From: Dimitar Dimitrov <dimitar@dinux.eu>
To: Lehua Ding <lehua.ding@rivai.ai>
Cc: gcc-patches@gcc.gnu.org, juzhe.zhong@rivai.ai,
shuo.chen@rivai.ai, jin.xia@rivai.ai, vmakarov@redhat.com,
richard.sandiford@arm.com
Subject: Re: [PATCH 2/4] df: Add DF_LIVE_SUBREG problem
Date: Thu, 25 Apr 2024 23:56:16 +0300 [thread overview]
Message-ID: <ZirDcLRsCF06K6TY@kendros> (raw)
In-Reply-To: <20240424100505.1301-3-lehua.ding@rivai.ai>
On Wed, Apr 24, 2024 at 06:05:03PM +0800, Lehua Ding wrote:
> This patch add a new DF problem, named DF_LIVE_SUBREG. This problem
> is extended from the DF_LR problem and support track the subreg liveness
> of multireg pseudo if these pseudo satisfy the following conditions:
>
> 1. the mode size greater than it's REGMODE_NATURAL_SIZE.
> 2. the reg is used in insns via subreg pattern.
>
> The main methods are as follows:
>
> 1. split bitmap in/out/def/use fileds to full_in/out/def/use and
> partial_in/out/def/use. If a pseudo need to be tracked it's subreg
> liveness, then it is recorded in partial_in/out/def/use fileds.
> Meantimes, there are range_in/out/def/use fileds which records the live
> range of the tracked pseudo.
> 2. in the df_live_subreg_finalize function, we move the tracked pseudo from
> the partial_in/out/def/use to full_in/out/def/use if the pseudo's live
> range is full.
Hi Lehua,
I'm not familiar with LRA, so my comments bellow could be totally off
point. Please treat them as mild suggestions.
>
> gcc/ChangeLog:
>
> * Makefile.in: Add subreg-live-range object file.
> * df-problems.cc (struct df_live_subreg_problem_data): Private struct
> for DF_LIVE_SUBREG problem.
> (df_live_subreg_get_bb_info): getting bb regs in/out data.
> (get_live_subreg_local_bb_info): getting bb regs use/def data.
> (multireg_p): checking is the regno a pseudo multireg.
> (need_track_subreg_p): checking is the regno need to be tracked.
> (init_range): getting the range of subreg rtx.
> (remove_subreg_range): removing use data for the reg/subreg rtx.
> (add_subreg_range): adding def/use data for the reg/subreg rtx.
> (df_live_subreg_free_bb_info): Free basic block df data.
> (df_live_subreg_alloc): Allocate and init df data.
> (df_live_subreg_reset): Reset the live in/out df data.
> (df_live_subreg_bb_local_compute): Compute basic block df data.
> (df_live_subreg_local_compute): Compute all basic blocks df data.
> (df_live_subreg_init): Init the in/out df data.
> (df_live_subreg_check_result): Assert the full and partial df data.
> (df_live_subreg_confluence_0): Confluence function for infinite loops.
> (df_live_subreg_confluence_n): Confluence function for normal edge.
> (df_live_subreg_transfer_function): Transfer function.
> (df_live_subreg_finalize): Finalize the all_in/all_out df data.
> (df_live_subreg_free): Free the df data.
> (df_live_subreg_top_dump): Dump top df data.
> (df_live_subreg_bottom_dump): Dump bottom df data.
> (df_live_subreg_add_problem): Add the DF_LIVE_SUBREG problem.
> * df.h (enum df_problem_id): Add DF_LIVE_SUBREG.
> (class subregs_live): Simple decalare.
> (class df_live_subreg_local_bb_info): New class for full/partial def/use
> df data.
> (class df_live_subreg_bb_info): New class for full/partial in/out
> df data.
> (df_live_subreg): getting the df_live_subreg data.
> (df_live_subreg_add_problem): Exported.
> (df_live_subreg_finalize): Ditto.
> (df_live_subreg_check_result): Ditto.
> (multireg_p): Ditto.
> (init_range): Ditto.
> (add_subreg_range): Ditto.
> (remove_subreg_range): Ditto.
> (df_get_subreg_live_in): Accessor the all_in df data.
> (df_get_subreg_live_out): Accessor the all_out df data.
> (df_get_subreg_live_full_in): Accessor the full_in df data.
