[Intel-gfx] [PATCH] drm/i915/psr : Add psr1 live status
vathsala nagaraju
vathsala.nagaraju at intel.com
Mon Apr 23 08:00:30 UTC 2018
On Saturday 21 April 2018 09:30 AM, Nagaraju, Vathsala wrote:
>
> -----Original Message-----
> From: Vivi, Rodrigo
> Sent: Friday, April 20, 2018 11:06 PM
> To: Nagaraju, Vathsala <vathsala.nagaraju at intel.com>
> Cc: intel-gfx at lists.freedesktop.org; Pandiyan, Dhinakaran <dhinakaran.pandiyan at intel.com>
> Subject: Re: [PATCH] drm/i915/psr : Add psr1 live status
>
> On Fri, Apr 20, 2018 at 03:06:03PM +0530, vathsala nagaraju wrote:
>> From: Vathsala Nagaraju <vathsala.nagaraju at intel.com>
>>
>> Prints live state of psr1.Extending the existing
>> PSR2 live state function to cover psr1.
>>
>> Tested on KBL with psr2 and psr1 panel.
> Does it really work?
>
> I mean... I heard DK complaining that any read to these MMIO in some gen9 platforms were triggering the PSR exit or something like that. So, is this really reliable?
https://patchwork.freedesktop.org/patch/218153/
https://patchwork.freedesktop.org/patch/218154/ "Writes to pipe related
registers will still cause HW to exit PSR."
Reading of PSR_STATUS is not causing the exit. Confirmed by reading pipe_event before to read and after read of psr_status reg.
Test case: pause the youtube video full screen.
Currenlty for psr1 , PSR_mask (bit 16 , mask display reg write ) is unset , there is continous psr_exit.
Or it is one of those info that will misslead users to file non existent bugs?
Google used this heavily for psr2 status during video playback etc, so far no has filed any non-existent bug using this interface.
This status is useful.
if you think it will confuse the user to file more bugs, we can add exit reason from PSR_EVENT register.
>> Cc: Rodrigo Vivi <rodrigo.vivi at intel.com>
>> Cc: Dhinakaran Pandiyan <dhinakaran.pandiyan at intel.com>
>>
>> Signed-off-by: Vathsala Nagaraju <vathsala.nagaraju at intel.com>
>> ---
>> drivers/gpu/drm/i915/i915_debugfs.c | 68 ++++++++++++++++++++++++-------------
>> drivers/gpu/drm/i915/i915_reg.h | 1 +
>> 2 files changed, 45 insertions(+), 24 deletions(-)
>>
>> diff --git a/drivers/gpu/drm/i915/i915_debugfs.c
>> b/drivers/gpu/drm/i915/i915_debugfs.c
>> index e0274f4..3056f04 100644
>> --- a/drivers/gpu/drm/i915/i915_debugfs.c
>> +++ b/drivers/gpu/drm/i915/i915_debugfs.c
>> @@ -2580,25 +2580,42 @@ static int i915_guc_log_relay_release(struct inode *inode, struct file *file)
>> .release = i915_guc_log_relay_release, };
>>
>> -static const char *psr2_live_status(u32 val) -{
>> - static const char * const live_status[] = {
>> - "IDLE",
>> - "CAPTURE",
>> - "CAPTURE_FS",
>> - "SLEEP",
>> - "BUFON_FW",
>> - "ML_UP",
>> - "SU_STANDBY",
>> - "FAST_SLEEP",
>> - "DEEP_SLEEP",
>> - "BUF_ON",
>> - "TG_ON"
>> - };
>> -
>> - val = (val & EDP_PSR2_STATUS_STATE_MASK) >> EDP_PSR2_STATUS_STATE_SHIFT;
>> - if (val < ARRAY_SIZE(live_status))
>> - return live_status[val];
>> +static const char *psr_live_status(bool is_psr2_enabled, u32 val) {
>> + if (is_psr2_enabled) {
>> + static const char * const live_status[] = {
>> + "IDLE",
>> + "CAPTURE",
>> + "CAPTURE_FS",
>> + "SLEEP",
>> + "BUFON_FW",
>> + "ML_UP",
>> + "SU_STANDBY",
>> + "FAST_SLEEP",
>> + "DEEP_SLEEP",
>> + "BUF_ON",
>> + "TG_ON"
>> + };
>> + val = (val & EDP_PSR2_STATUS_STATE_MASK) >>
>> + EDP_PSR2_STATUS_STATE_SHIFT;
>> + if (val < ARRAY_SIZE(live_status))
>> + return live_status[val];
>> + } else {
>> + static const char * const live_status[] = {
>> + "IDLE",
>> + "SRDONACK",
>> + "SRDENT",
>> + "BUFOFF",
>> + "BUFON",
>> + "AUXACK",
>> + "SRDOFFACK",
>> + "SRDENT_ON",
>> + };
>> + val = (val & EDP_PSR_STATUS_STATE_MASK) >>
>> + EDP_PSR_STATUS_STATE_SHIFT;
>> + if (val < ARRAY_SIZE(live_status))
>> + return live_status[val];
>> + }
>>
>> return "unknown";
>> }
>> @@ -2611,6 +2628,7 @@ static int i915_edp_psr_status(struct seq_file *m, void *data)
>> enum pipe pipe;
>> bool enabled = false;
>> bool sink_support;
>> + u32 psr_status;
>>
>> if (!HAS_PSR(dev_priv))
>> return -ENODEV;
>> @@ -2678,12 +2696,14 @@ static int i915_edp_psr_status(struct seq_file
>> *m, void *data)
>>
>> seq_printf(m, "Performance_Counter: %u\n", psrperf);
>> }
>> - if (dev_priv->psr.psr2_enabled) {
>> - u32 psr2 = I915_READ(EDP_PSR2_STATUS);
>>
>> - seq_printf(m, "EDP_PSR2_STATUS: %x [%s]\n",
>> - psr2, psr2_live_status(psr2));
>> - }
>> + psr_status = (dev_priv->psr.psr2_enabled) ? I915_READ(EDP_PSR2_STATUS) :
>> + I915_READ(EDP_PSR_STATUS);
>> + seq_printf(m, "EDP_PSR%s_STATUS: %x [%s]\n",
>> + dev_priv->psr.psr2_enabled ? "2" : "1",
>> + psr_status,
>> + psr_live_status(dev_priv->psr.psr2_enabled, psr_status));
>> +
>> mutex_unlock(&dev_priv->psr.lock);
>>
>> intel_runtime_pm_put(dev_priv);
>> diff --git a/drivers/gpu/drm/i915/i915_reg.h
>> b/drivers/gpu/drm/i915/i915_reg.h index fb10602..c9598b4 100644
>> --- a/drivers/gpu/drm/i915/i915_reg.h
>> +++ b/drivers/gpu/drm/i915/i915_reg.h
>> @@ -4058,6 +4058,7 @@ enum {
>> #define EDP_PSR_STATUS_SENDING_TP2_TP3 (1<<8)
>> #define EDP_PSR_STATUS_SENDING_TP1 (1<<4)
>> #define EDP_PSR_STATUS_IDLE_MASK 0xf
>> +#define EDP_PSR_STATUS_STATE_SHIFT 29
>>
>> #define EDP_PSR_PERF_CNT _MMIO(dev_priv->psr_mmio_base + 0x44)
>> #define EDP_PSR_PERF_CNT_MASK 0xffffff
>> --
>> 1.9.1
>>
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx at lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/intel-gfx
More information about the Intel-gfx
mailing list