mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Andrew Cooper <andrew.cooper3@citrix.com>
To: "Stefano Stabellini" <sstabellini@kernel.org>,
	"Roger Pau Monné" <roger.pau@citrix.com>
Cc: <jgross@suse.com>, <linux-kernel@vger.kernel.org>,
	<xen-devel@lists.xen.org>,
	Stefano Stabellini <stefano@aporeto.com>,
	<boris.ostrovsky@oracle.com>
Subject: Re: [Xen-devel] [PATCH v4 07/18] xen/pvcalls: implement socket command
Date: Thu, 22 Jun 2017 19:36:16 +0100	[thread overview]
Message-ID: <0ae24b76-03f5-5adf-5437-2c61f934502b@citrix.com> (raw)
In-Reply-To: <alpine.DEB.2.10.1706221118110.12819@sstabellini-ThinkPad-X260>

On 22/06/17 19:29, Stefano Stabellini wrote:
> On Thu, 22 Jun 2017, Roger Pau Monné wrote:
>> On Wed, Jun 21, 2017 at 01:16:56PM -0700, Stefano Stabellini wrote:
>>> On Tue, 20 Jun 2017, Roger Pau Monné wrote:
>>>> On Thu, Jun 15, 2017 at 12:09:36PM -0700, Stefano Stabellini wrote:
>>>>> Just reply with success to the other end for now. Delay the allocation
>>>>> of the actual socket to bind and/or connect.
>>>>>
>>>>> Signed-off-by: Stefano Stabellini <stefano@aporeto.com>
>>>>> CC: boris.ostrovsky@oracle.com
>>>>> CC: jgross@suse.com
>>>>> ---
>>>>>  drivers/xen/pvcalls-back.c | 27 +++++++++++++++++++++++++++
>>>>>  1 file changed, 27 insertions(+)
>>>>>
>>>>> diff --git a/drivers/xen/pvcalls-back.c b/drivers/xen/pvcalls-back.c
>>>>> index 437c2ad..953458b 100644
>>>>> --- a/drivers/xen/pvcalls-back.c
>>>>> +++ b/drivers/xen/pvcalls-back.c
>>>>> @@ -12,12 +12,17 @@
>>>>>   * GNU General Public License for more details.
>>>>>   */
>>>>>  
>>>>> +#include <linux/inet.h>
>>>>>  #include <linux/kthread.h>
>>>>>  #include <linux/list.h>
>>>>>  #include <linux/radix-tree.h>
>>>>>  #include <linux/module.h>
>>>>>  #include <linux/semaphore.h>
>>>>>  #include <linux/wait.h>
>>>>> +#include <net/sock.h>
>>>>> +#include <net/inet_common.h>
>>>>> +#include <net/inet_connection_sock.h>
>>>>> +#include <net/request_sock.h>
>>>>>  
>>>>>  #include <xen/events.h>
>>>>>  #include <xen/grant_table.h>
>>>>> @@ -54,6 +59,28 @@ struct pvcalls_fedata {
>>>>>  static int pvcalls_back_socket(struct xenbus_device *dev,
>>>>>  		struct xen_pvcalls_request *req)
>>>>>  {
>>>>> +	struct pvcalls_fedata *fedata;
>>>>> +	int ret;
>>>>> +	struct xen_pvcalls_response *rsp;
>>>>> +
>>>>> +	fedata = dev_get_drvdata(&dev->dev);
>>>>> +
>>>>> +	if (req->u.socket.domain != AF_INET ||
>>>>> +	    req->u.socket.type != SOCK_STREAM ||
>>>>> +	    (req->u.socket.protocol != IPPROTO_IP &&
>>>>> +	     req->u.socket.protocol != AF_INET))
>>>>> +		ret = -EAFNOSUPPORT;
>>>> Sorry for jumping into this out of the blue, but shouldn't all the
>>>> constants used above be part of the protocol? AF_INET/SOCK_STREAM/...
>>>> are all part of POSIX, but their specific value is not defined in the
>>>> standard, hence we should have XEN_AF_INET/XEN_SOCK_STREAM/... Or am I
>>>> just missing something?
>>> The values of these constants for the pvcalls protocol are defined by
>>> docs/misc/pvcalls.markdown under "Socket families and address format".
>>>
>>> They happen to be the same as the ones defined by Linux as AF_INET,
>>> SOCK_STREAM, etc, so in Linux I am just using those, but that is just an
>>> implementation detail internal to the Linux kernel driver. What is
>>> important from the protocol ABI perspective are the values defined by
>>> docs/misc/pvcalls.markdown.
>> Oh I see. I still think this should be part of the public pvcalls.h
>> header, and that the error codes should be the ones defined in
>> public/errno.h (or else also added to the pvcalls header).
> This was done differently in the past, but now that we have a formal
> process, a person in charge of new PV drivers reviews, and design
> documents with clearly spelled out ABIs, I consider the design docs
> under docs/misc as the official specification. We don't need headers
> anymore, they are redundant. In fact, we cannot have two specifications,
> and the design docs are certainly the official ones (we don't want the
> specs to be written as header files in C). To me, the headers under
> xen/include/public/io/ are optional helpers. It doesn't matter what's in
> there, or if frontends and backends use them or not.
>
> There is really an argument for removing those headers, because they
> might get out of sync with the spec by mistake, and in those cases, then
> we really end up with two specifications for the same protocol. I would
> be in favor of `git rm'ing all files under xen/include/public/io/ for
> which we have a complete design doc under docs/misc.