> (df_get_subreg_live_full_out): Accessor the full_out df data.
> (df_get_subreg_live_partial_in): Accessor the partial_in df data.
> (df_get_subreg_live_partial_out): Accessor the partial_out df data.
> (df_get_subreg_live_range_in): Accessor the range_in df data.
> (df_get_subreg_live_range_out): Accessor the range_out df data.
> * regs.h (get_nblocks): Get the blocks of mode.
> * sbitmap.cc (bitmap_full_p): sbitmap predicator.
> (bitmap_same_p): sbitmap predicator.
> (test_full): test bitmap_full_p.
> (test_same): test bitmap_same_p.
> (sbitmap_cc_tests): Add test_full and test_same.
> * sbitmap.h (bitmap_full_p): Exported.
> (bitmap_same_p): Ditto.
> * timevar.def (TV_DF_LIVE_SUBREG): add DF_LIVE_SUBREG timevar.
> * subreg-live-range.cc: New file.
> * subreg-live-range.h: New file.
> ---
> gcc/Makefile.in | 1 +
> gcc/df-problems.cc | 855 ++++++++++++++++++++++++++++++++++++++-
> gcc/df.h | 155 +++++++
> gcc/regs.h | 5 +
> gcc/sbitmap.cc | 98 +++++
> gcc/sbitmap.h | 2 +
> gcc/subreg-live-range.cc | 53 +++
> gcc/subreg-live-range.h | 206 ++++++++++
> gcc/timevar.def | 1 +
> 9 files changed, 1375 insertions(+), 1 deletion(-)
> create mode 100644 gcc/subreg-live-range.cc
> create mode 100644 gcc/subreg-live-range.h
>
> diff --git a/gcc/Makefile.in b/gcc/Makefile.in
> index a74761b7ab3..e195238f6ab 100644
> --- a/gcc/Makefile.in
> +++ b/gcc/Makefile.in
> @@ -1684,6 +1684,7 @@ OBJS = \
> store-motion.o \
> streamer-hooks.o \
> stringpool.o \
> + subreg-live-range.o \
> substring-locations.o \
> target-globals.o \
> targhooks.o \
> diff --git a/gcc/df-problems.cc b/gcc/df-problems.cc
> index 88ee0dd67fc..93c5b52146c 100644
> --- a/gcc/df-problems.cc
> +++ b/gcc/df-problems.cc
> @@ -28,6 +28,7 @@ along with GCC; see the file COPYING3. If not see
> #include "target.h"
> #include "rtl.h"
> #include "df.h"
> +#include "subreg-live-range.h"
> #include "memmodel.h"
> #include "tm_p.h"
> #include "insn-config.h"
> @@ -1344,8 +1345,860 @@ df_lr_verify_transfer_functions (void)
> bitmap_clear (&all_blocks);
> }
>
> +/*----------------------------------------------------------------------------
> + REGISTER AND SUBREGS LIVES
> + Like DF_LR, but include tracking subreg liveness. Currently used to provide
Nit: Should have two spaces after the dot, per GNU coding style. I'd suggest
to run the contrib/check_GNU_style.py script on your patches.
> + subreg liveness related information to the register allocator. The subreg
> + information is currently tracked for registers that satisfy the following
> + conditions:
> + 1. REG is a pseudo register
> + 2. MODE_SIZE > UNIT_SIZE
> + 3. MODE_SIZE is a multiple of UNIT_SIZE
> + 4. REG is used via subreg pattern
> + Assuming: MODE = the machine mode of the REG
> + MODE_SIZE = GET_MODE_SIZE (MODE)
> + UNIT_SIZE = REGMODE_NATURAL_SIZE (MODE)
> + Condition 3 is currently strict, maybe it can be removed in the future, but
> + for now it is sufficient.
> + ----------------------------------------------------------------------------*/
> +
> +/* These two empty data are used as default data in case the user does not turn
> + * on the track-subreg-liveness feature. */
> +bitmap_head empty_bitmap;
> +subregs_live empty_live;
These names seem a bit too short for global variables. Perhaps tuck
them in a namespace?
