && TREE_CONSTANT (exp)
> > +      && tree_fits_uhwi_p (TYPE_SIZE_UNIT (type))
> > +      && can_move_by_pieces (tree_to_uhwi (TYPE_SIZE_UNIT (type)),
> > +                            TYPE_ALIGN (type))
> > +      && ! mostly_zeros_p (exp))
> > +    return NULL_RTX;
> > +
> >    /* Handle calls that pass values in multiple non-contiguous
> >       locations.  The Irix 6 ABI has examples of this.  */
> >    if (target == 0 || ! safe_from_p (target, exp, 1)
> >
> >
> > OK with that change.
> >
>
> Note however (I've been playing with the previous version)
> that the test case
>
> FAIL: gcc.target/i386/pr90773-25.c scan-assembler-times vmovdqu[\\\\t ]%ymm[0-9]+, \\\\(%[^,]+\\\\) 1
> FAIL: gcc.target/i386/pr90773-25.c scan-assembler-times vmovdqu[\\\\t ]%ymm[0-9]+, 32\\\\(%[^,]+\\\\) 1
>
> fails for --target_board=unix
>
> $ grep movdqu pr90773-25.s
>         vmovdqu %xmm0, (%rdi)
>         vmovdqu %xmm1, 16(%rdi)
>         vmovdqu %xmm2, 32(%rdi)
>         vmovdqu %xmm3, 48(%rdi)
>
> while the test expects %ymm
> /* { dg-final { scan-assembler-times "vmovdqu\[\\t \]%ymm\[0-9\]+, \\(%\[\^,\]+\\)" 1 } } */
> /* { dg-final { scan-assembler-times "vmovdqu\[\\t \]%ymm\[0-9\]+, 32\\(%\[\^,\]+\\)" 1 } } */
>
> and
>
> FAIL: gcc.target/i386/pr90773-24.c scan-assembler-times movups[\\\\t ]%xmm[0-9]+, \\\\(%[^,]+\\\\) 1
> FAIL: gcc.target/i386/pr90773-24.c scan-assembler-times movups[\\\\t ]%xmm[0-9]+, 16\\\\(%[^,]+\\\\) 1
> FAIL: gcc.target/i386/pr90773-24.c scan-assembler-times movups[\\\\t ]%xmm[0-9]+, 32\\\\(%[^,]+\\\\) 1
> FAIL: gcc.target/i386/pr90773-24.c scan-assembler-times movups[\\\\t ]%xmm[0-9]+, 48\\\\(%[^,]+\\\\) 1
> FAIL: gcc.target/i386/pr90773-25.c scan-assembler-times vmovdqu[\\\\t ]%ymm[0-9]+, \\\\(%[^,]+\\\\) 1
> FAIL: gcc.target/i386/pr90773-25.c scan-assembler-times vmovdqu[\\\\t ]%ymm[0-9]+, 32\\\\(%[^,]+\\\\) 1
> FAIL: gcc.target/i386/pr90773-26.c scan-assembler-times pxor[\\\\t ]%xmm[0-9]+, %xmm[0-9]+ 1
> FAIL: gcc.target/i386/pr90773-26.c scan-assembler-times movups[\\\\t ]%xmm[0-9]+, \\\\(%[^,]+\\\\) 1
> FAIL: gcc.target/i386/pr90773-26.c scan-assembler-times movups[\\\\t ]%xmm[0-9]+, 16\\\\(%[^,]+\\\\) 1
> FAIL: gcc.target/i386/pr90773-26.c scan-assembler-times movups[\\\\t ]%xmm[0-9]+, 32\\\\(%[^,]+\\\\) 1
> FAIL: gcc.target/i386/pr90773-26.c scan-assembler-times movups[\\\\t ]%xmm[0-9]+, 48\\\\(%[^,]+\\\\) 1
>
> fails for --target_board=unix/-m32
>

The whole patch set is needed.   My users/hjl/pieces/hook branch is at

https://gitlab.com/x86-gcc/gcc/-/tree/users/hjl/pieces/hook

I got

[hjl@gnu-cfl-2 testsuite]$
/export/build/gnu/tools-build/gcc-gitlab-debug/build-x86_64-linux/gcc/xgcc
-B/export/build/gnu/tools-build/gcc-gitlab-debug/build-x86_64-linux/gcc/
/export/gnu/import/git/gitlab/x86-gcc/gcc/testsuite/gcc.target/i386/pr90773-25.c
-fdiagnostics-plain-output -O2 -march=skylake -ffat-lto-objects
-fno-ident -S -o pr90773-25.s
[hjl@gnu-cfl-2 testsuite]$ cat pr90773-25.s
.file "pr90773-25.c"
.text
.p2align 4
.globl foo
.type foo, @function
foo:
.LFB0:
.cfi_startproc
vpxor %xmm0, %xmm0, %xmm0
vmovdqu %xmm0, (%rdi)
vmovdqu %xmm0, 16(%rdi)
vmovdqu %xmm0, 32(%rdi)
vmovdqu %xmm0, 48(%rdi)
ret
.cfi_endproc
.LFE0:
.size foo, .-foo
.globl array
.section .rodata
.align 32
.type array, @object
.size array, 64
array:
.zero 64
.section .note.GNU-stack,"",@progbits
[hjl@gnu-cfl-2 testsuite]$

-- 
H.J.