[resend] drm/i915: Check incoming alignment for unfenced buffers (on i915gm)

[resend] drm/i915: Check incoming alignment for unfenced buffers (on i915gm)

[Chris Wilson]

In case the object has changed tiling between calls to execbuf, we need
to check if the existing offset inside the GTT matches the new tiling
constraint. We even need to do this for "unfenced" tiled objects, where
the 3D commands use an implied fence and so the object still needs to
match the physical fence restrictions on alignment (only required for
gen2 and early gen3).

In commit 2889caa92321 ("drm/i915: Eliminate lots of iterations over
the execobjects array"), the idea was to remove the second guessing and
only set the NEEDS_MAP flag when required. However, the entire check
for an unusable offset for fencing was removed and not just the
secondary check. I.e.

	/* avoid costly ping-pong once a batch bo ended up non-mappable */
        if (entry->flags & __EXEC_OBJECT_NEEDS_MAP &&
            !i915_vma_is_map_and_fenceable(vma))
                return !only_mappable_for_reloc(entry->flags);

was entirely removed as the ping-pong between execbuf passes was fixed,
but its primary purpose in forcing unaligned unfenced access to be
rebound was forgotten.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=103502
Fixes: 2889caa92321 ("drm/i915: Eliminate lots of iterations over the execobjects array")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
---
 drivers/gpu/drm/i915/i915_gem_execbuffer.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index 3d7190764f10..5b8213b1a8c7 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -343,6 +343,10 @@ eb_vma_misplaced(const struct drm_i915_gem_exec_object2 *entry,
 	    (vma->node.start + vma->node.size - 1) >> 32)
 		return true;
 
+	if (flags & __EXEC_OBJECT_NEEDS_MAP &&
+	    !i915_vma_is_map_and_fenceable(vma))
+		return true;
+
 	return false;
 }
 
-- 
2.15.0.rc2

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

✓ Fi.CI.BAT: success for drm/i915: Check incoming alignment for unfenced buffers (on i915gm) (rev2)

[Patchwork]

== Series Details ==

Series: drm/i915: Check incoming alignment for unfenced buffers (on i915gm) (rev2)
URL   : https://patchwork.freedesktop.org/series/32828/
State : success

== Summary ==

Series 32828v2 drm/i915: Check incoming alignment for unfenced buffers (on i915gm)
https://patchwork.freedesktop.org/api/1.0/series/32828/revisions/2/mbox/

Test pm_rpm:
        Subgroup basic-rte:
                skip       -> PASS       (fi-hsw-4770r) fdo#103522
Test drv_module_reload:
        Subgroup basic-reload-inject:
                incomplete -> DMESG-WARN (fi-cfl-s) fdo#103206

fdo#103522 https://bugs.freedesktop.org/show_bug.cgi?id=103522
fdo#103206 https://bugs.freedesktop.org/show_bug.cgi?id=103206

