All of lore.kernel.org
 help / color / mirror / Atom feed
From: Eric Blake <eblake@redhat.com>
To: "Philippe Mathieu-Daudé" <f4bug@amsat.org>,
	"Alistair Francis" <alistair.francis@xilinx.com>,
	"Peter Maydell" <peter.maydell@linaro.org>,
	"Paolo Bonzini" <pbonzini@redhat.com>,
	"Kevin Wolf" <kwolf@redhat.com>,
	"Eduardo Habkost" <ehabkost@redhat.com>,
	"Daniel P . Berrange" <berrange@redhat.com>,
	"Stefan Hajnoczi" <stefanha@redhat.com>
Cc: qemu-devel@nongnu.org,
	"Edgar E . Iglesias" <edgar.iglesias@xilinx.com>,
	"Markus Armbruster" <armbru@redhat.com>,
	"Thomas Huth" <thuth@redhat.com>,
	"Marc-André Lureau" <marcandre.lureau@redhat.com>
Subject: Re: [Qemu-devel] [PATCH v2 1/4] sdbus: add a QMP command to access a SDBus
Date: Fri, 5 Jan 2018 09:29:43 -0600	[thread overview]
Message-ID: <58860a06-3d3f-ab2a-90ea-dd56dfbe9d79@redhat.com> (raw)
In-Reply-To: <20180103214925.16677-2-f4bug@amsat.org>

[-- Attachment #1: Type: text/plain, Size: 5158 bytes --]

On 01/03/2018 03:49 PM, Philippe Mathieu-Daudé wrote:
> Use Base64 to serialize the binary blobs in JSON.
> So far at most 512 bytes will be transfered, which result

s/transfered/transferred/

> in a 684 bytes payload.
> Since this command is intented for qtesting, it is acceptable.

s/intented/intended/

Might be worth mentioning the actual command name,
x-debug-sdbus-command, in the commit message to make future git log
trawling easier.

> 
> Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
> ---

> +##
> +# @SDBusCommandResponse:
> +#
> +# SD Bus command response.
> +#
> +# @base64: the command response encoded as a Base64 string, if any (optional)

s/ (optional)//, now that the documentation engine automatically takes
care of that.

Even if there is no response, isn't the empty string "" more accurate
than omitting the string altogether?  In other words, I'm not sure why
you made the 'base64' member optional.

> +#
> +# Since: 2.11

2.12

> +##
> +{ 'struct': 'SDBusCommandResponse', 'data': {'*base64': 'str'} }
> +
> +##
> +# @x-debug-sdbus-command:
> +#
> +# Execute a command on a SD Bus return the response (if any).
> +#

Maybe mention that this command is only intended for use during unit
testing (that information is already implicit from the x-debug prefix,
but stating it explicitly doesn't hurt).

> +# @qom-path: the SD Bus path
> +# @command: the SD protocol command to execute in the bus
> +# @arg: a 64-bit command argument (optional)
> +# @crc: the command/argument CRC (optional)
> +#
> +# Returns: the response of the command encoded as a Base64 string
> +#
> +# Since: 2.11

2.12

> +#
> +# -> { "execute": "x-debug-sdbus-command",
> +#      "arguments": { "qom-path": "/machine/unattached/device[32]/sd.0",
> +#                     "command": 0x01

That's invalid JSON (which does not understand hex numbers).  You need
"command": 1

> +#      }
> +#    }
> +# <- { "return": {'base64': 'A='} }
> +#
> +##
> +{ 'command': 'x-debug-sdbus-command',
> +  'data': { 'qom-path': 'str',
> +            'command': 'uint8',
> +            '*arg': 'uint64',
> +            '*crc': 'uint16' },
> +  'returns': 'SDBusCommandResponse'
> +}
> diff --git a/hw/sd/sdbus-qmp.c b/hw/sd/sdbus-qmp.c
> new file mode 100644
> index 0000000000..8c4b6f2aee
> --- /dev/null
> +++ b/hw/sd/sdbus-qmp.c
> @@ -0,0 +1,65 @@
> +/*
> + * SD card bus QMP debugging interface (for QTesting).
> + *
> + * Copyright (c) 2017 ?

The question mark in a copyright line is not right.  I don't know what
you meant to use, but unless you were doing the work on behalf of an
employer, your own name is probably correct.  You could include 2018 now
if you wanted.


> +SDBusCommandResponse *qmp_x_debug_sdbus_command(const char *qom_path,
> +                                                uint8_t command,
> +                                                bool has_arg, uint64_t arg,
> +                                                bool has_crc, uint16_t crc,
> +                                                Error **errp)
> +{
> +    uint8_t response[16 + 1];
> +    SDBusCommandResponse *res;
> +    bool ambiguous = false;
> +    Object *obj;
> +    SDBus *sdbus;
> +    int sz;
> +
> +    obj = object_resolve_path(qom_path, &ambiguous);
> +    if (obj == NULL) {

I don't know if the style 'if (!obj) {' is any more prevalent; but it
doesn't really matter.

