From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: stable@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
patches@lists.linux.dev, Kevin Tian <kevin.tian@intel.com>,
Nicolin Chen <nicolinc@nvidia.com>,
Jason Gunthorpe <jgg@nvidia.com>, Sasha Levin <sashal@kernel.org>
Subject: [PATCH 6.16 24/41] iommufd: WARN if an object is aborted with an elevated refcount
Date: Fri, 10 Oct 2025 15:16:12 +0200 [thread overview]
Message-ID: <20251010131334.294671518@linuxfoundation.org> (raw)
In-Reply-To: <20251010131333.420766773@linuxfoundation.org>
6.16-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jason Gunthorpe <jgg@nvidia.com>
[ Upstream commit 53d0584eeb2c85a46c83656246d61a89558d74b3 ]
If something holds a refcount then it is at risk of UAFing. For abort
paths we expect the caller to never share the object with a parallel
thread and to clean up any refcounts it obtained on its own.
Add the missing dec inside iommufd_hwpt_paging_alloc() during error unwind
by making iommufd_hw_pagetable_attach/detach() proper pairs.
Link: https://patch.msgid.link/r/2-v1-02cd136829df+31-iommufd_syz_fput_jgg@nvidia.com
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
Reviewed-by: Nicolin Chen <nicolinc@nvidia.com>
Tested-by: Nicolin Chen <nicolinc@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
drivers/iommu/iommufd/device.c | 3 ++-
drivers/iommu/iommufd/iommufd_private.h | 3 +--
drivers/iommu/iommufd/main.c | 4 ++++
3 files changed, 7 insertions(+), 3 deletions(-)
diff --git a/drivers/iommu/iommufd/device.c b/drivers/iommu/iommufd/device.c
index 86244403b5320..674f9f244f7b4 100644
--- a/drivers/iommu/iommufd/device.c
+++ b/drivers/iommu/iommufd/device.c
@@ -661,6 +661,8 @@ iommufd_hw_pagetable_detach(struct iommufd_device *idev, ioasid_t pasid)
iopt_remove_reserved_iova(&hwpt_paging->ioas->iopt, idev->dev);
mutex_unlock(&igroup->lock);
+ iommufd_hw_pagetable_put(idev->ictx, hwpt);
+
/* Caller must destroy hwpt */
return hwpt;
}
@@ -1007,7 +1009,6 @@ void iommufd_device_detach(struct iommufd_device *idev, ioasid_t pasid)
hwpt = iommufd_hw_pagetable_detach(idev, pasid);
if (!hwpt)
return;
- iommufd_hw_pagetable_put(idev->ictx, hwpt);
refcount_dec(&idev->obj.users);
}
EXPORT_SYMBOL_NS_GPL(iommufd_device_detach, "IOMMUFD");
diff --git a/drivers/iommu/iommufd/iommufd_private.h b/drivers/iommu/iommufd/iommufd_private.h
index 9ccc83341f321..e68d8d63076a8 100644
--- a/drivers/iommu/iommufd/iommufd_private.h
+++ b/drivers/iommu/iommufd/iommufd_private.h
@@ -390,9 +390,8 @@ static inline void iommufd_hw_pagetable_put(struct iommufd_ctx *ictx,
if (hwpt->obj.type == IOMMUFD_OBJ_HWPT_PAGING) {
struct iommufd_hwpt_paging *hwpt_paging = to_hwpt_paging(hwpt);
- lockdep_assert_not_held(&hwpt_paging->ioas->mutex);
-
if (hwpt_paging->auto_domain) {
+ lockdep_assert_not_held(&hwpt_paging->ioas->mutex);
iommufd_object_put_and_try_destroy(ictx, &hwpt->obj);
return;
}
