public inbox for gcc-patches@gcc.gnu.org
 help / color / mirror / Atom feed
From: Richard Sandiford <richard.sandiford@arm.com>
To: HAO CHEN GUI <guihaoc@linux.ibm.com>
Cc: gcc-patches <gcc-patches@gcc.gnu.org>,
	 Segher Boessenkool <segher@kernel.crashing.org>,
	 David <dje.gcc@gmail.com>,
	 Peter Bergner <bergner@linux.ibm.com>,
	 "Kewen.Lin" <linkw@linux.ibm.com>
Subject: Re: [PATCH-1v4, expand] Enable vector mode for compare_by_pieces [PR111449]
Date: Fri, 20 Oct 2023 09:49:14 +0100	[thread overview]
Message-ID: <mpt8r7xn0fp.fsf@arm.com> (raw)
In-Reply-To: <992fc509-51d6-3dbd-ee9d-393378227fc5@linux.ibm.com> (HAO CHEN GUI's message of "Fri, 20 Oct 2023 15:22:52 +0800")

HAO CHEN GUI <guihaoc@linux.ibm.com> writes:
> Hi,
>   Vector mode instructions are efficient for compare on some targets.
> This patch enables vector mode for compare_by_pieces. Two help
> functions are added to check if vector mode is available for certain
> by pieces operations and if if optabs exists for the mode and certain
> by pieces operations. One member is added in class op_by_pieces_d to
> record the type of operations.
>
>   The test case is in the second patch which is rs6000 specific.
>
>   Compared to last version, the main change is to add a target hook
> check - scalar_mode_supported_p when retrieving the available scalar
> modes. The mode which is not supported for a target should be skipped.
> (e.g. TImode on ppc). Also some function names and comments are refined
> according to reviewer's advice.
>
>   Bootstrapped and tested on x86 and powerpc64-linux BE and LE with no
> regressions.
>
> Thanks
> Gui Haochen
>
> ChangeLog
> Expand: Enable vector mode for by pieces compares
>
> Vector mode compare instructions are efficient for equality compare on
> rs6000. This patch refactors the codes of by pieces operation to enable
> vector mode for compare.
>
> gcc/
> 	PR target/111449
> 	* expr.cc (can_use_qi_vectors): New function to return true if
> 	we know how to implement OP using vectors of bytes.
> 	(qi_vector_mode_supported_p): New function to check if optabs
> 	exists for the mode and certain by pieces operations.
> 	(widest_fixed_size_mode_for_size): Replace the second argument
> 	with the type of by pieces operations.  Call can_use_qi_vectors
> 	and qi_vector_mode_supported_p to do the check.  Call
> 	scalar_mode_supported_p to check if the scalar mode is supported.
> 	(by_pieces_ninsns): Pass the type of by pieces operation to
> 	widest_fixed_size_mode_for_size.
> 	(class op_by_pieces_d): Remove m_qi_vector_mode.  Add m_op to
> 	record the type of by pieces operations.
> 	(op_by_pieces_d::op_by_pieces_d): Change last argument to the
> 	type of by pieces operations, initialize m_op with it.  Pass
> 	m_op to function widest_fixed_size_mode_for_size.
> 	(op_by_pieces_d::get_usable_mode): Pass m_op to function
> 	widest_fixed_size_mode_for_size.
> 	(op_by_pieces_d::smallest_fixed_size_mode_for_size): Call
> 	can_use_qi_vectors and qi_vector_mode_supported_p to do the
> 	check.
> 	(op_by_pieces_d::run): Pass m_op to function
> 	widest_fixed_size_mode_for_size.
> 	(move_by_pieces_d::move_by_pieces_d): Set m_op to MOVE_BY_PIECES.
> 	(store_by_pieces_d::store_by_pieces_d): Set m_op with the op.
> 	(can_store_by_pieces): Pass the type of by pieces operations to
> 	widest_fixed_size_mode_for_size.
> 	(clear_by_pieces): Initialize class store_by_pieces_d with
> 	CLEAR_BY_PIECES.
> 	(compare_by_pieces_d::compare_by_pieces_d): Set m_op to
> 	COMPARE_BY_PIECES.

OK, thanks.  And thanks for your patience.