> +        if (ambiguous) {
> +            error_setg(errp, "Path '%s' is ambiguous", qom_path);
> +        } else {
> +            error_set(errp, ERROR_CLASS_DEVICE_NOT_FOUND,
> +                      "Device '%s' not found", qom_path);
> +        }
> +        return NULL;
> +    }
> +    sdbus = (SDBus *)object_dynamic_cast(obj, TYPE_SD_BUS);

Is the cast still necessary, or does object_dynamic_cast() return void*
so that you can omit the cast?

> +    if (sdbus == NULL) {
> +        error_set(errp, ERROR_CLASS_GENERIC_ERROR,
> +                  "Device '%s' not a sd-bus", qom_path);
> +        return NULL;
> +    }
> +
> +    res = g_new0(SDBusCommandResponse, 1);
> +    sz = sdbus_do_command(sdbus,
> +                          &(SDRequest){ command, arg, has_crc ? crc : -1 },

It's probably safer to use specific initializer assignments, as in:

&(SDRequest){ .cmd = command, .arg = arg, ...


> +++ b/stubs/qmp_sdbus.c
> @@ -0,0 +1,12 @@
> +#include "qemu/osdep.h"
> +#include "qmp-commands.h"
> +#include "hw/sd/sd.h"
> +
> +SDBusCommandResponse *qmp_x_debug_sdbus_command(const char *qom_path,
> +                                                uint8_t command,
> +                                                bool has_arg, uint64_t arg,
> +                                                bool has_crc, uint16_t crc,
> +                                                Error **errp)
> +{
> +    return NULL;

In addition to returning NULL, the stub should set errp.

-- 
Eric Blake, Principal Software Engineer
Red Hat, Inc.           +1-919-301-3266
Virtualization:  qemu.org | libvirt.org


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 619 bytes --]

  parent reply	other threads:[~2018-01-05 15:30 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-01-03 21:49 [Qemu-devel] [RFC PATCH v2 0/4] sdbus: testing sdcards Philippe Mathieu-Daudé
2018-01-03 21:49 ` [Qemu-devel] [PATCH v2 1/4] sdbus: add a QMP command to access a SDBus Philippe Mathieu-Daudé
2018-01-05 15:11   ` Stefan Hajnoczi
2018-01-05 15:29   ` Eric Blake [this message]
2018-01-05 16:06     ` Philippe Mathieu-Daudé
2018-01-05 16:10       ` Eric Blake
2018-01-05 21:28     ` Philippe Mathieu-Daudé
2019-03-08 16:11   ` Philippe Mathieu-Daudé
2019-03-11 11:49     ` Thomas Huth
2019-03-11 13:43       ` Eduardo Habkost
2019-03-11 13:48         ` Peter Maydell
2019-03-11 15:52           ` Markus Armbruster
2018-01-03 21:49 ` [Qemu-devel] [RFC PATCH v2 2/4] libqos: add a sdbus API Philippe Mathieu-Daudé
2018-01-05 15:18   ` Stefan Hajnoczi
2018-01-03 21:49 ` [Qemu-devel] [RFC PATCH v2 3/4] libqos: implement sdbus QMP driver Philippe Mathieu-Daudé
2018-01-05 15:25   ` Stefan Hajnoczi
2018-01-03 21:49 ` [Qemu-devel] [RFC PATCH v2 4/4] tests: add some sdcard qtest Philippe Mathieu-Daudé
2018-01-05 15:31   ` Stefan Hajnoczi
2018-01-05 15:55     ` Philippe Mathieu-Daudé
2018-01-08 14:32       ` Stefan Hajnoczi
2018-01-03 22:17 ` [Qemu-devel] [RFC PATCH v2 0/4] sdbus: testing sdcards no-reply
2018-01-05 15:32 ` Stefan Hajnoczi

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=58860a06-3d3f-ab2a-90ea-dd56dfbe9d79@redhat.com \
    --to=eblake@redhat.com \
    --cc=alistair.francis@xilinx.com \
    --cc=armbru@redhat.com \
    --cc=berrange@redhat.com \
    --cc=edgar.iglesias@xilinx.com \
    --cc=ehabkost@redhat.com \
    --cc=f4bug@amsat.org \
    --cc=kwolf@redhat.com \
    --cc=marcandre.lureau@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=peter.maydell@linaro.org \
    --cc=qemu-devel@nongnu.org \
    --cc=stefanha@redhat.com \
    --cc=thuth@redhat.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.