The code tracks a counter which is the number of events until the next
sample. On context switch in, it loads the saved counter. On context
switch out, it tries to calculate a new saved counter.
Problems:
1. The saved counter was shared by all threads in a process. However, this
means that all threads would be initially loaded with the same saved
counter. However, that could result in sampling more often than once every
X number of events.
2. The calculation to determine a new saved counter was backwards. It
added when it should have subtracted, and subtracted when it should have
added. Assume a single-threaded process with a reload count of 1000 events.
Assuming the counter on context switch in was 100 and the counter on context
switch out was 50 (meaning the thread has "consumed" 50 more events), the
code would calculate a new saved counter of 150 (instead of the proper 50).
Fix:
1. As soon as the saved counter is used to initialize a monitor for a
thread on context switch in, set the saved counter to the reload count.
That way, subsequent threads to use the saved counter will get the full
reload count, assuring we sample at least once every X number of events
(across all threads).
2. Change the calculation of the saved counter. Due to the change to the
saved counter in #1, we simply need to add (modulo the reload count) the
remaining counter time we retrieve from the CPU when a thread is context
switched out.
Differential Revision: https://reviews.freebsd.org/D4122
Approved by: gnn (mentor)
MFC after: 1 month
Sponsored by: Juniper Networks
Changes to the code to gather user stacks:
* Delay setting pmc_cpumask until we actually have the stack.
* When recording user stack traces, only walk the portion of the ring
that should have samples for us.
Sponsored by: Juniper Networks
Approved by: gnn (mentor)
MFC after: 1 month
Currently, there is a single pm_stalled flag that tracks whether a
performance monitor was "stalled" due to insufficent ring buffer
space for samples. However, because the same performance monitor
can run on multiple processes or threads at the same time, a single
pm_stalled flag that impacts them all seems insufficient.
In particular, you can hit corner cases where the code fails to stop
performance monitors during a context switch out, because it thinks
the performance monitor is already stopped. However, in reality,
it may be that only the monitor running on a different CPU was stalled.
This patch attempts to fix that behavior by tracking on a per-CPU basis
whether a PM desires to run and whether it is "stalled". This lets the
code make better decisions about when to stop PMs and when to try to
restart them. Ideally, we should avoid the case where the code fails
to stop a PM during a context switch out.
Sponsored by: Juniper Networks
Reviewed by: jhb
Approved by: gnn (mentor)
Differential Revision: https://reviews.freebsd.org/D4124
according to the Cortex-A8 TRM r3p2 section 3.2.49.
The A8 list differs from the "ARM-v7 common" list, given the A8
was an earlier model.
There is still more work to be done for other Cortex-Ax version as
andrew points out, but I am just trying to fix A8 for now for teaching.
MFC after: 2 weeks
Sponsored by: DARPA/AFRL
Obtained from: Cambridge/L41
Reviewed by: andrew
Differential Revision: https://reviews.freebsd.org/D3876
in the parent, we will inherit the pmcids but cannot execute any operations
on them in the child. The reason for this is that pmc_find_pmc() only
tries to find the current process on the owners hash list, but given the
child does not own the attachment, we cannot find it.
Thus, in case the initial lookup fails, try to find the pmc_process state
affiliated with the child process, lookup the pmc from there using the
row index, and get the owner process from that pmc.
Then continue as normal and lookup the pmc context of the owner (process).
This allows us to call, e.g., pmc_start() in the child process before we
start the work there, but to collect the accumulated results later in
the parent.
Sponsored by: DARPA,AFRL
Obtained from: L41
Tested by: rwatson, L41
MFC after: 4 weeks
Reviewed by: gnn
Differential Revision: https://reviews.freebsd.org/D2052
When providing memory map information to userland, populate the vnode pointer
for tmpfs files. Set the memory mapping to appear as a vnode type, to match
FreeBSD 9 behavior.
This fixes the use of tmpfs files with the dtrace pid provider,
procstat -v, procfs, linprocfs, pmc (pmcstat), and ptrace (PT_VM_ENTRY).
Submitted by: Eric Badger <eric@badgerio.us> (initial revision)
Obtained from: Dell Inc.
PR: 198431
MFC after: 2 weeks
Reviewed by: jhb
Approved by: kib (mentor)
In both cases, the the effect of the bug was that a very small positive
number was written to the counter. This means that a large number of
events needed to occur before the next sampling interrupt would trigger.
