Message-ID: <1440800980.8932.66.camel@edumazet-glaptop2.roam.corp.google.com>
Subject: Re: [PATCH RFC V2 2/2] net: Optimize snmp stat aggregation by
 walking all the percpu data at once
From: Eric Dumazet <eric.dumazet@gmail.com>
To: Joe Perches <joe@perches.com>
Cc: David Miller <davem@davemloft.net>, 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
Date: Fri, 28 Aug 2015 15:29:40 -0700
In-Reply-To: <1440797172.11525.161.camel@perches.com>
References: <1440610653-14210-3-git-send-email-raghavendra.kt@linux.vnet.ibm.com>
	 <20150827.113823.214019265460582055.davem@davemloft.net>
	 <55E00238.10909@linux.vnet.ibm.com>
	 <20150828.112413.424099339331017970.davem@davemloft.net>
	 <1440789654.11525.137.camel@perches.com>
	 <1440794020.8932.45.camel@edumazet-glaptop2.roam.corp.google.com>
	 <1440795211.11525.146.camel@perches.com>
	 <1440795336.8932.46.camel@edumazet-glaptop2.roam.corp.google.com>
	 <1440796183.11525.153.camel@perches.com>
	 <1440796487.8932.48.camel@edumazet-glaptop2.roam.corp.google.com>
	 <1440797172.11525.161.camel@perches.com>
Content-Type: text/plain; charset="UTF-8"
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1157
Lines: 29

On Fri, 2015-08-28 at 14:26 -0700, Joe Perches wrote:

> Always a possibility, but I don't think so.
> 
> > put_unaligned is happening on a space allocated from rtnetlink skb, not
> > the temp space needed to perform the per cpu folding.
> 
> That's why I suggested changing the snmp_fill_stats arguments.
> 
> If the naturally aligned allocated u64 array is used and then
> copied as a block to the rtnetlink skb, I believe there's no
> alignment issue that would require put_unaligned.

1) u64 array[XX] on stack is naturally aligned,
kzalloc() wont improve this at all. Not sure what you believe.

2) put_unaligned() is basically a normal memory write on x86.
 memcpy(dst,src,...) will have a problem anyway on arches that care,
because src & dst wont have same alignment.

288 bytes on stack in a leaf function in this path is totally fine, it
is not like we're calling ext4/xfs/nfs code after this point.


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/