From: "Govindapillai, Vinod" <vinod.govindapillai@intel.com>
To: "ville.syrjala@linux.intel.com" <ville.syrjala@linux.intel.com>
Cc: "intel-xe@lists.freedesktop.org" <intel-xe@lists.freedesktop.org>,
"Saarinen, Jani" <jani.saarinen@intel.com>,
"Nikula, Jani" <jani.nikula@intel.com>,
"intel-gfx@lists.freedesktop.org"
<intel-gfx@lists.freedesktop.org>,
"Syrjala, Ville" <ville.syrjala@intel.com>
Subject: Re: [PATCH 8/8] drm/i915/debugfs: add dbuf alloc status as part of i915_ddb_info
Date: Wed, 20 Nov 2024 22:30:06 +0000 [thread overview]
Message-ID: <9b098d16eea5da3fb561964953260e929dd77608.camel@intel.com> (raw)
In-Reply-To: <Zyvx49k6xg6vA83T@intel.com>
On Thu, 2024-11-07 at 00:46 +0200, Ville Syrjälä wrote:
> On Tue, Nov 05, 2024 at 09:16:00AM +0200, Vinod Govindapillai wrote:
> > > From xe3 onwards, there is a provision to define and
> > use min ddb and interim ddb allocations for async flip
> > use case. Add the dbuf allocation status as part of
> > i915_ddb_info as well to show if min or interim ddb
> > is being used.
> >
> > Bspec: 72053
> > Signed-off-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
> > ---
> > .../drm/i915/display/intel_display_debugfs.c | 23 ++++++++++++++++---
> > .../i915/display/skl_universal_plane_regs.h | 1 +
> > 2 files changed, 21 insertions(+), 3 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/i915/display/intel_display_debugfs.c
> > b/drivers/gpu/drm/i915/display/intel_display_debugfs.c
> > index 11aff485d8fa..bce4a1ab05c0 100644
> > --- a/drivers/gpu/drm/i915/display/intel_display_debugfs.c
> > +++ b/drivers/gpu/drm/i915/display/intel_display_debugfs.c
> > @@ -39,6 +39,7 @@
> > #include "intel_pps.h"
> > #include "intel_psr.h"
> > #include "intel_psr_regs.h"
> > +#include "skl_universal_plane_regs.h"
> > #include "intel_vdsc.h"
> > #include "intel_wm.h"
> >
> > @@ -688,9 +689,24 @@ static int i915_shared_dplls_info(struct seq_file *m, void *unused)
> > return 0;
> > }
> >
> > +static u32 dbuf_alloc_status(struct intel_display *display,
> > + enum pipe pipe, enum plane_id plane_id)
> > +{
> > + u32 val = 0;
> > +
> > + if (DISPLAY_VER(display) >= 30) {
> > + u32 reg = intel_de_read(display,
> > + PLANE_MIN_BUF_CFG(pipe, plane_id));
> > + val = REG_FIELD_GET(PLANE_DBUF_ALLOC_STATUS_MASK, reg);
> > + }
> > +
> > + return val;
> > +}
> > +
> > static int i915_ddb_info(struct seq_file *m, void *unused)
> > {
> > struct drm_i915_private *dev_priv = node_to_i915(m->private);
> > + struct intel_display *display = &dev_priv->display;
> > struct skl_ddb_entry *entry;
> > struct intel_crtc *crtc;
> >
> > @@ -699,7 +715,7 @@ static int i915_ddb_info(struct seq_file *m, void *unused)
> >
> > drm_modeset_lock_all(&dev_priv->drm);
> >
> > - seq_printf(m, "%-15s%8s%8s%8s\n", "", "Start", "End", "Size");
> > + seq_printf(m, "%-15s%8s%8s%8s%16s\n", "", "Start", "End", "Size", "Alloc Status");
>
> This guy is meant ot just print the software state. The hardware might
> not even be awake here. So this doesn't belong here. It would be better
> to add it to the intel_watermark tool instead.
Okay. I was using this to verify this feature. With async flip IGT test cases, I can check this
ddb_info in a loop to see if this dbuf allocation status is being changed.
So could there be any use if this be part of a new debugfs entry?