diff --git a/drivers/iommu/iommufd/main.c b/drivers/iommu/iommufd/main.c
index 62a3469bbd37e..2b26747ac2021 100644
--- a/drivers/iommu/iommufd/main.c
+++ b/drivers/iommu/iommufd/main.c
@@ -62,6 +62,10 @@ void iommufd_object_abort(struct iommufd_ctx *ictx, struct iommufd_object *obj)
old = xas_store(&xas, NULL);
xa_unlock(&ictx->objects);
WARN_ON(old != XA_ZERO_ENTRY);
+
+ if (WARN_ON(!refcount_dec_and_test(&obj->users)))
+ return;
+
kfree(obj);
}
--
2.51.0
next prev parent reply other threads:[~2025-10-10 13:19 UTC|newest]
Thread overview: 52+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-10-10 13:15 [PATCH 6.16 00/41] 6.16.12-rc1 review Greg Kroah-Hartman
2025-10-10 13:15 ` [PATCH 6.16 01/41] ALSA: hda/tas2781: Fix the order of TAS2781 calibrated-data Greg Kroah-Hartman
2025-10-10 13:15 ` [PATCH 6.16 02/41] drm/amdgpu: Enable MES lr_compute_wa by default Greg Kroah-Hartman
2025-10-10 13:15 ` [PATCH 6.16 03/41] wifi: rtw89: mcc: stop TX during MCC prepare Greg Kroah-Hartman
2025-10-10 13:15 ` [PATCH 6.16 04/41] wifi: rtw89: fix use-after-free in rtw89_core_tx_kick_off_and_wait() Greg Kroah-Hartman
2025-10-10 13:15 ` [PATCH 6.16 05/41] USB: serial: option: add SIMCom 8230C compositions Greg Kroah-Hartman
2025-10-10 13:15 ` [PATCH 6.16 06/41] Bluetooth: btusb: Add USB ID 2001:332a for D-Link AX9U rev. A1 Greg Kroah-Hartman
2025-10-10 13:15 ` [PATCH 6.16 07/41] wifi: rtlwifi: rtl8192cu: Dont claim USB ID 07b8:8188 Greg Kroah-Hartman
2025-10-10 13:15 ` [PATCH 6.16 08/41] wifi: rtl8xxxu: " Greg Kroah-Hartman
2025-10-10 13:15 ` [PATCH 6.16 09/41] rust: drm: fix `srctree/` links Greg Kroah-Hartman
2025-10-10 13:15 ` [PATCH 6.16 10/41] rust: block: " Greg Kroah-Hartman
2025-10-10 13:15 ` [PATCH 6.16 11/41] rust: pci: fix incorrect platform reference in PCI driver probe doc comment Greg Kroah-Hartman
2025-10-10 13:16 ` [PATCH 6.16 12/41] gpiolib: acpi: Ignore touchpad wakeup on GPD G1619-05 Greg Kroah-Hartman
2025-10-10 13:16 ` [PATCH 6.16 13/41] ASoC: rt712: avoid skipping the blind write Greg Kroah-Hartman
2025-10-10 13:16 ` [PATCH 6.16 14/41] ASoC: amd: acp: Adjust pdm gain value Greg Kroah-Hartman
2025-10-10 13:16 ` [PATCH 6.16 15/41] dm-integrity: limit MAX_TAG_SIZE to 255 Greg Kroah-Hartman
2025-10-10 13:16 ` [PATCH 6.16 16/41] platform/x86/amd/pmc: Add MECHREVO Yilong15Pro to spurious_8042 list Greg Kroah-Hartman
2025-10-10 13:16 ` [PATCH 6.16 17/41] platform/x86: oxpec: Add support for OneXPlayer X1Pro EVA-02 Greg Kroah-Hartman
2025-10-10 13:16 ` [PATCH 6.16 18/41] perf subcmd: avoid crash in exclude_cmds when excludes is empty Greg Kroah-Hartman
2025-10-10 13:16 ` [PATCH 6.16 19/41] drm/amdgpu/gfx11: Add Cleaner Shader Support for GFX11.0.1/11.0.4 GPUs Greg Kroah-Hartman
2025-10-10 13:16 ` [PATCH 6.16 20/41] platform/x86/amd/pmf: Support new ACPI ID AMDI0108 Greg Kroah-Hartman
2025-10-10 13:16 ` [PATCH 6.16 21/41] ASoC: rt5682s: Adjust SAR ADC button mode to fix noise issue Greg Kroah-Hartman
