All of lore.kernel.org
 help / color / mirror / Atom feed
From: Denis Kenzior <denkenz@gmail.com>
To: ofono@ofono.org
Subject: Re: [PATCH 1/2] unit: add new test-rilmodem-sms
Date: Tue, 17 Nov 2015 21:28:20 -0600	[thread overview]
Message-ID: <564BF054.2050101@gmail.com> (raw)
In-Reply-To: <1447811116-11970-2-git-send-email-espy@canonical.com>

[-- Attachment #1: Type: text/plain, Size: 5682 bytes --]

Hi Tony,

On 11/17/2015 07:45 PM, Tony Espy wrote:
> ---
>   unit/rilmodem-test-server.c | 198 ++++++++++++++++++++++++++++++++
>   unit/rilmodem-test-server.h |  40 +++++++
>   unit/test-rilmodem-sms.c    | 267 ++++++++++++++++++++++++++++++++++++++++++++
>   3 files changed, 505 insertions(+)
>   create mode 100644 unit/rilmodem-test-server.c
>   create mode 100644 unit/rilmodem-test-server.h
>   create mode 100644 unit/test-rilmodem-sms.c
>

<snip>

> +void rilmodem_test_server_create(ConnectFunc connect,
> +				const struct rilmodem_test_data *test_data,
> +				void *data)
> +{
> +	GIOChannel *io;
> +	struct sockaddr_un addr;
> +	int retval;
> +
> +	g_assert(server_sk == 0);
> +
> +	connect_func = connect;
> +	rtd = test_data;
> +
> +	server_sk = socket(AF_UNIX, SOCK_STREAM, 0);
> +	g_assert(server_sk);
> +
> +	memset(&addr, 0, sizeof(addr));
> +	addr.sun_family = AF_UNIX;
> +	strncpy(addr.sun_path, RIL_SERVER_SOCK_PATH, sizeof(addr.sun_path) - 1);
> +
> +	/* Unlink any existing socket for this session */
> +	unlink(addr.sun_path);
> +
> +	retval = bind(server_sk, (struct sockaddr *) &addr, sizeof(addr));
> +	g_assert(retval >= 0);
> +
> +	retval = listen(server_sk, 0);
> +	g_assert(retval >= 0);
> +
> +	io = g_io_channel_unix_new(server_sk);
> +	g_assert(io != NULL);
> +
> +	g_io_channel_set_close_on_unref(io, TRUE);
> +	g_io_channel_set_flags(io, G_IO_FLAG_NONBLOCK, NULL);

Not sure you really need the non-block flag for a socket you will call 
accept on once.  But okay.

> +
> +	g_io_add_watch_full(io,	G_PRIORITY_DEFAULT,
> +				G_IO_IN | G_IO_HUP | G_IO_ERR | G_IO_NVAL,
> +				on_socket_connected, data, NULL);
> +
> +	g_io_channel_unref(io);
> +}

<snip>

> diff --git a/unit/test-rilmodem-sms.c b/unit/test-rilmodem-sms.c
> new file mode 100644
> index 0000000..0e3bff3
> --- /dev/null
> +++ b/unit/test-rilmodem-sms.c

<snip>

> +/* TODO: Did these get removed upstream? */
> +

Whats this about?

> +/* Declarations && Re-implementations of core functions. */
> +void ril_sms_exit(void);
> +void ril_sms_init(void);
> +
> +struct ofono_sms {
> +	void *driver_data;
> +};
> +
> +struct ofono_sms *ofono_sms_create(struct ofono_modem *modem,
> +					unsigned int vendor,
> +					const char *driver,
> +					void *data)
> +{
> +	struct rilmodem_sms_data *rsd = data;
> +	struct ofono_sms *sms = g_new0(struct ofono_sms, 1);
> +	int retval;
> +
> +	retval = smsdriver->probe(sms, OFONO_RIL_VENDOR_AOSP, rsd->ril);
> +	g_assert(retval == 0);
> +
> +	return sms;
> +}
> +
> +int ofono_sms_driver_register(const struct ofono_sms_driver *d)
> +{
> +	if (smsdriver == NULL)
> +		smsdriver = d;
> +
> +	return 0;
> +}
> +
> +void ofono_sms_set_data(struct ofono_sms *sms, void *data)
> +{
> +	sms->driver_data = data;
> +}
> +
> +void *ofono_sms_get_data(struct ofono_sms *sms)
> +{
> +	return sms->driver_data;
> +}
> +
> +void ofono_sms_register(struct ofono_sms *sms)
> +{
> +	;

Can we just make these empty functions instead of the ';' thing?

