From: Joe Perches <joe@perches.com>
To: David Miller <davem@davemloft.net>
Cc: raghavendra.kt@linux.vnet.ibm.com, edumazet@google.com,
kuznet@ms2.inr.ac.ru, jmorris@namei.org, yoshfuji@linux-ipv6.org,
kaber@trash.net, jiri@resnulli.us, hannes@stressinduktion.org,
tom@herbertland.com, azhou@nicira.com, ebiederm@xmission.com,
ipm@chirality.org.uk, nicolas.dichtel@6wind.com,
serge.hallyn@canonical.com, netdev@vger.kernel.org,
linux-kernel@vger.kernel.org, anton@au1.ibm.com,
nacc@linux.vnet.ibm.com, srikar@linux.vnet.ibm.com
Subject: Re: [PATCH RFC V2 2/2] net: Optimize snmp stat aggregation by walking all the percpu data at once
Date: Fri, 28 Aug 2015 12:20:54 -0700 [thread overview]
Message-ID: <1440789654.11525.137.camel@perches.com> (raw)
In-Reply-To: <20150828.112413.424099339331017970.davem@davemloft.net>
On Fri, 2015-08-28 at 11:24 -0700, David Miller wrote:
> From: Raghavendra K T <raghavendra.kt@linux.vnet.ibm.com>
> Date: Fri, 28 Aug 2015 12:09:52 +0530
>
> > On 08/28/2015 12:08 AM, David Miller wrote:
> >> From: Raghavendra K T <raghavendra.kt@linux.vnet.ibm.com>
> >> Date: Wed, 26 Aug 2015 23:07:33 +0530
> >>
> >>> @@ -4641,10 +4647,12 @@ static inline void __snmp6_fill_stats64(u64
> >>> *stats, void __percpu *mib,
> >>> static void snmp6_fill_stats(u64 *stats, struct inet6_dev *idev, int
> >>> attrtype,
> >>> int bytes)
> >>> {
> >>> + u64 buff[IPSTATS_MIB_MAX] = {0,};
> >>> +
> ...
> > hope you wanted to know the overhead than to change the current
> > patch. please let me know..
>
> I want you to change that variable initializer to an explicit memset().
>
> The compiler is emitting a memset() or similar _anyways_.
>
> Not because it will have any impact at all upon performance, but because
> of how it looks to people trying to read and understand the code.
I don't read it as particularly different.
There are > 100 uses of the not quite a memset initialization
style using "= { <0,> }" in net/
$ git grep -E "=\s*\{\s*0?\s*,?\s*\}" net | wc -l
138
There is a difference though if a struct is copied to
user-space as a {} initialization only guarantees that
struct members are initialized to 0 where memset also
zeros any alignment padding.
Maybe a checkpatch rule like this?
---
scripts/checkpatch.pl | 7 +++++++
1 file changed, 7 insertions(+)
diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index e14dcdb..f79e5c9 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -3237,6 +3237,13 @@ sub process {
next;
}
+# check for non-global initializations that could be memset
+ if ($realfile =~ m@^(drivers/net/|net/)@ &&
+ $sline =~ /^.\s+$Declare\s*$Ident\s*=\s*\{\s*0?\s*,?\s*\}/) {
+ CHK("BRACE_INITIALIZATION",
+ "Prefer an explicit memset to a declaration initialization\n" . $herecurr);
+ }
+
# check for initialisation to aggregates open brace on the next line
if ($line =~ /^.\s*{/ &&
$prevline =~ /(?:^|[^=])=\s*$/) {
next prev parent reply other threads:[~2015-08-28 19:21 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-08-26 17:37 [PATCH RFC V2 0/2] Optimize the snmp stat aggregation for large cpus Raghavendra K T
2015-08-26 17:37 ` [PATCH RFC V2 1/2] net: Introduce helper functions to get the per cpu data Raghavendra K T
2015-08-26 17:37 ` [PATCH RFC V2 2/2] net: Optimize snmp stat aggregation by walking all the percpu data at once Raghavendra K T
2015-08-27 18:38 ` David Miller
2015-08-28 6:39 ` Raghavendra K T
2015-08-28 18:24 ` David Miller
2015-08-28 19:20 ` Joe Perches [this message]
2015-08-28 20:33 ` Eric Dumazet
2015-08-28 20:53 ` Joe Perches
2015-08-28 20:55 ` Eric Dumazet
2015-08-28 21:09 ` Joe Perches
2015-08-28 21:14 ` Eric Dumazet
2015-08-28 21:26 ` Joe Perches
2015-08-28 22:29 ` Eric Dumazet
2015-08-28 23:12 ` Joe Perches
2015-08-29 0:06 ` Eric Dumazet
2015-08-29 0:35 ` Joe Perches
2015-08-29 0:59 ` Eric Dumazet
2015-08-29 2:57 ` Raghavendra K T
2015-08-29 3:26 ` Eric Dumazet
2015-08-29 7:52 ` Raghavendra K T
2015-08-29 5:11 ` David Miller
2015-08-29 7:53 ` Raghavendra K T
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=1440789654.11525.137.camel@perches.com \
--to=joe@perches.com \
--cc=anton@au1.ibm.com \
--cc=azhou@nicira.com \
--cc=davem@davemloft.net \
--cc=ebiederm@xmission.com \
--cc=edumazet@google.com \
--cc=hannes@stressinduktion.org \
--cc=ipm@chirality.org.uk \
--cc=jiri@resnulli.us \
--cc=jmorris@namei.org \
--cc=kaber@trash.net \
--cc=kuznet@ms2.inr.ac.ru \
--cc=linux-kernel@vger.kernel.org \
--cc=nacc@linux.vnet.ibm.com \
--cc=netdev@vger.kernel.org \
--cc=nicolas.dichtel@6wind.com \
--cc=raghavendra.kt@linux.vnet.ibm.com \
--cc=serge.hallyn@canonical.com \
--cc=srikar@linux.vnet.ibm.com \
--cc=tom@herbertland.com \
--cc=yoshfuji@linux-ipv6.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