linux.git/arch/x86/kernel/cpu/intel.c, branch v6.0

x86/bus_lock: Don't assume the init value of DEBUGCTLMSR.BUS_LOCK_DETECT to be zero

2022-08-02T11:42:00+00:00

It's possible that this kernel has been kexec'd from a kernel that
enabled bus lock detection, or (hypothetically) BIOS/firmware has set
DEBUGCTLMSR_BUS_LOCK_DETECT.

Disable bus lock detection explicitly if not wanted.

Fixes: ebb1064e7c2e ("x86/traps: Handle #DB for bus lock")
Signed-off-by: Chenyi Qiang 
Signed-off-by: Ingo Molnar 
Reviewed-by: Tony Luck 
Link: https://lore.kernel.org/r/20220802033206.21333-1-chenyi.qiang@intel.com

Merge tag 'x86_cpu_for_v6.0_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

2022-08-01T16:49:29+00:00

Pull x86 cpu updates from Borislav Petkov:

 - Remove the vendor check when selecting MWAIT as the default idle
   state

 - Respect idle=nomwait when supplied on the kernel cmdline

 - Two small cleanups

* tag 'x86_cpu_for_v6.0_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/cpu: Use MSR_IA32_MISC_ENABLE constants
  x86: Fix comment for X86_FEATURE_ZEN
  x86: Remove vendor checks from prefer_mwait_c1_over_halt
  x86: Handle idle=nomwait cmdline properly for x86_idle

x86/cpu: Use MSR_IA32_MISC_ENABLE constants

2022-07-19T18:53:10+00:00

Instead of the magic numbers 1<<11 and 1<<12 use the constants
from msr-index.h.  This makes it obvious where those bits
of MSR_IA32_MISC_ENABLE are consumed (and in fact that Linux
consumes them at all) to simple minds that grep for
MSR_IA32_MISC_ENABLE_.*_UNAVAIL.

Signed-off-by: Paolo Bonzini 
Signed-off-by: Borislav Petkov 
Link: https://lore.kernel.org/r/20220719174714.2410374-1-pbonzini@redhat.com

Merge tag 'platform-drivers-x86-v5.19-1' of git://git.kernel.org/pub/scm/linux/kernel/git/pdx86/platform-drivers-x86

2022-05-24T03:38:39+00:00

Pull x86 platform driver updates from Hans de Goede:
 "This includes some small changes to kernel/stop_machine.c and arch/x86
  which are deps of the new Intel IFS support.

  Highlights:

   - New drivers:
       - Intel "In Field Scan" (IFS) support
       - Winmate FM07/FM07P buttons
       - Mellanox SN2201 support

   -  AMD PMC driver enhancements

   -  Lots of various other small fixes and hardware-id additions"

* tag 'platform-drivers-x86-v5.19-1' of git://git.kernel.org/pub/scm/linux/kernel/git/pdx86/platform-drivers-x86: (54 commits)
  platform/x86/intel/ifs: Add CPU_SUP_INTEL dependency
  platform/x86: intel_cht_int33fe: Set driver data
  platform/x86: intel-hid: fix _DSM function index handling
  platform/x86: toshiba_acpi: use kobj_to_dev()
  platform/x86: samsung-laptop: use kobj_to_dev()
  platform/x86: gigabyte-wmi: Add support for Z490 AORUS ELITE AC and X570 AORUS ELITE WIFI
  tools/power/x86/intel-speed-select: Fix warning for perf_cap.cpu
  tools/power/x86/intel-speed-select: Display error on turbo mode disabled
  Documentation: In-Field Scan
  platform/x86/intel/ifs: add ABI documentation for IFS
  trace: platform/x86/intel/ifs: Add trace point to track Intel IFS operations
  platform/x86/intel/ifs: Add IFS sysfs interface
  platform/x86/intel/ifs: Add scan test support
  platform/x86/intel/ifs: Authenticate and copy to secured memory
  platform/x86/intel/ifs: Check IFS Image sanity
  platform/x86/intel/ifs: Read IFS firmware image
  platform/x86/intel/ifs: Add stub driver for In-Field Scan
  stop_machine: Add stop_core_cpuslocked() for per-core operations
  x86/msr-index: Define INTEGRITY_CAPABILITIES MSR
  x86/microcode/intel: Expose collect_cpu_info_early() for IFS
  ...

Merge tag 'x86_misc_for_v5.19_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

2022-05-24T02:32:59+00:00

Pull misc x86 updates from Borislav Petkov:
 "A variety of fixes which don't fit any other tip bucket:

   - Remove unnecessary function export

   - Correct asm constraint

   - Fix __setup handlers retval"

* tag 'x86_misc_for_v5.19_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/mm: Cleanup the control_va_addr_alignment() __setup handler
  x86: Fix return value of __setup handlers
  x86/delay: Fix the wrong asm constraint in delay_loop()
  x86/amd_nb: Unexport amd_cache_northbridges()

x86/microcode/intel: Expose collect_cpu_info_early() for IFS

2022-05-12T13:35:29+00:00

IFS is a CPU feature that allows a binary blob, similar to microcode,
to be loaded and consumed to perform low level validation of CPU
circuitry. In fact, it carries the same Processor Signature
(family/model/stepping) details that are contained in Intel microcode
blobs.

In support of an IFS driver to trigger loading, validation, and running
of these tests blobs, make the functionality of cpu_signatures_match()
and collect_cpu_info_early() available outside of the microcode driver.

