From: Denis Efremov <efremov@linux.com>
To: Andy Whitcroft <apw@canonical.com>, Joe Perches <joe@perches.com>
Cc: Andrew Morton <akpm@linux-foundation.org>, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2] checkpatch: check for nested unlikely calls
Date: Wed, 28 Aug 2019 18:36:53 +0300 [thread overview]
Message-ID: <9655ddea-a98d-073c-f130-e9345517f44f@linux.com> (raw)
In-Reply-To: <20190828133256.13630-1-efremov@linux.com>
On 8/28/19 4:32 PM, Denis Efremov wrote:
> IS_ERR, IS_ERR_OR_NULL, IS_ERR_VALUE already contain unlikely optimization
> internally. Thus, there is no point in calling these functions under
> likely/unlikely.
It looks like this rule could be extended with this list:
CHECK_DATA_CORRUPTION
GLOCK_BUG_ON
SPIN_BUG_ON
RWLOCK_BUG_ON
DCCP_BUG_ON
GEM_BUG_ON
BUG_ON
WARN
WARN_TAINT
WARN_ON_ONCE
WARN_ONCE
WARN_TAINT_ONCE
WARN_ON_SMP
However, grep shows that maybe only BUG_ON, WARN_ON, WARN, WARN_ON_ONCE worth checking:
git grep 'likely(\s*\(CHECK_DATA_CORRUPTION\|GLOCK_BUG_ON\|SPIN_BUG_ON\|RWLOCK_BUG_ON\|DCCP_BUG_ON\|GEM_BUG_ON\|BUG_ON\|WARN\|WARN_TAINT\|WARN_ON_ONCE\|WARN_ONCE\|WARN_TAINT_ONCE\|WARN_ON_SMP\)' .
drivers/gpu/drm/msm/disp/mdp5/mdp5_ctl.c: if (unlikely(WARN_ON(!mixer))) {
drivers/gpu/drm/msm/disp/mdp5/mdp5_ctl.c: if (unlikely(WARN_ON(ctl_cfg->count > MAX_CTL))) {
drivers/gpu/drm/msm/disp/mdp_format.c: if (unlikely(WARN_ON(type >= CSC_MAX)))
drivers/net/ethernet/mellanox/mlx5/core/en_accel/ktls_tx.c: if (unlikely(WARN_ON_ONCE(tls_ctx->netdev != netdev)))
drivers/net/wimax/i2400m/tx.c: if (unlikely(WARN_ON(pad_buf == NULL
drivers/xen/events/events_base.c: if (unlikely(WARN(irq >= nr_irqs, "Invalid irq %d!\n", irq)))
fs/open.c: if (unlikely(WARN_ON(!f->f_op))) {
fs/xfs/xfs_buf.c: if (unlikely(WARN_ON(!bp->b_ops || !bp->b_ops->magic[idx])))
fs/xfs/xfs_buf.c: if (unlikely(WARN_ON(!bp->b_ops || !bp->b_ops->magic16[idx])))
> +# nested likely/unlikely calls
if ($line =~ /\b(?:(?:un)?likely)\s*\(\s*!?\s*((?:IS_ERR(?:_OR_NULL|_VALUE)?)|(?:BUG_ON|WARN(?:_ON(?:_ONCE)?)?)))/) {
or maybe even to match all possible WARNs:
if ($line =~ /\b(?:(?:un)?likely)\s*\(\s*!?\s*((?:IS_ERR(?:_OR_NULL|_VALUE)?)|(?:BUG_ON|WARN)))/) {
> + WARN("LIKELY_MISUSE",
> + "nested (un)?likely calls, unlikely already used in $1 internally\n" . $herecurr);
> + }
Any suggestions for v3?
Thanks,
Denis
prev parent reply other threads:[~2019-08-28 15:36 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-08-27 16:55 [PATCH] checkpatch: check for nested (un)?likely calls Denis Efremov
2019-08-27 17:21 ` Joe Perches
2019-08-27 17:32 ` Denis Efremov
2019-08-28 13:32 ` [PATCH v2] checkpatch: check for nested unlikely calls Denis Efremov
2019-08-28 15:36 ` Denis Efremov [this message]
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=9655ddea-a98d-073c-f130-e9345517f44f@linux.com \
--to=efremov@linux.com \
--cc=akpm@linux-foundation.org \
--cc=apw@canonical.com \
--cc=joe@perches.com \
--cc=linux-kernel@vger.kernel.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox