Storage Performance Development Kit (SPDK)
 help / color / mirror / Atom feed
From: Anand Subramanian <anand.s at samsung.com>
To: spdk@lists.01.org
Subject: Re: [SPDK] Testing out the compress/reduce flow
Date: Tue, 13 Nov 2018 00:11:21 +0000	[thread overview]
Message-ID: <b52c70e8c36a4ca1aaa0093a1bc30640@samsung.com> (raw)
In-Reply-To: 4C1DA182-3CA0-4594-9411-3DF020398324@intel.com

[-- Attachment #1: Type: text/plain, Size: 4590 bytes --]

Hi Jim, Paul,

Thanks for letting me know. Sounds good. I think initially it would really help to have the pmem interface calls abstracted in the sense that we should ideally be able to run the UTs without a pm device. And for the most part, it would help to be able to test it out with a ramdisk or something (without a real pm device). 

It would really help if you could put out some steps for a UT setup similar to what you folks use, at least for starters before we try other configurations.

Look fwd to the reviews.

Cheers,
Anand

-----Original Message-----
From: SPDK [mailto:spdk-bounces(a)lists.01.org] On Behalf Of Harris, James R
Sent: Monday, November 12, 2018 3:41 PM
To: Storage Performance Development Kit <spdk(a)lists.01.org>
Subject: Re: [SPDK] Testing out the compress/reduce flow

Hi Anand,

I think the main thing right now is just reviewing patches as they are posted.  I have a series of patches I've been working on locally that I've been trying to get out for review a few patches at a time.  But after Paul and I discussed some of the interfaces last week, we're going to change how the pmem file is opened.  libreduce is going to do that now rather than the user of the library.  I'm reworking those interfaces now and will have 5 hours on planes tomorrow to wrap up those patches and get them out to GerritHub for review.

-Jim


On 11/12/18, 4:07 PM, "SPDK on behalf of Luse, Paul E" <spdk-bounces(a)lists.01.org on behalf of paul.e.luse(a)intel.com> wrote:

    Hi Anand,
    
    Thanks for asking! Yes, there's always ways to help :)  As you know Jim is doing the SPDK reduce library so I'll let him comment on what might make sense there.  I'm doing the vbdev module and I'm probably just a few days away from being able to use some help with unit test. Right now I'm in the middle of getting the path app<-->vbdev<-->dpdk compressdev<-->bdev> coded and there will likely be a little throw away work there because once it interfaces with the reduce lib a lot of things will change however we want to make sure, since both compressdev API and reduclibe API are new, that we can test them individually through the vbdev first.  With Jim's library we'll even be able to test the complete path without doing any real compression, etc.  
    
    Stay tuned and don't ever hesitate to hit us up on the mailing list.  I'll send a note out later this week and let you know the best time to start -  initially it will be a patch based on the vbdev compress patch that start by adding the basic framework for UT and can cover the non-compression centric functions. Won't be long!
    
    Thanks
    Paul
    
    -----Original Message-----
    From: SPDK [mailto:spdk-bounces(a)lists.01.org] On Behalf Of Anand Subramanian
    Sent: Monday, November 12, 2018 2:42 PM
    To: spdk(a)lists.01.org
    Subject: [SPDK] Testing out the compress/reduce flow
    
    Hi Jim, Paul,
    
    Is there any way to help test out the basic flows for the md or the data for the compress/reduce work so far?
    
    Is there an easy way to setup a testbed for this without pm (some steps to maybe repro the compress lvol or to setup what you folks have would help)? Maybe that will help w.r.t. some initial testing and/or fixing of some of the initial low-hanging issues?
    
    Thanks,
    Anand
    _______________________________________________
    SPDK mailing list
    SPDK(a)lists.01.org
    https://urldefense.proofpoint.com/v2/url?u=https-3A__lists.01.org_mailman_listinfo_spdk&d=DwIGaQ&c=JfeWlBa6VbDyTXraMENjy_b_0yKWuqQ4qY-FPhxK4x8w-TfgRBDyeV4hVQQBEgL2&r=WDvM24o7tXcmRRL6eRTRZUTaIuukZ9NPcFO9eKhUBVs&m=d1rIGsz4g34kRr5kmRLZlJH1D9x9Kie2K6dEnORFiOw&s=mGa0QuxmhKd0govK6zdAej4AHhKaw94vg_xiLYQTOYM&e=
    _______________________________________________
    SPDK mailing list
    SPDK(a)lists.01.org
    https://urldefense.proofpoint.com/v2/url?u=https-3A__lists.01.org_mailman_listinfo_spdk&d=DwIGaQ&c=JfeWlBa6VbDyTXraMENjy_b_0yKWuqQ4qY-FPhxK4x8w-TfgRBDyeV4hVQQBEgL2&r=WDvM24o7tXcmRRL6eRTRZUTaIuukZ9NPcFO9eKhUBVs&m=d1rIGsz4g34kRr5kmRLZlJH1D9x9Kie2K6dEnORFiOw&s=mGa0QuxmhKd0govK6zdAej4AHhKaw94vg_xiLYQTOYM&e=
    

_______________________________________________
SPDK mailing list
SPDK(a)lists.01.org
https://urldefense.proofpoint.com/v2/url?u=https-3A__lists.01.org_mailman_listinfo_spdk&d=DwIGaQ&c=JfeWlBa6VbDyTXraMENjy_b_0yKWuqQ4qY-FPhxK4x8w-TfgRBDyeV4hVQQBEgL2&r=WDvM24o7tXcmRRL6eRTRZUTaIuukZ9NPcFO9eKhUBVs&m=d1rIGsz4g34kRr5kmRLZlJH1D9x9Kie2K6dEnORFiOw&s=mGa0QuxmhKd0govK6zdAej4AHhKaw94vg_xiLYQTOYM&e=

             reply	other threads:[~2018-11-13  0:11 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-11-13  0:11 Anand Subramanian [this message]
  -- strict thread matches above, loose matches on Subject: below --
2018-11-13  0:21 [SPDK] Testing out the compress/reduce flow Luse, Paul E
2018-11-13  0:18 Anand Subramanian
2018-11-13  0:16 Luse, Paul E
2018-11-12 23:40 Harris, James R
2018-11-12 23:24 Luse, Paul E
2018-11-12 23:07 Luse, Paul E
2018-11-12 21:41 Anand Subramanian

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=b52c70e8c36a4ca1aaa0093a1bc30640@samsung.com \
    --to=spdk@lists.01.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox