public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org, Dave Airlie <airlied@redhat.com>,
	Chris Wilson <chris@chris-wilson.co.uk>,
	Tvrtko Ursulin <tvrtko.ursulin@intel.com>,
	Rodrigo Vivi <rodrigo.vivi@intel.com>
Subject: [PATCH 5.6 06/73] drm/i915/gem: Hold obj->vma.lock over for_each_ggtt_vma()
Date: Mon,  4 May 2020 19:57:09 +0200	[thread overview]
Message-ID: <20200504165503.088299523@linuxfoundation.org> (raw)
In-Reply-To: <20200504165501.781878940@linuxfoundation.org>

From: Chris Wilson <chris@chris-wilson.co.uk>

commit f524a774a4ff702bdfbfc094c9dd463ee623252b upstream.

While the ggtt vma are protected by their object lifetime, the list
continues until it hits a non-ggtt vma, and that vma is not protected
and may be freed as we inspect it. Hence, we require the obj->vma.lock
to protect the list as we iterate.

An example of forgetting to hold the obj->vma.lock is

[1642834.464973] general protection fault, probably for non-canonical address 0xdead000000000122: 0000 [#1] SMP PTI
[1642834.464977] CPU: 3 PID: 1954 Comm: Xorg Not tainted 5.6.0-300.fc32.x86_64 #1
[1642834.464979] Hardware name: LENOVO 20ARS25701/20ARS25701, BIOS GJET94WW (2.44 ) 09/14/2017
[1642834.465021] RIP: 0010:i915_gem_object_set_tiling+0x2c0/0x3e0 [i915]
[1642834.465024] Code: 8b 84 24 18 01 00 00 f6 c4 80 74 59 49 8b 94 24 a0 00 00 00 49 8b 84 24 e0 00 00 00 49 8b 74 24 10 48 8b 92 30 01 00 00 89 c7 <80> ba 0a 06 00 00 03 0f 87 86 00 00 00 ba 00 00 08 00 b9 00 00 10
[1642834.465025] RSP: 0018:ffffa98780c77d60 EFLAGS: 00010282
[1642834.465028] RAX: ffff8d232bfb2578 RBX: 0000000000000002 RCX: ffff8d25873a0000
[1642834.465029] RDX: dead000000000122 RSI: fffff0af8ac6e408 RDI: 000000002bfb2578
[1642834.465030] RBP: ffff8d25873a0000 R08: ffff8d252bfb5638 R09: 0000000000000000
[1642834.465031] R10: 0000000000000000 R11: ffff8d252bfb5640 R12: ffffa987801cb8f8
[1642834.465032] R13: 0000000000001000 R14: ffff8d233e972e50 R15: ffff8d233e972d00
[1642834.465034] FS:  00007f6a3d327f00(0000) GS:ffff8d25926c0000(0000) knlGS:0000000000000000
[1642834.465036] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[1642834.465037] CR2: 00007f6a2064d000 CR3: 00000002fb57c001 CR4: 00000000001606e0
[1642834.465038] Call Trace:
[1642834.465083]  i915_gem_set_tiling_ioctl+0x122/0x230 [i915]
[1642834.465121]  ? i915_gem_object_set_tiling+0x3e0/0x3e0 [i915]
[1642834.465151]  drm_ioctl_kernel+0x86/0xd0 [drm]
[1642834.465156]  ? avc_has_perm+0x3b/0x160
[1642834.465178]  drm_ioctl+0x206/0x390 [drm]
[1642834.465216]  ? i915_gem_object_set_tiling+0x3e0/0x3e0 [i915]
[1642834.465221]  ? selinux_file_ioctl+0x122/0x1c0
[1642834.465226]  ? __do_munmap+0x24b/0x4d0
[1642834.465231]  ksys_ioctl+0x82/0xc0
[1642834.465235]  __x64_sys_ioctl+0x16/0x20
[1642834.465238]  do_syscall_64+0x5b/0xf0
[1642834.465243]  entry_SYSCALL_64_after_hwframe+0x44/0xa9
[1642834.465245] RIP: 0033:0x7f6a3d7b047b
[1642834.465247] Code: 0f 1e fa 48 8b 05 1d aa 0c 00 64 c7 00 26 00 00 00 48 c7 c0 ff ff ff ff c3 66 0f 1f 44 00 00 f3 0f 1e fa b8 10 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d ed a9 0c 00 f7 d8 64 89 01 48
[1642834.465249] RSP: 002b:00007ffe71adba28 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
[1642834.465251] RAX: ffffffffffffffda RBX: 000055f99048fa40 RCX: 00007f6a3d7b047b
[1642834.465253] RDX: 00007ffe71adba30 RSI: 00000000c0106461 RDI: 000000000000000e
[1642834.465254] RBP: 0000000000000002 R08: 000055f98f3f1798 R09: 0000000000000002
[1642834.465255] R10: 0000000000001000 R11: 0000000000000246 R12: 0000000000000080
[1642834.465257] R13: 000055f98f3f1690 R14: 00000000c0106461 R15: 00007ffe71adba30

Now to take the spinlock during the list iteration, we need to break it
down into two phases. In the first phase under the lock, we cannot sleep
and so must defer the actual work to a second list, protected by the
ggtt->mutex.

