xref: /freebsd/share/man/man4/hwpmc.4 (revision 89240e0ab60a29a47c9af9f952a09a7f2e1bb72c)

.\" Copyright (c) 2003-2008 Joseph Koshy
.\" Copyright (c) 2007,2023 The FreeBSD Foundation
.\"
.\" Portions of this software were developed by A. Joseph Koshy under
.\" sponsorship from the FreeBSD Foundation and Google, Inc.
.\"
.\" Portions of this documentation were written by Mitchell Horne
.\" under sponsorship from the FreeBSD Foundation.
.\"
.\" Redistribution and use in source and binary forms, with or without
.\" modification, are permitted provided that the following conditions
.\" are met:
.\" 1. Redistributions of source code must retain the above copyright
.\"    notice, this list of conditions and the following disclaimer.
.\" 2. Redistributions in binary form must reproduce the above copyright
.\"    notice, this list of conditions and the following disclaimer in the
.\"    documentation and/or other materials provided with the distribution.
.\"
.\" THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
.\" ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
.\" SUCH DAMAGE.
.\"
.\" $FreeBSD$
.\"
.Dd July 8, 2023
.Dt HWPMC 4
.Os
.Sh NAME
.Nm hwpmc
.Nd "Hardware Performance Monitoring Counter support"
.Sh SYNOPSIS
The following option must be present in the kernel configuration file:
.Bd -ragged -offset indent
.Cd "options HWPMC_HOOKS"
.Ed
.Pp
Additionally, for i386 systems:
.Bd -ragged -offset indent
.Cd "device apic"
.Ed
.Pp
To load the driver as a module at boot time:
.Bd -literal -offset indent
sysrc kld_list+=hwpmc
.Ed
.Pp
Alternatively, to compile the driver into the kernel:
.Bd -ragged -offset indent
.Cd "device hwpmc"
.Ed
.Pp
To enable debugging features
.Po see
.Sx DEBUGGING
.Pc :
.Bd -ragged -offset indent
.Cd "options KTR"
.Cd "options KTR_COMPILE=(KTR_SUBSYS)"
.Cd "options KTR_MASK=(KTR_SUBSYS)"
.Cd "options HWPMC_DEBUG"
.Ed
.Sh DESCRIPTION
The
.Nm
driver virtualizes the hardware performance monitoring facilities in
modern CPUs and provides support for using these facilities from
user level processes.
.Pp
The driver supports multi-processor systems.
.Pp
PMCs are allocated using the
.Dv PMC_OP_PMCALLOCATE
request.
A successful
.Dv PMC_OP_PMCALLOCATE
request will return a handle to the requesting process.
Subsequent operations on the allocated PMC use this handle to denote
the specific PMC.
A process that has successfully allocated a PMC is termed an
.Dq "owner process" .
.Pp
PMCs may be allocated with process or system scope.
.Bl -tag -width ".Em Process-scope"
.It Em "Process-scope"
The PMC is active only when a thread belonging
to a process it is attached to is scheduled on a CPU.
.It Em "System-scope"
The PMC operates independently of processes and
measures hardware events for the system as a whole.
.El
.Pp
PMCs may be allocated for counting or for sampling:
.Bl -tag -width ".Em Counting"
.It Em Counting
In counting modes, the PMCs count hardware events.
These counts are retrievable using the
.Dv PMC_OP_PMCREAD
system call on all architectures.
Some architectures offer faster methods of reading these counts.
.It Em Sampling
In sampling modes, the PMCs are configured to sample the CPU
instruction pointer (and optionally to capture the call chain leading
up to the sampled instruction pointer) after a configurable number of
hardware events have been observed.
Instruction pointer samples and call chain records are usually
directed to a log file for subsequent analysis.
.El
.Pp
Scope and operational mode are orthogonal; a PMC may thus be
configured to operate in one of the following four modes:
.Bl -tag -width indent
.It Process-scope, counting
These PMCs count hardware events whenever a thread in their attached process is
scheduled on a CPU.
These PMCs normally count from zero, but the initial count may be
set using the
.Dv PMC_OP_SETCOUNT
operation.
Applications can read the value of the PMC anytime using the
.Dv PMC_OP_PMCRW
operation.
.It Process-scope, sampling
These PMCs sample the target processes instruction pointer after they
have seen the configured number of hardware events.
The PMCs only count events when a thread belonging to their attached
process is active.
