Message-ID: <CAJd=RBAvvzK=TXwDaEjq2t+uEuP2PSi6zaUj7EW4UbL_AUsJAg@mail.gmail.com>
Date:	Tue, 7 Feb 2012 20:45:18 +0800
From:	Hillf Danton <dhillf@...il.com>
To:	Mel Gorman <mgorman@...e.de>
Cc:	Andrew Morton <akpm@...ux-foundation.org>,
	Linux-MM <linux-mm@...ck.org>,
	Linux-Netdev <netdev@...r.kernel.org>,
	LKML <linux-kernel@...r.kernel.org>,
	David Miller <davem@...emloft.net>, Neil Brown <neilb@...e.de>,
	Peter Zijlstra <a.p.zijlstra@...llo.nl>
Subject: Re: [PATCH 00/15] Swap-over-NBD without deadlocking V8

On Tue, Feb 7, 2012 at 6:56 AM, Mel Gorman <mgorman@...e.de> wrote:
>
> The core issue is that network block devices do not use mempools like normal
> block devices do. As the host cannot control where they receive packets from,
> they cannot reliably work out in advance how much memory they might need.
>
>
> Patch 1 serialises access to min_free_kbytes. It's not strictly needed
>        by this series but as the series cares about watermarks in
>        general, it's a harmless fix. It could be merged independently.
>
>
Any light shed on tuning min_free_kbytes for every day work?


> Patch 2 adds knowledge of the PFMEMALLOC reserves to SLAB and SLUB to
>        preserve access to pages allocated under low memory situations
>        to callers that are freeing memory.
>
> Patch 3 introduces __GFP_MEMALLOC to allow access to the PFMEMALLOC
>        reserves without setting PFMEMALLOC.
>
> Patch 4 opens the possibility for softirqs to use PFMEMALLOC reserves
>        for later use by network packet processing.
>
> Patch 5 ignores memory policies when ALLOC_NO_WATERMARKS is set.
>
> Patches 6-11 allows network processing to use PFMEMALLOC reserves when
>        the socket has been marked as being used by the VM to clean
>        pages. If packets are received and stored in pages that were
>        allocated under low-memory situations and are unrelated to
>        the VM, the packets are dropped.
>
> Patch 12 is a micro-optimisation to avoid a function call in the
>        common case.
>
> Patch 13 tags NBD sockets as being SOCK_MEMALLOC so they can use
>        PFMEMALLOC if necessary.
>
If it is feasible to bypass hang by tuning min_mem_kbytes, things may
become simpler if NICs are also tagged. Sock buffers, pre-allocated if
necessary just after NICs are turned on, are not handed back to kmem
cache but queued on local lists which are maintained by NIC driver, based
the on the info of min_mem_kbytes or similar, for tagged NICs.
Upside is no changes in VM core. Downsides?


> Patch 14 notes that it is still possible for the PFMEMALLOC reserve
>        to be depleted. To prevent this, direct reclaimers get
>        throttled on a waitqueue if 50% of the PFMEMALLOC reserves are
>        depleted.  It is expected that kswapd and the direct reclaimers
>        already running will clean enough pages for the low watermark
>        to be reached and the throttled processes are woken up.
>
> Patch 15 adds a statistic to track how often processes get throttled
>
>
> For testing swap-over-NBD, a machine was booted with 2G of RAM with a
> swapfile backed by NBD. 8*NUM_CPU processes were started that create
> anonymous memory mappings and read them linearly in a loop. The total
> size of the mappings were 4*PHYSICAL_MEMORY to use swap heavily under
> memory pressure. Without the patches, the machine locks up within
> minutes and runs to completion with them applied.
>
>
While testing, what happens if the network wire is plugged off over
three minutes?

Thanks
Hillf
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Powered by blists - more mailing lists