Also, since these must remain empty, shouldn't they be declared as const?
namespace df {
const bitmap_head empty_bitmap;
const subregs_live empty_live;
}
> +
> +/* Private data for live_subreg problem. */
> +struct df_live_subreg_problem_data
> +{
> + /* Record registers that need to track subreg liveness. */
> + bitmap_head tracked_regs;
> + /* An obstack for the bitmaps we need for this problem. */
> + bitmap_obstack live_subreg_bitmaps;
> +};
> +
> +/* Helper functions */
> +
> +static df_live_subreg_bb_info *
> +df_live_subreg_get_bb_info (unsigned int index)
> +{
> + if (index < df_live_subreg->block_info_size)
> + return &(
> + (class df_live_subreg_bb_info *) df_live_subreg->block_info)[index];
Please consider using C++ instead:
return &static_cast<df_live_subreg_bb_info *>(df_live_subreg->block_info)[index];
> + else
> + return NULL;
Use nullptr instead?
> +}
> +
> +static df_live_subreg_local_bb_info *
> +get_live_subreg_local_bb_info (unsigned int bb_index)
> +{
> + return df_live_subreg_get_bb_info (bb_index);
> +}
> +
> +/* Return true if regno is a multireg. */
> +bool
> +multireg_p (int regno)
> +{
> + if (regno < FIRST_PSEUDO_REGISTER)
> + return false;
> + rtx regno_rtx = regno_reg_rtx[regno];
> + machine_mode reg_mode = GET_MODE (regno_rtx);
> + poly_int64 total_size = GET_MODE_SIZE (reg_mode);
> + poly_int64 natural_size = REGMODE_NATURAL_SIZE (reg_mode);
> + return maybe_gt (total_size, natural_size)
> + && multiple_p (total_size, natural_size);
> +}
> +
> +/* Return true if the REGNO need be track with subreg liveness. */
> +
> +static bool
> +need_track_subreg_p (unsigned regno)
> +{
> + struct df_live_subreg_problem_data *problem_data
> + = (struct df_live_subreg_problem_data *) df_live_subreg->problem_data;
Here auto could be used:
auto problem_data = (struct df_live_subreg_problem_data *) df_live_subreg->problem_data;
> + return bitmap_bit_p (&problem_data->tracked_regs, regno);
> +}
> +
> +/* Return the subreg_range of REF. */
Is this comment valid? I see neither a REF parameter, nor a returned
value. Did you perhaps mean:
/* Fill RANGE with the subreg range for OP in REGMODE_NATURAL_SIZE
granularity. */
> +void
> +init_range (rtx op, sbitmap range)
> +{
> + rtx reg = SUBREG_P (op) ? SUBREG_REG (op) : op;
> + machine_mode reg_mode = GET_MODE (reg);
> +
> + if (!read_modify_subreg_p (op))
> + {
> + bitmap_set_range (range, 0, get_nblocks (reg_mode));
> + return;
> + }
> +
> + rtx subreg = op;
> + machine_mode subreg_mode = GET_MODE (subreg);
> + poly_int64 offset = SUBREG_BYTE (subreg);
> + int nblocks = get_nblocks (reg_mode);
> + poly_int64 unit_size = REGMODE_NATURAL_SIZE (reg_mode);
> + poly_int64 subreg_size = GET_MODE_SIZE (subreg_mode);
> + poly_int64 left = offset + subreg_size;
> +
> + int subreg_start = -1;
> + int subreg_nblocks = -1;
> + for (int i = 0; i < nblocks; i += 1)
> + {
> + poly_int64 right = unit_size * (i + 1);
> + if (subreg_start < 0 && maybe_lt (offset, right))
> + subreg_start = i;
> + if (subreg_nblocks < 0 && maybe_le (left, right))
> + {
> + subreg_nblocks = i + 1 - subreg_start;
> + break;
> + }
> + }
> + gcc_assert (subreg_start >= 0 && subreg_nblocks > 0);
> +
> + bitmap_set_range (range, subreg_start, subreg_nblocks);
> +}
> +
> +/* Remove RANGE from BB_INFO's use data. */
> +void
> +remove_subreg_range (df_live_subreg_local_bb_info *bb_info, unsigned int regno,
