All of lore.kernel.org
 help / color / mirror / Atom feed
From: Sasha Levin <sashal@kernel.org>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org
Cc: Stephane Eranian <eranian@google.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Sasha Levin <sashal@kernel.org>,
	mingo@redhat.com, acme@kernel.org, tglx@linutronix.de,
	bp@alien8.de, dave.hansen@linux.intel.com, x86@kernel.org,
	linux-perf-users@vger.kernel.org
Subject: [PATCH AUTOSEL 5.16 15/52] perf/x86/rapl: fix AMD event handling
Date: Thu,  3 Feb 2022 15:29:09 -0500	[thread overview]
Message-ID: <20220203202947.2304-15-sashal@kernel.org> (raw)
In-Reply-To: <20220203202947.2304-1-sashal@kernel.org>

From: Stephane Eranian <eranian@google.com>

[ Upstream commit 0036fb00a756a2f6e360d44e2e3d2200a8afbc9b ]

The RAPL events exposed under /sys/devices/power/events should only reflect
what the underlying hardware actually support. This is how it works on Intel
RAPL and Intel core/uncore PMUs in general.
But on AMD, this was not the case. All possible RAPL events were advertised.

This is what it showed on an AMD Fam17h:
$ ls /sys/devices/power/events/
energy-cores        energy-gpu          energy-pkg          energy-psys
energy-ram          energy-cores.scale  energy-gpu.scale    energy-pkg.scale
energy-psys.scale   energy-ram.scale    energy-cores.unit   energy-gpu.unit
energy-pkg.unit     energy-psys.unit    energy-ram.unit

Yet, on AMD Fam17h, only energy-pkg is supported.

This patch fixes the problem. Given the way perf_msr_probe() works, the
amd_rapl_msrs[] table has to have all entries filled out and in particular
the group field, otherwise perf_msr_probe() defaults to making the event
visible.

With the patch applied, the kernel now only shows was is actually supported:

$ ls /sys/devices/power/events/
energy-pkg  energy-pkg.scale  energy-pkg.unit

The patch also uses the RAPL_MSR_MASK because only the 32-bits LSB of the
RAPL counters are relevant when reading power consumption.

Signed-off-by: Stephane Eranian <eranian@google.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20220105185659.643355-1-eranian@google.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 arch/x86/events/rapl.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/arch/x86/events/rapl.c b/arch/x86/events/rapl.c
index 85feafacc445d..77e3a47af5ad5 100644
--- a/arch/x86/events/rapl.c
+++ b/arch/x86/events/rapl.c
@@ -536,11 +536,14 @@ static struct perf_msr intel_rapl_spr_msrs[] = {
  * - perf_msr_probe(PERF_RAPL_MAX)
  * - want to use same event codes across both architectures
  */
-static struct perf_msr amd_rapl_msrs[PERF_RAPL_MAX] = {
-	[PERF_RAPL_PKG]  = { MSR_AMD_PKG_ENERGY_STATUS,  &rapl_events_pkg_group,   test_msr },
+static struct perf_msr amd_rapl_msrs[] = {
+	[PERF_RAPL_PP0]  = { 0, &rapl_events_cores_group, 0, false, 0 },
+	[PERF_RAPL_PKG]  = { MSR_AMD_PKG_ENERGY_STATUS,  &rapl_events_pkg_group,   test_msr, false, RAPL_MSR_MASK },
+	[PERF_RAPL_RAM]  = { 0, &rapl_events_ram_group,   0, false, 0 },
+	[PERF_RAPL_PP1]  = { 0, &rapl_events_gpu_group,   0, false, 0 },
+	[PERF_RAPL_PSYS] = { 0, &rapl_events_psys_group,  0, false, 0 },
 };
 
-
 static int rapl_cpu_offline(unsigned int cpu)
 {
 	struct rapl_pmu *pmu = cpu_to_rapl_pmu(cpu);
-- 
2.34.1


  parent reply	other threads:[~2022-02-03 20:30 UTC|newest]

