linux.git/arch/x86/kernel/cpu/vmware.c, branch v6.6 Linux kernel source tree sched/clock/x86: Mark sched_clock() noinstr 2023-01-31T14:01:47+00:00 Peter Zijlstra peterz@infradead.org 2023-01-26T15:08:36+00:00 8739c6811572b087decd561f96382087402cc343 In order to use sched_clock() from noinstr code, mark it and all it's implenentations noinstr. The whole pvclock thing (used by KVM/Xen) is a bit of a pain, since it calls out to watchdogs, create a pvclock_clocksource_read_nowd() variant doesn't do that and can be noinstr. Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by: Ingo Molnar <mingo@kernel.org> Link: https://lore.kernel.org/r/20230126151323.702003578@infradead.org 
In order to use sched_clock() from noinstr code, mark it and all it's
implenentations noinstr.

The whole pvclock thing (used by KVM/Xen) is a bit of a pain,
since it calls out to watchdogs, create a
pvclock_clocksource_read_nowd() variant doesn't do that and can be
noinstr.

Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Link: https://lore.kernel.org/r/20230126151323.702003578@infradead.org
x86/vmware: Use BIT() macro for shifting 2022-06-22T09:23:14+00:00 Shreenidhi Shedi sshedi@vmware.com 2022-06-01T10:18:20+00:00 4745ca43104b422354f06dc814d3f13661f217af VMWARE_CMD_VCPU_RESERVED is bit 31 and that would mean undefined behavior when shifting an int but the kernel is built with -fno-strict-overflow which will wrap around using two's complement. Use the BIT() macro to improve readability and avoid any potential overflow confusion because it uses an unsigned long. [ bp: Clarify commit message. ] Signed-off-by: Shreenidhi Shedi <sshedi@vmware.com> Signed-off-by: Borislav Petkov <bp@suse.de> Reviewed-by: Srivatsa S. Bhat (VMware) <srivatsa@csail.mit.edu> Link: https://lore.kernel.org/r/20220601101820.535031-1-sshedi@vmware.com 
VMWARE_CMD_VCPU_RESERVED is bit 31 and that would mean undefined
behavior when shifting an int but the kernel is built with
-fno-strict-overflow which will wrap around using two's complement.

Use the BIT() macro to improve readability and avoid any potential
overflow confusion because it uses an unsigned long.

  [ bp: Clarify commit message. ]

Signed-off-by: Shreenidhi Shedi <sshedi@vmware.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Srivatsa S. Bhat (VMware) <srivatsa@csail.mit.edu>
Link: https://lore.kernel.org/r/20220601101820.535031-1-sshedi@vmware.com
Merge tag 'x86_vmware_for_v5.13' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip 2021-04-26T16:13:43+00:00 Linus Torvalds torvalds@linux-foundation.org 2021-04-26T16:13:43+00:00 47e12f14108170c36af9a0ee2271abb3dfb73e0e Pull x86 vmware guest update from Borislav Petkov: "Have vmware guests skip the refined TSC calibration when the TSC frequency has been retrieved from the hypervisor" * tag 'x86_vmware_for_v5.13' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: x86/vmware: Avoid TSC recalibration when frequency is known 
Pull x86 vmware guest update from Borislav Petkov:
 "Have vmware guests skip the refined TSC calibration when the TSC
  frequency has been retrieved from the hypervisor"

* tag 'x86_vmware_for_v5.13' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/vmware: Avoid TSC recalibration when frequency is known
x86/vmware: Avoid TSC recalibration when frequency is known 2021-03-28T19:11:43+00:00 Alexey Makhalov amakhalov@vmware.com 2021-01-05T00:47:52+00:00 0b4a285e2c65c2c9449c6eccb87298e385213e7b When the TSC frequency is known because it is retrieved from the hypervisor, skip TSC refined calibration by setting X86_FEATURE_TSC_KNOWN_FREQ. Signed-off-by: Alexey Makhalov <amakhalov@vmware.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de> Link: https://lore.kernel.org/r/20210105004752.131069-1-amakhalov@vmware.com 
When the TSC frequency is known because it is retrieved from the
hypervisor, skip TSC refined calibration by setting X86_FEATURE_TSC_KNOWN_FREQ.