Add an "intel_" prefix and drop the "_early" suffix from
collect_cpu_info_early() and EXPORT_SYMBOL_GPL() it. Add
declaration to x86 

Make cpu_signatures_match() an inline function in x86 ,
and also give it an "intel_" prefix.

No functional change intended.

Reviewed-by: Dan Williams 
Signed-off-by: Jithu Joseph 
Co-developed-by: Tony Luck 
Signed-off-by: Tony Luck 
Reviewed-by: Thomas Gleixner 
Acked-by: Borislav Petkov 
Reviewed-by: Greg Kroah-Hartman 
Link: https://lore.kernel.org/r/20220506225410.1652287-2-tony.luck@intel.com
Signed-off-by: Hans de Goede

x86: Fix return value of __setup handlers

2022-05-04T14:47:57+00:00

__setup() handlers should return 1 to obsolete_checksetup() in
init/main.c to indicate that the boot option has been handled. A return
of 0 causes the boot option/value to be listed as an Unknown kernel
parameter and added to init's (limited) argument (no '=') or environment
(with '=') strings. So return 1 from these x86 __setup handlers.

Examples:

  Unknown kernel command line parameters "apicpmtimer
    BOOT_IMAGE=/boot/bzImage-517rc8 vdso=1 ring3mwait=disable", will be
    passed to user space.

  Run /sbin/init as init process
   with arguments:
     /sbin/init
     apicpmtimer
   with environment:
     HOME=/
     TERM=linux
     BOOT_IMAGE=/boot/bzImage-517rc8
     vdso=1
     ring3mwait=disable

Fixes: 2aae950b21e4 ("x86_64: Add vDSO for x86-64 with gettimeofday/clock_gettime/getcpu")
Fixes: 77b52b4c5c66 ("x86: add "debugpat" boot option")
Fixes: e16fd002afe2 ("x86/cpufeature: Enable RING3MWAIT for Knights Landing")
Fixes: b8ce33590687 ("x86_64: convert to clock events")
Reported-by: Igor Zhbanov 
Signed-off-by: Randy Dunlap 
Signed-off-by: Borislav Petkov 
Link: https://lore.kernel.org/r/64644a2f-4a20-bab3-1e15-3b2cdd0defe3@omprussia.ru
Link: https://lore.kernel.org/r/20220314012725.26661-1-rdunlap@infradead.org

x86/split_lock: Enable the split lock feature on Raptor Lake

2022-05-04T10:11:25+00:00

Raptor Lake supports the split lock detection feature. Add it to
the split_lock_cpu_ids[] array.

Signed-off-by: Tony Luck 
Signed-off-by: Borislav Petkov 
Acked-by: Thomas Gleixner 
Link: https://lore.kernel.org/r/20220427231059.293086-1-tony.luck@intel.com

x86/split-lock: Remove unused TIF_SLD bit

2022-04-27T13:43:39+00:00

Changes to the "warn" mode of split lock handling mean that TIF_SLD is
never set.

Remove the bit, and the functions that use it.

Signed-off-by: Tony Luck 
Signed-off-by: Thomas Gleixner 
Link: https://lore.kernel.org/r/20220310204854.31752-3-tony.luck@intel.com

x86/split_lock: Make life miserable for split lockers

2022-04-27T13:43:38+00:00

In https://lore.kernel.org/all/87y22uujkm.ffs@tglx/ Thomas
said:

  Its's simply wishful thinking that stuff gets fixed because of a
  WARN_ONCE(). This has never worked. The only thing which works is to
  make stuff fail hard or slow it down in a way which makes it annoying
  enough to users to complain.

He was talking about WBINVD. But it made me think about how we use the
split lock detection feature in Linux.

Existing code has three options for applications:

 1) Don't enable split lock detection (allow arbitrary split locks)
 2) Warn once when a process uses split lock, but let the process
    keep running with split lock detection disabled
 3) Kill process that use split locks

Option 2 falls into the "wishful thinking" territory that Thomas warns does
nothing. But option 3 might not be viable in a situation with legacy
applications that need to run.

Hence make option 2 much stricter to "slow it down in a way which makes
it annoying".

Primary reason for this change is to provide better quality of service to
the rest of the applications running on the system. Internal testing shows
that even with many processes splitting locks, performance for the rest of
the system is much more responsive.

The new "warn" mode operates like this.  When an application tries to
execute a bus lock the #AC handler.

 1) Delays (interruptibly) 10 ms before moving to next step.

 2) Blocks (interruptibly) until it can get the semaphore
	If interrupted, just return. Assume the signal will either
	kill the task, or direct execution away from the instruction
	that is trying to get the bus lock.
 3) Disables split lock detection for the current core
 4) Schedules a work queue to re-enable split lock detect in 2 jiffies
 5) Returns

The work queue that re-enables split lock detection also releases the
semaphore.

There is a corner case where a CPU may be taken offline while split lock
detection is disabled. A CPU hotplug handler handles this case.

Old behaviour was to only print the split lock warning on the first
occurrence of a split lock from a task. Preserve that by adding a flag to
the task structure that suppresses subsequent split lock messages from that
task.

Signed-off-by: Tony Luck 
Signed-off-by: Thomas Gleixner 
Link: https://lore.kernel.org/r/20220310204854.31752-2-tony.luck@intel.com