From: "Christian König" <deathsimple@vodafone.de>
To: Alex Deucher <alexdeucher@gmail.com>, dri-devel@lists.freedesktop.org
Cc: Alex Deucher <alexander.deucher@amd.com>, stable@vger.kernel.org
Subject: Re: [PATCH] drm/radeon/pm: handle powered down GPUs properly in sysfs/debugfs
Date: Fri, 21 Feb 2014 16:52:24 +0100 [thread overview]
Message-ID: <53077638.7020709@vodafone.de> (raw)
In-Reply-To: <1392997806-12588-1-git-send-email-alexander.deucher@amd.com>
Am 21.02.2014 16:50, schrieb Alex Deucher:
> When we power down the dGPU on a PX system, bail if the
> user tried to adjust the power state or check the temperature.
> The GPU is powered down, so it doesn't make sense to actually
> do anything. We could power up the dGPU to complete the
> operation, but it would just be undone again as soon as it was
> completed as the card would be powered down again. Return 0
> for temperature and return -EINVAL for other interfaces.
>
> bug:
> https://bugzilla.kernel.org/show_bug.cgi?id=70651
>
> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
> Cc: stable@vger.kernel.org
Reviewed-by: Christian König <christian.koenig@amd.com>
> ---
> drivers/gpu/drm/radeon/radeon_pm.c | 29 ++++++++++++++++++++++++++---
> 1 file changed, 26 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/gpu/drm/radeon/radeon_pm.c b/drivers/gpu/drm/radeon/radeon_pm.c
> index 8e8153e..6f20bb0 100644
> --- a/drivers/gpu/drm/radeon/radeon_pm.c
> +++ b/drivers/gpu/drm/radeon/radeon_pm.c
> @@ -67,6 +67,11 @@ int radeon_pm_get_type_index(struct radeon_device *rdev,
>
> void radeon_pm_acpi_event_handler(struct radeon_device *rdev)
> {
> + struct drm_device *ddev = rdev->ddev;
> +
> + if (ddev->switch_power_state == DRM_SWITCH_POWER_OFF)
> + return;
> +
> if ((rdev->pm.pm_method == PM_METHOD_DPM) && rdev->pm.dpm_enabled) {
> mutex_lock(&rdev->pm.mutex);
> if (power_supply_is_system_supplied() > 0)
> @@ -361,6 +366,9 @@ static ssize_t radeon_set_pm_profile(struct device *dev,
> struct drm_device *ddev = dev_get_drvdata(dev);
> struct radeon_device *rdev = ddev->dev_private;
>
> + if (ddev->switch_power_state == DRM_SWITCH_POWER_OFF)
> + return -EINVAL;
> +
> mutex_lock(&rdev->pm.mutex);
> if (rdev->pm.pm_method == PM_METHOD_PROFILE) {
> if (strncmp("default", buf, strlen("default")) == 0)
> @@ -410,7 +418,8 @@ static ssize_t radeon_set_pm_method(struct device *dev,
> struct radeon_device *rdev = ddev->dev_private;
>
> /* we don't support the legacy modes with dpm */
> - if (rdev->pm.pm_method == PM_METHOD_DPM) {
> + if ((rdev->pm.pm_method == PM_METHOD_DPM) ||
> + (ddev->switch_power_state == DRM_SWITCH_POWER_OFF)) {
> count = -EINVAL;
> goto fail;
> }
> @@ -459,6 +468,11 @@ static ssize_t radeon_set_dpm_state(struct device *dev,
> struct drm_device *ddev = dev_get_drvdata(dev);
> struct radeon_device *rdev = ddev->dev_private;
>
> + if (ddev->switch_power_state == DRM_SWITCH_POWER_OFF) {
> + count = -EINVAL;
> + goto fail;
> + }
> +
> mutex_lock(&rdev->pm.mutex);
> if (strncmp("battery", buf, strlen("battery")) == 0)
> rdev->pm.dpm.user_state = POWER_STATE_TYPE_BATTERY;
> @@ -500,6 +514,9 @@ static ssize_t radeon_set_dpm_forced_performance_level(struct device *dev,
> enum radeon_dpm_forced_level level;
> int ret = 0;
>
> + if (ddev->switch_power_state == DRM_SWITCH_POWER_OFF)
> + return -EINVAL;
> +
> mutex_lock(&rdev->pm.mutex);
> if (strncmp("low", buf, strlen("low")) == 0) {
> level = RADEON_DPM_FORCED_LEVEL_LOW;
> @@ -538,9 +555,13 @@ static ssize_t radeon_hwmon_show_temp(struct device *dev,
> char *buf)
> {
> struct radeon_device *rdev = dev_get_drvdata(dev);
> + struct drm_device *ddev = rdev->ddev;
> int temp;
>
> - if (rdev->asic->pm.get_temperature)
> + /* return 0 if PX card is off */
> + if (ddev->switch_power_state == DRM_SWITCH_POWER_OFF)
> + temp = 0;
> + else if (rdev->asic->pm.get_temperature)
> temp = radeon_get_temperature(rdev);
> else
> temp = 0;
> @@ -1579,7 +1600,9 @@ static int radeon_debugfs_pm_info(struct seq_file *m, void *data)
> struct drm_device *dev = node->minor->dev;
> struct radeon_device *rdev = dev->dev_private;
>
> - if (rdev->pm.dpm_enabled) {
> + if (dev->switch_power_state == DRM_SWITCH_POWER_OFF) {
> + seq_printf(m, "Chip powered off\n");
> + } else if (rdev->pm.dpm_enabled) {
> mutex_lock(&rdev->pm.mutex);
> if (rdev->asic->dpm.debugfs_print_current_performance_level)
> radeon_dpm_debugfs_print_current_performance_level(rdev, m);
prev parent reply other threads:[~2014-02-21 15:52 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-02-21 15:50 [PATCH] drm/radeon/pm: handle powered down GPUs properly in sysfs/debugfs Alex Deucher
2014-02-21 15:52 ` Christian König [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=53077638.7020709@vodafone.de \
--to=deathsimple@vodafone.de \
--cc=alexander.deucher@amd.com \
--cc=alexdeucher@gmail.com \
--cc=dri-devel@lists.freedesktop.org \
--cc=stable@vger.kernel.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