From: Lyude Paul <lyude@redhat.com>
To: Paul Kocialkowski <paul.kocialkowski@linux.intel.com>,
intel-gfx@lists.freedesktop.org
Subject: Re: [PATCH i-g-t v4 1/3] tests/chamelium: Update connector state without reprobe when possible
Date: Thu, 06 Jul 2017 12:02:08 -0400 [thread overview]
Message-ID: <1499356928.20388.17.camel@redhat.com> (raw)
In-Reply-To: <20170704133319.5130-1-paul.kocialkowski@linux.intel.com>
On Tue, 2017-07-04 at 16:33 +0300, Paul Kocialkowski wrote:
> This renames the reprobe_connector function to update_connector and
> ensures that full reprobe of the connector is only done when really
> necessary (e.g. when changing the EDID).
>
> A full reprobe takes time and is not required for updating the
> connector
> state. Thus, this allows executing tests faster.
After thinking on this for a while I still have to NAK it unfortunately
:(. While this stuff will work for i915 since it's not really following
the expected DRM HPD behavior there's not much guarantee it'll work for
anything else. I think we need to have a more well-defined hotplug
event model in DRM before we consider adding patches like this.
>
> Signed-off-by: Paul Kocialkowski <paul.kocialkowski@linux.intel.com>
> ---
> tests/chamelium.c | 34 ++++++++++++++++++----------------
> 1 file changed, 18 insertions(+), 16 deletions(-)
>
> diff --git a/tests/chamelium.c b/tests/chamelium.c
> index b412c6a7..346018a5 100644
> --- a/tests/chamelium.c
> +++ b/tests/chamelium.c
> @@ -103,13 +103,14 @@ require_connector_present(data_t *data,
> unsigned int type)
> }
>
> static drmModeConnection
> -reprobe_connector(data_t *data, struct chamelium_port *port)
> +update_connector(data_t *data, struct chamelium_port *port, bool
> reprobe)
> {
> drmModeConnector *connector;
> drmModeConnection status;
>
> - igt_debug("Reprobing %s...\n",
> chamelium_port_get_name(port));
> - connector = chamelium_port_get_connector(data->chamelium,
> port, true);
> + igt_debug("Updating %s...\n",
> chamelium_port_get_name(port));
> + connector = chamelium_port_get_connector(data->chamelium,
> port,
> + reprobe);
> igt_assert(connector);
> status = connector->connection;
>
> @@ -119,7 +120,7 @@ reprobe_connector(data_t *data, struct
> chamelium_port *port)
>
> static void
> wait_for_connector(data_t *data, struct chamelium_port *port,
> - drmModeConnection status)
> + drmModeConnection status, bool reprobe)
> {
> bool finished = false;
>
> @@ -132,7 +133,7 @@ wait_for_connector(data_t *data, struct
> chamelium_port *port,
> * that hpd events work in the event that hpd doesn't work
> on the system
> */
> igt_until_timeout(HOTPLUG_TIMEOUT) {
> - if (reprobe_connector(data, port) == status) {
> + if (update_connector(data, port, reprobe) == status)
> {
> finished = true;
> return;
> }
> @@ -151,11 +152,12 @@ reset_state(data_t *data, struct chamelium_port
> *port)
> chamelium_reset(data->chamelium);
>
> if (port) {
> - wait_for_connector(data, port,
> DRM_MODE_DISCONNECTED);
> + wait_for_connector(data, port,
> DRM_MODE_DISCONNECTED, false);
> } else {
> for (p = 0; p < data->port_count; p++) {
> port = data->ports[p];
> - wait_for_connector(data, port,
> DRM_MODE_DISCONNECTED);
> + wait_for_connector(data, port,
> DRM_MODE_DISCONNECTED,
> + false);
> }
> }
> }
> @@ -175,7 +177,7 @@ test_basic_hotplug(data_t *data, struct
> chamelium_port *port, int toggle_count)
> /* Check if we get a sysfs hotplug event */
> chamelium_plug(data->chamelium, port);
> igt_assert(igt_hotplug_detected(mon,
> HOTPLUG_TIMEOUT));
> - igt_assert_eq(reprobe_connector(data, port),
> + igt_assert_eq(update_connector(data, port, false),