Richard

> patch.diff
> diff --git a/gcc/expr.cc b/gcc/expr.cc
> index 2c9930ec674..ad5f9dd8ec2 100644
> --- a/gcc/expr.cc
> +++ b/gcc/expr.cc
> @@ -988,18 +988,44 @@ alignment_for_piecewise_move (unsigned int max_pieces, unsigned int align)
>    return align;
>  }
>
> -/* Return the widest QI vector, if QI_MODE is true, or integer mode
> -   that is narrower than SIZE bytes.  */
> +/* Return true if we know how to implement OP using vectors of bytes.  */
> +static bool
> +can_use_qi_vectors (by_pieces_operation op)
> +{
> +  return (op == COMPARE_BY_PIECES
> +	  || op == SET_BY_PIECES
> +	  || op == CLEAR_BY_PIECES);
> +}
> +
> +/* Return true if optabs exists for the mode and certain by pieces
> +   operations.  */
> +static bool
> +qi_vector_mode_supported_p (fixed_size_mode mode, by_pieces_operation op)
> +{
> +  if ((op == SET_BY_PIECES || op == CLEAR_BY_PIECES)
> +      && optab_handler (vec_duplicate_optab, mode) != CODE_FOR_nothing)
> +    return true;
> +
> +  if (op == COMPARE_BY_PIECES
> +      && optab_handler (mov_optab, mode) != CODE_FOR_nothing
> +      && can_compare_p (EQ, mode, ccp_jump))
> +    return true;
>
> +  return false;
> +}
> +
> +/* Return the widest mode that can be used to perform part of an
> +   operation OP on SIZE bytes.  Try to use QI vector modes where
> +   possible.  */
>  static fixed_size_mode
> -widest_fixed_size_mode_for_size (unsigned int size, bool qi_vector)
> +widest_fixed_size_mode_for_size (unsigned int size, by_pieces_operation op)
>  {
>    fixed_size_mode result = NARROWEST_INT_MODE;
>
>    gcc_checking_assert (size > 1);
>
>    /* Use QI vector only if size is wider than a WORD.  */
> -  if (qi_vector && size > UNITS_PER_WORD)
> +  if (can_use_qi_vectors (op) && size > UNITS_PER_WORD)
>      {
>        machine_mode mode;
>        fixed_size_mode candidate;
> @@ -1009,8 +1035,7 @@ widest_fixed_size_mode_for_size (unsigned int size, bool qi_vector)
>  	  {
>  	    if (GET_MODE_SIZE (candidate) >= size)
>  	      break;
> -	    if (optab_handler (vec_duplicate_optab, candidate)
> -		!= CODE_FOR_nothing)
> +	    if (qi_vector_mode_supported_p (candidate, op))
>  	      result = candidate;
>  	  }
>
> @@ -1019,9 +1044,14 @@ widest_fixed_size_mode_for_size (unsigned int size, bool qi_vector)
>      }
>
>    opt_scalar_int_mode tmode;
> +  scalar_int_mode mode;
>    FOR_EACH_MODE_IN_CLASS (tmode, MODE_INT)
> -    if (GET_MODE_SIZE (tmode.require ()) < size)
> -      result = tmode.require ();
> +    {
> +      mode = tmode.require ();
> +      if (GET_MODE_SIZE (mode) < size
> +	  && targetm.scalar_mode_supported_p (mode))
> +      result = mode;
> +    }
>
>    return result;
>  }
> @@ -1061,8 +1091,7 @@ by_pieces_ninsns (unsigned HOST_WIDE_INT l, unsigned int align,
>      {
>        /* NB: Round up L and ALIGN to the widest integer mode for
>  	 MAX_SIZE.  */
> -      mode = widest_fixed_size_mode_for_size (max_size,
> -					      op == SET_BY_PIECES);
> +      mode = widest_fixed_size_mode_for_size (max_size, op);
>        if (optab_handler (mov_optab, mode) != CODE_FOR_nothing)
>  	{
>  	  unsigned HOST_WIDE_INT up = ROUND_UP (l, GET_MODE_SIZE (mode));
> @@ -1076,8 +1105,7 @@ by_pieces_ninsns (unsigned HOST_WIDE_INT l, unsigned int align,
>
>    while (max_size > 1 && l > 0)
>      {
> -      mode = widest_fixed_size_mode_for_size (max_size,
> -					      op == SET_BY_PIECES);
> +      mode = widest_fixed_size_mode_for_size (max_size, op);
>        enum insn_code icode;
>
>        unsigned int modesize = GET_MODE_SIZE (mode);
> @@ -1317,8 +1345,8 @@ class op_by_pieces_d
>    bool m_push;
