qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
To: Cornelia Huck <cohuck@redhat.com>
Cc: qemu-devel@nongnu.org, peter.maydell@linaro.org,
	yuval.shaia@oracle.com, pjp@fedoraproject.org
Subject: Re: [Qemu-devel] [PATCH PULL 02/31] contrib/rdmacm-mux: Add implementation of RDMA User MAD multiplexer
Date: Mon, 7 Jan 2019 19:28:10 +0200	[thread overview]
Message-ID: <bed8bcb6-67ca-b8c5-ffbe-684241fe2b70@gmail.com> (raw)
In-Reply-To: <20190103113414.503eed8b.cohuck@redhat.com>



On 1/3/19 12:34 PM, Cornelia Huck wrote:
> On Sat, 22 Dec 2018 11:50:07 +0200
> Marcel Apfelbaum <marcel.apfelbaum@gmail.com> wrote:
>
>> From: Yuval Shaia <yuval.shaia@oracle.com>
>>
>> RDMA MAD kernel module (ibcm) disallow more than one MAD-agent for a
>> given MAD class.
>> This does not go hand-by-hand with qemu pvrdma device's requirements
>> where each VM is MAD agent.
>> Fix it by adding implementation of RDMA MAD multiplexer service which on
>> one hand register as a sole MAD agent with the kernel module and on the
>> other hand gives service to more than one VM.
>>
>> Design Overview:
>> Reviewed-by: Shamir Rabinovitch <shamir.rabinovitch@oracle.com>
>> ----------------
>> A server process is registered to UMAD framework (for this to work the
>> rdma_cm kernel module needs to be unloaded) and creates a unix socket to
>> listen to incoming request from clients.
>> A client process (such as QEMU) connects to this unix socket and
>> registers with its own GID.
>>
>> TX:
>> ----
>> When client needs to send rdma_cm MAD message it construct it the same
>> way as without this multiplexer, i.e. creates a umad packet but this
>> time it writes its content to the socket instead of calling umad_send().
>> The server, upon receiving such a message fetch local_comm_id from it so
>> a context for this session can be maintain and relay the message to UMAD
>> layer by calling umad_send().
>>
>> RX:
>> ----
>> The server creates a worker thread to process incoming rdma_cm MAD
>> messages. When an incoming message arrived (umad_recv()) the server,
>> depending on the message type (attr_id) looks for target client by
>> either searching in gid->fd table or in local_comm_id->fd table. With
>> the extracted fd the server relays to incoming message to the client.
>>
>> Signed-off-by: Yuval Shaia <yuval.shaia@oracle.com>
>> Reviewed-by: Shamir Rabinovitch <shamir.rabinovitch@oracle.com>
>> Signed-off-by: Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
>>
>> Signed-off-by: Marcel Apfelbaum <marcel.apfelbaum@gmail.com>
>> ---
>>   MAINTAINERS                      |   1 +
>>   Makefile                         |   3 +
>>   Makefile.objs                    |   1 +
>>   contrib/rdmacm-mux/Makefile.objs |   4 +
>>   contrib/rdmacm-mux/main.c        | 798 +++++++++++++++++++++++++++++++
>>   contrib/rdmacm-mux/rdmacm-mux.h  |  61 +++
>>   6 files changed, 868 insertions(+)
>>   create mode 100644 contrib/rdmacm-mux/Makefile.objs
>>   create mode 100644 contrib/rdmacm-mux/main.c
>>   create mode 100644 contrib/rdmacm-mux/rdmacm-mux.h
> This commit breaks building with clang on Fedora 28 for me; I get a
> bunch of
>

Hi Cornelia,
Thanks for spotting the issue.

> clang-6.0: error: -libumad: 'linker' input unused [-Werror,-Wunused-command-line-argument]

Is really strange, the rdma-mux is part of the contrib directory
and is not even compiled by default.

> Explicitly passing --disable-pvrdma to configure disables the offending
> code.
>
> Let me know if you need more information.

Can you please send the exact steps you are using to configure and 
compile QEMU,
are you compiling on a x86 machine?