Even with very frequently occurring events like clock cycles wrapping all
the way around could take a long time. Both bugs occurred when updating
the saved reload count for an outgoing thread on a context switch.
First, the counter-independent code compares the current reload count
against the count set when the thread switched in and generates a delta
to apply to the saved count. If this delta causes the reload counter
to go negative, it would add a full reload interval to wrap it around to
a positive value. The fix is to add the full reload interval if the
resulting counter is zero.
Second, occasionally the raw counter value read during a context switch
has actually wrapped, but an interrupt has not yet triggered. In this
case the existing logic would return a very large reload count (e.g.
2^48 - 2 if the counter had overflowed by a count of 2). This was seen
both for fixed-function and programmable counters on an E5-2643.
Workaround this case by returning a reload count of zero.
PR: 198149
Differential Revision: https://reviews.freebsd.org/D2557
Reviewed by: emaste
MFC after: 1 week
Sponsored by: Norse Corp, Inc.
This removes one of the frequent causes of ABI breakage when new CPU
types are added to hwpmc(4).
Differential Revision: https://reviews.freebsd.org/D2586
Reviewed by: davide, emaste, gnn (earlier version)
MFC after: 2 weeks
This supports e500v1, e500v2, and e500mc. Tested only on e500v2, but the
performance counters are identical across all, with e500mc having some
additional events.
Relnotes: Yes
These are similar to the mips24k performance counters - some are
available on perfcnt0/3, some are available on perfcnt1/4.
However, the events aren't all the same.
* Add the events, named the same as from Linux oprofile.
* Verify they're the same as "MIPS32(R) 74KTM Processor Core Family
Software User's Manual"; Document Number: MD00519; Revision 01.05.
* Rename INSTRUCTIONS to something else, so it doesn't clash with
the alias INSTRUCTIONS. I'll try to tidy this up later; there
are a few other aliases to add and shuffle around.
Tested:
* QCA9558 SoC (AP135 board) - MIPS74Kc core (no FPU.)
* make universe; where it didn't fail for other reasons.
TODO:
* It'd be nice to support the four performance counters
in at least this hardware, rather than just two.
Reviewed by: bsdimp ("looks good; don't break world".)
common (autogenerated) versions. Removes extra vertical space,
and makes it easier to grep for usage throughout the tree.
Conditionally compile only for arm6 [1] (yes sounds odd but is right).
Submitted by: andrew [1]
Reviewed by: gnn, andrew (ian earlier version I think)
Differential Revision: https://reviews.freebsd.org/D2159
Obtained from: Cambridge/L41
Sponsored by: DARPA, AFRL
the PMC_IN_KERNEL() macro definition.
Add missing macros to extract the return address (LR) from the trapframe.
Discussed with: andrew
Obtained from: Cambridge/L41
Sponsored by: DARPA, AFRL
MFC after: 2 weeks
not on head.. otherwise the file pointer will be NULL and when
you try to do something with it you will crash. Make the #else
be the old capabilites, and then remove the erroneous ifdefs for
11.
MFC after: 1 week (with the other MFC I was going to do until the panic)
The MEM_UOPS_RETIRED actually work the same way as the Sandy
Bridge counters, but the counters were documented in a different
way and that seemed to cause the Ivy Bridge counters to be
implemented incorrectly. Use the same counter definitions as
Sandy Bridge. While I'm here, rename the counters to match
what's documented in the datasheet.
Differential Revision: https://reviews.freebsd.org/D1590
MFC after: 1 month
Sponsored by: Sandvine Inc.
On Sandy Bridge and later, to count branch-related events you
have to or together a mask indicating the type of branch
instruction to count (e.g. direct jump, branch, etc) and a bits
indicating whether to count taken and not-taken branches. The
current counter definitions where defining this bits individually,
so the counters never worked and always just counted 0.
Fix the counter definitions to instead contain the proper
combination of masks. Also update the man pages to reflect the
new counters.
Differential Revision: https://reviews.freebsd.org/D1587
MFC after: 1 month
Sponsored by: Sandvine Inc.
A couple of pmc counters did not work because there were being
restricted to the wrong PMC unit. I've verified that these
counters now work and match the documented restrictions.
Differential Revision: https://reviews.freebsd.org/D1586
MFC after: 1 month
Sponsored by: Sandvine Inc
go back through HASWELL, IVY_BRIDGE, IVY_BRIDGE_XEON and SANDY_BRIDGE
to straighten out all the missing PMCs. We also add a new pmc tool
pmcstudy, this allows one to run the various formulas from
the documents "Using Intel Vtune Amplifier XE on XXX Generation platforms" for
IB/SB and Haswell. The tool also allows one to postulate your own
formulas with any of the various PMC's. At some point I will enahance
this to work with Brendan Gregg's flame-graphs so we can flamegraph
various PMC interactions. Note the manual page also needs some
work (lots of work) but gnn has committed to help me with that ;-)
Reviewed by: gnn
MFC after:1 month
Sponsored by: Netflix Inc.
consistent with pmc_destroy_owner_descriptor(). Also be sure to destroy
PMCs if a process exits or execs without explicitly releasing them.
Reviewed by: bz, gnn
MFC after: 2 weeks
Sponsored by: EMC / Isilon Storage Division
Differential Revision: https://reviews.freebsd.org/D958
events we have actually counted 'Branch Instruction Retired' when people
asked for 'Unhalted core cycles' using the 'unhalted-core-cycles' event mask
mnemonic.
Reviewed by: jimharris
Discussed with: gnn, rwatson
MFC after: 3 days
Sponsored by: DARPA/AFRL
spaces, rather than a split address, we actually can't check for being within
the kernel's address range. Instead, do what other backtraces do, and use
trapexit()/asttrapexit() as the stack sentinel.
MFC after: 3 weeks
This includes:
o All directories named *ia64*
o All files named *ia64*
o All ia64-specific code guarded by __ia64__
o All ia64-specific makefile logic
o Mention of ia64 in comments and documentation
This excludes:
o Everything under contrib/
o Everything under crypto/
o sys/xen/interface
o sys/sys/elf_common.h
Discussed at: BSDcan
The array index for the callchain is getting double-incremented -- both in the
loop and the storing. It should only be incremented in one location.
Also, constrain the stack pointer range check.
MFC after: 2 weeks
These changes prevent sysctl(8) from returning proper output,
such as:
1) no output from sysctl(8)
2) erroneously returning ENOMEM with tools like truss(1)
or uname(1)
truss: can not get etype: Cannot allocate memory
there is an environment variable which shall initialize the SYSCTL
during early boot. This works for all SYSCTL types both statically and
dynamically created ones, except for the SYSCTL NODE type and SYSCTLs
which belong to VNETs. A new flag, CTLFLAG_NOFETCH, has been added to
be used in the case a tunable sysctl has a custom initialisation
function allowing the sysctl to still be marked as a tunable. The
kernel SYSCTL API is mostly the same, with a few exceptions for some
special operations like iterating childrens of a static/extern SYSCTL
node. This operation should probably be made into a factored out
common macro, hence some device drivers use this. The reason for
changing the SYSCTL API was the need for a SYSCTL parent OID pointer
and not only the SYSCTL parent OID list pointer in order to quickly
generate the sysctl path. The motivation behind this patch is to avoid
parameter loading cludges inside the OFED driver subsystem. Instead of
adding special code to the OFED driver subsystem to post-load tunables
into dynamically created sysctls, we generalize this in the kernel.
Other changes:
- Corrected a possibly incorrect sysctl name from "hw.cbb.intr_mask"
to "hw.pcic.intr_mask".
- Removed redundant TUNABLE statements throughout the kernel.
- Some minor code rewrites in connection to removing not needed
TUNABLE statements.
- Added a missing SYSCTL_DECL().
- Wrapped two very long lines.
- Avoid malloc()/free() inside sysctl string handling, in case it is
called to initialize a sysctl from a tunable, hence malloc()/free() is
not ready when sysctls from the sysctl dataset are registered.
- Bumped FreeBSD version to indicate SYSCTL API change.
MFC after: 2 weeks
Sponsored by: Mellanox Technologies
Core i7 and Westmere processors, the uncore PMC subsystem is
completely different from the uncore PMC on smaller versions of CPUs.
Disable existing uncore hwpmc code for EX, otherwise non-existing MSRs
are accessed.
The cores PMCs seems to be identical for non-EX and EX, according to
the SDM.
Reviewed by: davide, fabient
Sponsored by: The FreeBSD Foundation
MFC after: 2 weeks
further refinement is required as some device drivers intended to be
portable over FreeBSD versions rely on __FreeBSD_version to decide whether
to include capability.h.
MFC after: 3 weeks