>    /* True if targetm.overlap_op_by_pieces_p () returns true.  */
>    bool m_overlap_op_by_pieces;
> -  /* True if QI vector mode can be used.  */
> -  bool m_qi_vector_mode;
> +  /* The type of operation that we're performing.  */
> +  by_pieces_operation m_op;
>
>    /* Virtual functions, overriden by derived classes for the specific
>       operation.  */
> @@ -1331,7 +1359,7 @@ class op_by_pieces_d
>   public:
>    op_by_pieces_d (unsigned int, rtx, bool, rtx, bool, by_pieces_constfn,
>  		  void *, unsigned HOST_WIDE_INT, unsigned int, bool,
> -		  bool = false);
> +		  by_pieces_operation);
>    void run ();
>  };
>
> @@ -1349,11 +1377,11 @@ op_by_pieces_d::op_by_pieces_d (unsigned int max_pieces, rtx to,
>  				void *from_cfn_data,
>  				unsigned HOST_WIDE_INT len,
>  				unsigned int align, bool push,
> -				bool qi_vector_mode)
> +				by_pieces_operation op)
>    : m_to (to, to_load, NULL, NULL),
>      m_from (from, from_load, from_cfn, from_cfn_data),
>      m_len (len), m_max_size (max_pieces + 1),
> -    m_push (push), m_qi_vector_mode (qi_vector_mode)
> +    m_push (push), m_op (op)
>  {
>    int toi = m_to.get_addr_inc ();
>    int fromi = m_from.get_addr_inc ();
> @@ -1375,8 +1403,7 @@ op_by_pieces_d::op_by_pieces_d (unsigned int max_pieces, rtx to,
>      {
>        /* Find the mode of the largest comparison.  */
>        fixed_size_mode mode
> -	= widest_fixed_size_mode_for_size (m_max_size,
> -					   m_qi_vector_mode);
> +	= widest_fixed_size_mode_for_size (m_max_size, m_op);
>
>        m_from.decide_autoinc (mode, m_reverse, len);
>        m_to.decide_autoinc (mode, m_reverse, len);
> @@ -1401,7 +1428,7 @@ op_by_pieces_d::get_usable_mode (fixed_size_mode mode, unsigned int len)
>        if (len >= size && prepare_mode (mode, m_align))
>  	break;
>        /* widest_fixed_size_mode_for_size checks SIZE > 1.  */
> -      mode = widest_fixed_size_mode_for_size (size, m_qi_vector_mode);
> +      mode = widest_fixed_size_mode_for_size (size, m_op);
>      }
>    while (1);
>    return mode;
> @@ -1414,7 +1441,7 @@ fixed_size_mode
>  op_by_pieces_d::smallest_fixed_size_mode_for_size (unsigned int size)
>  {
>    /* Use QI vector only for > size of WORD.  */
> -  if (m_qi_vector_mode && size > UNITS_PER_WORD)
> +  if (can_use_qi_vectors (m_op) && size > UNITS_PER_WORD)
>      {
>        machine_mode mode;
>        fixed_size_mode candidate;
> @@ -1427,8 +1454,7 @@ op_by_pieces_d::smallest_fixed_size_mode_for_size (unsigned int size)
>  	      break;
>
>  	    if (GET_MODE_SIZE (candidate) >= size
> -		&& (optab_handler (vec_duplicate_optab, candidate)
> -		    != CODE_FOR_nothing))
> +		&& qi_vector_mode_supported_p (candidate, m_op))
>  	      return candidate;
>  	  }
>      }
> @@ -1451,7 +1477,7 @@ op_by_pieces_d::run ()
>
>    /* widest_fixed_size_mode_for_size checks M_MAX_SIZE > 1.  */
>    fixed_size_mode mode
> -    = widest_fixed_size_mode_for_size (m_max_size, m_qi_vector_mode);
> +    = widest_fixed_size_mode_for_size (m_max_size, m_op);
>    mode = get_usable_mode (mode, length);
>
>    by_pieces_prev to_prev = { nullptr, mode };
> @@ -1516,8 +1542,7 @@ op_by_pieces_d::run ()
>        else
>  	{
>  	  /* widest_fixed_size_mode_for_size checks SIZE > 1.  */
> -	  mode = widest_fixed_size_mode_for_size (size,
> -						  m_qi_vector_mode);
> +	  mode = widest_fixed_size_mode_for_size (size, m_op);
>  	  mode = get_usable_mode (mode, length);
>  	}
>      }
> @@ -1543,7 +1568,7 @@ class move_by_pieces_d : public op_by_pieces_d
>    move_by_pieces_d (rtx to, rtx from, unsigned HOST_WIDE_INT len,
>  		    unsigned int align)
