From: Ben Hutchings <ben@decadent.org.uk>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org
Cc: akpm@linux-foundation.org, Denis Kirjanov <kda@linux-powerpc.org>,
"Sean Christopherson" <sean.j.christopherson@intel.com>,
"Paolo Bonzini" <pbonzini@redhat.com>,
"Jun Nakajima" <jun.nakajima@intel.com>
Subject: [PATCH 3.16 58/99] KVM: x86: Don't let userspace set host-reserved cr4 bits
Date: Wed, 20 May 2020 15:14:26 +0100 [thread overview]
Message-ID: <lsq.1589984008.288512231@decadent.org.uk> (raw)
In-Reply-To: <lsq.1589984008.673931885@decadent.org.uk>
3.16.84-rc1 review patch. If anyone has any objections, please let me know.
------------------
From: Sean Christopherson <sean.j.christopherson@intel.com>
commit b11306b53b2540c6ba068c4deddb6a17d9f8d95b upstream.
Calculate the host-reserved cr4 bits at runtime based on the system's
capabilities (using logic similar to __do_cpuid_func()), and use the
dynamically generated mask for the reserved bit check in kvm_set_cr4()
instead using of the static CR4_RESERVED_BITS define. This prevents
userspace from "enabling" features in cr4 that are not supported by the
system, e.g. by ignoring KVM_GET_SUPPORTED_CPUID and specifying a bogus
CPUID for the vCPU.
Allowing userspace to set unsupported bits in cr4 can lead to a variety
of undesirable behavior, e.g. failed VM-Enter, and in general increases
KVM's attack surface. A crafty userspace can even abuse CR4.LA57 to
induce an unchecked #GP on a WRMSR.
On a platform without LA57 support:
KVM_SET_CPUID2 // CPUID_7_0_ECX.LA57 = 1
KVM_SET_SREGS // CR4.LA57 = 1
KVM_SET_MSRS // KERNEL_GS_BASE = 0x0004000000000000
KVM_RUN
leads to a #GP when writing KERNEL_GS_BASE into hardware:
unchecked MSR access error: WRMSR to 0xc0000102 (tried to write 0x0004000000000000)
at rIP: 0xffffffffa00f239a (vmx_prepare_switch_to_guest+0x10a/0x1d0 [kvm_intel])
Call Trace:
kvm_arch_vcpu_ioctl_run+0x671/0x1c70 [kvm]
kvm_vcpu_ioctl+0x36b/0x5d0 [kvm]
do_vfs_ioctl+0xa1/0x620
ksys_ioctl+0x66/0x70
__x64_sys_ioctl+0x16/0x20
do_syscall_64+0x4c/0x170
entry_SYSCALL_64_after_hwframe+0x44/0xa9
RIP: 0033:0x7fc08133bf47
Note, the above sequence fails VM-Enter due to invalid guest state.
Userspace can allow VM-Enter to succeed (after the WRMSR #GP) by adding
a KVM_SET_SREGS w/ CR4.LA57=0 after KVM_SET_MSRS, in which case KVM will
technically leak the host's KERNEL_GS_BASE into the guest. But, as
KERNEL_GS_BASE is a userspace-defined value/address, the leak is largely
benign as a malicious userspace would simply be exposing its own data to
the guest, and attacking a benevolent userspace would require multiple
bugs in the userspace VMM.