The desired frequency of sampling is set using the
.Dv PMC_OP_SETCOUNT
operation prior to starting the PMC.
Log files are configured using the
.Dv PMC_OP_CONFIGURELOG
operation.
.It System-scope, counting
These PMCs count hardware events seen by them independent of the
processes that are executing.
The current count on these PMCs can be read using the
.Dv PMC_OP_PMCRW
request.
These PMCs normally count from zero, but the initial count may be
set using the
.Dv PMC_OP_SETCOUNT
operation.
.It System-scope, sampling
These PMCs will periodically sample the instruction pointer of the CPU
they are allocated on, and will write the sample to a log for further
processing.
The desired frequency of sampling is set using the
.Dv PMC_OP_SETCOUNT
operation prior to starting the PMC.
Log files are configured using the
.Dv PMC_OP_CONFIGURELOG
operation.
.Pp
System-wide statistical sampling can only be enabled by a process with
super-user privileges.
.El
.Pp
Processes are allowed to allocate as many PMCs as the hardware and
current operating conditions permit.
Processes may mix allocations of system-wide and process-private
PMCs.
Multiple processes may be using PMCs simultaneously.
.Pp
Allocated PMCs are started using the
.Dv PMC_OP_PMCSTART
operation, and stopped using the
.Dv PMC_OP_PMCSTOP
operation.
Stopping and starting a PMC is permitted at any time the owner process
has a valid handle to the PMC.
.Pp
Process-private PMCs need to be attached to a target process before
they can be used.
Attaching a process to a PMC is done using the
.Dv PMC_OP_PMCATTACH
operation.
An already attached PMC may be detached from its target process
using the converse
.Dv PMC_OP_PMCDETACH
operation.
Issuing a
.Dv PMC_OP_PMCSTART
operation on an as yet unattached PMC will cause it to be attached
to its owner process.
The following rules determine whether a given process may attach
a PMC to another target process:
.Bl -bullet -compact
.It
A non-jailed process with super-user privileges is allowed to attach
to any other process in the system.
.It
Other processes are only allowed to attach to targets that they would
be able to attach to for debugging (as determined by
.Xr p_candebug 9 ) .
.El
.Pp
PMCs are released using
.Dv PMC_OP_PMCRELEASE .
After a successful
.Dv PMC_OP_PMCRELEASE
operation the handle to the PMC will become invalid.
.Ss Modifier Flags
The
.Dv PMC_OP_PMCALLOCATE
operation supports the following flags that modify the behavior
of an allocated PMC:
.Bl -tag -width indent
.It Dv PMC_F_CALLCHAIN
This modifier informs sampling PMCs to record a callchain when
capturing a sample.
The maximum depth to which call chains are recorded is specified
by the
.Va "kern.hwpmc.callchaindepth"
kernel tunable.
.It Dv PMC_F_DESCENDANTS
This modifier is valid only for a PMC being allocated in process-private
mode.
It signifies that the PMC will track hardware events for its
target process and the target's current and future descendants.
.It Dv PMC_F_LOG_PROCCSW
This modifier is valid only for a PMC being allocated in process-private
mode.
When this modifier is present, at every context switch,
.Nm
will log a record containing the number of hardware events
seen by the target process when it was scheduled on the CPU.
.It Dv PMC_F_LOG_PROCEXIT
This modifier is valid only for a PMC being allocated in process-private
mode.
With this modifier present,
.Nm
will maintain per-process counts for each target process attached to
a PMC.
At process exit time, a record containing the target process' PID and
the accumulated per-process count for that process will be written to the
configured log file.
.El
.Pp
Modifiers
.Dv PMC_F_LOG_PROCEXIT
and
.Dv PMC_F_LOG_PROCCSW
may be used in combination with modifier
.Dv PMC_F_DESCENDANTS
to track the behavior of complex pipelines of processes.
PMCs with modifiers
.Dv PMC_F_LOG_PROCEXIT
and
.Dv PMC_F_LOG_PROCCSW
cannot be started until their owner process has configured a log file.
.Ss Signals
The
.Nm
driver may deliver signals to processes that have allocated PMCs:
.Bl -tag -width ".Dv SIGBUS"
.It Dv SIGIO
A
.Dv PMC_OP_PMCRW
operation was attempted on a process-private PMC that does not have
attached target processes.
