From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
stable@vger.kernel.org, Qiuhao Li <qiuhao@sysec.org>,
Gaoning Pan <pgn@zju.edu.cn>, Yongkang Jia <kangel@zju.edu.cn>,
Sean Christopherson <seanjc@google.com>,
Paolo Bonzini <pbonzini@redhat.com>
Subject: [PATCH 5.15 31/66] KVM: x86: avoid calling x86 emulator without a decoded instruction
Date: Fri, 3 Jun 2022 19:43:11 +0200 [thread overview]
Message-ID: <20220603173821.553626389@linuxfoundation.org> (raw)
In-Reply-To: <20220603173820.663747061@linuxfoundation.org>
From: Sean Christopherson <seanjc@google.com>
commit fee060cd52d69c114b62d1a2948ea9648b5131f9 upstream.
Whenever x86_decode_emulated_instruction() detects a breakpoint, it
returns the value that kvm_vcpu_check_breakpoint() writes into its
pass-by-reference second argument. Unfortunately this is completely
bogus because the expected outcome of x86_decode_emulated_instruction
is an EMULATION_* value.
Then, if kvm_vcpu_check_breakpoint() does "*r = 0" (corresponding to
a KVM_EXIT_DEBUG userspace exit), it is misunderstood as EMULATION_OK
and x86_emulate_instruction() is called without having decoded the
instruction. This causes various havoc from running with a stale
emulation context.
The fix is to move the call to kvm_vcpu_check_breakpoint() where it was
before commit 4aa2691dcbd3 ("KVM: x86: Factor out x86 instruction
emulation with decoding") introduced x86_decode_emulated_instruction().
The other caller of the function does not need breakpoint checks,
because it is invoked as part of a vmexit and the processor has already
checked those before executing the instruction that #GP'd.
This fixes CVE-2022-1852.
Reported-by: Qiuhao Li <qiuhao@sysec.org>
Reported-by: Gaoning Pan <pgn@zju.edu.cn>
Reported-by: Yongkang Jia <kangel@zju.edu.cn>
Fixes: 4aa2691dcbd3 ("KVM: x86: Factor out x86 instruction emulation with decoding")
Cc: stable@vger.kernel.org
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20220311032801.3467418-2-seanjc@google.com>
[Rewrote commit message according to Qiuhao's report, since a patch
already existed to fix the bug. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
arch/x86/kvm/x86.c | 31 +++++++++++++++++++------------
1 file changed, 19 insertions(+), 12 deletions(-)
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -7846,7 +7846,7 @@ int kvm_skip_emulated_instruction(struct
}
EXPORT_SYMBOL_GPL(kvm_skip_emulated_instruction);
-static bool kvm_vcpu_check_breakpoint(struct kvm_vcpu *vcpu, int *r)
+static bool kvm_vcpu_check_code_breakpoint(struct kvm_vcpu *vcpu, int *r)
{
if (unlikely(vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP) &&
(vcpu->arch.guest_debug_dr7 & DR7_BP_EN_MASK)) {
@@ -7915,25 +7915,23 @@ static bool is_vmware_backdoor_opcode(st
}
/*
- * Decode to be emulated instruction. Return EMULATION_OK if success.
+ * Decode an instruction for emulation. The caller is responsible for handling
+ * code breakpoints. Note, manually detecting code breakpoints is unnecessary
+ * (and wrong) when emulating on an intercepted fault-like exception[*], as
+ * code breakpoints have higher priority and thus have already been done by
+ * hardware.
+ *
+ * [*] Except #MC, which is higher priority, but KVM should never emulate in
+ * response to a machine check.
*/
int x86_decode_emulated_instruction(struct kvm_vcpu *vcpu, int emulation_type,
void *insn, int insn_len)
{
- int r = EMULATION_OK;
struct x86_emulate_ctxt *ctxt = vcpu->arch.emulate_ctxt;
+ int r;
init_emulate_ctxt(vcpu);
- /*
- * We will reenter on the same instruction since we do not set
- * complete_userspace_io. This does not handle watchpoints yet,
- * those would be handled in the emulate_ops.
