linux-kernel - Re: [PATCH v6] numa: make node_to_cpumask_map() NUMA_NO

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-prev] [thread-next>] [day] [month] [year] [list]
Date:   Thu, 10 Oct 2019 14:07:21 +0800
From:   Yunsheng Lin <linyunsheng@...wei.com>
To:     Robin Murphy <robin.murphy@....com>,
        Peter Zijlstra <peterz@...radead.org>
CC:     Michal Hocko <mhocko@...nel.org>, <catalin.marinas@....com>,
        <will@...nel.org>, <mingo@...hat.com>, <bp@...en8.de>,
        <rth@...ddle.net>, <ink@...assic.park.msu.ru>,
        <mattst88@...il.com>, <benh@...nel.crashing.org>,
        <paulus@...ba.org>, <mpe@...erman.id.au>,
        <heiko.carstens@...ibm.com>, <gor@...ux.ibm.com>,
        <borntraeger@...ibm.com>, <ysato@...rs.sourceforge.jp>,
        <dalias@...c.org>, <davem@...emloft.net>, <ralf@...ux-mips.org>,
        <paul.burton@...s.com>, <jhogan@...nel.org>,
        <jiaxun.yang@...goat.com>, <chenhc@...ote.com>,
        <akpm@...ux-foundation.org>, <rppt@...ux.ibm.com>,
        <anshuman.khandual@....com>, <tglx@...utronix.de>, <cai@....pw>,
        <linux-arm-kernel@...ts.infradead.org>,
        <linux-kernel@...r.kernel.org>, <hpa@...or.com>, <x86@...nel.org>,
        <dave.hansen@...ux.intel.com>, <luto@...nel.org>,
        <len.brown@...el.com>, <axboe@...nel.dk>, <dledford@...hat.com>,
        <jeffrey.t.kirsher@...el.com>, <linux-alpha@...r.kernel.org>,
        <naveen.n.rao@...ux.vnet.ibm.com>, <mwb@...ux.vnet.ibm.com>,
        <linuxppc-dev@...ts.ozlabs.org>, <linux-s390@...r.kernel.org>,
        <linux-sh@...r.kernel.org>, <sparclinux@...r.kernel.org>,
        <tbogendoerfer@...e.de>, <linux-mips@...r.kernel.org>,
        <rafael@...nel.org>, <gregkh@...uxfoundation.org>
Subject: Re: [PATCH v6] numa: make node_to_cpumask_map() NUMA_NO_NODE aware

On 2019/10/9 20:25, Robin Murphy wrote:
> On 2019-10-08 9:38 am, Yunsheng Lin wrote:
>> On 2019/9/25 18:41, Peter Zijlstra wrote:
>>> On Wed, Sep 25, 2019 at 05:14:20PM +0800, Yunsheng Lin wrote:
>>>>  From the discussion above, It seems making the node_to_cpumask_map()
>>>> NUMA_NO_NODE aware is the most feasible way to move forwad.
>>>
>>> That's still wrong.
>>
>> Hi, Peter
>>
>> It seems this has trapped in the dead circle.
>>
>>  From my understanding, NUMA_NO_NODE which means not node numa preference
>> is the state to describe the node of virtual device or the physical device
>> that has equal distance to all cpu.
>>
>> We can be stricter if the device does have a nearer node, but we can not
>> deny that a device does not have a node numa preference or node affinity,
>> which also means the control or data buffer can be allocated at the node where
>> the process is running.
>>
>> As you has proposed, making it -2 and have dev_to_node() warn if the device does
>> have a nearer node and not set by the fw is a way to be stricter.
>>
>> But I think maybe being stricter is not really relevant to NUMA_NO_NODE, because
>> we does need a state to describe the device that have equal distance to all node,
>> even if it is not physically scalable.
>>
>> Any better suggestion to move this forward?
> 
> FWIW (since this is in my inbox), it sounds like the fundamental issue is that NUMA_NO_NODE is conflated for at least two different purposes, so trying to sort that out would be a good first step. AFAICS we have genuine "don't care" cases like alloc_pages_node(), where if the producer says it doesn't matter then the consumer is free to make its own judgement on what to do, and fundamentally different "we expect this thing to have an affinity but it doesn't, so we can't say what's appropriate" cases which could really do with some separate indicator like "NUMA_INVALID_NODE".
> 
> The tricky part is then bestowed on the producers to decide whether they can downgrade "invalid" to "don't care". You can technically build 'a device' whose internal logic is distributed between nodes and thus appears to have equal affinity - interrupt controllers, for example, may have per-CPU or per-node interfaces that end up looking like that - so although it's unlikely it's not outright nonsensical. Similarly a 'device' that's actually emulated behind a firmware call interface may well effectively have no real affinity.

We may set node of the physical device to NUMA_INVALID_NODE when fw does not
provide one.

But what do we do about NUMA_INVALID_NODE when alloc_pages_node() is called
with nid being NUMA_INVALID_NODE?

If we change the node to default one(like node 0) when node of device is
NUMA_INVALID_NODE in device_add(), how do we know the default one(like node 0)
is the right one to choose?

>From the privous disccusion, the below seems not get to consensus yet:
1) Do we need a state like NUMA_NO_NODE to describe that the device does not
   have any numa preference?

2) What do we do if the fw does not provide a node for the device? Should
   we guess and pick one for it and how do we do the guessing? Or leave it
   as it is and handle it as NUMA_NO_NODE?

The point of adding another state like NUMA_INVALID_NODE seems to catch
the case and give a warning above it when the device does have a nearer
node and the fw does not provide one, and alloc_pages_node() still need to
handle it as NUMA_NO_NODE?

If the above is true, then maybe we can move forward with the above goal.

Thanks very much for the suggestion.

> 



> Robin.
> 
> .
>