.It Dv SIGBUS
The
.Nm
driver is being unloaded from the kernel.
.El
.Ss PMC ROW DISPOSITIONS
A PMC row is defined as the set of PMC resources at the same hardware
address in the CPUs in a system.
Since process scope PMCs need to move between CPUs following their
target threads, allocation of a process scope PMC reserves all PMCs in
a PMC row for use only with process scope PMCs.
Accordingly a PMC row will be in one of the following dispositions:
.Bl -tag -width ".Dv PMC_DISP_STANDALONE" -compact
.It Dv PMC_DISP_FREE
Hardware counters in this row are free and may be use to satisfy
either of system scope or process scope allocation requests.
.It Dv PMC_DISP_THREAD
Hardware counters in this row are in use by process scope PMCs
and are only available for process scope allocation requests.
.It Dv PMC_DISP_STANDALONE
Some hardware counters in this row have been administratively
disabled or are in use by system scope PMCs.
Non-disabled hardware counters in such a row may be used
for satisfying system scope allocation requests.
No process scope PMCs will use hardware counters in this row.
.El
.Sh COMPATIBILITY
The API and ABI documented in this manual page may change in the future.
This interface is intended to be consumed by the
.Xr pmc 3
library; other consumers are unsupported.
Applications targeting PMCs should use the
.Xr pmc 3
library API.
.Sh PROGRAMMING API
The
.Nm
driver operates using a system call number that is dynamically
allotted to it when it is loaded into the kernel.
.Pp
The
.Nm
driver supports the following operations:
.Bl -tag -width indent
.It Dv PMC_OP_CONFIGURELOG
Configure a log file for PMCs that require a log file.
The
.Nm
driver will write log data to this file asynchronously.
If it encounters an error, logging will be stopped and the error code
encountered will be saved for subsequent retrieval by a
.Dv PMC_OP_FLUSHLOG
request.
.It Dv PMC_OP_FLUSHLOG
Transfer buffered log data inside
.Nm
to a configured output file.
This operation returns to the caller after the write operation
has returned.
The returned error code reflects any pending error state inside
.Nm .
.It Dv PMC_OP_GETCPUINFO
Retrieve information about the highest possible CPU number for the system,
and the number of hardware performance monitoring counters available per CPU.
.It Dv PMC_OP_GETDRIVERSTATS
Retrieve module statistics (for analyzing the behavior of
.Nm
itself).
.It Dv PMC_OP_GETMODULEVERSION
Retrieve the version number of API.
.It Dv PMC_OP_GETPMCINFO
Retrieve information about the current state of the PMCs on a
given CPU.
.It Dv PMC_OP_PMCADMIN
Set the administrative state (i.e., whether enabled or disabled) for
the hardware PMCs managed by the
.Nm
driver.
The invoking process needs to possess the
.Dv PRIV_PMC_MANAGE
privilege.
.It Dv PMC_OP_PMCALLOCATE
Allocate and configure a PMC.
On successful allocation, a handle to the PMC (a 32 bit value)
is returned.
.It Dv PMC_OP_PMCATTACH
Attach a process mode PMC to a target process.
The PMC will be active whenever a thread in the target process is
scheduled on a CPU.
.Pp
If the
.Dv PMC_F_DESCENDANTS
flag had been specified at PMC allocation time, then the PMC is
attached to all current and future descendants of the target process.
.It Dv PMC_OP_PMCDETACH
Detach a PMC from its target process.
.It Dv PMC_OP_PMCRELEASE
Release a PMC.
.It Dv PMC_OP_PMCRW
Read and write a PMC.
This operation is valid only for PMCs configured in counting modes.
.It Dv PMC_OP_SETCOUNT
Set the initial count (for counting mode PMCs) or the desired sampling
rate (for sampling mode PMCs).
.It Dv PMC_OP_PMCSTART
Start a PMC.
.It Dv PMC_OP_PMCSTOP
Stop a PMC.
.It Dv PMC_OP_WRITELOG
Insert a timestamped user record into the log file.
.El
.Ss i386 Specific API
Some i386 family CPUs support the RDPMC instruction which allows a
user process to read a PMC value without needing to invoke a
.Dv PMC_OP_PMCRW
operation.
On such CPUs, the machine address associated with an allocated PMC is
retrievable using the
.Dv PMC_OP_PMCX86GETMSR
system call.
