stable.vger.kernel.org archive mirror
 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, Wei Huang <wei.huang2@amd.com>,
	Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>,
	Joerg Roedel <jroedel@suse.de>, Sasha Levin <sashal@kernel.org>
Subject: [PATCH 5.10 39/63] iommu/amd: Relocate GAMSup check to early_enable_iommus
Date: Fri, 24 Sep 2021 14:44:39 +0200	[thread overview]
Message-ID: <20210924124335.613031622@linuxfoundation.org> (raw)
In-Reply-To: <20210924124334.228235870@linuxfoundation.org>

From: Wei Huang <wei.huang2@amd.com>

[ Upstream commit c3811a50addd23b9bb5a36278609ee1638debcf6 ]

Currently, iommu_init_ga() checks and disables IOMMU VAPIC support
(i.e. AMD AVIC support in IOMMU) when GAMSup feature bit is not set.
However it forgets to clear IRQ_POSTING_CAP from the previously set
amd_iommu_irq_ops.capability.

This triggers an invalid page fault bug during guest VM warm reboot
if AVIC is enabled since the irq_remapping_cap(IRQ_POSTING_CAP) is
incorrectly set, and crash the system with the following kernel trace.

    BUG: unable to handle page fault for address: 0000000000400dd8
    RIP: 0010:amd_iommu_deactivate_guest_mode+0x19/0xbc
    Call Trace:
     svm_set_pi_irte_mode+0x8a/0xc0 [kvm_amd]
     ? kvm_make_all_cpus_request_except+0x50/0x70 [kvm]
     kvm_request_apicv_update+0x10c/0x150 [kvm]
     svm_toggle_avic_for_irq_window+0x52/0x90 [kvm_amd]
     svm_enable_irq_window+0x26/0xa0 [kvm_amd]
     vcpu_enter_guest+0xbbe/0x1560 [kvm]
     ? avic_vcpu_load+0xd5/0x120 [kvm_amd]
     ? kvm_arch_vcpu_load+0x76/0x240 [kvm]
     ? svm_get_segment_base+0xa/0x10 [kvm_amd]
     kvm_arch_vcpu_ioctl_run+0x103/0x590 [kvm]
     kvm_vcpu_ioctl+0x22a/0x5d0 [kvm]
     __x64_sys_ioctl+0x84/0xc0
     do_syscall_64+0x33/0x40
     entry_SYSCALL_64_after_hwframe+0x44/0xae

Fixes by moving the initializing of AMD IOMMU interrupt remapping mode
(amd_iommu_guest_ir) earlier before setting up the
amd_iommu_irq_ops.capability with appropriate IRQ_POSTING_CAP flag.

[joro:	Squashed the two patches and limited
	check_features_on_all_iommus() to CONFIG_IRQ_REMAP
	to fix a compile warning.]

Signed-off-by: Wei Huang <wei.huang2@amd.com>
Co-developed-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
Signed-off-by: Suravee Suthikulpanit <suravee.suthikulpanit@amd.com>
Link: https://lore.kernel.org/r/20210820202957.187572-2-suravee.suthikulpanit@amd.com
Link: https://lore.kernel.org/r/20210820202957.187572-3-suravee.suthikulpanit@amd.com
Fixes: 8bda0cfbdc1a ("iommu/amd: Detect and initialize guest vAPIC log")
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/iommu/amd/init.c | 31 ++++++++++++++++++++++++-------
 1 file changed, 24 insertions(+), 7 deletions(-)

diff --git a/drivers/iommu/amd/init.c b/drivers/iommu/amd/init.c
index fa57986c2309..28de889aa516 100644
--- a/drivers/iommu/amd/init.c
+++ b/drivers/iommu/amd/init.c
@@ -298,6 +298,22 @@ int amd_iommu_get_num_iommus(void)
 	return amd_iommus_present;
 }
 
+#ifdef CONFIG_IRQ_REMAP
+static bool check_feature_on_all_iommus(u64 mask)
+{
+	bool ret = false;
+	struct amd_iommu *iommu;
+
+	for_each_iommu(iommu) {
+		ret = iommu_feature(iommu, mask);
+		if (!ret)
+			return false;
+	}
+
+	return true;
+}
+#endif
+
 /*
  * For IVHD type 0x11/0x40, EFR is also available via IVHD.
  * Default to IVHD EFR since it is available sooner
@@ -854,13 +870,6 @@ static int iommu_init_ga(struct amd_iommu *iommu)
 	int ret = 0;
 
 #ifdef CONFIG_IRQ_REMAP
-	/* Note: We have already checked GASup from IVRS table.
-	 *       Now, we need to make sure that GAMSup is set.
-	 */
-	if (AMD_IOMMU_GUEST_IR_VAPIC(amd_iommu_guest_ir) &&
-	    !iommu_feature(iommu, FEATURE_GAM_VAPIC))
-		amd_iommu_guest_ir = AMD_IOMMU_GUEST_IR_LEGACY_GA;
-
 	ret = iommu_init_ga_log(iommu);
 #endif /* CONFIG_IRQ_REMAP */
 
@@ -2396,6 +2405,14 @@ static void early_enable_iommus(void)
 	}
 
 #ifdef CONFIG_IRQ_REMAP
+	/*
+	 * Note: We have already checked GASup from IVRS table.
+	 *       Now, we need to make sure that GAMSup is set.
+	 */
+	if (AMD_IOMMU_GUEST_IR_VAPIC(amd_iommu_guest_ir) &&
+	    !check_feature_on_all_iommus(FEATURE_GAM_VAPIC))
+		amd_iommu_guest_ir = AMD_IOMMU_GUEST_IR_LEGACY_GA;
+
 	if (AMD_IOMMU_GUEST_IR_VAPIC(amd_iommu_guest_ir))
 		amd_iommu_irq_ops.capability |= (1 << IRQ_POSTING_CAP);
 #endif
-- 
2.33.0




  parent reply	other threads:[~2021-09-24 13:12 UTC|newest]