Cc: Jun Nakajima <jun.nakajima@intel.com>
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
[bwh: Backported to 3.16:
- PKE, LA57, and UMIP are totally unsupported and already included in
CR4_RESERVED_BITS
- Adjust context]
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
---
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -82,6 +82,8 @@ u64 __read_mostly efer_reserved_bits = ~
static u64 __read_mostly efer_reserved_bits = ~((u64)EFER_SCE);
#endif
+static u64 __read_mostly cr4_reserved_bits = CR4_RESERVED_BITS;
+
#define VM_STAT(x) offsetof(struct kvm, stat.x), KVM_STAT_VM
#define VCPU_STAT(x) offsetof(struct kvm_vcpu, stat.x), KVM_STAT_VCPU
@@ -660,13 +662,32 @@ int kvm_set_xcr(struct kvm_vcpu *vcpu, u
}
EXPORT_SYMBOL_GPL(kvm_set_xcr);
+static u64 kvm_host_cr4_reserved_bits(struct cpuinfo_x86 *c)
+{
+ u64 reserved_bits = CR4_RESERVED_BITS;
+
+ if (!cpu_has(c, X86_FEATURE_XSAVE))
+ reserved_bits |= X86_CR4_OSXSAVE;
+
+ if (!cpu_has(c, X86_FEATURE_SMEP))
+ reserved_bits |= X86_CR4_SMEP;
+
+ if (!cpu_has(c, X86_FEATURE_SMAP))
+ reserved_bits |= X86_CR4_SMAP;
+
+ if (!cpu_has(c, X86_FEATURE_FSGSBASE))
+ reserved_bits |= X86_CR4_FSGSBASE;
+
+ return reserved_bits;
+}
+
int kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4)
{
unsigned long old_cr4 = kvm_read_cr4(vcpu);
unsigned long pdptr_bits = X86_CR4_PGE | X86_CR4_PSE | X86_CR4_PAE |
X86_CR4_SMEP | X86_CR4_SMAP;
- if (cr4 & CR4_RESERVED_BITS)
+ if (cr4 & cr4_reserved_bits)
return 1;
if (!guest_cpuid_has_xsave(vcpu) && (cr4 & X86_CR4_OSXSAVE))
@@ -7220,6 +7241,8 @@ int kvm_arch_hardware_setup(void)
if (r != 0)
return r;
+ cr4_reserved_bits = kvm_host_cr4_reserved_bits(&boot_cpu_data);
+
kvm_init_msr_list();
return 0;
}
next prev parent reply other threads:[~2020-05-20 14:24 UTC|newest]
Thread overview: 112+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-05-20 14:13 [PATCH 3.16 00/99] 3.16.84-rc1 review Ben Hutchings
2020-05-20 14:13 ` [PATCH 3.16 01/99] fs/namespace.c: fix mountpoint reference counter race Ben Hutchings
2020-05-20 14:13 ` [PATCH 3.16 02/99] propagate_one(): mnt_set_mountpoint() needs mount_lock Ben Hutchings
2020-05-20 14:13 ` [PATCH 3.16 03/99] spi: spi-dw: Add lock protect dw_spi rx/tx to prevent concurrent calls Ben Hutchings
2020-05-20 14:13 ` [PATCH 3.16 04/99] padata: Remove unused but set variables Ben Hutchings
2020-05-20 14:13 ` [PATCH 3.16 05/99] padata: avoid race in reordering Ben Hutchings
2020-05-20 14:13 ` [PATCH 3.16 06/99] padata: get_next is never NULL Ben Hutchings
2020-05-20 14:13 ` [PATCH 3.16 07/99] padata: ensure the reorder timer callback runs on the correct CPU Ben Hutchings
2020-05-20 14:13 ` [PATCH 3.16 08/99] padata: ensure padata_do_serial() " Ben Hutchings
2020-05-20 14:13 ` [PATCH 3.16 09/99] padata: Replace delayed timer with immediate workqueue in padata_reorder Ben Hutchings
2020-05-20 14:13 ` [PATCH 3.16 10/99] padata: initialize pd->cpu with effective cpumask Ben Hutchings
2020-05-20 14:13 ` [PATCH 3.16 11/99] padata: Remove broken queue flushing Ben Hutchings
2020-05-20 14:13 ` [PATCH 3.16 12/99] padata: purge get_cpu and reorder_via_wq from padata_do_serial Ben Hutchings
2020-05-20 14:13 ` [PATCH 3.16 13/99] crypto: pcrypt - Fix user-after-free on module unload Ben Hutchings
2020-05-20 14:13 ` [PATCH 3.16 14/99] crypto: pcrypt - Do not clear MAY_SLEEP flag in original request Ben Hutchings
2020-05-20 14:13 ` [PATCH 3.16 15/99] padata: always acquire cpu_hotplug_lock before pinst->lock Ben Hutchings
2020-05-20 14:13 ` [PATCH 3.16 16/99] crypto: af_alg - Use bh_lock_sock in sk_destruct Ben Hutchings
2020-05-20 14:13 ` [PATCH 3.16 17/99] crypto: api - Check spawn->alg under lock in crypto_drop_spawn Ben Hutchings
2020-05-20 14:13 ` [PATCH 3.16 18/99] crypto: api - Fix race condition in crypto_spawn_alg Ben Hutchings
2020-05-20 14:13 ` [PATCH 3.16 19/99] mmc: spi: Toggle SPI polarity, do not hardcode it Ben Hutchings
2020-05-20 14:13 ` [PATCH 3.16 20/99] reiserfs: Fix memory leak of journal device string Ben Hutchings
2020-05-20 14:13 ` [PATCH 3.16 21/99] reiserfs: Fix spurious unlock in reiserfs_fill_super() error handling Ben Hutchings
2020-05-20 14:13 ` [PATCH 3.16 22/99] ath9k: fix storage endpoint lookup Ben Hutchings
2020-05-20 14:13 ` [PATCH 3.16 23/99] rsi: fix use-after-free on failed probe and unbind Ben Hutchings
2020-05-20 14:13 ` [PATCH 3.16 24/99] brcmfmac: Fix use after free in brcmf_sdio_readframes() Ben Hutchings
2020-05-20 14:13 ` [PATCH 3.16 25/99] brcmfmac: abort and release host after error Ben Hutchings