.Bl -tag -width indent
.It Dv PMC_OP_PMCX86GETMSR
Retrieve the MSR (machine specific register) number associated with
the given PMC handle.
.Pp
The PMC needs to be in process-private mode and allocated without the
.Dv PMC_F_DESCENDANTS
modifier flag, and should be attached only to its owner process at the
time of the call.
.El
.Ss amd64 Specific API
AMD64 CPUs support the RDPMC instruction which allows a
user process to read a PMC value without needing to invoke a
.Dv PMC_OP_PMCRW
operation.
The machine address associated with an allocated PMC is
retrievable using the
.Dv PMC_OP_PMCX86GETMSR
system call.
.Bl -tag -width indent
.It Dv PMC_OP_PMCX86GETMSR
Retrieve the MSR (machine specific register) number associated with
the given PMC handle.
.Pp
The PMC needs to be in process-private mode and allocated without the
.Dv PMC_F_DESCENDANTS
modifier flag, and should be attached only to its owner process at the
time of the call.
.El
.Sh SYSCTL VARIABLES AND LOADER TUNABLES
The behavior of
.Nm
is influenced by the following
.Xr sysctl 8
and
.Xr loader 8
tunables:
.Bl -tag -width indent
.It Va kern.hwpmc.callchaindepth Pq integer, read-only
The maximum number of call chain records to capture per sample.
The default is 8.
.It Va kern.hwpmc.debugflags Pq string, read-write
(Only available if the
.Nm
driver was compiled with
.Fl DDEBUG . )
Control the verbosity of debug messages from the
.Nm
driver.
.It Va kern.hwpmc.hashsize Pq integer, read-only
The number of rows in the hash tables used to keep track of owner and
target processes.
The default is 16.
.It Va kern.hwpmc.logbuffersize Pq integer, read-only
The size in kilobytes of each log buffer used by
.Nm Ns 's
logging function.
The default buffer size is 4KB.
.It Va kern.hwpmc.mincount Pq integer, read-write
The minimum sampling rate for sampling mode PMCs.
The default count is 1000 events.
.It Va kern.hwpmc.mtxpoolsize Pq integer, read-only
The size of the spin mutex pool used by the PMC driver.
The default is 32.
.It Va kern.hwpmc.nbuffers_pcpu Pq integer, read-only
The number of log buffers used by
.Nm
for logging.
The default is 64.
.It Va kern.hwpmc.nsamples Pq integer, read-only
The number of entries in the per-CPU ring buffer used during sampling.
The default is 512.
.It Va security.bsd.unprivileged_syspmcs Pq boolean, read-write
If set to non-zero, allow unprivileged processes to allocate system-wide
PMCs.
The default value is 0.
.It Va security.bsd.unprivileged_proc_debug Pq boolean, read-write
If set to 0, the
.Nm
driver will only allow privileged processes to attach PMCs to other
processes.
.El
.Pp
These variables may be set in the kernel environment using
.Xr kenv 1
before
.Nm
is loaded.
.Sh IMPLEMENTATION NOTES
.Ss SMP Symmetry
The kernel driver requires all physical CPUs in an SMP system to have
identical performance monitoring counter hardware.
.Ss Sparse CPU Numbering
On platforms that sparsely number CPUs and which support hot-plugging
of CPUs, requests that specify non-existent or disabled CPUs will fail
with an error.
Applications allocating system-scope PMCs need to be aware of
the possibility of such transient failures.
.Ss x86 TSC Handling
Historically, on the x86 architecture,
.Fx
has permitted user processes running at a processor CPL of 3 to
read the TSC using the RDTSC instruction.
The
.Nm
driver preserves this behavior.
.Ss Intel P4/HTT Handling
On CPUs with HTT support, Intel P4 PMCs are capable of qualifying
only a subset of hardware events on a per-logical CPU basis.
Consequently, if HTT is enabled on a system with Intel Pentium P4
PMCs, then the
.Nm
driver will reject allocation requests for process-private PMCs that
request counting of hardware events that cannot be counted separately
for each logical CPU.
.Sh DIAGNOSTICS
.Bl -diag
.It "hwpmc: [class/npmc/capabilities]..."
Announce the presence of
.Va npmc
PMCs of class
.Va class ,
with capabilities described by bit string
.Va capabilities .
.It "hwpmc: kernel version (0x%x) does not match module version (0x%x)."
