mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Rasmus Villemoes <linux@rasmusvillemoes.dk>
To: Roland Dreier <roland@kernel.org>
Cc: Sean Hefty <sean.hefty@intel.com>,
	Hal Rosenstock <hal.rosenstock@gmail.com>,
	linux-rdma@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH 1/2] infiniband: Help gcc generate better code for ocrdma_srq_toggle_bit
Date: Sat, 31 Jan 2015 00:00:02 +0100	[thread overview]
Message-ID: <87twz7ho25.fsf@rasmusvillemoes.dk> (raw)
In-Reply-To: <1421419196-4659-1-git-send-email-linux@rasmusvillemoes.dk> (Rasmus Villemoes's message of "Fri, 16 Jan 2015 15:39:55 +0100")

ping

On Fri, Jan 16 2015, Rasmus Villemoes <linux@rasmusvillemoes.dk> wrote:

> gcc emits a surprising amount of code in order to flip a bit. One
> would think that a single instruction is enough.
>
> $ scripts/bloat-o-meter /tmp/ocrdma_verbs.o drivers/infiniband/hw/ocrdma/ocrdma_verbs.o
> add/remove: 0/0 grow/shrink: 0/3 up/down: 0/-142 (-142)
> function                                     old     new   delta
> ocrdma_post_srq_recv                         498     460     -38
> ocrdma_poll_cq                              2010    1962     -48
> ocrdma_discard_cqes                          495     439     -56
>
> All three calls of ocrdma_srq_toggle_bit happen within spinlocks, so
> saving a few useless instructions might be worthwhile.
>
> Signed-off-by: Rasmus Villemoes <linux@rasmusvillemoes.dk>
> ---
>  drivers/infiniband/hw/ocrdma/ocrdma_verbs.c | 5 +----
>  1 file changed, 1 insertion(+), 4 deletions(-)
>
> diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c
> index fb8d8c4dfbb9..eff11e6c6183 100644
> --- a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c
> +++ b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c
> @@ -1484,10 +1484,7 @@ static void ocrdma_srq_toggle_bit(struct ocrdma_srq *srq, int idx)
>  	int i = idx / 32;
>  	unsigned int mask = (1 << (idx % 32));
>  
> -	if (srq->idx_bit_fields[i] & mask)
> -		srq->idx_bit_fields[i] &= ~mask;
> -	else
> -		srq->idx_bit_fields[i] |= mask;
> +	srq->idx_bit_fields[i] ^= mask;
>  }
>  
>  static int ocrdma_hwq_free_cnt(struct ocrdma_qp_hwq_info *q)

  parent reply	other threads:[~2015-01-30 23:00 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-01-16 14:39 Rasmus Villemoes
2015-01-16 14:39 ` [PATCH 2/2] infiniband: Use unsigned for bit index Rasmus Villemoes
2015-02-02 12:03   ` Selvin Xavier
2015-01-30 23:00 ` Rasmus Villemoes [this message]
2015-01-31 11:07   ` [PATCH 1/2] infiniband: Help gcc generate better code for ocrdma_srq_toggle_bit Yann Droneaud
2015-02-02 12:58     ` Selvin Xavier
2015-02-02 12:04 ` Selvin Xavier

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=87twz7ho25.fsf@rasmusvillemoes.dk \
    --to=linux@rasmusvillemoes.dk \
    --cc=hal.rosenstock@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-rdma@vger.kernel.org \
    --cc=roland@kernel.org \
    --cc=sean.hefty@intel.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

all inboxes | Powered by JetHome®