> +}
> +
> +void ofono_sms_driver_unregister(const struct ofono_sms_driver *d)
> +{
> +	;
> +}
> +
> +void ofono_sms_deliver_notify(struct ofono_sms *sms, const unsigned char *pdu,
> +				int len, int tpdu_len)
> +{
> +	;
> +}
> +
> +void ofono_sms_status_notify(struct ofono_sms *sms, const unsigned char *pdu,
> +				int len, int tpdu_len)
> +{
> +	;
> +}
> +
> +static void server_connect_cb(gpointer data)
> +{
> +	struct rilmodem_sms_data *rsd = data;
> +	const struct sms_data *sd = rsd->test_data;
> +
> +	/* This causes local impl of _create() to call driver's probe func. */
> +	rsd->sms = ofono_sms_create(NULL, OFONO_RIL_VENDOR_AOSP,
> +							"rilmodem", rsd);
> +
> +	/* add_idle doesn't work, read blocks main loop!!! */
> +	g_assert(sd->start_func(rsd) == FALSE);
> +}
> +
> +#if BYTE_ORDER == LITTLE_ENDIAN
> +

Can we keep the number of these #if conditions to a minimum?  Would it 
suffice to simply #ifdef this out in main.c?  Or better yet, disable 
from the build completely.

> +/*
> + * This unit test:
> + *  - does some test data setup
> + *  - configures a dummy server socket
> + *  - creates a new gril client instance
> + *    - triggers a connect to the dummy
> + *      server socket
> + *  - starts a mainloop
> + */
> +static void test_sms_func(gconstpointer data)
> +{
> +	const struct sms_data *sd = data;
> +	struct rilmodem_sms_data *rsd;
> +
> +	ril_sms_init();
> +
> +	rsd = g_new0(struct rilmodem_sms_data, 1);
> +
> +	rsd->test_data = sd;
> +
> +	rilmodem_test_server_create(&server_connect_cb, &sd->rtd, rsd);
> +
> +	rsd->ril = g_ril_new(RIL_SERVER_SOCK_PATH, OFONO_RIL_VENDOR_AOSP);
> +	g_assert(rsd->ril != NULL);
> +
> +	mainloop = g_main_loop_new(NULL, FALSE);
> +
> +	g_main_loop_run(mainloop);
> +	g_main_loop_unref(mainloop);
> +
> +	smsdriver->remove(rsd->sms);
> +	g_ril_unref(rsd->ril);
> +	g_free(rsd);
> +
> +	rilmodem_test_server_close();
> +
> +	ril_sms_exit();
> +}
> +
> +#endif
> +
> +int main(int argc, char **argv)
> +{
> +	g_test_init(&argc, &argv, NULL);
> +
> +/*
> + * As all our architectures are little-endian except for
> + * PowerPC, and the Binder wire-format differs slightly
> + * depending on endian-ness, the following guards against test
> + * failures when run on PowerPC.
> + */
> +#if BYTE_ORDER == LITTLE_ENDIAN
> +	g_test_add_data_func("/testrilmodemsms/sca_query/valid/1",
> +					&testdata_sca_query_valid_1,
> +					test_sms_func);
> +
> +#endif
> +	return g_test_run();
> +}
>

Regards,
-Denis

  reply	other threads:[~2015-11-18  3:28 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-11-18  1:45 [PATCH 0/2] New rilmodem sms test Tony Espy
2015-11-18  1:45 ` [PATCH 1/2] unit: add new test-rilmodem-sms Tony Espy
2015-11-18  3:28   ` Denis Kenzior [this message]
2015-11-18  1:45 ` [PATCH 2/2] build: add support for test-rilmodem-sms Tony Espy
2015-11-18  3:31   ` Denis Kenzior
     [not found] <1447810524-11752-1-git-send-email-espy@canonical.com>
2015-11-18  1:35 ` [PATCH 1/2] unit: add new test-rilmodem-sms Tony Espy

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=564BF054.2050101@gmail.com \
    --to=denkenz@gmail.com \
    --cc=ofono@ofono.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 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.