Attachment 85158 Details for Bug 68171 – Hold a reference whilst shrinking the objects

[patch] Hold a reference whilst shrinking the objects

0001-drm-i915-Hold-an-object-reference-whilst-we-shrink-i.patch (text/plain), 8.68 KB, created by Chris Wilson on 2013-09-04 00:02:30 UTC

(hide)

Description:

Filename:

MIME Type:

Creator: Chris Wilson

Created: 2013-09-04 00:02:30 UTC

Size: 8.68 KB

patch

obsolete

>From 5d1d5231bbc73f2ad0600727fa83619402f75800 Mon Sep 17 00:00:00 2001
>From: Chris Wilson <chris@chris-wilson.co.uk>
>Date: Wed, 4 Sep 2013 00:17:36 +0100
>Subject: [PATCH] drm/i915: Hold an object reference whilst we shrink it
>
>Whilst running the shrinker, we need to hold a reference as we unbind
>the objects, or else we may end up waiting for and retiring requests,
>which in turn may result in this object being freed.
>
>This is very similar to the eviction code which also has to be very
>careful to keep a reference to its objects as it retires and unbinds
>them.
>
>Another similarity, that Ben pointed out, is that as we may call
>retire-requests, the unbound_list is outside of our control. We must
>only process a single element of that list at a time, that is we can not
>rely on the "safe" next pointer being valid after a call to
>i915_vma_unbind().
>
>Sep  3 00:12:19 x-hswu33 kernel: [22027.290905] BUG: unable to handle kernel NULL pointer dereference at 0000000000000008
>Sep  3 00:12:19 x-hswu33 kernel: [22027.290971] IP: [<ffffffffa0082892>] i915_gem_gtt_finish_object+0x68/0xbd [i915]
>Sep  3 00:12:19 x-hswu33 kernel: [22027.291047] PGD 758d3067 PUD ac0d6067 PMD 0
>Sep  3 00:12:19 x-hswu33 kernel: [22027.291085] Oops: 0000 [#1] SMP
>Sep  3 00:12:19 x-hswu33 kernel: [22027.291110] Modules linked in: dm_mod snd_hda_codec_realtek iTCO_wdt iTCO_vendor_support pcspkr snd_hda_intel i2c_i801 snd_hda_codec snd_hwdep snd_pcm snd_page_alloc snd_timer snd lpc_ich mfd_core soundcore battery ac option usb_wwan usbserial uvcvideo videobuf2_vmalloc videobuf2_memops videobuf2_core videodev i915 video button drm_kms_helper drm acpi_cpufreq mperf freq_table
>Sep  3 00:12:19 x-hswu33 kernel: [22027.291392] CPU: 1 PID: 16835 Comm: fbo-maxsize Not tainted 3.11.0-rc7_nightlytop_8fdad4_20130902_+ #7977
>Sep  3 00:12:19 x-hswu33 kernel: [22027.291460] task: ffff8800712106d0 ti: ffff880028e4a000 task.ti: ffff880028e4a000
>Sep  3 00:12:19 x-hswu33 kernel: [22027.291518] RIP: 0010:[<ffffffffa0082892>]  [<ffffffffa0082892>] i915_gem_gtt_finish_object+0x68/0xbd [i915]
>Sep  3 00:12:19 x-hswu33 kernel: [22027.291602] RSP: 0018:ffff880028e4b9e8  EFLAGS: 00010246
>Sep  3 00:12:19 x-hswu33 kernel: [22027.291640] RAX: 0000000000000000 RBX: ffff880145734000 RCX: ffff880145735328
>Sep  3 00:12:19 x-hswu33 kernel: [22027.291689] RDX: ffff8801457353fc RSI: 0000000000000000 RDI: ffff88007597cc00
>Sep  3 00:12:19 x-hswu33 kernel: [22027.291738] RBP: ffff88007597cc00 R08: 0000000000000001 R09: ffff88014f257f00
>Sep  3 00:12:19 x-hswu33 kernel: [22027.291787] R10: ffffea0001d65f00 R11: 0000000000bba60b R12: ffff880149e5b000
>Sep  3 00:12:19 x-hswu33 kernel: [22027.291837] R13: ffff880145734001 R14: ffff88007597ccc8 R15: ffff88007597cc00
