public inbox for linux-rdma@vger.kernel.org
 help / color / mirror / Atom feed
From: Romulo Goncalves <R.A.Goncalves-rh8NL+sEX9E@public.gmane.org>
To: Tziporet Koren <tziporet-VPRAkNaXOzVS1MOuV/RT9w@public.gmane.org>
Cc: linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
Subject: Re: Support for Memory windows.
Date: Thu, 18 Nov 2010 13:45:19 +0100	[thread overview]
Message-ID: <4CE51FDF.10705@cwi.nl> (raw)
In-Reply-To: <4CE280BD.8080505-VPRAkNaXOzVS1MOuV/RT9w@public.gmane.org>

On 16/11/10 14:01, Tziporet Koren wrote:
>   On 11/15/2010 5:06 PM, Romulo Goncalves wrote:
>> Hello,
>>
>> For my PhD project I have been using RDMA over inifiniband.
>> Until now I have used memory regions, but now I would like to also
>> define memory windows.
>>
>> The steps taken are the following ones:
>> Register a memory region.
>> Alloc a window
>> bind the window to the registered memory region.
>>
>> The memory region registration worked as expected, however, the call
>> of the function to alloc the memory window did not.
>> It seems there is not an implementation for the functions to alloc,
>> bind, and dealloc a memory window.
>>
>> I looked into verbs.h and I found the reference to the functions in
>> the context_ops:
>>
>> struct ibv_context_ops {
>> ....
>>           struct ibv_mw * (*alloc_mw)(struct ibv_pd *pd, enum
>> ibv_mw_type type);
>>           int                     (*bind_mw)(struct ibv_qp *qp, struct
>> ibv_mw *mw, struct ibv_mw_bind *mw_bind);
>>           int                     (*dealloc_mw)(struct ibv_mw *mw);
>> ....
>> }
>>
>> However, I do not see their declaration or implementation anywhere else.
>> Could be the version of my openfabrics does not support memory windows?
>>
> Memory Windows have not been implemented
>
> Tziporet
That's bad news for me.
Is there any plans to implement it in a near future?

Romulo
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

  parent reply	other threads:[~2010-11-18 12:45 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-11-15 15:06 Support for Memory windows Romulo Goncalves
     [not found] ` <4CE14C82.9020907-rh8NL+sEX9E@public.gmane.org>
2010-11-16 13:01   ` Tziporet Koren
     [not found]     ` <4CE280BD.8080505-VPRAkNaXOzVS1MOuV/RT9w@public.gmane.org>
2010-11-18 12:45       ` Romulo Goncalves [this message]
     [not found]         ` <4CE51FDF.10705-rh8NL+sEX9E@public.gmane.org>
2010-11-18 12:49           ` Tziporet Koren

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=4CE51FDF.10705@cwi.nl \
    --to=r.a.goncalves-rh8nl+sex9e@public.gmane.org \
    --cc=linux-rdma-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=tziporet-VPRAkNaXOzVS1MOuV/RT9w@public.gmane.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