- */
- if (!(emulation_type & EMULTYPE_SKIP) &&
- kvm_vcpu_check_breakpoint(vcpu, &r))
- return r;
-
r = x86_decode_insn(ctxt, insn, insn_len, emulation_type);
trace_kvm_emulate_insn_start(vcpu);
@@ -7966,6 +7964,15 @@ int x86_emulate_instruction(struct kvm_v
if (!(emulation_type & EMULTYPE_NO_DECODE)) {
kvm_clear_exception_queue(vcpu);
+ /*
+ * Return immediately if RIP hits a code breakpoint, such #DBs
+ * are fault-like and are higher priority than any faults on
+ * the code fetch itself.
+ */
+ if (!(emulation_type & EMULTYPE_SKIP) &&
+ kvm_vcpu_check_code_breakpoint(vcpu, &r))
+ return r;
+
r = x86_decode_emulated_instruction(vcpu, emulation_type,
insn, insn_len);
if (r != EMULATION_OK) {
next prev parent reply other threads:[~2022-06-03 17:54 UTC|newest]
Thread overview: 73+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-06-03 17:42 [PATCH 5.15 00/66] 5.15.45-rc1 review Greg Kroah-Hartman
2022-06-03 17:42 ` [PATCH 5.15 01/66] ALSA: usb-audio: Dont get sample rate for MCT Trigger 5 USB-to-HDMI Greg Kroah-Hartman
2022-06-03 17:42 ` [PATCH 5.15 02/66] pinctrl: sunxi: fix f1c100s uart2 function Greg Kroah-Hartman
2022-06-03 17:42 ` [PATCH 5.15 03/66] KVM: arm64: Dont hypercall before EL2 init Greg Kroah-Hartman
2022-06-03 17:42 ` [PATCH 5.15 04/66] percpu_ref_init(): clean ->percpu_count_ref on failure Greg Kroah-Hartman
2022-06-03 17:42 ` [PATCH 5.15 05/66] net: af_key: check encryption module availability consistency Greg Kroah-Hartman
2022-06-03 17:42 ` [PATCH 5.15 06/66] nfc: pn533: Fix buggy cleanup order Greg Kroah-Hartman
2022-06-03 17:42 ` [PATCH 5.15 07/66] net: ftgmac100: Disable hardware checksum on AST2600 Greg Kroah-Hartman
2022-06-03 17:42 ` [PATCH 5.15 08/66] i2c: ismt: Provide a DMA buffer for Interrupt Cause Logging Greg Kroah-Hartman
2022-06-03 17:42 ` [PATCH 5.15 09/66] drivers: i2c: thunderx: Allow driver to work with ACPI defined TWSI controllers Greg Kroah-Hartman
2022-06-03 17:42 ` [PATCH 5.15 10/66] netfilter: nf_tables: disallow non-stateful expression in sets earlier Greg Kroah-Hartman
2022-06-03 17:42 ` [PATCH 5.15 11/66] i2c: ismt: prevent memory corruption in ismt_access() Greg Kroah-Hartman
2022-06-03 17:42 ` [PATCH 5.15 12/66] assoc_array: Fix BUG_ON during garbage collect Greg Kroah-Hartman
2022-06-03 17:42 ` [PATCH 5.15 13/66] pipe: make poll_usage boolean and annotate its access Greg Kroah-Hartman
2022-06-03 17:42 ` [PATCH 5.15 14/66] pipe: Fix missing lock in pipe_resize_ring() Greg Kroah-Hartman
2022-06-03 17:42 ` [PATCH 5.15 15/66] net: ipa: compute proper aggregation limit Greg Kroah-Hartman
2022-06-03 17:42 ` [PATCH 5.15 16/66] drm/i915: Fix -Wstringop-overflow warning in call to intel_read_wm_latency() Greg Kroah-Hartman
2022-06-03 17:42 ` [PATCH 5.15 17/66] exfat: check if cluster num is valid Greg Kroah-Hartman
2022-06-03 17:42 ` [PATCH 5.15 18/66] exfat: fix referencing wrong parent directory information after renaming Greg Kroah-Hartman
2022-06-03 17:42 ` [PATCH 5.15 19/66] lib/crypto: add prompts back to crypto libraries Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 20/66] crypto: drbg - prepare for more fine-grained tracking of seeding state Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 21/66] crypto: drbg - track whether DRBG was seeded with !rng_is_initialized() Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 22/66] crypto: drbg - move dynamic ->reseed_threshold adjustments to __drbg_seed() Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 23/66] crypto: drbg - make reseeding from get_random_bytes() synchronous Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 24/66] netfilter: nf_tables: sanitize nft_set_desc_concat_parse() Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 25/66] netfilter: nf_tables: hold mutex on netns pre_exit path Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 26/66] netfilter: nf_tables: double hook unregistration in netns path Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 27/66] netfilter: conntrack: re-fetch conntrack after insertion Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 28/66] KVM: PPC: Book3S HV: fix incorrect NULL check on list iterator Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 29/66] x86/kvm: Alloc dummy async #PF token outside of raw spinlock Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 30/66] x86, kvm: use correct GFP flags for preemption disabled Greg Kroah-Hartman
2022-06-03 17:43 ` Greg Kroah-Hartman [this message]