>Sep  3 00:12:19 x-hswu33 kernel: [22027.291887] FS:  00007ff5bc919740(0000) GS:ffff88014f240000(0000) knlGS:0000000000000000
>Sep  3 00:12:19 x-hswu33 kernel: [22027.291943] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>Sep  3 00:12:19 x-hswu33 kernel: [22027.291983] CR2: 0000000000000008 CR3: 0000000028f4c000 CR4: 00000000001407e0
>Sep  3 00:12:19 x-hswu33 kernel: [22027.292032] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
>Sep  3 00:12:19 x-hswu33 kernel: [22027.292081] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
>Sep  3 00:12:19 x-hswu33 kernel: [22027.292130] Stack:
>Sep  3 00:12:19 x-hswu33 kernel: [22027.292145]  0000000000000000 ffff88007597cc00 ffff8801440d6840 0000000000000000
>Sep  3 00:12:19 x-hswu33 kernel: [22027.292202]  ffff880145734000 ffffffffa007c854 0000000000000010 ffff88007597c900
>Sep  3 00:12:19 x-hswu33 kernel: [22027.292261]  0000000000018000 00000000004a1201 ffff88007597cc60 ffffffffa007d183
>Sep  3 00:12:19 x-hswu33 kernel: [22027.292317] Call Trace:
>Sep  3 00:12:19 x-hswu33 kernel: [22027.292350]  [<ffffffffa007c854>] ? i915_vma_unbind+0xe2/0x1d1 [i915]
>Sep  3 00:12:19 x-hswu33 kernel: [22027.292410]  [<ffffffffa007d183>] ? __i915_gem_shrink+0xf1/0x162 [i915]
>Sep  3 00:12:19 x-hswu33 kernel: [22027.292464]  [<ffffffffa007d2ee>] ? i915_gem_object_get_pages_gtt+0xfa/0x303 [i915]
>Sep  3 00:12:19 x-hswu33 kernel: [22027.292527]  [<ffffffffa00795f4>] ? i915_gem_object_get_pages+0x54/0x89 [i915]
>Sep  3 00:12:19 x-hswu33 kernel: [22027.292586]  [<ffffffffa007cbda>] ? i915_gem_object_pin+0x238/0x5ce [i915]
>Sep  3 00:12:19 x-hswu33 kernel: [22027.292638]  [<ffffffff812cba5f>] ? __sg_page_iter_next+0x2b/0x58
>Sep  3 00:12:19 x-hswu33 kernel: [22027.292694]  [<ffffffffa0082056>] ? gen6_ppgtt_insert_entries+0xf2/0x114 [i915]
>Sep  3 00:12:19 x-hswu33 kernel: [22027.292754]  [<ffffffffa007fe4b>] ? i915_gem_execbuffer_reserve_vma.isra.13+0x79/0x18d [i915]
>Sep  3 00:12:19 x-hswu33 kernel: [22027.292820]  [<ffffffffa008017c>] ? i915_gem_execbuffer_reserve+0x21d/0x347 [i915]
>Sep  3 00:12:19 x-hswu33 kernel: [22027.292881]  [<ffffffffa0080bfb>] ? i915_gem_do_execbuffer.isra.17+0x4f3/0xe61 [i915]
>Sep  3 00:12:19 x-hswu33 kernel: [22027.292943]  [<ffffffffa00795f4>] ? i915_gem_object_get_pages+0x54/0x89 [i915]
>Sep  3 00:12:19 x-hswu33 kernel: [22027.293002]  [<ffffffffa007e405>] ? i915_gem_pwrite_ioctl+0x743/0x7a5 [i915]
>Sep  3 00:12:19 x-hswu33 kernel: [22027.293060]  [<ffffffffa0081a46>] ? i915_gem_execbuffer2+0x15e/0x1e4 [i915]
>Sep  3 00:12:19 x-hswu33 kernel: [22027.293123]  [<ffffffffa000e20d>] ? drm_ioctl+0x2a5/0x3c4 [drm]
>Sep  3 00:12:19 x-hswu33 kernel: [22027.293173]  [<ffffffffa00818e8>] ? i915_gem_execbuffer+0x37f/0x37f [i915]
>Sep  3 00:12:19 x-hswu33 kernel: [22027.293224]  [<ffffffff816f64c0>] ? __do_page_fault+0x3ab/0x449
>Sep  3 00:12:19 x-hswu33 kernel: [22027.293269]  [<ffffffff810be3da>] ? do_mmap_pgoff+0x2b2/0x341
>Sep  3 00:12:19 x-hswu33 kernel: [22027.293317]  [<ffffffff810e49be>] ? vfs_ioctl+0x1e/0x31
>Sep  3 00:12:19 x-hswu33 kernel: [22027.293354]  [<ffffffff810e5194>] ? do_vfs_ioctl+0x3ad/0x3ef
>Sep  3 00:12:19 x-hswu33 kernel: [22027.293396]  [<ffffffff810e5224>] ? SyS_ioctl+0x4e/0x7e