Signed-off-by: Alexey Makhalov <amakhalov@vmware.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20210105004752.131069-1-amakhalov@vmware.com
x86/paravirt: Switch time pvops functions to use static_call() 2021-03-11T15:17:52+00:00 Juergen Gross jgross@suse.com 2021-03-11T14:23:09+00:00 a0e2bf7cb7006b5a58ee81f4da4fe575875f2781 The time pvops functions are the only ones left which might be used in 32-bit mode and which return a 64-bit value. Switch them to use the static_call() mechanism instead of pvops, as this allows quite some simplification of the pvops implementation. Signed-off-by: Juergen Gross <jgross@suse.com> Signed-off-by: Borislav Petkov <bp@suse.de> Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org> Link: https://lkml.kernel.org/r/20210311142319.4723-5-jgross@suse.com 
The time pvops functions are the only ones left which might be
used in 32-bit mode and which return a 64-bit value.

Switch them to use the static_call() mechanism instead of pvops, as
this allows quite some simplification of the pvops implementation.

Signed-off-by: Juergen Gross <jgross@suse.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20210311142319.4723-5-jgross@suse.com
KVM: SVM: Add GHCB accessor functions for retrieving fields 2020-12-14T16:09:32+00:00 Tom Lendacky thomas.lendacky@amd.com 2020-12-10T17:09:39+00:00 0f60bde15ee11d03b6143f567cf840d30bf1b588 Update the GHCB accessor functions to add functions for retrieve GHCB fields by name. Update existing code to use the new accessor functions. Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com> Message-Id: <664172c53a5fb4959914e1a45d88e805649af0ad.1607620209.git.thomas.lendacky@amd.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> 
Update the GHCB accessor functions to add functions for retrieve GHCB
fields by name. Update existing code to use the new accessor functions.

Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
Message-Id: <664172c53a5fb4959914e1a45d88e805649af0ad.1607620209.git.thomas.lendacky@amd.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
x86/vmware: Add VMware-specific handling for VMMCALL under SEV-ES 2020-09-09T09:33:20+00:00 Doug Covelli dcovelli@vmware.com 2020-09-07T13:16:05+00:00 1a222de8dcfb903d039810b0823570ee0be4e6c6 Add VMware-specific handling for #VC faults caused by VMMCALL instructions. Signed-off-by: Doug Covelli <dcovelli@vmware.com> Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com> [ jroedel@suse.de: - Adapt to different paravirt interface ] Co-developed-by: Joerg Roedel <jroedel@suse.de> Signed-off-by: Joerg Roedel <jroedel@suse.de> Signed-off-by: Borislav Petkov <bp@suse.de> Link: https://lkml.kernel.org/r/20200907131613.12703-65-joro@8bytes.org 
Add VMware-specific handling for #VC faults caused by VMMCALL
instructions.

Signed-off-by: Doug Covelli <dcovelli@vmware.com>
Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
[ jroedel@suse.de: - Adapt to different paravirt interface ]
Co-developed-by: Joerg Roedel <jroedel@suse.de>
Signed-off-by: Joerg Roedel <jroedel@suse.de>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lkml.kernel.org/r/20200907131613.12703-65-joro@8bytes.org
x86/vmware: Use bool type for vmw_sched_clock 2020-03-24T09:29:22+00:00 Alexey Makhalov amakhalov@vmware.com 2020-03-23T19:57:07+00:00 8fefe9dacdb0a1347d3dac871bb1bba3cbc32945 To be aligned with other bool variables. Signed-off-by: Alexey Makhalov <amakhalov@vmware.com> Signed-off-by: Borislav Petkov <bp@suse.de> Reviewed-by: Thomas Gleixner <tglx@linutronix.de> Link: https://lkml.kernel.org/r/20200323195707.31242-6-amakhalov@vmware.com 
To be aligned with other bool variables.