Thanks,
Marcel

  reply	other threads:[~2019-01-07 17:28 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-12-22  9:50 [Qemu-devel] [PATCH PULL 00/31] RDMA queue Marcel Apfelbaum
2018-12-22  9:50 ` [Qemu-devel] [PATCH PULL 01/31] hw/pvrdma: Check the correct return value Marcel Apfelbaum
2018-12-22  9:50 ` [Qemu-devel] [PATCH PULL 02/31] contrib/rdmacm-mux: Add implementation of RDMA User MAD multiplexer Marcel Apfelbaum
2019-01-03 10:34   ` Cornelia Huck
2019-01-07 17:28     ` Marcel Apfelbaum [this message]
2019-01-07 17:54       ` Cornelia Huck
2019-01-08  8:41         ` Marcel Apfelbaum
2019-01-08  9:18           ` Cornelia Huck
2019-01-12  9:46             ` Marcel Apfelbaum
2018-12-22  9:50 ` [Qemu-devel] [PATCH PULL 03/31] hw/rdma: Add ability to force notification without re-arm Marcel Apfelbaum
2018-12-22  9:50 ` [Qemu-devel] [PATCH PULL 04/31] hw/rdma: Return qpn 1 if ibqp is NULL Marcel Apfelbaum
2018-12-22  9:50 ` [Qemu-devel] [PATCH PULL 05/31] hw/rdma: Abort send-op if fail to create addr handler Marcel Apfelbaum
2018-12-22  9:50 ` [Qemu-devel] [PATCH PULL 06/31] hw/rdma: Add support for MAD packets Marcel Apfelbaum
2018-12-22  9:50 ` [Qemu-devel] [PATCH PULL 07/31] hw/pvrdma: Make function reset_device return void Marcel Apfelbaum
2018-12-22  9:50 ` [Qemu-devel] [PATCH PULL 08/31] hw/pvrdma: Make default pkey 0xFFFF Marcel Apfelbaum
2018-12-22  9:50 ` [Qemu-devel] [PATCH PULL 09/31] hw/pvrdma: Set the correct opcode for recv completion Marcel Apfelbaum
2018-12-22  9:50 ` [Qemu-devel] [PATCH PULL 10/31] hw/pvrdma: Set the correct opcode for send completion Marcel Apfelbaum
2018-12-22  9:50 ` [Qemu-devel] [PATCH PULL 11/31] qapi: Define new QMP message for pvrdma Marcel Apfelbaum
2018-12-22  9:50 ` [Qemu-devel] [PATCH PULL 12/31] hw/pvrdma: Add support to allow guest to configure GID table Marcel Apfelbaum
2018-12-22  9:50 ` [Qemu-devel] [PATCH PULL 13/31] vmxnet3: Move some definitions to header file Marcel Apfelbaum
2018-12-22  9:50 ` [Qemu-devel] [PATCH PULL 14/31] hw/pvrdma: Make sure PCI function 0 is vmxnet3 Marcel Apfelbaum
2018-12-22  9:50 ` [Qemu-devel] [PATCH PULL 15/31] hw/rdma: Initialize node_guid from vmxnet3 mac address Marcel Apfelbaum
2018-12-22  9:50 ` [Qemu-devel] [PATCH PULL 16/31] hw/pvrdma: Make device state depend on Ethernet function state Marcel Apfelbaum
2018-12-22  9:50 ` [Qemu-devel] [PATCH PULL 17/31] hw/pvrdma: Fill all CQE fields Marcel Apfelbaum
2018-12-22  9:50 ` [Qemu-devel] [PATCH PULL 18/31] hw/pvrdma: Fill error code in command's response Marcel Apfelbaum
2018-12-22  9:50 ` [Qemu-devel] [PATCH PULL 19/31] hw/rdma: Remove unneeded code that handles more that one port Marcel Apfelbaum
2018-12-22  9:50 ` [Qemu-devel] [PATCH PULL 20/31] vl: Introduce shutdown_notifiers Marcel Apfelbaum
2018-12-22  9:50 ` [Qemu-devel] [PATCH PULL 21/31] hw/pvrdma: Clean device's resource when system is shutdown Marcel Apfelbaum
2018-12-22  9:50 ` [Qemu-devel] [PATCH PULL 22/31] hw/rdma: Do not use bitmap_zero_extend to free bitmap Marcel Apfelbaum
2018-12-22  9:50 ` [Qemu-devel] [PATCH PULL 23/31] hw/rdma: Do not call rdma_backend_del_gid on an empty gid Marcel Apfelbaum
2018-12-22  9:50 ` [Qemu-devel] [PATCH PULL 24/31] docs: Update pvrdma device documentation Marcel Apfelbaum
2018-12-22  9:50 ` [Qemu-devel] [PATCH PULL 25/31] pvrdma: release device resources in case of an error Marcel Apfelbaum
2018-12-22  9:50 ` [Qemu-devel] [PATCH PULL 26/31] rdma: check num_sge does not exceed MAX_SGE Marcel Apfelbaum
2018-12-22  9:50 ` [Qemu-devel] [PATCH PULL 27/31] pvrdma: add uar_read routine Marcel Apfelbaum
2018-12-22  9:50 ` [Qemu-devel] [PATCH PULL 28/31] pvrdma: check number of pages when creating rings Marcel Apfelbaum
2018-12-22  9:50 ` [Qemu-devel] [PATCH PULL 29/31] pvrdma: release ring object in case of an error Marcel Apfelbaum
2018-12-22  9:50 ` [Qemu-devel] [PATCH PULL 30/31] rdma: remove unused VENDOR_ERR_NO_SGE macro Marcel Apfelbaum
2018-12-22  9:50 ` [Qemu-devel] [PATCH PULL 31/31] pvrdma: check return value from pvrdma_idx_ring_has_ routines Marcel Apfelbaum
2018-12-22 13:59 ` [Qemu-devel] [PATCH PULL 00/31] RDMA queue Peter Maydell
2018-12-22 14:30   ` Marcel Apfelbaum

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=bed8bcb6-67ca-b8c5-ffbe-684241fe2b70@gmail.com \
    --to=marcel.apfelbaum@gmail.com \
    --cc=cohuck@redhat.com \
    --cc=peter.maydell@linaro.org \
    --cc=pjp@fedoraproject.org \
    --cc=qemu-devel@nongnu.org \
    --cc=yuval.shaia@oracle.com \
    /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;
as well as URLs for NNTP newsgroup(s).