>
> >
> > for_each_intel_crtc(&dev_priv->drm, crtc) {
> > struct intel_crtc_state *crtc_state =
> > @@ -711,9 +727,10 @@ static int i915_ddb_info(struct seq_file *m, void *unused)
> >
> > for_each_plane_id_on_crtc(crtc, plane_id) {
> > entry = &crtc_state->wm.skl.plane_ddb[plane_id];
> > - seq_printf(m, " Plane%-8d%8u%8u%8u\n", plane_id + 1,
> > + seq_printf(m, " Plane%-8d%8u%8u%8u%8u\n", plane_id + 1,
> > entry->start, entry->end,
> > - skl_ddb_entry_size(entry));
> > + skl_ddb_entry_size(entry),
> > + dbuf_alloc_status(display, pipe, plane_id));
> > }
> >
> > entry = &crtc_state->wm.skl.plane_ddb[PLANE_CURSOR];
> > diff --git a/drivers/gpu/drm/i915/display/skl_universal_plane_regs.h
> > b/drivers/gpu/drm/i915/display/skl_universal_plane_regs.h
> > index 65a5482fae60..53550356430d 100644
> > --- a/drivers/gpu/drm/i915/display/skl_universal_plane_regs.h
> > +++ b/drivers/gpu/drm/i915/display/skl_universal_plane_regs.h
> > @@ -390,6 +390,7 @@
> >
> > #define PLANE_AUTO_MIN_DBUF_EN REG_BIT(31)
> > #define PLANE_MIN_DDB_BLOCKS_MASK REG_GENMASK(27, 16)
> > +#define PLANE_DBUF_ALLOC_STATUS_MASK REG_GENMASK(15, 14)
>
> Do these bits actually mean something?
This gives the dbuf allocation status. Whenever HW uses the min_ddb configuration this bit is
updated and I had verified this.
BR
Vinod
>
> > #define PLANE_INTERIM_DDB_BLOCKS_MASK REG_GENMASK(11, 0)
> >
> > /* tgl+ */
> > --
> > 2.34.1
>
next prev parent reply other threads:[~2024-11-20 22:30 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-11-05 7:15 [PATCH 0/8] use hw support for min/interim ddb allocation for async flip Vinod Govindapillai
2024-11-05 7:15 ` [PATCH 1/8] drm/i915/display: update intel_enabled_dbuf_slices_mask to use intel_display Vinod Govindapillai
2024-11-05 8:52 ` Jani Nikula
2024-11-05 9:07 ` Govindapillai, Vinod
2024-11-05 7:15 ` [PATCH 2/8] drm/i9i5/display: use intel_display in intel_de_read calls of skl_watermark.c Vinod Govindapillai
2024-11-05 8:58 ` Jani Nikula
2024-11-05 9:03 ` Jani Nikula
2024-11-05 7:15 ` [PATCH 3/8] drm/i915/display: update use_minimal_wm0_only to use intel_display Vinod Govindapillai
2024-11-05 9:08 ` Jani Nikula
2024-11-06 14:06 ` Ville Syrjälä
2024-11-06 15:49 ` Govindapillai, Vinod
2024-11-05 7:15 ` [PATCH 4/8] drm/i915/display: update use_min_ddb " Vinod Govindapillai
2024-11-05 7:15 ` [PATCH 5/8] drm/i915/display: update skl_plane_wm_equals " Vinod Govindapillai
2024-11-05 9:09 ` Jani Nikula
2024-11-05 7:15 ` [PATCH 6/8] drm/i915/display: update to plane_wm register access function Vinod Govindapillai
2024-11-05 7:15 ` [PATCH 7/8] drm/i915/xe3: Use hw support for min/interim ddb allocations for async flip Vinod Govindapillai
2024-11-06 22:45 ` Ville Syrjälä
2024-11-20 22:26 ` Govindapillai, Vinod
2024-11-05 7:16 ` [PATCH 8/8] drm/i915/debugfs: add dbuf alloc status as part of i915_ddb_info Vinod Govindapillai
2024-11-06 22:46 ` Ville Syrjälä
2024-11-20 22:30 ` Govindapillai, Vinod [this message]
2024-11-05 8:37 ` ✓ CI.Patch_applied: success for use hw support for min/interim ddb allocation for async flip Patchwork
2024-11-05 8:37 ` ✗ CI.checkpatch: warning " Patchwork
2024-11-05 8:38 ` ✓ CI.KUnit: success " Patchwork
2024-11-05 8:50 ` ✓ CI.Build: " Patchwork
2024-11-05 8:52 ` ✓ CI.Hooks: " Patchwork
2024-11-05 8:54 ` ✗ CI.checksparse: warning " Patchwork
2024-11-05 9:15 ` ✓ CI.BAT: success " Patchwork
2024-11-06 8:53 ` ✗ CI.FULL: failure " Patchwork
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=9b098d16eea5da3fb561964953260e929dd77608.camel@intel.com \
--to=vinod.govindapillai@intel.com \
--cc=intel-gfx@lists.freedesktop.org \
--cc=intel-xe@lists.freedesktop.org \
--cc=jani.nikula@intel.com \
--cc=jani.saarinen@intel.com \
--cc=ville.syrjala@intel.com \
--cc=ville.syrjala@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox