netdev - [PATCH] alloc_percpu() fails to allocate percpu data

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Date:	Thu, 21 Feb 2008 19:00:03 +0100
From:	Eric Dumazet <dada1@...mosbay.com>
To:	"David S. Miller" <davem@...emloft.net>,
	Andrew Morton <akpm@...ux-foundation.org>
Cc:	linux kernel <linux-kernel@...r.kernel.org>,
	netdev@...r.kernel.org, Christoph Lameter <clameter@....com>,
	"Zhang, Yanmin" <yanmin_zhang@...ux.intel.com>
Subject: [PATCH] alloc_percpu() fails to allocate percpu data

Some oprofile results obtained while using tbench on a 2x2 cpu machine 
were very surprising.

For example, loopback_xmit() function was using high number of cpu 
cycles to perform
the statistic updates, supposed to be real cheap since they use percpu data

        pcpu_lstats = netdev_priv(dev);
        lb_stats = per_cpu_ptr(pcpu_lstats, smp_processor_id());
        lb_stats->packets++;  /* HERE : serious contention */
        lb_stats->bytes += skb->len;

struct pcpu_lstats is a small structure containing two longs. It appears 
that on my 32bits platform,
alloc_percpu(8) allocates a single cache line,  instead of giving to 
each cpu a separate
cache line.

Using the following patch gave me impressive boost in various benchmarks 
( 6 % in tbench)
(all percpu_counters hit this bug too)

Long term fix (ie >= 2.6.26) would be to let each CPU allocate their own 
block of memory, so that we
dont need to roudup sizes to L1_CACHE_BYTES, or merging the SGI stuff of 
course...

Note : SLUB vs SLAB is important here to *show* the improvement, since 
they dont have the same minimum
allocation sizes (8 bytes vs 32 bytes).
This could very well explain regressions some guys reported when they 
switched to SLUB.

Signed-off-by: Eric Dumazet <dada1@...mosbay.com>

 mm/allocpercpu.c |   15 ++++++++++++++-
 1 files changed, 14 insertions(+), 1 deletion(-)

View attachment "percpu_populate.patch" of type "text/plain" (1230 bytes)