>      : op_by_pieces_d (MOVE_MAX_PIECES, to, false, from, true, NULL,
> -		      NULL, len, align, PUSHG_P (to))
> +		      NULL, len, align, PUSHG_P (to), MOVE_BY_PIECES)
>    {
>    }
>    rtx finish_retmode (memop_ret);
> @@ -1632,15 +1657,16 @@ move_by_pieces (rtx to, rtx from, unsigned HOST_WIDE_INT len,
>  class store_by_pieces_d : public op_by_pieces_d
>  {
>    insn_gen_fn m_gen_fun;
> +
>    void generate (rtx, rtx, machine_mode) final override;
>    bool prepare_mode (machine_mode, unsigned int) final override;
>
>   public:
>    store_by_pieces_d (rtx to, by_pieces_constfn cfn, void *cfn_data,
>  		     unsigned HOST_WIDE_INT len, unsigned int align,
> -		     bool qi_vector_mode)
> +		     by_pieces_operation op)
>      : op_by_pieces_d (STORE_MAX_PIECES, to, false, NULL_RTX, true, cfn,
> -		      cfn_data, len, align, false, qi_vector_mode)
> +		      cfn_data, len, align, false, op)
>    {
>    }
>    rtx finish_retmode (memop_ret);
> @@ -1729,8 +1755,8 @@ can_store_by_pieces (unsigned HOST_WIDE_INT len,
>        max_size = STORE_MAX_PIECES + 1;
>        while (max_size > 1 && l > 0)
>  	{
> -	  fixed_size_mode mode
> -	    = widest_fixed_size_mode_for_size (max_size, memsetp);
> +	  auto op = memsetp ? SET_BY_PIECES : STORE_BY_PIECES;
> +	  auto mode = widest_fixed_size_mode_for_size (max_size, op);
>
>  	  icode = optab_handler (mov_optab, mode);
>  	  if (icode != CODE_FOR_nothing
> @@ -1793,7 +1819,7 @@ store_by_pieces (rtx to, unsigned HOST_WIDE_INT len,
>  		 optimize_insn_for_speed_p ()));
>
>    store_by_pieces_d data (to, constfun, constfundata, len, align,
> -			  memsetp);
> +			  memsetp ? SET_BY_PIECES : STORE_BY_PIECES);
>    data.run ();
>
>    if (retmode != RETURN_BEGIN)
> @@ -1814,7 +1840,7 @@ clear_by_pieces (rtx to, unsigned HOST_WIDE_INT len, unsigned int align)
>    /* Use builtin_memset_read_str to support vector mode broadcast.  */
>    char c = 0;
>    store_by_pieces_d data (to, builtin_memset_read_str, &c, len, align,
> -			  true);
> +			  CLEAR_BY_PIECES);
>    data.run ();
>  }
>
> @@ -1832,12 +1858,13 @@ class compare_by_pieces_d : public op_by_pieces_d
>    void generate (rtx, rtx, machine_mode) final override;
>    bool prepare_mode (machine_mode, unsigned int) final override;
>    void finish_mode (machine_mode) final override;
> +
>   public:
>    compare_by_pieces_d (rtx op0, rtx op1, by_pieces_constfn op1_cfn,
>  		       void *op1_cfn_data, HOST_WIDE_INT len, int align,
>  		       rtx_code_label *fail_label)
>      : op_by_pieces_d (COMPARE_MAX_PIECES, op0, true, op1, true, op1_cfn,
> -		      op1_cfn_data, len, align, false)
> +		      op1_cfn_data, len, align, false, COMPARE_BY_PIECES)
>    {
>      m_fail_label = fail_label;
>    }

  reply	other threads:[~2023-10-20  8:49 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-10-20  7:22 HAO CHEN GUI
2023-10-20  8:49 ` Richard Sandiford [this message]
2023-10-23  1:29   ` HAO CHEN GUI
2023-10-24  8:42     ` Jiang, Haochen
2023-10-24  8:49       ` Jiang, Haochen
2023-10-24  9:28         ` HAO CHEN GUI
2023-10-25  8:32         ` HAO CHEN GUI
2023-10-25  8:40           ` Richard Sandiford
2023-10-25  8:47             ` Jiang, Haochen
2023-10-26  8:57               ` Jiang, Haochen

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=mpt8r7xn0fp.fsf@arm.com \
    --to=richard.sandiford@arm.com \
    --cc=bergner@linux.ibm.com \
    --cc=dje.gcc@gmail.com \
    --cc=gcc-patches@gcc.gnu.org \
    --cc=guihaoc@linux.ibm.com \
    --cc=linkw@linux.ibm.com \
    --cc=segher@kernel.crashing.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
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).