2025-10-10 13:16 ` [PATCH 6.16 22/41] btrfs: ref-verify: handle damaged extent root tree Greg Kroah-Hartman
2025-10-10 13:16 ` [PATCH 6.16 23/41] netfs: Prevent duplicate unlocking Greg Kroah-Hartman
2025-10-10 13:16 ` Greg Kroah-Hartman [this message]
2025-10-10 13:16 ` [PATCH 6.16 25/41] can: hi311x: fix null pointer dereference when resuming from sleep before interface was enabled Greg Kroah-Hartman
2025-10-10 13:16 ` [PATCH 6.16 26/41] can: rcar_canfd: Fix controller mode setting Greg Kroah-Hartman
2025-10-10 13:16 ` [PATCH 6.16 27/41] platform/x86/amd/pmc: Add Stellaris Slim Gen6 AMD to spurious 8042 quirks list Greg Kroah-Hartman
2025-10-10 13:16 ` [PATCH 6.16 28/41] hid: fix I2C read buffer overflow in raw_event() for mcp2221 Greg Kroah-Hartman
2025-10-10 13:16 ` [PATCH 6.16 29/41] nvmem: layouts: fix automatic module loading Greg Kroah-Hartman
2025-10-10 13:16 ` [PATCH 6.16 30/41] drivers/misc/amd-sbi/Kconfig: select REGMAP_I2C Greg Kroah-Hartman
2025-10-10 13:16 ` [PATCH 6.16 31/41] binder: fix double-free in dbitmap Greg Kroah-Hartman
2025-10-10 13:16 ` [PATCH 6.16 32/41] serial: stm32: allow selecting console when the driver is module Greg Kroah-Hartman
2025-10-10 13:16 ` [PATCH 6.16 33/41] staging: axis-fifo: fix maximum TX packet length check Greg Kroah-Hartman
2025-10-10 13:16 ` [PATCH 6.16 34/41] staging: axis-fifo: fix TX handling on copy_from_user() failure Greg Kroah-Hartman
2025-10-10 13:16 ` [PATCH 6.16 35/41] staging: axis-fifo: flush RX FIFO on read errors Greg Kroah-Hartman
2025-10-10 13:16 ` [PATCH 6.16 36/41] driver core: faux: Set power.no_pm for faux devices Greg Kroah-Hartman
2025-10-10 13:16 ` [PATCH 6.16 37/41] driver core/PM: Set power.no_callbacks along with power.no_pm Greg Kroah-Hartman
2025-10-10 13:16 ` [PATCH 6.16 38/41] crypto: rng - Ensure set_ent is always present Greg Kroah-Hartman
2025-10-10 13:16 ` [PATCH 6.16 39/41] net/9p: fix double req put in p9_fd_cancelled Greg Kroah-Hartman
2025-10-10 13:16 ` [PATCH 6.16 40/41] KVM: x86: Dont (re)check L1 intercepts when completing userspace I/O Greg Kroah-Hartman
2025-10-10 13:16 ` [PATCH 6.16 41/41] ring buffer: Propagate __rb_map_vma return value to caller Greg Kroah-Hartman
2025-10-10 17:15 ` [PATCH 6.16 00/41] 6.16.12-rc1 review Jon Hunter
2025-10-10 19:48 ` Justin Forbes
2025-10-10 22:20 ` Shuah Khan
2025-10-11 7:32 ` Pavel Machek
2025-10-11 9:03 ` Naresh Kamboju
2025-10-11 10:51 ` Mark Brown
2025-10-11 11:37 ` Ron Economos
2025-10-11 16:56 ` Brett A C Sheffield
2025-10-11 20:34 ` Peter Schneider
2025-10-12 9:35 ` Miguel Ojeda
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=20251010131334.294671518@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=jgg@nvidia.com \
--cc=kevin.tian@intel.com \
--cc=nicolinc@nvidia.com \
--cc=patches@lists.linux.dev \
--cc=sashal@kernel.org \
--cc=stable@vger.kernel.org \
/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;
as well as URLs for NNTP newsgroup(s).