Signed-off-by: Alexey Makhalov <amakhalov@vmware.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lkml.kernel.org/r/20200323195707.31242-6-amakhalov@vmware.com
x86/vmware: Enable steal time accounting 2020-03-24T09:06:27+00:00 Alexey Makhalov amakhalov@vmware.com 2020-03-23T19:57:06+00:00 e73a8f38f82dd1c41b70a06556bea7dc250cc384 Set paravirt_steal_rq_enabled if steal clock present. paravirt_steal_rq_enabled is used in sched/core.c to adjust task progress by offsetting stolen time. Use 'no-steal-acc' off switch (share same name with KVM) to disable steal time accounting. Signed-off-by: Alexey Makhalov <amakhalov@vmware.com> Signed-off-by: Borislav Petkov <bp@suse.de> Reviewed-by: Thomas Hellstrom <thellstrom@vmware.com> Reviewed-by: Thomas Gleixner <tglx@linutronix.de> Link: https://lkml.kernel.org/r/20200323195707.31242-5-amakhalov@vmware.com 
Set paravirt_steal_rq_enabled if steal clock present.
paravirt_steal_rq_enabled is used in sched/core.c to adjust task
progress by offsetting stolen time. Use 'no-steal-acc' off switch (share
same name with KVM) to disable steal time accounting.

Signed-off-by: Alexey Makhalov <amakhalov@vmware.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lkml.kernel.org/r/20200323195707.31242-5-amakhalov@vmware.com
x86/vmware: Add steal time clock support for VMware guests 2020-03-24T09:04:51+00:00 Alexey Makhalov amakhalov@vmware.com 2020-03-23T19:57:05+00:00 ab02bb3f55f58e7608a88188000c3353398ebe3b Steal time is the amount of CPU time needed by a guest virtual machine that is not provided by the host. Steal time occurs when the host allocates this CPU time elsewhere, for example, to another guest. Steal time can be enabled by adding the VM configuration option stealclock.enable = "TRUE". It is supported by VMs that run hardware version 13 or newer. Introduce the VMware steal time infrastructure. The high level code (such as enabling, disabling and hot-plug routines) was derived from KVM. [ Tomer: use READ_ONCE macros and 32bit guests support. ] [ bp: Massage. ] Co-developed-by: Tomer Zeltzer <tomerr90@gmail.com> Signed-off-by: Alexey Makhalov <amakhalov@vmware.com> Signed-off-by: Tomer Zeltzer <tomerr90@gmail.com> Signed-off-by: Borislav Petkov <bp@suse.de> Reviewed-by: Thomas Hellstrom <thellstrom@vmware.com> Reviewed-by: Thomas Gleixner <tglx@linutronix.de> Link: https://lkml.kernel.org/r/20200323195707.31242-4-amakhalov@vmware.com 
Steal time is the amount of CPU time needed by a guest virtual machine
that is not provided by the host. Steal time occurs when the host
allocates this CPU time elsewhere, for example, to another guest.

Steal time can be enabled by adding the VM configuration option
stealclock.enable = "TRUE". It is supported by VMs that run hardware
version 13 or newer.

Introduce the VMware steal time infrastructure. The high level code
(such as enabling, disabling and hot-plug routines) was derived from KVM.

 [ Tomer: use READ_ONCE macros and 32bit guests support. ]
 [ bp: Massage. ]

Co-developed-by: Tomer Zeltzer <tomerr90@gmail.com>
Signed-off-by: Alexey Makhalov <amakhalov@vmware.com>
Signed-off-by: Tomer Zeltzer <tomerr90@gmail.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Thomas Hellstrom <thellstrom@vmware.com>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lkml.kernel.org/r/20200323195707.31242-4-amakhalov@vmware.com