2022-06-03 17:43 ` [PATCH 5.15 32/66] KVM: x86: Drop WARNs that assert a triple fault never "escapes" from L2 Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 33/66] KVM: SVM: Use kzalloc for sev ioctl interfaces to prevent kernel data leak Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 34/66] crypto: caam - fix i.MX6SX entropy delay value Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 35/66] crypto: ecrdsa - Fix incorrect use of vli_cmp Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 36/66] zsmalloc: fix races between asynchronous zspage free and page migration Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 37/66] ALSA: usb-audio: Workaround for clock setup on TEAC devices Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 38/66] ALSA: usb-audio: Add missing ep_idx in fixed EP quirks Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 39/66] ALSA: usb-audio: Configure sync endpoints before data Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 40/66] Bluetooth: hci_qca: Use del_timer_sync() before freeing Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 41/66] ARM: dts: s5pv210: Correct interrupt name for bluetooth in Aries Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 42/66] dm integrity: fix error code in dm_integrity_ctr() Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 43/66] dm crypt: make printing of the key constant-time Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 44/66] dm stats: add cond_resched when looping over entries Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 45/66] dm verity: set DM_TARGET_IMMUTABLE feature flag Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 46/66] raid5: introduce MD_BROKEN Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 47/66] fs/ntfs3: validate BOOT sectors_per_clusters Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 48/66] HID: multitouch: Add support for Google Whiskers Touchpad Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 49/66] HID: multitouch: add quirks to enable Lenovo X12 trackpoint Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 50/66] x86/sgx: Disconnect backing page references from dirty status Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 51/66] x86/sgx: Mark PCMD page as dirty when modifying contents Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 52/66] x86/sgx: Obtain backing storage page with enclave mutex held Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 53/66] x86/sgx: Fix race between reclaimer and page fault handler Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 54/66] x86/sgx: Ensure no data in PCMD page after truncate Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 55/66] media: i2c: imx412: Fix reset GPIO polarity Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 56/66] media: i2c: imx412: Fix power_off ordering Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 57/66] tpm: Fix buffer access in tpm2_get_tpm_pt() Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 58/66] tpm: ibmvtpm: Correct the return value in tpm_ibmvtpm_probe() Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 59/66] docs: submitting-patches: Fix crossref to The canonical patch format Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 60/66] NFS: Memory allocation failures are not server fatal errors Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 61/66] NFSD: Fix possible sleep during nfsd4_release_lockowner() Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 62/66] bpf: Fix potential array overflow in bpf_trampoline_get_progs() Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 63/66] bpf: Enlarge offset check value to INT_MAX in bpf_skb_{load,store}_bytes Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 64/66] bpf: Fix excessive memory allocation in stack_map_alloc() Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 65/66] bpf: Reject writes for PTR_TO_MAP_KEY in check_helper_mem_access Greg Kroah-Hartman
2022-06-03 17:43 ` [PATCH 5.15 66/66] bpf: Check PTR_TO_MEM | MEM_RDONLY " Greg Kroah-Hartman
2022-06-04 0:34 ` [PATCH 5.15 00/66] 5.15.45-rc1 review Fox Chen
2022-06-04 12:37 ` Sudip Mukherjee
2022-06-04 17:11 ` Naresh Kamboju
2022-06-04 21:19 ` Guenter Roeck
2022-06-05 3:37 ` Bagas Sanjaya
2022-06-05 4:15 ` Ron Economos
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=20220603173821.553626389@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=kangel@zju.edu.cn \
--cc=linux-kernel@vger.kernel.org \
--cc=pbonzini@redhat.com \
--cc=pgn@zju.edu.cn \
--cc=qiuhao@sysec.org \
--cc=seanjc@google.com \
--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