The module loading process failed because a version mismatch was detected
between the currently executing kernel and the module being loaded.
.It "hwpmc: this kernel has not been compiled with 'options HWPMC_HOOKS'."
The module loading process failed because the currently executing kernel
was not configured with the required configuration option
.Dv HWPMC_HOOKS .
.It "hwpmc: tunable hashsize=%d must be greater than zero."
A negative value was supplied for tunable
.Va kern.hwpmc.hashsize .
.It "hwpmc: tunable logbuffersize=%d must be greater than zero."
A negative value was supplied for tunable
.Va kern.hwpmc.logbuffersize .
.It "hwpmc: tunable nlogbuffers=%d must be greater than zero."
A negative value was supplied for tunable
.Va kern.hwpmc.nlogbuffers .
.It "hwpmc: tunable nsamples=%d out of range."
The value for tunable
.Va kern.hwpmc.nsamples
was negative or greater than 65535.
.El
.Sh DEBUGGING
The
.Nm
module can be configured to record trace entries using the
.Xr ktr 4
interface.
This is useful for debugging the driver's functionality, primarily during
development.
This debugging functionality is not enabled by default, and requires
recompiling the kernel and
.Nm
module after adding the following to the kernel config:
.Bd -literal -offset indent
.Cd options KTR
.Cd options KTR_COMPILE=(KTR_SUBSYS)
.Cd options KTR_MASK=(KTR_SUBSYS)
.Cd options HWPMC_DEBUG
.Ed
.Pp
This alone is not enough to enable tracing; one must also configure the
.Va kern.hwpmc.debugflags
.Xr sysctl 8
variable, which provides fine-grained control over which types of events are
logged to the trace buffer.
.Pp
.Nm
trace events are grouped by 'major' and 'minor' flag types.
The major flag names are as follows:
.Pp
.Bl -tag -width "sampling" -compact -offset indent
.It cpu
CPU events
.It csw
Context switch events
.It logging
Logging events
.It md
Machine-dependent/class-dependent events
.It module
Miscellaneous events
.It owner
PMC owner events
.It pmc
PMC management events
.It process
Process events
.It sampling
Sampling events
.El
.Pp
The minor flags for each major flag group can vary.
The individual minor flag names are:
.Bd -ragged -offset indent
allocaterow,
allocate,
attach,
bind,
config,
exec,
exit,
find,
flush,
fork,
getbuf,
hook,
init,
intr,
linktarget,
mayberemove,
ops,
read,
register,
release,
remove,
sample,
scheduleio,
select,
signal,
swi,
swo,
start,
stop,
syscall,
unlinktarget,
write
.Ed
.Pp
The
.Va kern.hwpmc.debugflags
variable is a string with a custom format.
The string should contain a space-separated list of event specifiers.
Each event specifier consists of the major flag name, followed by an equal sign
(=), followed by a comma-separated list of minor event types.
To track all events for a major group, an asterisk (*) can be given instead of
minor event names.
.Pp
For example, to trace all allocation and release events, set
.Va debugflags
as follows:
.Bd -literal -offset indent
kern.hwpmc.debugflags="pmc=allocate,release md=allocate,release"
.Ed
.Pp
To trace all events in the process and context switch major flag groups:
.Bd -literal -offset indent
kern.hwpmc.debugflags="process=* csw=*"
.Ed
.Pp
To disable all trace events, set the variable to an empty string.
.Bd -literal -offset indent
kern.hwpmc.debugflags=""
.Ed
.Pp
Trace events are recorded by
.Xr ktr 4 ,
and can be inspected at run-time using the
.Xr ktrdump 8
utility, or at the
.Xr ddb 4
prompt after a panic with the 'show ktr' command.
.Sh ERRORS
A command issued to the
.Nm
driver may fail with the following errors:
.Bl -tag -width Er
.It Bq Er EAGAIN
Helper process creation failed for a
.Dv PMC_OP_CONFIGURELOG
request due to a temporary resource shortage in the kernel.
.It Bq Er EBUSY
A
.Dv PMC_OP_CONFIGURELOG
operation was requested while an existing log was active.
.It Bq Er EBUSY
A DISABLE operation was requested using the
.Dv PMC_OP_PMCADMIN
request for a set of hardware resources currently in use for
process-private PMCs.
.It Bq Er EBUSY
A
.Dv PMC_OP_PMCADMIN
operation was requested on an active system mode PMC.
