All of lore.kernel.org
 help / color / mirror / Atom feed
From: Niklas Cassel <Niklas.Cassel@wdc.com>
To: Ammar Faizi <ammarfaizi2@gnuweeb.org>
Cc: Jens Axboe <axboe@kernel.dk>,
	fio Mailing List <fio@vger.kernel.org>,
	GNU/Weeb Mailing List <gwml@vger.gnuweeb.org>
Subject: Re: [PATCH v2 3/6] stat: Handle `ENOMEM` case in `__show_run_stats()`
Date: Wed, 27 Apr 2022 09:35:01 +0000	[thread overview]
Message-ID: <YmkORZJIxvBOxHJN@x1-carbon> (raw)
In-Reply-To: <20220427091125.114146-4-ammarfaizi2@gnuweeb.org>

On Wed, Apr 27, 2022 at 04:11:22PM +0700, Ammar Faizi wrote:
> This handles memory allocation failure and several improvements.
> 
> 1) Change `malloc(n * size)` to `calloc(n, size)`. This is to avoid
>    multiplication on `malloc()` because it doesn't do overflow check.
>    Also, `calloc()` zeroes the allocated memory, so we can omit the
>    zeroing elements of the array inside the loop.
> 
> 2) Make sure we are calling `free()` properly if we fail. Use goto to
>    do this.
> 
> Signed-off-by: Ammar Faizi <ammarfaizi2@gnuweeb.org>
> ---
>  stat.c | 30 ++++++++++++++++++++++--------
>  1 file changed, 22 insertions(+), 8 deletions(-)
> 
> diff --git a/stat.c b/stat.c
> index 949af5ed..cc9be02e 100644
> --- a/stat.c
> +++ b/stat.c
> @@ -2429,7 +2429,11 @@ void __show_run_stats(void)
>  	struct buf_output output[FIO_OUTPUT_NR];
>  	struct flist_head **opt_lists;
>  
> -	runstats = malloc(sizeof(struct group_run_stats) * (groupid + 1));
> +	runstats = calloc(groupid + 1, sizeof(*runstats));

Hello Ammar,

here you allocate runstats with calloc.

> +	if (!runstats) {
> +		log_err("fio: failed to allocate runstats\n");
> +		return;
> +	}
>  
>  	for (i = 0; i < groupid + 1; i++)
>  		init_group_run_stat(&runstats[i]);

Here you call init_group_run_stat() on each runstats,
which calls memset(). Seems a bit excessive to clear
the memory to zero twice.

If you intend to modify init_group_run_stat(), be careful,
as it is also called by client.c


Kind regards,
Niklas


> @@ -2454,14 +2458,21 @@ void __show_run_stats(void)
>  		nr_ts++;
>  	}
>  
> -	threadstats = malloc(nr_ts * sizeof(struct thread_stat));
> -	opt_lists = malloc(nr_ts * sizeof(struct flist_head *));
> +	threadstats = calloc(nr_ts, sizeof(*threadstats));
> +	if (!threadstats) {
> +		log_err("fio: failed to allocate threadstats\n");
> +		goto out_free_runstats;
> +	}
>  
> -	for (i = 0; i < nr_ts; i++) {
> -		init_thread_stat(&threadstats[i]);
> -		opt_lists[i] = NULL;
> +	opt_lists = calloc(nr_ts, sizeof(*opt_lists));
> +	if (!opt_lists) {
> +		log_err("fio: failed to allocate opt_lists\n");
> +		goto out_free_threadstats;
>  	}
>  
> +	for (i = 0; i < nr_ts; i++)
> +		init_thread_stat(&threadstats[i]);
> +
>  	init_per_prio_stats(threadstats, nr_ts);
>  
>  	j = 0;
> @@ -2709,15 +2720,18 @@ void __show_run_stats(void)
>  	fio_idle_prof_cleanup();
>  
>  	log_info_flush();
> -	free(runstats);
>  
>  	/* free arrays allocated by sum_thread_stats(), if any */
>  	for (i = 0; i < nr_ts; i++) {
>  		ts = &threadstats[i];
>  		free_clat_prio_stats(ts);
>  	}
> -	free(threadstats);
> +
>  	free(opt_lists);
> +out_free_threadstats:
> +	free(threadstats);
> +out_free_runstats:
> +	free(runstats);
>  }
>  
>  int __show_running_run_stats(void)
> -- 
> Ammar Faizi
> 

  reply	other threads:[~2022-04-27 10:53 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-04-27  9:11 [PATCH v2 0/6] Small fio cleanups and fixes Ammar Faizi
2022-04-27  9:11 ` [PATCH v2 1/6] backend: Fix indentation Ammar Faizi
2022-04-27  9:11 ` [PATCH v2 2/6] cgroup: Handle `ENOMEM` case on `malloc()` call Ammar Faizi
2022-04-27  9:11 ` [PATCH v2 3/6] stat: Handle `ENOMEM` case in `__show_run_stats()` Ammar Faizi
2022-04-27  9:35   ` Niklas Cassel [this message]
2022-04-27  9:52     ` Ammar Faizi
2022-04-27  9:11 ` [PATCH v2 4/6] engines/net: Replace `malloc()` + `memset()` with `calloc()` Ammar Faizi
2022-04-27  9:11 ` [PATCH v2 5/6] json: Change `if (!strlen(str))` to `if (!str[0])` Ammar Faizi
2022-04-27  9:11 ` [PATCH v2 6/6] Makefile: Suppress `-Wimplicit-fallthrough` when compiling `lex.yy` Ammar Faizi

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=YmkORZJIxvBOxHJN@x1-carbon \
    --to=niklas.cassel@wdc.com \
    --cc=ammarfaizi2@gnuweeb.org \
    --cc=axboe@kernel.dk \
    --cc=fio@vger.kernel.org \
    --cc=gwml@vger.gnuweeb.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.