linux-kernel - Re: [PATCH v2] null_blk: add support for max open/active zone limit for zoned devices

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, 27 Aug 2020 09:14:50 -0700
From:   Randy Dunlap <rdunlap@...radead.org>
To:     Damien Le Moal <Damien.LeMoal@....com>,
        Niklas Cassel <Niklas.Cassel@....com>,
        Jens Axboe <axboe@...nel.dk>
Cc:     Johannes Thumshirn <Johannes.Thumshirn@....com>,
        "linux-block@...r.kernel.org" <linux-block@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v2] null_blk: add support for max open/active zone limit
 for zoned devices

On 8/27/20 8:04 AM, Damien Le Moal wrote:
> On 2020/08/27 23:51, Randy Dunlap wrote:
>> On 8/27/20 6:50 AM, Niklas Cassel wrote:
>>> Add support for user space to set a max open zone and a max active zone
>>> limit via configfs. By default, the default values are 0 == no limit.
>>
>> Hi,
>>
>> How does a user find out about how to use/set these limits?
> 
> For the setting part, this is for testing. So any value, even extreme ones (e.g.
> 1) would be OK to check that a software correctly handles write accesses to
> zones for a device that has open/active zone limitations. A more practical way
> is to reuse values of real devices. For instance, some SMR disks I use have a
> max open limit of 128 and max active 0 (there is no limit for active zones on
> SMR disks as ZBC/ZAC specifications do not define this concept).
> 
> Another example is our soon to come work on btrfs zone support which shows that
> at the very least 6 active zones are needed. So tests can be performed with that
> minimum to check the file system and that its block allocator does not go
> opening/activating too many zones.
> 
> For the using part, the above btrfs example is good: if the FS tries to allocate
> blocks in too many inactive zones at the same time without first filling out
> zones already active, it may exceed the limit and writes will fail. The FS must
> thus be aware of the limits and its block al;locator tuned to limit block
> allocations within a set of zones smaller than the maximum active limit.
> 
> Does this answer your question ?

Yes.  Thank you.

-- 
~Randy