.It Bq Er EBUSY
A
.Dv PMC_OP_PMCATTACH
operation was requested for a target process that already had another
PMC using the same hardware resources attached to it.
.It Bq Er EBUSY
A
.Dv PMC_OP_PMCRW
request writing a new value was issued on a PMC that was active.
.It Bq Er EBUSY
A
.Dv PMC_OP_PMCSETCOUNT
request was issued on a PMC that was active.
.It Bq Er EDOOFUS
A
.Dv PMC_OP_PMCSTART
operation was requested without a log file being configured for a
PMC allocated with
.Dv PMC_F_LOG_PROCCSW
and
.Dv PMC_F_LOG_PROCEXIT
modifiers.
.It Bq Er EDOOFUS
A
.Dv PMC_OP_PMCSTART
operation was requested on a system-wide sampling PMC without a log
file being configured.
.It Bq Er EEXIST
A
.Dv PMC_OP_PMCATTACH
request was reissued for a target process that already is the target
of this PMC.
.It Bq Er EFAULT
A bad address was passed in to the driver.
.It Bq Er EINVAL
An invalid PMC handle was specified.
.It Bq Er EINVAL
An invalid CPU number was passed in for a
.Dv PMC_OP_GETPMCINFO
operation.
.It Bq Er EINVAL
The
.Ar pm_flags
argument to a
.Dv PMC_OP_CONFIGURELOG
request contained unknown flags.
.It Bq Er EINVAL
A
.Dv PMC_OP_CONFIGURELOG
request to de-configure a log file was issued without a log file
being configured.
.It Bq Er EINVAL
A
.Dv PMC_OP_FLUSHLOG
request was issued without a log file being configured.
.It Bq Er EINVAL
An invalid CPU number was passed in for a
.Dv PMC_OP_PMCADMIN
operation.
.It Bq Er EINVAL
An invalid operation request was passed in for a
.Dv PMC_OP_PMCADMIN
operation.
.It Bq Er EINVAL
An invalid PMC ID was passed in for a
.Dv PMC_OP_PMCADMIN
operation.
.It Bq Er EINVAL
A suitable PMC matching the parameters passed in to a
.Dv PMC_OP_PMCALLOCATE
request could not be allocated.
.It Bq Er EINVAL
An invalid PMC mode was requested during a
.Dv PMC_OP_PMCALLOCATE
request.
.It Bq Er EINVAL
An invalid CPU number was specified during a
.Dv PMC_OP_PMCALLOCATE
request.
.It Bq Er EINVAL
A CPU other than
.Dv PMC_CPU_ANY
was specified in a
.Dv PMC_OP_PMCALLOCATE
request for a process-private PMC.
.It Bq Er EINVAL
A CPU number of
.Dv PMC_CPU_ANY
was specified in a
.Dv PMC_OP_PMCALLOCATE
request for a system-wide PMC.
.It Bq Er EINVAL
The
.Ar pm_flags
argument to an
.Dv PMC_OP_PMCALLOCATE
request contained unknown flags.
.It Bq Er EINVAL
(On Intel Pentium 4 CPUs with HTT support)
A
.Dv PMC_OP_PMCALLOCATE
request for a process-private PMC was issued for an event that does
not support counting on a per-logical CPU basis.
.It Bq Er EINVAL
A PMC allocated for system-wide operation was specified with a
.Dv PMC_OP_PMCATTACH
or
.Dv PMC_OP_PMCDETACH
request.
.It Bq Er EINVAL
The
.Ar pm_pid
argument to a
.Dv PMC_OP_PMCATTACH
or
.Dv PMC_OP_PMCDETACH
request specified an illegal process ID.
.It Bq Er EINVAL
A
.Dv PMC_OP_PMCDETACH
request was issued for a PMC not attached to the target process.
.It Bq Er EINVAL
Argument
.Ar pm_flags
to a
.Dv PMC_OP_PMCRW
request contained illegal flags.
.It Bq Er EINVAL
A
.Dv PMC_OP_PMCX86GETMSR
operation was requested for a PMC not in process-virtual mode, or
for a PMC that is not solely attached to its owner process, or for
a PMC that was allocated with flag
.Dv PMC_F_DESCENDANTS .
.It Bq Er EINVAL
A
.Dv PMC_OP_WRITELOG
request was issued for an owner process without a log file
configured.