+1.

Specifications should not be written in C.  The mess that is the net and
block protocol ABIs are perfect examples of why.

Its fine (and indeed recommended) to provide a header file which
describes the specified protocol, but the authoritative spec should be
in text from.

I would really prefer if more people started using ../docs/specs/.  The
migration v2 documents are currently lonely there...

~Andrew

  reply	other threads:[~2017-06-22 18:36 UTC|newest]

Thread overview: 43+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-06-15 19:09 [PATCH v4 00/18] introduce the Xen PV Calls backend Stefano Stabellini
2017-06-15 19:09 ` [PATCH v4 01/18] xen: introduce the pvcalls interface header Stefano Stabellini
2017-06-15 19:09   ` [PATCH v4 02/18] xen/pvcalls: introduce the pvcalls xenbus backend Stefano Stabellini
2017-06-15 19:09   ` [PATCH v4 03/18] xen/pvcalls: initialize the module and register the " Stefano Stabellini
2017-06-15 19:09   ` [PATCH v4 04/18] xen/pvcalls: xenbus state handling Stefano Stabellini
2017-06-15 19:09   ` [PATCH v4 05/18] xen/pvcalls: connect to a frontend Stefano Stabellini
2017-06-15 19:09   ` [PATCH v4 06/18] xen/pvcalls: handle commands from the frontend Stefano Stabellini
2017-06-15 19:09   ` [PATCH v4 07/18] xen/pvcalls: implement socket command Stefano Stabellini
2017-06-20 15:49     ` Boris Ostrovsky
2017-06-20 16:11     ` [Xen-devel] " Roger Pau Monné
2017-06-21 20:16       ` Stefano Stabellini
2017-06-22  8:26         ` Roger Pau Monné
2017-06-22 18:29           ` Stefano Stabellini
2017-06-22 18:36             ` Andrew Cooper [this message]
2017-06-22 18:38               ` Stefano Stabellini
2017-06-23  9:39             ` Roger Pau Monné
2017-06-15 19:09   ` [PATCH v4 08/18] xen/pvcalls: implement connect command Stefano Stabellini
2017-06-20 15:43     ` Boris Ostrovsky
2017-06-20 15:47       ` Boris Ostrovsky
2017-06-21 20:26         ` Stefano Stabellini
2017-06-21 20:25       ` Stefano Stabellini
2017-06-15 19:09   ` [PATCH v4 09/18] xen/pvcalls: implement bind command Stefano Stabellini
2017-06-15 19:09   ` [PATCH v4 10/18] xen/pvcalls: implement listen command Stefano Stabellini
2017-06-20 15:51     ` Boris Ostrovsky
2017-06-15 19:09   ` [PATCH v4 11/18] xen/pvcalls: implement accept command Stefano Stabellini
2017-06-20 16:02     ` Boris Ostrovsky
2017-06-21 20:29       ` Stefano Stabellini
2017-06-15 19:09   ` [PATCH v4 12/18] xen/pvcalls: implement poll command Stefano Stabellini
2017-06-20 16:43     ` Boris Ostrovsky
2017-06-21 20:39       ` Stefano Stabellini
2017-06-21 20:55         ` Boris Ostrovsky
2017-06-21 21:49           ` Stefano Stabellini
2017-06-15 19:09   ` [PATCH v4 13/18] xen/pvcalls: implement release command Stefano Stabellini
2017-06-20 17:15     ` Boris Ostrovsky
2017-06-21 21:33       ` Stefano Stabellini
2017-06-15 19:09   ` [PATCH v4 14/18] xen/pvcalls: disconnect and module_exit Stefano Stabellini
2017-06-15 19:09   ` [PATCH v4 15/18] xen/pvcalls: implement the ioworker functions Stefano Stabellini
2017-06-20 17:26     ` Boris Ostrovsky
2017-06-15 19:09   ` [PATCH v4 16/18] xen/pvcalls: implement read Stefano Stabellini
2017-06-21 22:37     ` Boris Ostrovsky
2017-06-21 23:32       ` Stefano Stabellini
2017-06-15 19:09   ` [PATCH v4 17/18] xen/pvcalls: implement write Stefano Stabellini
2017-06-15 19:09   ` [PATCH v4 18/18] xen: introduce a Kconfig option to enable the pvcalls backend Stefano Stabellini

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=0ae24b76-03f5-5adf-5437-2c61f934502b@citrix.com \
    --to=andrew.cooper3@citrix.com \
    --cc=boris.ostrovsky@oracle.com \
    --cc=jgross@suse.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=roger.pau@citrix.com \
    --cc=sstabellini@kernel.org \
    --cc=stefano@aporeto.com \
    --cc=xen-devel@lists.xen.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

all inboxes | Powered by JetHome®