>Sep  3 00:12:19 x-hswu33 kernel: [22027.293435]  [<ffffffff816f88d2>] ? system_call_fastpath+0x16/0x1b
>Sep  3 00:12:19 x-hswu33 kernel: [22027.293478] Code: 52 0c a0 48 c7 c6 22 30 0d a0 31 c0 e8 ef 00 f9 ff bf c6 a7 00 00 e8 90 5d 24 e1 f6 85 13 01 00 00 10 75 44 48 8b 85 18 01 00 00 <8b> 50 08 48 8b 30 49 8b 84 24 88 02 00 00 48 89 c7 48 81 c7 98
>Sep  3 00:12:19 x-hswu33 kernel: [22027.293678] RIP  [<ffffffffa0082892>] i915_gem_gtt_finish_object+0x68/0xbd [i915]
>Sep  3 00:12:19 x-hswu33 kernel: [22027.293746]  RSP <ffff880028e4b9e8>
>Sep  3 00:12:19 x-hswu33 kernel: [22027.293773] CR2: 0000000000000008
>
>Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=68171
>Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
>---
> drivers/gpu/drm/i915/i915_gem.c | 36 +++++++++++++++++++++++++++++-------
> 1 file changed, 29 insertions(+), 7 deletions(-)
>
>diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c
>index 8e414cf..d8ef227 100644
>--- a/drivers/gpu/drm/i915/i915_gem.c
>+++ b/drivers/gpu/drm/i915/i915_gem.c
>@@ -1839,6 +1839,7 @@ __i915_gem_shrink(struct drm_i915_private *dev_priv,
> {
> 	struct drm_i915_gem_object *obj, *next;
> 	unsigned long count = 0;
>+	struct list_head still_bound_list;
> 
> 	lockdep_assert_held(&dev_priv->dev->struct_mutex);
> 
>@@ -1853,23 +1854,44 @@ __i915_gem_shrink(struct drm_i915_private *dev_priv,
> 		}
> 	}
> 
>-	list_for_each_entry_safe(obj, next, &dev_priv->mm.bound_list,
>-				 global_list) {
>+	/* As we may completely rewrite the bound list whilst unbinding
>+	 * (due to retiring requests) we have to strictly process only
>+	 * one element of the list at the time, and recheck the list
>+	 * on every iteration.
>+	 */
>+	INIT_LIST_HEAD(&still_bound_list);
>+	while (count < target && !list_empty(&dev_priv->mm.bound_list)) {
> 		struct i915_vma *vma, *v;
> 
>-		if (!i915_gem_object_is_purgeable(obj) && purgeable_only)
>+		obj = list_first_entry(&dev_priv->mm.bound_list,
>+				       typeof(*obj), global_list);
>+
>+		if (!i915_gem_object_is_purgeable(obj) && purgeable_only) {
>+			list_move_tail(&obj->global_list, &still_bound_list);
> 			continue;
>+		}
>+
>+		/* Hold a reference whilst we unbind this object, as we may
>+		 * end up waiting for and retiring requests, which may result
>+		 * in this object being freed.
>+		 *
>+		 * Note that the shrinker and eviction is special as they operate
>+		 * on the inactive lists which reference limbo objects.
>+		 */
>+		drm_gem_object_reference(&obj->base);
> 
> 		list_for_each_entry_safe(vma, v, &obj->vma_list, vma_link)
> 			if (i915_vma_unbind(vma))
> 				break;
> 
>-		if (!i915_gem_object_put_pages(obj)) {
>+		if (i915_gem_object_put_pages(obj) == 0)
> 			count += obj->base.size >> PAGE_SHIFT;
>-			if (count >= target)
>-				return count;
>-		}
>+		else
>+			list_move_tail(&obj->global_list, &still_bound_list);
>+
>+		drm_gem_object_unreference(&obj->base);
> 	}
>+	list_splice(&still_bound_list, &dev_priv->mm.bound_list);
> 
> 	return count;
> }
>-- 
>1.8.4.rc3
>

Actions: View | Splinter Review

Attachments on bug 68171: 84120 | 84430 | 84431 | 84444 | 85095 | 85141 | 85148 | 85149 | 85153 | 85158 | 85194

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.