We also need to hold the spinlock during creation of a new vma to
serialise with updates of the tiling on the object.

Reported-by: Dave Airlie <airlied@redhat.com>
Fixes: 2850748ef876 ("drm/i915: Pull i915_vma_pin under the vm->mutex")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Dave Airlie <airlied@redhat.com>
Cc: <stable@vger.kernel.org> # v5.5+
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20200422072805.17340-1-chris@chris-wilson.co.uk
(cherry picked from commit cb593e5d2b6d3ad489669914d9fd1c64c7a4a6af)
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
 drivers/gpu/drm/i915/gem/i915_gem_tiling.c |   20 ++++++++++++++++++--
 drivers/gpu/drm/i915/i915_vma.c            |   10 ++++++----
 2 files changed, 24 insertions(+), 6 deletions(-)

--- a/drivers/gpu/drm/i915/gem/i915_gem_tiling.c
+++ b/drivers/gpu/drm/i915/gem/i915_gem_tiling.c
@@ -183,21 +183,35 @@ i915_gem_object_fence_prepare(struct drm
 			      int tiling_mode, unsigned int stride)
 {
 	struct i915_ggtt *ggtt = &to_i915(obj->base.dev)->ggtt;
-	struct i915_vma *vma;
+	struct i915_vma *vma, *vn;
+	LIST_HEAD(unbind);
 	int ret = 0;
 
 	if (tiling_mode == I915_TILING_NONE)
 		return 0;
 
 	mutex_lock(&ggtt->vm.mutex);
+
+	spin_lock(&obj->vma.lock);
 	for_each_ggtt_vma(vma, obj) {
+		GEM_BUG_ON(vma->vm != &ggtt->vm);
+
 		if (i915_vma_fence_prepare(vma, tiling_mode, stride))
 			continue;
 
+		list_move(&vma->vm_link, &unbind);
+	}
+	spin_unlock(&obj->vma.lock);
+
+	list_for_each_entry_safe(vma, vn, &unbind, vm_link) {
 		ret = __i915_vma_unbind(vma);
-		if (ret)
+		if (ret) {
+			/* Restore the remaining vma on an error */
+			list_splice(&unbind, &ggtt->vm.bound_list);
 			break;
+		}
 	}
+
 	mutex_unlock(&ggtt->vm.mutex);
 
 	return ret;
@@ -269,6 +283,7 @@ i915_gem_object_set_tiling(struct drm_i9
 	}
 	mutex_unlock(&obj->mm.lock);
 
+	spin_lock(&obj->vma.lock);
 	for_each_ggtt_vma(vma, obj) {
 		vma->fence_size =
 			i915_gem_fence_size(i915, vma->size, tiling, stride);
@@ -279,6 +294,7 @@ i915_gem_object_set_tiling(struct drm_i9
 		if (vma->fence)
 			vma->fence->dirty = true;
 	}
+	spin_unlock(&obj->vma.lock);
 
 	obj->tiling_and_stride = tiling | stride;
 	i915_gem_object_unlock(obj);
--- a/drivers/gpu/drm/i915/i915_vma.c
+++ b/drivers/gpu/drm/i915/i915_vma.c
@@ -158,16 +158,18 @@ vma_create(struct drm_i915_gem_object *o
 
 	GEM_BUG_ON(!IS_ALIGNED(vma->size, I915_GTT_PAGE_SIZE));
 
+	spin_lock(&obj->vma.lock);
+
 	if (i915_is_ggtt(vm)) {
 		if (unlikely(overflows_type(vma->size, u32)))
-			goto err_vma;
+			goto err_unlock;
 
 		vma->fence_size = i915_gem_fence_size(vm->i915, vma->size,
 						      i915_gem_object_get_tiling(obj),
 						      i915_gem_object_get_stride(obj));
 		if (unlikely(vma->fence_size < vma->size || /* overflow */
 			     vma->fence_size > vm->total))
-			goto err_vma;
+			goto err_unlock;
 
 		GEM_BUG_ON(!IS_ALIGNED(vma->fence_size, I915_GTT_MIN_ALIGNMENT));
 