> DRM_MODE_CONNECTED);
>
> igt_flush_hotplugs(mon);
> @@ -183,7 +185,7 @@ test_basic_hotplug(data_t *data, struct
> chamelium_port *port, int toggle_count)
> /* Now check if we get a hotplug from disconnection
> */
> chamelium_unplug(data->chamelium, port);
> igt_assert(igt_hotplug_detected(mon,
> HOTPLUG_TIMEOUT));
> - igt_assert_eq(reprobe_connector(data, port),
> + igt_assert_eq(update_connector(data, port, false),
> DRM_MODE_DISCONNECTED);
> }
>
> @@ -204,7 +206,7 @@ test_edid_read(data_t *data, struct
> chamelium_port *port,
>
> chamelium_port_set_edid(data->chamelium, port, edid_id);
> chamelium_plug(data->chamelium, port);
> - wait_for_connector(data, port, DRM_MODE_CONNECTED);
> + wait_for_connector(data, port, DRM_MODE_CONNECTED, true);
>
> igt_assert(kmstest_get_property(data->drm_fd, connector-
> >connector_id,
> DRM_MODE_OBJECT_CONNECTOR,
> "EDID", NULL,
> @@ -247,13 +249,13 @@ try_suspend_resume_hpd(data_t *data, struct
> chamelium_port *port,
>
> igt_assert(igt_hotplug_detected(mon, HOTPLUG_TIMEOUT));
> if (port) {
> - igt_assert_eq(reprobe_connector(data, port),
> connected ?
> + igt_assert_eq(update_connector(data, port, false),
> connected ?
> DRM_MODE_DISCONNECTED :
> DRM_MODE_CONNECTED);
> } else {
> for (p = 0; p < data->port_count; p++) {
> port = data->ports[p];
> - igt_assert_eq(reprobe_connector(data, port),
> connected ?
> - DRM_MODE_DISCONNECTED :
> + igt_assert_eq(update_connector(data, port,
> false),
> + connected ?
> DRM_MODE_DISCONNECTED :
> DRM_MODE_CONNECTED);
> }
>
> @@ -317,7 +319,7 @@ test_suspend_resume_edid_change(data_t *data,
> struct chamelium_port *port,
> /* First plug in the port */
> chamelium_port_set_edid(data->chamelium, port, edid_id);
> chamelium_plug(data->chamelium, port);
> - wait_for_connector(data, port, DRM_MODE_CONNECTED);
> + wait_for_connector(data, port, DRM_MODE_CONNECTED, true);
>
> igt_flush_hotplugs(mon);
>
> @@ -352,7 +354,7 @@ prepare_output(data_t *data,
> chamelium_port_set_edid(data->chamelium, port, data-
> >edid_id);
>
> chamelium_plug(data->chamelium, port);
> - wait_for_connector(data, port, DRM_MODE_CONNECTED);
> + wait_for_connector(data, port, DRM_MODE_CONNECTED, true);
>
> igt_display_init(display, data->drm_fd);
> output = igt_output_from_connector(display, connector);
> @@ -590,7 +592,7 @@ test_hpd_without_ddc(data_t *data, struct
> chamelium_port *port)
> chamelium_plug(data->chamelium, port);
>
> igt_assert(igt_hotplug_detected(mon, HOTPLUG_TIMEOUT));
> - igt_assert_eq(reprobe_connector(data, port),
> DRM_MODE_CONNECTED);
> + igt_assert_eq(update_connector(data, port, false),
> DRM_MODE_CONNECTED);
>
> igt_cleanup_hotplug(mon);
> }
--
Cheers,
Lyude
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
prev parent reply other threads:[~2017-07-06 16:02 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-07-04 13:33 [PATCH i-g-t v4 1/3] tests/chamelium: Update connector state without reprobe when possible Paul Kocialkowski
2017-07-04 13:33 ` [PATCH i-g-t v4 2/3] chamelium: Remove init reset duplicate in favor of per-test reset Paul Kocialkowski
2017-07-06 15:45 ` Lyude Paul
2017-07-04 13:33 ` [PATCH i-g-t v4 3/3] Make igtrc configuration common, with configurable suspend/resume delay Paul Kocialkowski
2017-07-06 15:57 ` Lyude Paul
2017-07-06 16:02 ` Lyude Paul [this message]
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=1499356928.20388.17.camel@redhat.com \
--to=lyude@redhat.com \
--cc=intel-gfx@lists.freedesktop.org \
--cc=paul.kocialkowski@linux.intel.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.