linux-f2fs-devel.lists.sourceforge.net archive mirror
 help / color / mirror / Atom feed
From: Dan Carpenter <dan.carpenter@oracle.com>
To: daehojeong@google.com
Cc: linux-f2fs-devel@lists.sourceforge.net
Subject: [f2fs-dev] [bug report] f2fs: introduce periodic iostat io latency traces
Date: Mon, 16 Aug 2021 10:35:59 +0300	[thread overview]
Message-ID: <20210816073559.GA12039@kili> (raw)

Hello Daeho Jeong,

The patch f4b05791dda9: "f2fs: introduce periodic iostat io latency
traces" from Aug 3, 2021, leads to the following
Smatch static checker warnings:

fs/f2fs/f2fs.h:3365 __update_iostat_latency() error: buffer overflow 'sbi->rd_sum_lat' 3 <= 3
fs/f2fs/f2fs.h:3366 __update_iostat_latency() error: buffer overflow 'sbi->rd_bio_cnt' 3 <= 3
fs/f2fs/f2fs.h:3367 __update_iostat_latency() error: buffer overflow 'sbi->rd_peak_lat' 3 <= 3
fs/f2fs/f2fs.h:3368 __update_iostat_latency() error: buffer overflow 'sbi->rd_peak_lat' 3 <= 3
fs/f2fs/f2fs.h:3370 __update_iostat_latency() error: buffer overflow 'sbi->wr_sum_lat[sync]' 3 <= 3
fs/f2fs/f2fs.h:3371 __update_iostat_latency() error: buffer overflow 'sbi->wr_bio_cnt[sync]' 3 <= 3
fs/f2fs/f2fs.h:3372 __update_iostat_latency() error: buffer overflow 'sbi->wr_peak_lat[sync]' 3 <= 3
fs/f2fs/f2fs.h:3373 __update_iostat_latency() error: buffer overflow 'sbi->wr_peak_lat[sync]' 3 <= 3

fs/f2fs/f2fs.h
    3348 static inline void __update_iostat_latency(struct bio_iostat_ctx *iostat_ctx,
    3349 				int rw, int sync)
    3350 {
    3351 	unsigned long ts_diff;
    3352 	unsigned int iotype = iostat_ctx->type;
    3353 	unsigned long flags;
    3354 	struct f2fs_sb_info *sbi = iostat_ctx->sbi;
    3355 
    3356 	if (!sbi->iostat_enable)
    3357 		return;
    3358 
    3359 	ts_diff = jiffies - iostat_ctx->submit_ts;
    3360 	if (iotype >= META_FLUSH)
                    ^^^^^^^^^^^^^^^^^^^^
This means the highest value of "iotype" is NR_PAGE_TYPE.

    3361 		iotype = META;
    3362 
    3363 	spin_lock_irqsave(&sbi->iostat_lat_lock, flags);
    3364 	if (rw == 0) {
--> 3365 		sbi->rd_sum_lat[iotype] += ts_diff;
                        ^^^^^^^^^^^^^^^^^^^^^^^
    3366 		sbi->rd_bio_cnt[iotype]++;
                        ^^^^^^^^^^^^^^^^^^^^^^
These arrays have NR_PAGE_TYPE elements so it's off by one.

    3367 		if (ts_diff > sbi->rd_peak_lat[iotype])
    3368 			sbi->rd_peak_lat[iotype] = ts_diff;
    3369 	} else {
    3370 		sbi->wr_sum_lat[sync][iotype] += ts_diff;
    3371 		sbi->wr_bio_cnt[sync][iotype]++;
    3372 		if (ts_diff > sbi->wr_peak_lat[sync][iotype])
    3373 			sbi->wr_peak_lat[sync][iotype] = ts_diff;
    3374 	}
    3375 	spin_unlock_irqrestore(&sbi->iostat_lat_lock, flags);
    3376 }

regards,
dan carpenter


_______________________________________________
Linux-f2fs-devel mailing list
Linux-f2fs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel

             reply	other threads:[~2021-08-16  7:36 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-08-16  7:35 Dan Carpenter [this message]
2021-08-17  4:09 ` [f2fs-dev] [bug report] f2fs: introduce periodic iostat io latency traces Daeho Jeong

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=20210816073559.GA12039@kili \
    --to=dan.carpenter@oracle.com \
    --cc=daehojeong@google.com \
    --cc=linux-f2fs-devel@lists.sourceforge.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).