qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Paolo Bonzini <pbonzini@redhat.com>
To: Aurelien Jarno <aurelien@aurel32.net>, qemu-devel@nongnu.org
Cc: Richard Henderson <rth@twiddle.net>
Subject: Re: [Qemu-devel] [PATCH for-2.5 01/10] tcg/optimize: optimize temps tracking
Date: Mon, 27 Jul 2015 10:21:40 +0200	[thread overview]
Message-ID: <55B5EA14.2050209@redhat.com> (raw)
In-Reply-To: <1437755447-10537-2-git-send-email-aurelien@aurel32.net>



On 24/07/2015 18:30, Aurelien Jarno wrote:
> The tcg_temp_info structure uses 24 bytes per temp. Now that we emulate
> vector registers on most guests, it's not uncommon to have more than 100
> used temps. This means we have initialize more than 2kB at least twice
> per TB, often more when there is a few goto_tb.
> 
> Instead used a TCGTempSet bit array to track which temps are in used in
> the current basic block. This means there are only around 16 bytes to
> initialize.
> 
> This improves the boot time of a MIPS guest on an x86-64 host by around
> 7% and moves out tcg_optimize from the the top of the profiler list.
> 
> Cc: Richard Henderson <rth@twiddle.net>
> Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
> ---
>  tcg/optimize.c | 32 ++++++++++++++++++++++----------
>  1 file changed, 22 insertions(+), 10 deletions(-)
> 
> diff --git a/tcg/optimize.c b/tcg/optimize.c
> index cd0e793..20e24b3 100644
> --- a/tcg/optimize.c
> +++ b/tcg/optimize.c
> @@ -50,6 +50,7 @@ struct tcg_temp_info {
>  };
>  
>  static struct tcg_temp_info temps[TCG_MAX_TEMPS];
> +static TCGTempSet temps_used;
>  
>  /* Reset TEMP's state to TCG_TEMP_UNDEF.  If TEMP only had one copy, remove
>     the copy flag from the left temp.  */
> @@ -67,6 +68,22 @@ static void reset_temp(TCGArg temp)
>      temps[temp].mask = -1;
>  }
>  
> +/* Reset all temporaries, given that there are NB_TEMPS of them.  */
> +static void reset_all_temps(int nb_temps)
> +{
> +    memset(&temps_used.l, 0, sizeof(long) * BITS_TO_LONGS(nb_temps));

You can use bitmap_zero here.

Paolo

> +}
> +
> +/* Initialize and activate a temporary.  */
> +static void init_temp_info(TCGArg temp)
> +{
> +    if (!test_bit(temp, temps_used.l)) {
> +        temps[temp].state = TCG_TEMP_UNDEF;
> +        temps[temp].mask = -1;
> +        set_bit(temp, temps_used.l);
> +    }
> +}
> +
>  static TCGOp *insert_op_before(TCGContext *s, TCGOp *old_op,
>                                  TCGOpcode opc, int nargs)
>  {
> @@ -98,16 +115,6 @@ static TCGOp *insert_op_before(TCGContext *s, TCGOp *old_op,
>      return new_op;
>  }
>  
> -/* Reset all temporaries, given that there are NB_TEMPS of them.  */
> -static void reset_all_temps(int nb_temps)
> -{
> -    int i;
> -    for (i = 0; i < nb_temps; i++) {
> -        temps[i].state = TCG_TEMP_UNDEF;
> -        temps[i].mask = -1;
> -    }
> -}
> -
>  static int op_bits(TCGOpcode op)
>  {
>      const TCGOpDef *def = &tcg_op_defs[op];
> @@ -606,6 +613,11 @@ void tcg_optimize(TCGContext *s)
>              nb_iargs = def->nb_iargs;
>          }
>  
> +        /* Initialize the temps that are going to be used */
> +        for (i = 0; i < nb_oargs + nb_iargs; i++) {
> +            init_temp_info(args[i]);
> +        }
> +
>          /* Do copy propagation */
>          for (i = nb_oargs; i < nb_oargs + nb_iargs; i++) {
>              if (temps[args[i]].state == TCG_TEMP_COPY) {
> 

  reply	other threads:[~2015-07-27  8:21 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-07-24 16:30 [Qemu-devel] [PATCH for-2.5 00/10] tcg: improve optimizer Aurelien Jarno
2015-07-24 16:30 ` [Qemu-devel] [PATCH for-2.5 01/10] tcg/optimize: optimize temps tracking Aurelien Jarno
2015-07-27  8:21   ` Paolo Bonzini [this message]
2015-07-27  9:09     ` Aurelien Jarno
2015-07-24 16:30 ` [Qemu-devel] [PATCH for-2.5 02/10] tcg/optimize: add temp_is_const and temp_is_copy functions Aurelien Jarno
2015-07-29 16:01   ` Alex Bennée
2015-07-29 16:25     ` Aurelien Jarno
2015-07-24 16:30 ` [Qemu-devel] [PATCH for-2.5 03/10] tcg/optimize: track const/copy status separately Aurelien Jarno
2015-07-27  8:25   ` Paolo Bonzini
2015-07-27  9:10     ` Aurelien Jarno
2015-07-29 16:10   ` Alex Bennée
2015-07-29 16:25     ` Aurelien Jarno
2015-07-24 16:30 ` [Qemu-devel] [PATCH for-2.5 04/10] tcg/optimize: allow constant to have copies Aurelien Jarno
2015-07-24 20:15   ` Richard Henderson
2015-07-24 22:56     ` Aurelien Jarno
2015-07-29 16:12   ` Alex Bennée
2015-07-29 16:27     ` Aurelien Jarno
2015-07-29 20:42       ` Alex Bennée
2015-07-30  7:46         ` Aurelien Jarno
2015-07-24 16:30 ` [Qemu-devel] [PATCH for-2.5 05/10] tcg: rename trunc_shr_i32 into trunc_shr_i64_i32 Aurelien Jarno
2015-07-31  6:31   ` Alex Bennée
2015-07-24 16:30 ` [Qemu-devel] [PATCH for-2.5 06/10] tcg: don't abuse TCG type in tcg_gen_trunc_shr_i64_i32 Aurelien Jarno
2015-07-31  7:32   ` Alex Bennée
2015-07-24 16:30 ` [Qemu-devel] [PATCH for-2.5 07/10] tcg: implement real ext_i32_i64 and extu_i32_i64 ops Aurelien Jarno
2015-07-31 16:01   ` Alex Bennée
2015-07-31 16:11     ` Richard Henderson
2015-07-24 16:30 ` [Qemu-devel] [PATCH for-2.5 08/10] tcg/optimize: add optimizations for " Aurelien Jarno
2015-07-24 16:30 ` [Qemu-devel] [PATCH for-2.5 09/10] tcg/optimize: do not remember garbage high bits for 32-bit ops Aurelien Jarno
2015-07-24 16:30 ` [Qemu-devel] [PATCH for-2.5 10/10] tcg: update README about size changing ops Aurelien Jarno
2015-07-31 16:02   ` Alex Bennée

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=55B5EA14.2050209@redhat.com \
    --to=pbonzini@redhat.com \
    --cc=aurelien@aurel32.net \
    --cc=qemu-devel@nongnu.org \
    --cc=rth@twiddle.net \
    /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 NNTP newsgroup(s).