Thread overview: 74+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-02-03 20:28 [PATCH AUTOSEL 5.16 01/52] NFS: change nfs_access_get_cached to only report the mask Sasha Levin
2022-02-03 20:28 ` [PATCH AUTOSEL 5.16 02/52] NFSv4 only print the label when its queried Sasha Levin
2022-02-03 20:28 ` [PATCH AUTOSEL 5.16 03/52] nfs: nfs4clinet: check the return value of kstrdup() Sasha Levin
2022-02-03 20:28 ` [PATCH AUTOSEL 5.16 04/52] NFSv4.1: Fix uninitialised variable in devicenotify Sasha Levin
2022-02-03 20:28 ` [PATCH AUTOSEL 5.16 05/52] netfilter: nf_conntrack_netbios_ns: fix helper module alias Sasha Levin
2022-02-03 21:46   ` Jakub Kicinski
2022-02-03 22:23     ` Florian Westphal
2022-02-11 14:41       ` Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 06/52] NFSv4 remove zero number of fs_locations entries error check Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 07/52] NFSv4 store server support for fs_location attribute Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 08/52] NFSv4.1 query for fs_location attr on a new file system Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 09/52] NFSv4 expose nfs_parse_server_name function Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 10/52] NFSv4 handle port presence in fs_location server string Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 11/52] SUNRPC allow for unspecified transport time in rpc_clnt_add_xprt Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 12/52] net/sunrpc: fix reference count leaks in rpc_sysfs_xprt_state_change Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 13/52] sunrpc: Fix potential race conditions in rpc_sysfs_xprt_state_change() Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 14/52] irqchip/realtek-rtl: Service all pending interrupts Sasha Levin
2022-02-03 20:29 ` Sasha Levin [this message]
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 16/52] x86/perf: Avoid warning for Arch LBR without XSAVE Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 17/52] sched/pelt: Relax the sync of runnable_sum with runnable_avg Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 18/52] sched: Avoid double preemption in __cond_resched_*lock*() Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 19/52] drm/vc4: Fix deadlock on DSI device attach error Sasha Levin
2022-02-03 20:29   ` Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 20/52] drm: panel-orientation-quirks: Add quirk for the 1Netbook OneXPlayer Sasha Levin
2022-02-03 20:29   ` Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 21/52] net: sched: Clarify error message when qdisc kind is unknown Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 22/52] powerpc/fixmap: Fix VM debug warning on unmap Sasha Levin
2022-02-03 20:29   ` Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 23/52] s390/module: test loading modules with a lot of relocations Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 24/52] arm64: Add Cortex-X2 CPU part definition Sasha Levin
2022-02-03 20:29   ` Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 25/52] arm64: errata: Update ARM64_ERRATUM_[2119858|2224489] with Cortex-X2 ranges Sasha Levin
2022-02-03 20:29   ` Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 26/52] scsi: target: iscsi: Make sure the np under each tpg is unique Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 27/52] scsi: ufs: ufshcd-pltfrm: Check the return value of devm_kstrdup() Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 28/52] scsi: qedf: Add stag_work to all the vports Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 29/52] scsi: qedf: Fix refcount issue when LOGO is received during TMF Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 30/52] scsi: qedf: Change context reset messages to ratelimited Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 31/52] scsi: pm8001: Fix bogus FW crash for maxcpus=1 Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 32/52] scsi: ufs: Use generic error code in ufshcd_set_dev_pwr_mode() Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 33/52] scsi: ufs: Treat link loss as fatal error Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 34/52] scsi: myrs: Fix crash in error case Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 35/52] net: stmmac: reduce unnecessary wakeups from eee sw timer Sasha Levin
2022-02-03 20:29   ` Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 36/52] PM: wakeup: simplify the output logic of pm_show_wakelocks() Sasha Levin
2022-02-04  8:29   ` Greg Kroah-Hartman
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 37/52] PM: hibernate: Remove register_nosave_region_late() Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 38/52] drm/amd/display: Correct MPC split policy for DCN301 Sasha Levin
2022-02-03 20:29   ` Sasha Levin
2022-02-03 20:29   ` Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 39/52] drm/amdgpu/display: adjust msleep limit in dp_wait_for_training_aux_rd_interval Sasha Levin
2022-02-03 20:29   ` Sasha Levin
2022-02-03 20:29   ` Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 40/52] drm/amdgpu/display: use msleep rather than udelay for long delays Sasha Levin
2022-02-03 20:29   ` Sasha Levin
2022-02-03 20:29   ` Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 41/52] usb: dwc2: gadget: don't try to disable ep0 in dwc2_hsotg_suspend Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 42/52] perf: Always wake the parent event Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 43/52] nvme-pci: add the IGNORE_DEV_SUBNQN quirk for Intel P4500/P4600 SSDs Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 44/52] MIPS: Fix build error due to PTR used in more places Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 45/52] net: stmmac: dwmac-sun8i: use return val of readl_poll_timeout() Sasha Levin
2022-02-03 20:29   ` Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 46/52] arm64: Add Cortex-A510 CPU part definition Sasha Levin
2022-02-03 20:29   ` Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 47/52] arm64: errata: Add detection for TRBE ignored system register writes Sasha Levin
2022-02-03 20:29   ` Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 48/52] arm64: errata: Add detection for TRBE invalid prohibited states Sasha Levin
2022-02-03 20:29   ` Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 49/52] arm64: errata: Add detection for TRBE trace data corruption Sasha Levin
2022-02-03 20:29   ` Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 50/52] arm64: cpufeature: List early Cortex-A510 parts as having broken dbm Sasha Levin
2022-02-03 20:29   ` Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 51/52] block: add bio_start_io_acct_time() to control start_time Sasha Levin
2022-02-03 20:29 ` [PATCH AUTOSEL 5.16 52/52] kasan: test: fix compatibility with FORTIFY_SOURCE Sasha Levin

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20220203202947.2304-15-sashal@kernel.org \
    --to=sashal@kernel.org \
    --cc=acme@kernel.org \
    --cc=bp@alien8.de \
    --cc=dave.hansen@linux.intel.com \
    --cc=eranian@google.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-perf-users@vger.kernel.org \
    --cc=mingo@redhat.com \
    --cc=peterz@infradead.org \
    --cc=stable@vger.kernel.org \
    --cc=tglx@linutronix.de \
    --cc=x86@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.