> + const_sbitmap range)
> +{
> + bitmap partial = &bb_info->partial_use;
> +
> + bb_info->range_use->remove_range (regno, range);
> + if (bb_info->range_use->empty_p (regno))
> + bitmap_clear_bit (partial, regno);
> +}
> +
> +/* Remove RANGE of REF from BB_INFO's use data. */
> +static void
> +remove_subreg_range (df_live_subreg_local_bb_info *bb_info, df_ref ref)
> +{
> + unsigned int regno = DF_REF_REGNO (ref);
> + machine_mode reg_mode = GET_MODE (DF_REF_REAL_REG (ref));
> +
> + if (need_track_subreg_p (regno))
> + {
> + auto_sbitmap range (get_nblocks (reg_mode));
> + init_range (DF_REF_REG (ref), range);
> + remove_subreg_range (bb_info, regno, range);
> + }
> + else
> + {
> + bitmap_clear_bit (&bb_info->full_use, regno);
> + gcc_assert (!bitmap_bit_p (&bb_info->partial_use, regno));
> + gcc_assert (!bitmap_bit_p (&bb_info->partial_def, regno));
> + }
> +}
> +
> +/* add RANGE to BB_INFO's def/use. If is_def is true, means for BB_INFO's def,
> + otherwise for BB_INFO's use. */
> +void
> +add_subreg_range (df_live_subreg_local_bb_info *bb_info, unsigned int regno,
> + const_sbitmap range, bool is_def)
Would it be more clear to split into two separate functions
instead of using a boolean flag? E.g.:
add_subreg_range_to_def (df_live_subreg_local_bb_info *bb_info, unsigned int regno,
const_sbitmap range)
add_subreg_range_to_use (df_live_subreg_local_bb_info *bb_info, unsigned int regno,
const_sbitmap range)
> +{
> + bitmap partial = is_def ? &bb_info->partial_def : &bb_info->partial_use;
> + subregs_live *range_live = is_def ? bb_info->range_def : bb_info->range_use;
> +
> + bitmap_set_bit (partial, regno);
> + range_live->add_range (regno, range);
> +}
> +
> +/* add RANGE of REF to BB_INFO def/use. If is_def is true, means for BB_INFO's
> + def, otherwise for BB_INFO's use. */
> +static void
> +add_subreg_range (df_live_subreg_local_bb_info *bb_info, df_ref ref,
> + bool is_def)
> +{
> + unsigned int regno = DF_REF_REGNO (ref);
> + machine_mode reg_mode = GET_MODE (DF_REF_REAL_REG (ref));
> +
> + if (need_track_subreg_p (regno))
> + {
> + auto_sbitmap range (get_nblocks (reg_mode));
> + init_range (DF_REF_REG (ref), range);
> + add_subreg_range (bb_info, regno, range, is_def);
> + }
> + else
> + {
> + bitmap full = is_def ? &bb_info->full_def : &bb_info->full_use;
> + bitmap partial = is_def ? &bb_info->partial_def : &bb_info->partial_use;
> + bitmap_set_bit (full, regno);
> + gcc_assert (!bitmap_bit_p (partial, regno));
> + }
> +}
> +
> +/* Free basic block info. */
> +
> +static void
> +df_live_subreg_free_bb_info (basic_block bb ATTRIBUTE_UNUSED, void *vbb_info)
> +{
> + df_live_subreg_bb_info *bb_info = (df_live_subreg_bb_info *) vbb_info;
Perhaps:
auto bb_info = static_cast<df_live_subreg_bb_info *> vbb_info;
> + if (bb_info)
> + {
> + delete bb_info->range_in;
> + bb_info->range_in = NULL;
> + delete bb_info->range_out;
> + bb_info->range_out = NULL;
> +
> + bitmap_clear (&bb_info->all_in);
> + bitmap_clear (&bb_info->full_in);
> + bitmap_clear (&bb_info->partial_in);
> + bitmap_clear (&bb_info->all_out);
> + bitmap_clear (&bb_info->full_out);
> + bitmap_clear (&bb_info->partial_out);
> + }
> +}
> +
> +/* Allocate or reset bitmaps for DF_LIVE_SUBREG blocks. The solution bits are
> + not touched unless the block is new. */
> +
> +static void
> +df_live_subreg_alloc (bitmap all_blocks ATTRIBUTE_UNUSED)
> +{