fi-bdw-5557u     total:289  pass:268  dwarn:0   dfail:0   fail:0   skip:21  time:436s
fi-bdw-gvtdvm    total:289  pass:265  dwarn:0   dfail:0   fail:0   skip:24  time:450s
fi-blb-e6850     total:289  pass:223  dwarn:1   dfail:0   fail:0   skip:65  time:374s
fi-bsw-n3050     total:289  pass:243  dwarn:0   dfail:0   fail:0   skip:46  time:526s
fi-bwr-2160      total:289  pass:183  dwarn:0   dfail:0   fail:0   skip:106 time:265s
fi-bxt-dsi       total:289  pass:259  dwarn:0   dfail:0   fail:0   skip:30  time:501s
fi-bxt-j4205     total:289  pass:260  dwarn:0   dfail:0   fail:0   skip:29  time:500s
fi-byt-j1900     total:289  pass:253  dwarn:1   dfail:0   fail:0   skip:35  time:492s
fi-byt-n2820     total:289  pass:249  dwarn:1   dfail:0   fail:0   skip:39  time:481s
fi-cfl-s         total:289  pass:253  dwarn:4   dfail:0   fail:0   skip:32  time:552s
fi-cnl-y         total:289  pass:262  dwarn:0   dfail:0   fail:0   skip:27  time:610s
fi-elk-e7500     total:289  pass:229  dwarn:0   dfail:0   fail:0   skip:60  time:423s
fi-gdg-551       total:289  pass:178  dwarn:1   dfail:0   fail:1   skip:109 time:247s
fi-glk-1         total:289  pass:261  dwarn:0   dfail:0   fail:0   skip:28  time:585s
fi-glk-dsi       total:289  pass:258  dwarn:0   dfail:0   fail:1   skip:30  time:489s
fi-hsw-4770      total:289  pass:262  dwarn:0   dfail:0   fail:0   skip:27  time:432s
fi-hsw-4770r     total:289  pass:262  dwarn:0   dfail:0   fail:0   skip:27  time:426s
fi-ilk-650       total:289  pass:228  dwarn:0   dfail:0   fail:0   skip:61  time:419s
fi-ivb-3520m     total:289  pass:260  dwarn:0   dfail:0   fail:0   skip:29  time:506s
fi-ivb-3770      total:289  pass:260  dwarn:0   dfail:0   fail:0   skip:29  time:460s
fi-kbl-7500u     total:289  pass:264  dwarn:1   dfail:0   fail:0   skip:24  time:488s
fi-kbl-7560u     total:289  pass:270  dwarn:0   dfail:0   fail:0   skip:19  time:567s
fi-kbl-7567u     total:289  pass:269  dwarn:0   dfail:0   fail:0   skip:20  time:476s
fi-kbl-r         total:289  pass:262  dwarn:0   dfail:0   fail:0   skip:27  time:581s
fi-pnv-d510      total:289  pass:222  dwarn:1   dfail:0   fail:0   skip:66  time:558s
fi-skl-6260u     total:289  pass:269  dwarn:0   dfail:0   fail:0   skip:20  time:455s
fi-skl-6600u     total:289  pass:262  dwarn:0   dfail:0   fail:0   skip:27  time:591s
fi-skl-6700hq    total:289  pass:263  dwarn:0   dfail:0   fail:0   skip:26  time:651s
fi-skl-6700k     total:289  pass:265  dwarn:0   dfail:0   fail:0   skip:24  time:529s
fi-skl-6770hq    total:289  pass:269  dwarn:0   dfail:0   fail:0   skip:20  time:498s
fi-skl-gvtdvm    total:289  pass:266  dwarn:0   dfail:0   fail:0   skip:23  time:458s
fi-snb-2520m     total:289  pass:250  dwarn:0   dfail:0   fail:0   skip:39  time:554s
fi-snb-2600      total:289  pass:249  dwarn:0   dfail:0   fail:0   skip:40  time:415s

dfe1410689e638c987b39e07288bc1951cb252f3 drm-tip: 2017y-10m-31d-09h-42m-59s UTC integration manifest
58f8a5b2b374 drm/i915: Check incoming alignment for unfenced buffers (on i915gm)

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_6275/
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

✗ Fi.CI.IGT: warning for drm/i915: Check incoming alignment for unfenced buffers (on i915gm) (rev2)

[Patchwork]

== Series Details ==

Series: drm/i915: Check incoming alignment for unfenced buffers (on i915gm) (rev2)
URL   : https://patchwork.freedesktop.org/series/32828/
State : warning

== Summary ==

Test kms_flip:
        Subgroup vblank-vs-dpms-suspend:
                pass       -> SKIP       (shard-hsw)
        Subgroup flip-vs-expired-vblank-interruptible:
                pass       -> FAIL       (shard-hsw) fdo#102887
Test kms_cursor_legacy:
        Subgroup flip-vs-cursor-legacy:
                pass       -> FAIL       (shard-hsw) fdo#102670
Test perf:
        Subgroup polling:
                pass       -> FAIL       (shard-hsw) fdo#102252
Test kms_frontbuffer_tracking:
        Subgroup fbc-modesetfrombusy:
                pass       -> SKIP       (shard-hsw)
Test drv_module_reload:
        Subgroup basic-reload:
                pass       -> DMESG-WARN (shard-hsw) fdo#102707

fdo#102887 https://bugs.freedesktop.org/show_bug.cgi?id=102887
fdo#102670 https://bugs.freedesktop.org/show_bug.cgi?id=102670
fdo#102252 https://bugs.freedesktop.org/show_bug.cgi?id=102252
fdo#102707 https://bugs.freedesktop.org/show_bug.cgi?id=102707

shard-hsw        total:2539 pass:1427 dwarn:2   dfail:0   fail:11  skip:1099 time:9398s