2020-05-20 14:13 ` [PATCH 3.16 26/99] brcmfmac: fix interface sanity check Ben Hutchings
2020-05-20 14:13 ` [PATCH 3.16 27/99] orinoco_usb: " Ben Hutchings
2020-05-20 14:13 ` [PATCH 3.16 28/99] rsi_91x_usb: " Ben Hutchings
2020-05-20 14:13 ` [PATCH 3.16 29/99] zd1211rw: fix storage endpoint lookup Ben Hutchings
2020-05-20 14:13 ` [PATCH 3.16 30/99] brcmfmac: Fix memory leak in brcmf_usbdev_qinit Ben Hutchings
2020-05-20 14:13 ` [PATCH 3.16 31/99] crypto: picoxcell - adjust the position of tasklet_init and fix missed tasklet_kill Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 32/99] scsi: qla2xxx: Fix mtcp dump collection failure Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 33/99] rtc: hym8563: Return -EINVAL if the time is known to be invalid Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 34/99] gianfar: Fix TX timestamping with a stacked DSA driver Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 35/99] pxa168fb: Fix the function used to release some memory in an error handling path Ben Hutchings
2020-05-21 14:09 ` Marion & Christophe JAILLET
2020-05-21 14:31 ` Marion & Christophe JAILLET
2020-05-21 20:28 ` Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 36/99] ALSA: sh: Fix compile warning wrt const Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 37/99] clk: tegra: Mark fuse clock as critical Ben Hutchings
2020-05-20 15:51 ` Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 38/99] ARM: tegra: Enable PLLP bypass during Tegra124 LP1 Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 39/99] media: iguanair: add sanity checks Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 40/99] media: iguanair: fix endpoint sanity check Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 41/99] ARM: dts: at91: sama5d3: fix maximum peripheral clock rates Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 42/99] ARM: dts: at91: sama5d3: define clock rate range for tcb1 Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 43/99] efi: Use early_mem*() instead of early_io*() Ben Hutchings
2020-05-20 15:53 ` Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 44/99] efi/x86: Map the entire EFI vendor string before copying it Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 45/99] PCI: Don't disable bridge BARs when assigning bus resources Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 46/99] power: supply: sbs-battery: Fix a signedness bug in sbs_get_battery_capacity() Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 47/99] dm space map common: fix to ensure new block isn't already in use Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 48/99] usb: dwc3: turn off VBUS when leaving host mode Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 49/99] usb: gadget: f_ncm: Use atomic_t to track in-flight request Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 50/99] usb: gadget: f_ecm: " Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 51/99] staging: wlan-ng: ensure error return is actually returned Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 52/99] nfs: NFS_SWAP should depend on SWAP Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 53/99] ubifs: Fix deadlock in concurrent bulk-read and writepage Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 54/99] x86/cpu: Update cached HLE state on write to TSX_CTRL_CPUID_CLEAR Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 55/99] jbd2: clear JBD2_ABORT flag before journal_reset to update log tail info when load journal Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 56/99] KVM: arm64: Only sign-extend MMIO up to register width Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 57/99] sparc32: fix struct ipc64_perm type definition Ben Hutchings
2020-05-20 14:14 ` Ben Hutchings [this message]
2020-05-20 14:14 ` [PATCH 3.16 59/99] KVM: nVMX: vmread should not set rflags to specify success in case of #PF Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 60/99] x86: kvm: avoid unused variable warning Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 61/99] KVM: x86/mmu: Apply max PA check for MMIO sptes to 32-bit KVM Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 62/99] USB: serial: ir-usb: add missing endpoint sanity check Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 63/99] USB: serial: ir-usb: fix link-speed handling Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 64/99] USB: serial: ir-usb: fix IrLAP framing Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 65/99] media: uvcvideo: Avoid cyclic entity chains due to malformed USB