> + struct df_live_subreg_problem_data *problem_data;
> + df_grow_bb_info (df_live_subreg);
> + if (df_live_subreg->problem_data)
> + problem_data
> + = (struct df_live_subreg_problem_data *) df_live_subreg->problem_data;
> + else
> + {
> + problem_data = XNEW (struct df_live_subreg_problem_data);
> + df_live_subreg->problem_data = problem_data;
> +
> + bitmap_obstack_initialize (&problem_data->live_subreg_bitmaps);
> + bitmap_initialize (&problem_data->tracked_regs,
> + &problem_data->live_subreg_bitmaps);
> + }
> +
> + bitmap_clear (&problem_data->tracked_regs);
> +
> + basic_block bb;
> + FOR_EACH_BB_FN (bb, cfun)
> + bitmap_set_bit (df_live_subreg->out_of_date_transfer_functions, bb->index);
> +
> + bitmap_set_bit (df_live_subreg->out_of_date_transfer_functions, ENTRY_BLOCK);
> + bitmap_set_bit (df_live_subreg->out_of_date_transfer_functions, EXIT_BLOCK);
> +
> + unsigned int bb_index;
> + bitmap_iterator bi;
> + EXECUTE_IF_SET_IN_BITMAP (df_live_subreg->out_of_date_transfer_functions, 0,
> + bb_index, bi)
> + {
> + /* Find the regs which we need to track it's subreg liveness. */
> + rtx_insn *insn;
> + df_ref use;
> + FOR_BB_INSNS (bb, insn)
> + {
> + if (!NONDEBUG_INSN_P (insn))
> + continue;
> +
> + df_insn_info *insn_info = DF_INSN_INFO_GET (insn);
> +
> + FOR_EACH_INSN_INFO_USE (use, insn_info)
> + {
> + unsigned int regno = DF_REF_REGNO (use);
> + /* A multireg which is used via subreg pattern. */
> + if (multireg_p (regno)
> + && DF_REF_FLAGS (use) & (DF_REF_SUBREG))
> + bitmap_set_bit (&problem_data->tracked_regs, regno);
> + }
> + }
> + }
> +
> + if (dump_file)
> + {
> + fprintf (dump_file, ";; regs need to be tracked subreg liveness: ");
> + df_print_regset (dump_file, &problem_data->tracked_regs);
> + }
> +
> + size_t n = bitmap_count_bits (&problem_data->tracked_regs);
> +
> + EXECUTE_IF_SET_IN_BITMAP (df_live_subreg->out_of_date_transfer_functions, 0,
> + bb_index, bi)
> + {
> + /* Clean global infos. */
> + df_live_subreg_bb_info *bb_info = df_live_subreg_get_bb_info (bb_index);
Here and for the rest of the call sites, I think auto is allowed to be used:
auto bb_info = df_live_subreg_get_bb_info (bb_index);
Regards,
Dimitar
next prev parent reply other threads:[~2024-04-25 20:56 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-04-24 10:05 [PATCH V2 0/4] Add DF_LIVE_SUBREG data and apply to IRA and LRA Lehua Ding
2024-04-24 10:05 ` [PATCH 1/4] df: Add -ftrack-subreg-liveness option Lehua Ding
2024-04-24 10:05 ` [PATCH 2/4] df: Add DF_LIVE_SUBREG problem Lehua Ding
2024-04-25 20:56 ` Dimitar Dimitrov [this message]
2024-05-08 2:46 ` Lehua Ding
2024-05-08 3:34 ` 陈硕
2024-05-08 15:57 ` Dimitar Dimitrov
2024-05-08 22:54 ` 钟居哲
2024-04-24 10:05 ` [PATCH 3/4] ira: Apply DF_LIVE_SUBREG data Lehua Ding
2024-04-24 10:05 ` [PATCH 4/4] lra: " Lehua Ding
-- strict thread matches above, loose matches on Subject: below --
2024-02-03 10:50 [PATCH 0/4] Add DF_LIVE_SUBREG data and apply to IRA and LRA Lehua Ding
2024-02-03 10:50 ` [PATCH 2/4] df: Add DF_LIVE_SUBREG problem Lehua Ding
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=ZirDcLRsCF06K6TY@kendros \
--to=dimitar@dinux.eu \
--cc=gcc-patches@gcc.gnu.org \
--cc=jin.xia@rivai.ai \
--cc=juzhe.zhong@rivai.ai \
--cc=lehua.ding@rivai.ai \
--cc=richard.sandiford@arm.com \
--cc=shuo.chen@rivai.ai \
--cc=vmakarov@redhat.com \
/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).