Thread overview: 72+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-09-24 12:44 [PATCH 5.10 00/63] 5.10.69-rc1 review Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 01/63] PCI: pci-bridge-emul: Add PCIe Root Capabilities Register Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 02/63] PCI: aardvark: Fix reporting CRS value Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 03/63] console: consume APC, DM, DCS Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 04/63] s390/pci_mmio: fully validate the VMA before calling follow_pte() Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 05/63] ARM: Qualify enabling of swiotlb_init() Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 06/63] ARM: 9077/1: PLT: Move struct plt_entries definition to header Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 07/63] ARM: 9078/1: Add warn suppress parameter to arm_gen_branch_link() Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 08/63] ARM: 9079/1: ftrace: Add MODULE_PLTS support Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 09/63] ARM: 9098/1: ftrace: MODULE_PLT: Fix build problem without DYNAMIC_FTRACE Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 10/63] [PATCH] Revert "net/mlx5: Register to devlink ingress VLAN filter trap" Greg Kroah-Hartman
2021-09-27  8:39   ` Tariq Toukan
2021-09-24 12:44 ` [PATCH 5.10 11/63] sctp: validate chunk size in __rcv_asconf_lookup Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 12/63] sctp: add param size validation for SCTP_PARAM_SET_PRIMARY Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 13/63] staging: rtl8192u: Fix bitwise vs logical operator in TranslateRxSignalStuff819xUsb() Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 14/63] coredump: fix memleak in dump_vma_snapshot() Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 15/63] um: virtio_uml: fix memory leak on init failures Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 16/63] dmaengine: acpi: Avoid comparison GSI with Linux vIRQ Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 17/63] perf test: Fix bpf test sample mismatch reporting Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 18/63] perf tools: Allow build-id with trailing zeros Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 19/63] thermal/drivers/exynos: Fix an error code in exynos_tmu_probe() Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 20/63] 9p/trans_virtio: Remove sysfs file on probe failure Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 21/63] prctl: allow to setup brk for et_dyn executables Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 22/63] nilfs2: use refcount_dec_and_lock() to fix potential UAF Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 23/63] profiling: fix shift-out-of-bounds bugs Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 24/63] PM: sleep: core: Avoid setting power.must_resume to false Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 25/63] pwm: lpc32xx: Dont modify HW state in .probe() after the PWM chip was registered Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 26/63] pwm: mxs: " Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 27/63] dmaengine: idxd: fix wq slot allocation index check Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 28/63] platform/chrome: sensorhub: Add trace events for sample Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 29/63] platform/chrome: cros_ec_trace: Fix format warnings Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 30/63] ceph: allow ceph_put_mds_session to take NULL or ERR_PTR Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 31/63] ceph: cancel delayed work instead of flushing on mdsc teardown Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 32/63] Kconfig.debug: drop selecting non-existing HARDLOCKUP_DETECTOR_ARCH Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 33/63] tools/bootconfig: Fix tracing_on option checking in ftrace2bconf.sh Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 34/63] thermal/core: Fix thermal_cooling_device_register() prototype Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 35/63] drm/amdgpu: Disable PCIE_DPM on Intel RKL Platform Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 36/63] drivers: base: cacheinfo: Get rid of DEFINE_SMP_CALL_CACHE_FUNCTION() Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 37/63] dma-buf: DMABUF_MOVE_NOTIFY should depend on DMA_SHARED_BUFFER Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 38/63] parisc: Move pci_dev_is_behind_card_dino to where it is used Greg Kroah-Hartman
2021-09-24 12:44 ` Greg Kroah-Hartman [this message]
2021-09-24 12:44 ` [PATCH 5.10 40/63] dmaengine: idxd: depends on !UML Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 41/63] dmaengine: sprd: Add missing MODULE_DEVICE_TABLE Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 42/63] dmaengine: ioat: depends on !UML Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 43/63] dmaengine: xilinx_dma: Set DMA mask for coherent APIs Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 44/63] ceph: request Fw caps before updating the mtime in ceph_write_iter Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 45/63] ceph: remove the capsnaps when removing caps Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 46/63] ceph: lockdep annotations for try_nonblocking_invalidate Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 47/63] btrfs: update the bdev time directly when closing Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 48/63] btrfs: fix lockdep warning while mounting sprout fs Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 49/63] nilfs2: fix memory leak in nilfs_sysfs_create_device_group Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 50/63] nilfs2: fix NULL pointer in nilfs_##name##_attr_release Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 51/63] nilfs2: fix memory leak in nilfs_sysfs_create_##name##_group Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 52/63] nilfs2: fix memory leak in nilfs_sysfs_delete_##name##_group Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 53/63] nilfs2: fix memory leak in nilfs_sysfs_create_snapshot_group Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 54/63] nilfs2: fix memory leak in nilfs_sysfs_delete_snapshot_group Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 55/63] habanalabs: add validity check for event ID received from F/W Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 56/63] pwm: img: Dont modify HW state in .remove() callback Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 57/63] pwm: rockchip: " Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 58/63] pwm: stm32-lp: " Greg Kroah-Hartman
2021-09-24 12:44 ` [PATCH 5.10 59/63] blk-throttle: fix UAF by deleteing timer in blk_throtl_exit() Greg Kroah-Hartman
2021-09-24 12:45 ` [PATCH 5.10 60/63] blk-mq: allow 4x BLK_MAX_REQUEST_COUNT at blk_plug for multiple_queues Greg Kroah-Hartman
2021-09-24 12:45 ` [PATCH 5.10 61/63] rtc: rx8010: select REGMAP_I2C Greg Kroah-Hartman
2021-09-24 12:45 ` [PATCH 5.10 62/63] sched/idle: Make the idle timer expire in hard interrupt context Greg Kroah-Hartman
2021-09-24 12:45 ` [PATCH 5.10 63/63] drm/nouveau/nvkm: Replace -ENOSYS with -ENODEV Greg Kroah-Hartman
2021-09-24 14:12 ` [PATCH 5.10 00/63] 5.10.69-rc1 review Daniel Díaz
2021-09-24 22:11   ` Florian Fainelli
2021-09-25 11:48   ` Greg Kroah-Hartman
2021-09-24 17:57 ` Jon Hunter
2021-09-24 18:05 ` Fox Chen
2021-09-24 21:49 ` Pavel Machek
2021-09-24 21:52 ` Shuah Khan

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=20210924124335.613031622@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=jroedel@suse.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=sashal@kernel.org \
    --cc=stable@vger.kernel.org \
    --cc=suravee.suthikulpanit@amd.com \
    --cc=wei.huang2@amd.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;
as well as URLs for NNTP newsgroup(s).