From: "H.J. Lu" <hjl.tools@gmail.com>
To: Uros Bizjak <ubizjak@gmail.com>
Cc: Ilya Enkovich <enkovich.gnu@gmail.com>, Jeff Law <law@redhat.com>,
"gcc-patches@gcc.gnu.org" <gcc-patches@gcc.gnu.org>
Subject: Re: [RFC, PR target/65105] Use vector instructions for scalar 64bit computations on 32bit target
Date: Tue, 29 Sep 2015 12:51:00 -0000 [thread overview]
Message-ID: <CAMe9rOozQ7zenjpmzbx5X4G=sN_X8bV2KW84wtZ_Q9+=qtO_Yg@mail.gmail.com> (raw)
In-Reply-To: <CAFULd4aBjzOq_TmbWgHP12+voh5SsHmure2LewkNo+LGLc-=-w@mail.gmail.com>
On Wed, Sep 23, 2015 at 3:29 AM, Uros Bizjak <ubizjak@gmail.com> wrote:
> On Wed, Sep 23, 2015 at 12:19 PM, Ilya Enkovich <enkovich.gnu@gmail.com> wrote:
>> On 14 Sep 17:50, Uros Bizjak wrote:
>>>
>>> +(define_insn_and_split "*zext<mode>_doubleword"
>>> + [(set (match_operand:DI 0 "register_operand" "=r")
>>> + (zero_extend:DI (match_operand:SWI24 1 "nonimmediate_operand" "rm")))]
>>> + "!TARGET_64BIT && TARGET_STV && TARGET_SSE2"
>>> + "#"
>>> + "&& reload_completed && GENERAL_REG_P (operands[0])"
>>> + [(set (match_dup 0) (zero_extend:SI (match_dup 1)))
>>> + (set (match_dup 2) (const_int 0))]
>>> + "split_double_mode (DImode, &operands[0], 1, &operands[0], &operands[2]);")
>>> +
>>> +(define_insn_and_split "*zextqi_doubleword"
>>> + [(set (match_operand:DI 0 "register_operand" "=r")
>>> + (zero_extend:DI (match_operand:QI 1 "nonimmediate_operand" "qm")))]
>>> + "!TARGET_64BIT && TARGET_STV && TARGET_SSE2"
>>> + "#"
>>> + "&& reload_completed && GENERAL_REG_P (operands[0])"
>>> + [(set (match_dup 0) (zero_extend:SI (match_dup 1)))
>>> + (set (match_dup 2) (const_int 0))]
>>> + "split_double_mode (DImode, &operands[0], 1, &operands[0], &operands[2]);")
>>> +
>>>
>>> Please put the above patterns together with other zero_extend
>>> patterns. You can also merge these two patterns using SWI124 mode
>>> iterator with <r> mode attribute as a register constraint. Also, no
>>> need to check for GENERAL_REG_P after reload, when "r" constraint is
>>> in effect:
>>>
>>> (define_insn_and_split "*zext<mode>_doubleword"
>>> [(set (match_operand:DI 0 "register_operand" "=r")
>>> (zero_extend:DI (match_operand:SWI124 1 "nonimmediate_operand" "<r>m")))]
>>> "!TARGET_64BIT && TARGET_STV && TARGET_SSE2"
>>> "#"
>>> "&& reload_completed"
>>> [(set (match_dup 0) (zero_extend:SI (match_dup 1)))
>>> (set (match_dup 2) (const_int 0))]
>>> "split_double_mode (DImode, &operands[0], 1, &operands[0], &operands[2]);")
>>
>> Register constraint doesn't affect split and I need GENERAL_REG_P to filter other registers case.
>
> OK.
>
>> I merged QI and HI cases of zext but made a separate pattern for SI case because it doesn't need zero_extend in resulting code. Bootstrapped and regtested for x86_64-unknown-linux-gnu.
>
> This change is OK.
>
> The patch LGTM, but please wait a couple of days if Jeff has some
> comment on algorithmic aspect of the patch.
>
> Thanks,
> Uros.
>
>>
>> Thanks,
>> Ilya
>> --
>> gcc/
>>
>> 2015-09-23 Ilya Enkovich <enkovich.gnu@gmail.com>
>>
>> * config/i386/i386.c: Include dbgcnt.h.
>> (has_non_address_hard_reg): New.
>> (convertible_comparison_p): New.
>> (scalar_to_vector_candidate_p): New.
>> (remove_non_convertible_regs): New.
>> (scalar_chain): New.
>> (scalar_chain::scalar_chain): New.
>> (scalar_chain::~scalar_chain): New.
>> (scalar_chain::add_to_queue): New.
>> (scalar_chain::mark_dual_mode_def): New.
>> (scalar_chain::analyze_register_chain): New.
>> (scalar_chain::add_insn): New.
>> (scalar_chain::build): New.
>> (scalar_chain::compute_convert_gain): New.
>> (scalar_chain::replace_with_subreg): New.
>> (scalar_chain::replace_with_subreg_in_insn): New.
>> (scalar_chain::emit_conversion_insns): New.
>> (scalar_chain::make_vector_copies): New.
>> (scalar_chain::convert_reg): New.
>> (scalar_chain::convert_op): New.
>> (scalar_chain::convert_insn): New.
>> (scalar_chain::convert): New.
>> (convert_scalars_to_vector): New.
>> (pass_data_stv): New.
>> (pass_stv): New.
>> (make_pass_stv): New.
>> (ix86_option_override): Created and register stv pass.
>> (flag_opts): Add -mstv.
>> (ix86_option_override_internal): Likewise.
>> * config/i386/i386.md (SWIM1248x): New.
>> (*movdi_internal): Add xmm to mem alternative for TARGET_STV.
>> (and<mode>3): Use SWIM1248x iterator instead of SWIM.
>> (*anddi3_doubleword): New.
>> (*zext<mode>_doubleword): New.
>> (*zextsi_doubleword): New.
>> (<code><mode>3): Use SWIM1248x iterator instead of SWIM.
>> (*<code>di3_doubleword): New.
>> * config/i386/i386.opt (mstv): New.
>> * dbgcnt.def (stv_conversion): New.
>>
This caused:
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=67761
--
H.J.
prev parent reply other threads:[~2015-09-29 12:11 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-06-19 13:45 Ilya Enkovich
2015-07-14 14:38 ` Ilya Enkovich
2015-08-03 20:52 ` Jeff Law
2015-08-21 13:55 ` Ilya Enkovich
2015-08-21 16:40 ` Jeff Law
2015-09-08 15:53 ` Ilya Enkovich
2015-09-09 8:20 ` Uros Bizjak
2015-09-09 8:31 ` Uros Bizjak
2015-09-14 12:45 ` Ilya Enkovich
2015-09-14 16:50 ` Uros Bizjak
2015-09-23 10:37 ` Ilya Enkovich
2015-09-23 10:46 ` Uros Bizjak
2015-09-29 12:51 ` H.J. Lu [this message]
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='CAMe9rOozQ7zenjpmzbx5X4G=sN_X8bV2KW84wtZ_Q9+=qtO_Yg@mail.gmail.com' \
--to=hjl.tools@gmail.com \
--cc=enkovich.gnu@gmail.com \
--cc=gcc-patches@gcc.gnu.org \
--cc=law@redhat.com \
--cc=ubizjak@gmail.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).