@@ -179,8 +181,6 @@ vma_create(struct drm_i915_gem_object *o
 		__set_bit(I915_VMA_GGTT_BIT, __i915_vma_flags(vma));
 	}
 
-	spin_lock(&obj->vma.lock);
-
 	rb = NULL;
 	p = &obj->vma.tree.rb_node;
 	while (*p) {
@@ -225,6 +225,8 @@ vma_create(struct drm_i915_gem_object *o
 
 	return vma;
 
+err_unlock:
+	spin_unlock(&obj->vma.lock);
 err_vma:
 	i915_vma_free(vma);
 	return ERR_PTR(-E2BIG);



  parent reply	other threads:[~2020-05-04 18:06 UTC|newest]

Thread overview: 87+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-05-04 17:57 [PATCH 5.6 00/73] 5.6.11-rc1 review Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 01/73] drm/scheduler: fix drm_sched_get_cleanup_job Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 02/73] dma-buf: Fix SET_NAME ioctl uapi Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 03/73] drm/amdgpu: invalidate L2 before SDMA IBs (v2) Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 04/73] drm/edid: Fix off-by-one in DispID DTD pixel clock Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 05/73] drm/amd/display: Fix green screen issue after suspend Greg Kroah-Hartman
2020-05-04 17:57 ` Greg Kroah-Hartman [this message]
2020-05-04 17:57 ` [PATCH 5.6 07/73] drm/i915/gt: Check cacheline is valid before acquiring Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 08/73] drm/qxl: qxl_release leak in qxl_draw_dirty_fb() Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 09/73] drm/qxl: qxl_release leak in qxl_hw_surface_alloc() Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 10/73] drm/qxl: qxl_release use after free Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 11/73] NFSv4.1: fix handling of backchannel binding in BIND_CONN_TO_SESSION Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 12/73] btrfs: fix transaction leak in btrfs_recover_relocation Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 13/73] btrfs: fix block group leak when removing fails Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 14/73] btrfs: fix partial loss of prealloc extent past i_size after fsync Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 15/73] btrfs: transaction: Avoid deadlock due to bad initialization timing of fs_info::journal_info Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 16/73] mmc: cqhci: Avoid false "cqhci: CQE stuck on" by not open-coding timeout loop Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 17/73] mmc: sdhci-xenon: fix annoying 1.8V regulator warning Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 18/73] mmc: sdhci-pci: Fix eMMC driver strength for BYT-based controllers Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 19/73] mmc: sdhci-msm: Enable host capabilities pertains to R1b response Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 20/73] mmc: meson-mx-sdio: Set MMC_CAP_WAIT_WHILE_BUSY Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 21/73] mmc: meson-mx-sdio: remove the broken ->card_busy() op Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 22/73] crypto: caam - fix the address of the last entry of S/G Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 23/73] ALSA: hda/realtek - Two front mics on a Lenovo ThinkCenter Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 24/73] ALSA: usb-audio: Correct a typo of NuPrime DAC-10 USB ID Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 25/73] ALSA: hda/hdmi: fix without unlocked before return Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 26/73] ALSA: line6: Fix POD HD500 audio playback Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 27/73] ALSA: pcm: oss: Place the plugin buffer overflow checks correctly Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 28/73] i2c: amd-mp2-pci: Fix Oops in amd_mp2_pci_init() error handling Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 29/73] x86/hyperv: Suspend/resume the VP assist page for hibernation Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 30/73] Drivers: hv: vmbus: Fix Suspend-to-Idle for Generation-2 VM Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 31/73] dlmfs_file_write(): fix the bogosity in handling non-zero *ppos Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 32/73] selinux: properly handle multiple messages in selinux_netlink_send() Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 33/73] IB/rdmavt: Always return ERR_PTR from rvt_create_mmap_info() Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 34/73] PM: ACPI: Output correct message on target power state Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 35/73] PM: hibernate: Freeze kernel threads in software_resume() Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 36/73] dm verity fec: fix hash block number in verity_fec_decode Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 37/73] dm writecache: fix data corruption when reloading the target Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 