linux-kernel - Re: [RFC 1/4] mm: Define coherent device memory node

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:   Wed, 30 Nov 2016 17:16:04 +0530
From:   Anshuman Khandual <khandual@...ux.vnet.ibm.com>
To:     Dave Hansen <dave.hansen@...el.com>, linux-kernel@...r.kernel.org,
        linux-mm@...ck.org
Cc:     mhocko@...e.com, vbabka@...e.cz, mgorman@...e.de,
        minchan@...nel.org, aneesh.kumar@...ux.vnet.ibm.com,
        bsingharora@...il.com, srikar@...ux.vnet.ibm.com,
        haren@...ux.vnet.ibm.com, jglisse@...hat.com
Subject: Re: [RFC 1/4] mm: Define coherent device memory node

On 11/29/2016 11:27 PM, Dave Hansen wrote:
> On 11/22/2016 06:19 AM, Anshuman Khandual wrote:
>> @@ -393,6 +393,9 @@ enum node_states {
>>  	N_MEMORY = N_HIGH_MEMORY,
>>  #endif
>>  	N_CPU,		/* The node has one or more cpus */
>> +#ifdef CONFIG_COHERENT_DEVICE
>> +	N_COHERENT_DEVICE,
>> +#endif
>>  	NR_NODE_STATES
>>  };
> 
> Don't we really want this to be N_MEMORY_ISOLATED?  Or, better yet,

Sure, If we move from a CDM description to a purely node isolation one.
I am still thinking through this.

> N_MEMORY_UNISOLATED so that we can just drop the bitmap in for N_MEMORY

Did not get that, N_MEMORY_UNISOLATED for the system RAM nodes which are
not isolated ? Then where the isolated/CDM nodes go in ?

> and not have to do any bit manipulation operations at runtime.
>