descriptors Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 66/99] KVM: PPC: Book3S HV: Uninit vCPU if vcore creation fails Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 67/99] KVM: PPC: Book3S PR: Free shared page if mmu initialization fails Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 68/99] KVM: x86: Free wbinvd_dirty_mask if vCPU creation fails Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 69/99] tracing: Fix very unlikely race of registering two stat tracers Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 70/99] tracing: Fix tracing_stat return values in error handling paths Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 71/99] jbd2: switch to use jbd2_journal_abort() when failed to submit the commit record Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 72/99] ext4, jbd2: ensure panic when aborting with zero errno Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 73/99] iwlegacy: ensure loop counter addr does not wrap and cause an infinite loop Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 74/99] CIFS: Fix task struct use-after-free on reconnect Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 75/99] net_sched: ematch: reject invalid TCF_EM_SIMPLE Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 76/99] KVM: x86: Protect x86_decode_insn from Spectre-v1/L1TF attacks Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 77/99] KVM: x86: Refactor picdev_write() to prevent " Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 78/99] KVM: x86: Protect ioapic_read_indirect() from " Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 79/99] KVM: x86: Protect ioapic_write_indirect() " Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 80/99] KVM: x86: Protect kvm_lapic_reg_write() " Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 81/99] kvm: x86: use macros to compute bank MSRs Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 82/99] KVM: x86: Protect MSR-based index computations from Spectre-v1/L1TF attacks in x86.c Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 83/99] KVM: x86: Protect DR-based index computations from Spectre-v1/L1TF attacks Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 84/99] KVM: Check for a bad hva before dropping into the ghc slow path Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 85/99] of: Add OF_DMA_DEFAULT_COHERENT & select it on powerpc Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 86/99] Btrfs: fix race between adding and putting tree mod seq elements and nodes Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 87/99] mm/mempolicy.c: fix out of bounds write in mpol_parse_str() Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 88/99] media/v4l2-core: set pages dirty upon releasing DMA buffers Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 89/99] tcp: clear tp->total_retrans in tcp_disconnect() Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 90/99] ALSA: dummy: Fix PCM format loop in proc output Ben Hutchings
2020-05-20 14:14 ` [PATCH 3.16 91/99] clocksource: Prevent double add_timer_on() for watchdog_timer Ben Hutchings
2020-05-20 14:15 ` [PATCH 3.16 92/99] cls_rsvp: fix rsvp_policy Ben Hutchings
2020-05-20 14:15 ` [PATCH 3.16 93/99] kconfig: fix broken dependency in randconfig-generated .config Ben Hutchings
2020-05-20 14:15 ` [PATCH 3.16 94/99] nfs: use kmap/kunmap directly Ben Hutchings
2020-05-20 14:15 ` [PATCH 3.16 95/99] NFS: Fix memory leaks and corruption in readdir Ben Hutchings
2020-05-20 14:15 ` [PATCH 3.16 96/99] NFS: Directory page cache pages need to be locked when read Ben Hutchings
2020-05-20 14:15 ` [PATCH 3.16 97/99] cifs: fail i/o on soft mounts if sessionsetup errors out Ben Hutchings
2020-05-20 14:15 ` [PATCH 3.16 98/99] bonding/alb: properly access headers in bond_alb_xmit() Ben Hutchings
2020-05-20 14:15 ` [PATCH 3.16 99/99] sunrpc: expiry_time should be seconds not timeval Ben Hutchings
2020-05-20 21:23 ` [PATCH 3.16 00/99] 3.16.84-rc1 review Guenter Roeck
2020-05-21 2:47 ` Chen-Yu Tsai
2020-05-21 7:40 ` Guenter Roeck
2020-05-21 20:22 ` Ben Hutchings
2020-05-21 20:20 ` Ben Hutchings
2020-05-21 22:37 ` Guenter Roeck
2020-05-22 0:00 ` Ben Hutchings
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=lsq.1589984008.288512231@decadent.org.uk \
--to=ben@decadent.org.uk \
--cc=akpm@linux-foundation.org \
--cc=jun.nakajima@intel.com \
--cc=kda@linux-powerpc.org \
--cc=linux-kernel@vger.kernel.org \
--cc=pbonzini@redhat.com \
--cc=sean.j.christopherson@intel.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;
as well as URLs for NNTP newsgroup(s).