38/73] dm multipath: use updated MPATHF_QUEUE_IO on mapping for bio-based mpath Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 39/73] ARM: dts: imx6qdl-sr-som-ti: indicate powering off wifi is safe Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 40/73] block: remove the bd_openers checks in blk_drop_partitions Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 41/73] scsi: qla2xxx: set UNLOADING before waiting for session deletion Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 42/73] scsi: qla2xxx: check UNLOADING before posting async work Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 43/73] RDMA/mlx5: Set GRH fields in query QP on RoCE Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 44/73] RDMA/uverbs: Fix a race with disassociate and exit_mmap() Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 45/73] RDMA/mlx4: Initialize ib_spec on the stack Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 46/73] RDMA/siw: Fix potential siw_mem refcnt leak in siw_fastreg_mr() Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 47/73] RDMA/core: Prevent mixed use of FDs between shared ufiles Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 48/73] RDMA/core: Fix overwriting of uobj in case of error Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 49/73] RDMA/core: Fix race between destroy and release FD object Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 50/73] RDMA/cm: Fix ordering of xa_alloc_cyclic() in ib_create_cm_id() Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 51/73] RDMA/cm: Fix an error check in cm_alloc_id_priv() Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 52/73] i2c: iproc: generate stop event for slave writes Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 53/73] dmaengine: hisilicon: Fix build error without PCI_MSI Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 54/73] vfio: avoid possible overflow in vfio_iommu_type1_pin_pages Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 55/73] vfio/type1: Fix VA->PA translation for PFNMAP VMAs in vaddr_get_pfn() Greg Kroah-Hartman
2020-05-04 17:57 ` [PATCH 5.6 56/73] iommu/qcom: Fix local_base status check Greg Kroah-Hartman
2020-05-04 17:58 ` [PATCH 5.6 57/73] dmaengine: ti: k3-psil: fix deadlock on error path Greg Kroah-Hartman
2020-05-04 17:58 ` [PATCH 5.6 58/73] dmaengine: fix channel index enumeration Greg Kroah-Hartman
2020-05-04 17:58 ` [PATCH 5.6 59/73] scsi: target/iblock: fix WRITE SAME zeroing Greg Kroah-Hartman
2020-05-04 17:58 ` [PATCH 5.6 60/73] iommu: Properly export iommu_group_get_for_dev() Greg Kroah-Hartman
2020-05-04 17:58 ` [PATCH 5.6 61/73] iommu/vt-d: Use right Kconfig option name Greg Kroah-Hartman
2020-05-04 19:42   ` Joe Perches
2020-05-04 17:58 ` [PATCH 5.6 62/73] iommu/amd: Fix legacy interrupt remapping for x2APIC-enabled system Greg Kroah-Hartman
2020-05-04 17:58 ` [PATCH 5.6 63/73] i2c: aspeed: Avoid i2c interrupt status clear race condition Greg Kroah-Hartman
2020-05-04 17:58 ` [PATCH 5.6 64/73] ALSA: opti9xx: shut up gcc-10 range warning Greg Kroah-Hartman
2020-05-04 17:58 ` [PATCH 5.6 65/73] Fix use after free in get_tree_bdev() Greg Kroah-Hartman
2020-05-04 17:58 ` [PATCH 5.6 66/73] nvme: prevent double free in nvme_alloc_ns() error handling Greg Kroah-Hartman
2020-05-04 17:58 ` [PATCH 5.6 67/73] drm/i915/selftests: Fix i915_address_space refcnt leak Greg Kroah-Hartman
2020-05-04 17:58 ` [PATCH 5.6 68/73] nfs: Fix potential posix_acl refcnt leak in nfs3_set_acl Greg Kroah-Hartman
2020-05-04 17:58 ` [PATCH 5.6 69/73] dmaengine: dmatest: Fix iteration non-stop logic Greg Kroah-Hartman
2020-05-04 17:58 ` [PATCH 5.6 70/73] drm/i915: Use proper fault mask in interrupt postinstall too Greg Kroah-Hartman
2020-05-04 17:58 ` [PATCH 5.6 71/73] dmaengine: dmatest: Fix process hang when reading wait parameter Greg Kroah-Hartman
2020-05-04 17:58 ` [PATCH 5.6 72/73] arm64: vdso: Add -fasynchronous-unwind-tables to cflags Greg Kroah-Hartman
2020-05-04 17:58 ` [PATCH 5.6 73/73] io_uring: statx must grab the file table for valid fd Greg Kroah-Hartman
2020-05-05  8:38 ` [PATCH 5.6 00/73] 5.6.11-rc1 review Jon Hunter
2020-05-05  9:18   ` Greg Kroah-Hartman
2020-05-05 14:27 ` Naresh Kamboju
2020-05-05 18:12   ` Greg Kroah-Hartman
2020-05-05 15:25 ` shuah
2020-05-05 15:30   ` shuah
2020-05-05 15:36     ` Takashi Iwai
2020-05-05 15:43       ` shuah
2020-05-05 16:19         ` shuah
2020-05-05 16:59           ` Greg Kroah-Hartman
2020-05-05 15:45 ` Guenter Roeck
2020-05-05 18:12   ` Greg Kroah-Hartman

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=20200504165503.088299523@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=airlied@redhat.com \
    --cc=chris@chris-wilson.co.uk \
    --cc=linux-kernel@vger.kernel.org \
    --cc=rodrigo.vivi@intel.com \
    --cc=stable@vger.kernel.org \
    --cc=tvrtko.ursulin@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