.It Bq Er ENOMEM
The system was not able to allocate kernel memory.
.It Bq Er ENOSYS
(On i386 and amd64 architectures)
A
.Dv PMC_OP_PMCX86GETMSR
operation was requested for hardware that does not support reading
PMCs directly with the RDPMC instruction.
.It Bq Er ENXIO
A
.Dv PMC_OP_GETPMCINFO
operation was requested for an absent or disabled CPU.
.It Bq Er ENXIO
A
.Dv PMC_OP_PMCALLOCATE
operation specified allocation of a system-wide PMC on an absent or
disabled CPU.
.It Bq Er ENXIO
A
.Dv PMC_OP_PMCSTART
or
.Dv PMC_OP_PMCSTOP
request was issued for a system-wide PMC that was allocated on a CPU
that is currently absent or disabled.
.It Bq Er EOPNOTSUPP
A
.Dv PMC_OP_PMCALLOCATE
request was issued for PMC capabilities not supported
by the specified PMC class.
.It Bq Er EOPNOTSUPP
(i386 architectures)
A sampling mode PMC was requested on a CPU lacking an APIC.
.It Bq Er EPERM
A
.Dv PMC_OP_PMCADMIN
request was issued by a process without super-user
privilege or by a jailed super-user process.
.It Bq Er EPERM
A
.Dv PMC_OP_PMCATTACH
operation was issued for a target process that the current process
does not have permission to attach to.
.It Bq Er EPERM
(i386 and amd64 architectures)
A
.Dv PMC_OP_PMCATTACH
operation was issued on a PMC whose MSR has been retrieved using
.Dv PMC_OP_PMCX86GETMSR .
.It Bq Er ESRCH
A process issued a PMC operation request without having allocated any
PMCs.
.It Bq Er ESRCH
A process issued a PMC operation request after the PMC was detached
from all of its target processes.
.It Bq Er ESRCH
A
.Dv PMC_OP_PMCATTACH
or
.Dv PMC_OP_PMCDETACH
request specified a non-existent process ID.
.It Bq Er ESRCH
The target process for a
.Dv PMC_OP_PMCDETACH
operation is not being monitored by
.Nm .
.El
.Sh SEE ALSO
.Xr kenv 1 ,
.Xr pmc 3 ,
.Xr pmclog 3 ,
.Xr ddb 4 ,
.Xr ktr 4 ,
.Xr kldload 8 ,
.Xr ktrdump 8 ,
.Xr pmccontrol 8 ,
.Xr pmcstat 8 ,
.Xr sysctl 8 ,
.Xr kproc_create 9 ,
.Xr p_candebug 9
.Sh HISTORY
The
.Nm
driver first appeared in
.Fx 6.0 .
.Sh AUTHORS
The
.Nm
driver was written by
.An Joseph Koshy Aq Mt jkoshy@FreeBSD.org .
.Sh BUGS
The driver samples the state of the kernel's logical processor support
at the time of initialization (i.e., at module load time).
On CPUs supporting logical processors, the driver could misbehave if
logical processors are subsequently enabled or disabled while the
driver is active.
.Pp
On the i386 architecture, the driver requires that the local APIC on the
CPU be enabled for sampling mode to be supported.
Many single-processor motherboards keep the APIC disabled in BIOS; on
such systems
.Nm
will not support sampling PMCs.
.Sh SECURITY CONSIDERATIONS
PMCs may be used to monitor the actual behavior of the system on hardware.
In situations where this constitutes an undesirable information leak,
the following options are available:
.Bl -enum
.It
Set the
.Xr sysctl 8
tunable
.Va security.bsd.unprivileged_syspmcs
to 0.
This ensures that unprivileged processes cannot allocate system-wide
PMCs and thus cannot observe the hardware behavior of the system
as a whole.
This tunable may also be set at boot time using
.Xr loader 8 ,
or with
.Xr kenv 1
prior to loading the
.Nm
driver into the kernel.
.It
Set the
.Xr sysctl 8
tunable
.Va security.bsd.unprivileged_proc_debug
to 0.
This will ensure that an unprivileged process cannot attach a PMC
to any process other than itself and thus cannot observe the hardware
behavior of other processes with the same credentials.
.El
.Pp
System administrators should note that on IA-32 platforms
.Fx
makes the content of the IA-32 TSC counter available to all processes
via the RDTSC instruction.