DPDK patches and discussions
 help / color / mirror / Atom feed
From: Dmitry Kozlyuk <dmitry.kozliuk@gmail.com>
To: Narcisa Ana Maria Vasile <navasile@linux.microsoft.com>,
	talshn@nvidia.com
Cc: dev@dpdk.org, thomas@monjalon.net, ocardona@microsoft.com,
	pallavi.kadam@intel.com, dmitrym@microsoft.com,
	Narcisa Vasile <navasile@microsoft.com>,
	stable@dpdk.org
Subject: Re: [dpdk-dev] [PATCH v2] eal/windows: ensure all enabled CPUs are counted
Date: Wed, 4 Aug 2021 17:56:59 +0300	[thread overview]
Message-ID: <20210804175659.41bd53d7@sovereign> (raw)
In-Reply-To: <1625018165-27103-1-git-send-email-navasile@linux.microsoft.com>

2021-06-29 18:56 (UTC-0700), Narcisa Ana Maria Vasile:
> From: Narcisa Vasile <navasile@microsoft.com>
> 
> rte_cpuset_t describes a set of CPUs by using an array of masks
> named '_bits'. Each element in the '_bits' array represents
> a bit mask, with each bit corresponding to a CPU.
> The maximum number of CPUs is given by 'CPU_SETSIZE'.
> The number of bit masks is computed using '_NUM_SETS(CPU_SETSIZE)'.
> 
> count_cpu() should count the number of CPUs enabled in the set 's'.
> Currently, it iterates through the number of masks in the
> set 's', instead of iterating through all the bits in all the masks.
> For example, if '_NUM_SETS(CPU_SETSIZE)' returns 2,
> which means there are 2 bit masks: _bits[0] and _bits[1],
> count_cpu() would only check if CPUs '0' and '1' are enabled.
> The correct behavior is to iterate through all the CPUs in the set and
> count the ones that are enabled.
> 
> This patch fixes count_cpu() to ensure all the bits in all the masks
> are checked to compute the correct number of CPUs enabled in 's'.
> 
> Fixes: e8428a9d89f1 ("eal/windows: add some basic functions and macros")
> Cc: pallavi.kadam@intel.com
> Cc: stable@dpdk.org
> 
> Signed-off-by: Narcisa Vasile <navasile@microsoft.com>
> ---
> v2:
>   * Fix commit message.
> 
>  lib/eal/windows/include/sched.h | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/lib/eal/windows/include/sched.h b/lib/eal/windows/include/sched.h
> index ff572b5dcb..bc31cc8465 100644
> --- a/lib/eal/windows/include/sched.h
> +++ b/lib/eal/windows/include/sched.h
> @@ -49,7 +49,7 @@ count_cpu(rte_cpuset_t *s)
>  	unsigned int _i;
>  	int count = 0;
>  
> -	for (_i = 0; _i < _NUM_SETS(CPU_SETSIZE); _i++)
> +	for (_i = 0; _i < CPU_SETSIZE; _i++)
>  		if (CPU_ISSET(_i, s) != 0LL)
>  			count++;
>  	return count;

Sorry, I lost track of this patch.
Great and thorough explanation, but I don't think it needs to be so long.
Suggestion:

	On Windows, -l/--lcores EAL option was unable to process CPU sets
	containing CPUs other than 0 and 1, because CPU_COUNT() macro
	only checked these CPUs in the set. Fix CPU_COUNT() by enumerating
	all possible CPU indices.

Whatever message is taken in the end,
Acked-by: Dmitry Kozlyuk <dmitry.kozliuk@gmail.com>

  reply	other threads:[~2021-08-04 14:57 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-06-25  0:27 [dpdk-dev] [PATCH] eal/windows: ensure all the CPUs in the set are checked Narcisa Ana Maria Vasile
2021-06-25  8:36 ` Dmitry Kozlyuk
2021-06-30  2:02   ` Narcisa Ana Maria Vasile
2021-06-30  1:56 ` [dpdk-dev] [PATCH v2] eal/windows: ensure all enabled CPUs are counted Narcisa Ana Maria Vasile
2021-08-04 14:56   ` Dmitry Kozlyuk [this message]
2021-08-04 20:40   ` Kadam, Pallavi
2021-08-18 13:46   ` [dpdk-dev] [PATCH v3] " Narcisa Ana Maria Vasile
2021-10-08 23:56     ` Narcisa Ana Maria Vasile
2021-10-11 18:48     ` [dpdk-dev] [dpdk-stable] " Thomas Monjalon

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=20210804175659.41bd53d7@sovereign \
    --to=dmitry.kozliuk@gmail.com \
    --cc=dev@dpdk.org \
    --cc=dmitrym@microsoft.com \
    --cc=navasile@linux.microsoft.com \
    --cc=navasile@microsoft.com \
    --cc=ocardona@microsoft.com \
    --cc=pallavi.kadam@intel.com \
    --cc=stable@dpdk.org \
    --cc=talshn@nvidia.com \
    --cc=thomas@monjalon.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).