LIBPFM(3)

💖 LIBPFM(3) — the Linux Programmer's Manual glow-up 💖

okay bestie, sit down, hydrate, because we're diving into libpfm_intel_slm and honestly? she's serving. this one lives in the Linux Programmer's Manual, section 3, and she's all about giving you the tea on Intel's Silvermont core PMU. let's go ✨

🌸 NAME 🌸

libpfm_intel_slm — support for Intel Silvermont core PMU. that's it, that's the whole vibe. 💅

🎀 SYNOPSIS 🎀

#include <perfmon/pfmlib.h>

PMU name: slm

PMU desc: Intel Silvermont

like, that's the whole incantation. include the header, know your PMU's name and description, and you're basically in. 🦋

💫 DESCRIPTION 💫

The library supports the Intel Silvermont core PMU. full stop, periodt. she's supported, she's here, she's ready to count things for you. 👑

✨ MODIFIERS ✨

bestie these are the little letters you sprinkle after your event and they CHANGE EVERYTHING. The following modifiers are supported on Intel Silvermont processors:

• u — Measure at user level which includes privilege levels 1, 2, 3. This corresponds to PFM_PLM3. This is a boolean modifier. 🫶
• k — Measure at kernel level which includes privilege level 0. This corresponds to PFM_PLM0. This is a boolean modifier.
• i — Invert the meaning of the event. The counter will now count cycles in which the event is not occurring. This is a boolean modifier. 🙈 (not me counting everything EXCEPT the thing, iconic)
• e — Enable edge detection, i.e., count only when there is a state transition from no occurrence of the event to at least one occurrence. This modifier must be combined with a counter mask modifier (m) with a value greater or equal to one. This is a boolean modifier. 🔥
• c — Set the counter mask value. The mask acts as a threshold. The counter will count the number of cycles in which the number of occurrences of the event is greater or equal to the threshold. This is an integer modifier with values in the range [0:255]. ⭐

lowkey important: e does NOT work solo. she needs m with a value >= 1 or she simply will not function. no cap.

🌈 OFFCORE_RESPONSE events 🌈

Intel Silvermont provides two offcore_response events: OFFCORE_RESPONSE_0 and OFFCORE_RESPONSE_1. two of them. a duo. main character energy, both of them. 👯

Those events need special treatment in the performance monitoring infrastructure because each event uses an extra register to store some settings. Thus, in case multiple offcore_response events are monitored simultaneously, the kernel needs to manage the sharing of that extra register. (yes, the kernel is playing referee. let her cook. 🧚)

The offcore_response events are exposed as a normal event by the library. The extra settings are exposed as regular umasks. The library takes care of encoding the events according to the underlying kernel interface.

On Intel Silvermont, the umasks are divided into three categories: request, supplier and snoop. The user must provide at least one umask for each category. The categories are shown in the umask descriptions. The library provides a default umask per category if not provided by the user. (so if you forget, she's got you. she's not gonna leave you hanging. 💜)

There is also the special response umask called ANY_RESPONSE. When this umask is used then it overrides any supplier and snoop umasks. In other words, users can specify either ANY_RESPONSE OR any combinations of supplier + snoops.

In case no supplier or snoop is specified, the library defaults to using ANY_RESPONSE.

For instance, the following are valid event selections:

OFFCORE_RESPONSE_0:DMND_DATA_RD:ANY_RESPONSE

OFFCORE_RESPONSE_0:ANY_REQUEST

OFFCORE_RESPONSE_0:ANY_RFO:LLC_HITM:SNOOP_ANY

But the following are illegal:

OFFCORE_RESPONSE_0:ANY_RFO:NON_DRAM:ANY_RESPONSE

OFFCORE_RESPONSE_0:ANY_RFO:L2_HIT:SNOOP_ANY:ANY_RESPONSE

🚨 do NOT try to be cute and mix ANY_RESPONSE with supplier/snoop umasks. it's giving invalid. she said either/or and she MEANT it. 💀

🫶 AUTHORS 🫶

Stephane Eranian <eranian@gmail.com>

and that's the whole page, bestie!! 🎀 this icon of a manpage is from November, 2013, and it belongs to LIBPFM(3). go forth and profile responsibly, ily, bye!! 💖✨🦋