== Logs ==

For more details see: https://intel-gfx-ci.01.org/tree/drm-tip/Patchwork_6275/shards.html
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [resend] drm/i915: Check incoming alignment for unfenced buffers (on i915gm)

[Joonas Lahtinen]

On Tue, 2017-10-31 at 10:36 +0000, Chris Wilson wrote:
> In case the object has changed tiling between calls to execbuf, we need
> to check if the existing offset inside the GTT matches the new tiling
> constraint. We even need to do this for "unfenced" tiled objects, where
> the 3D commands use an implied fence and so the object still needs to
> match the physical fence restrictions on alignment (only required for
> gen2 and early gen3).
> 
> In commit 2889caa92321 ("drm/i915: Eliminate lots of iterations over
> the execobjects array"), the idea was to remove the second guessing and
> only set the NEEDS_MAP flag when required. However, the entire check
> for an unusable offset for fencing was removed and not just the
> secondary check. I.e.
> 
> 	/* avoid costly ping-pong once a batch bo ended up non-mappable */
>         if (entry->flags & __EXEC_OBJECT_NEEDS_MAP &&
>             !i915_vma_is_map_and_fenceable(vma))
>                 return !only_mappable_for_reloc(entry->flags);
> 
> was entirely removed as the ping-pong between execbuf passes was fixed,
> but its primary purpose in forcing unaligned unfenced access to be
> rebound was forgotten.
> 
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=103502
> Fixes: 2889caa92321 ("drm/i915: Eliminate lots of iterations over the execobjects array")
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>

Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>

Regards, Joonas
-- 
Joonas Lahtinen
Open Source Technology Center
Intel Corporation
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [resend] drm/i915: Check incoming alignment for unfenced buffers (on i915gm)

[Chris Wilson]

Quoting Joonas Lahtinen (2017-11-01 13:17:11)
> On Tue, 2017-10-31 at 10:36 +0000, Chris Wilson wrote:
> > In case the object has changed tiling between calls to execbuf, we need
> > to check if the existing offset inside the GTT matches the new tiling
> > constraint. We even need to do this for "unfenced" tiled objects, where
> > the 3D commands use an implied fence and so the object still needs to
> > match the physical fence restrictions on alignment (only required for
> > gen2 and early gen3).
> > 
> > In commit 2889caa92321 ("drm/i915: Eliminate lots of iterations over
> > the execobjects array"), the idea was to remove the second guessing and
> > only set the NEEDS_MAP flag when required. However, the entire check
> > for an unusable offset for fencing was removed and not just the
> > secondary check. I.e.
> > 
> >       /* avoid costly ping-pong once a batch bo ended up non-mappable */
> >         if (entry->flags & __EXEC_OBJECT_NEEDS_MAP &&
> >             !i915_vma_is_map_and_fenceable(vma))
> >                 return !only_mappable_for_reloc(entry->flags);
> > 
> > was entirely removed as the ping-pong between execbuf passes was fixed,
> > but its primary purpose in forcing unaligned unfenced access to be
> > rebound was forgotten.
> > 
> > Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=103502
> > Fixes: 2889caa92321 ("drm/i915: Eliminate lots of iterations over the execobjects array")
> > Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> > Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
> 
> Reviewed-by: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>

Ta. So I'd been pondering how to catch this in CI. In theory, this could
be caught by gem_render_tiled_blit, but it firstly requires checking
that the rendercopy routines use the implicit fence instructions and
secondary we need lots of repetitions with interleaved set-tiling to try
and cause an address conflict. And then we have only one machine in the
farm that is even susceptible to this bug...
-Chris
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx