LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: "Suzuki K. Poulose" <suzuki.poulose@arm.com>
To: will.deacon@arm.com
Cc: Nicolas Pitre <nico@linaro.org>,
	Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com>,
	Kukjin Kim <kgene@kernel.org>,
	Abhilash Kesavan <a.kesavan@samsung.com>,
	Arnd Bergmann <arnd@arndb.de>,
	devicetree@vger.kernel.org, linux-kernel@vger.kernel.org,
	Liviu Dudau <liviu.dudau@arm.com>,
	Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>,
	Olof Johansson <olof@lixom.net>, Pawel Moll <pawel.moll@arm.com>,
	Punit Agrawal <punit.agrawal@arm.com>,
	Sudeep Holla <sudeep.holla@arm.com>,
	Catalin Marinas <catalin.marinas@arm.com>,
	"Suzuki K. Poulose" <suzuki.poulose@arm.com>
Subject: [PATCH 5/5] arm-cci: Fix CCI PMU event validation
Date: Wed, 18 Mar 2015 12:24:42 +0000	[thread overview]
Message-ID: <1426681482-23118-6-git-send-email-suzuki.poulose@arm.com> (raw)
In-Reply-To: <1426681482-23118-1-git-send-email-suzuki.poulose@arm.com>

From: "Suzuki K. Poulose" <suzuki.poulose@arm.com>

We mask the event with the CCI_PMU_EVENT_MASK, before passing
the config to pmu_validate_hw_event(), which causes extra bits
to be ignored and qualifies an invalid event code as valid.

e.g,
 $ perf stat -a -C 0 -e CCI_400/config=0x1ff,name=cycles/ sleep 1
   Performance counter stats for 'system wide':

         506951142      cycles

       1.013879626 seconds time elapsed

where, cycles has an event coding of 0xff. This patch also removes
the unnecessary 'event' mask in pmu_write_register, since the config_base
is set by the pmu code after the event is validated.

Changes since V3:
 - Fix type for CCI_PMU_EVENT_MASK - Suggested-by: Will Deacon.

Signed-off-by: Suzuki K. Poulose <suzuki.poulose@arm.com>
Acked-by: Punit Agrawal <punit.agrawal@arm.com>
Reviewed-by: Will Deacon <will.deacon@arm.com>
---
 drivers/bus/arm-cci.c |   12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/drivers/bus/arm-cci.c b/drivers/bus/arm-cci.c
index 054df84..b854125 100644
--- a/drivers/bus/arm-cci.c
+++ b/drivers/bus/arm-cci.c
@@ -81,7 +81,7 @@ static const struct of_device_id arm_cci_matches[] = {
 
 #define CCI_PMU_CNTR_MASK	((1ULL << 32) -1)
 
-#define CCI_PMU_EVENT_MASK		0xff
+#define CCI_PMU_EVENT_MASK		0xffUL
 #define CCI_PMU_EVENT_SOURCE(event)	((event >> 5) & 0x7)
 #define CCI_PMU_EVENT_CODE(event)	(event & 0x1f)
 
@@ -179,12 +179,15 @@ enum cci400_perf_events {
 #define CCI_REV_R1_MASTER_PORT_MIN_EV	0x00
 #define CCI_REV_R1_MASTER_PORT_MAX_EV	0x11
 
-static int pmu_validate_hw_event(u8 hw_event)
+static int pmu_validate_hw_event(unsigned long hw_event)
 {
 	u8 ev_source = CCI_PMU_EVENT_SOURCE(hw_event);
 	u8 ev_code = CCI_PMU_EVENT_CODE(hw_event);
 	int if_type;
 
+	if (hw_event & ~CCI_PMU_EVENT_MASK)
+		return -ENOENT;
+
 	switch (ev_source) {
 	case CCI_PORT_S0:
 	case CCI_PORT_S1:
@@ -258,7 +261,6 @@ static void pmu_enable_counter(int idx)
 
 static void pmu_set_event(int idx, unsigned long event)
 {
-	event &= CCI_PMU_EVENT_MASK;
 	pmu_write_register(event, idx, CCI_PMU_EVT_SEL);
 }
 
@@ -275,7 +277,7 @@ static int pmu_get_event_idx(struct cci_pmu_hw_events *hw, struct perf_event *ev
 {
 	struct cci_pmu *cci_pmu = to_cci_pmu(event->pmu);
 	struct hw_perf_event *hw_event = &event->hw;
-	unsigned long cci_event = hw_event->config_base & CCI_PMU_EVENT_MASK;
+	unsigned long cci_event = hw_event->config_base;
 	int idx;
 
 	if (cci_event == CCI_PMU_CYCLES) {
@@ -296,7 +298,7 @@ static int pmu_get_event_idx(struct cci_pmu_hw_events *hw, struct perf_event *ev
 static int pmu_map_event(struct perf_event *event)
 {
 	int mapping;
-	u8 config = event->attr.config & CCI_PMU_EVENT_MASK;
+	unsigned long config = event->attr.config;
 
 	if (event->attr.type < PERF_TYPE_MAX)
 		return -ENOENT;
-- 
1.7.9.5



  parent reply	other threads:[~2015-03-18 12:25 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-03-18 12:24 [PATCHv4 0/5] arm-cci400: PMU monitoring support on ARM64 Suzuki K. Poulose
2015-03-18 12:24 ` [PATCH 1/5] arm-cci: Rearrange code for splitting PMU vs driver code Suzuki K. Poulose
2015-03-18 12:24 ` [PATCH 2/5] arm-cci: Abstract the CCI400 PMU speicific definitions Suzuki K. Poulose
2015-03-18 12:24 ` [PATCH 3/5] arm-cci: Get rid of secure transactions for PMU driver Suzuki K. Poulose
2015-03-18 12:24 ` [PATCH 4/5] arm-cci: Split the code for PMU vs driver support Suzuki K. Poulose
2015-03-18 12:24 ` Suzuki K. Poulose [this message]
  -- strict thread matches above, loose matches on Subject: below --
2015-03-10 15:18 [PATCHv3 0/5] arm-cci400: PMU monitoring support on ARM64 Suzuki K. Poulose
2015-03-10 15:18 ` [PATCH 5/5] arm-cci: Fix CCI PMU event validation Suzuki K. Poulose
2015-03-17 18:52   ` Will Deacon
2015-03-02 11:29 [PATCH v2 0/5] arm-cci400: PMU monitoring support on ARM64 Suzuki K. Poulose
2015-03-02 11:29 ` [PATCH 5/5] arm-cci: Fix CCI PMU event validation Suzuki K. Poulose

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=1426681482-23118-6-git-send-email-suzuki.poulose@arm.com \
    --to=suzuki.poulose@arm.com \
    --cc=a.kesavan@samsung.com \
    --cc=arnd@arndb.de \
    --cc=b.zolnierkie@samsung.com \
    --cc=catalin.marinas@arm.com \
    --cc=devicetree@vger.kernel.org \
    --cc=kgene@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=liviu.dudau@arm.com \
    --cc=lorenzo.pieralisi@arm.com \
    --cc=nico@linaro.org \
    --cc=olof@lixom.net \
    --cc=pawel.moll@arm.com \
    --cc=punit.agrawal@arm.com \
    --cc=sudeep.holla@arm.com \
    --cc=will.deacon@arm.com \
    --subject='Re: [PATCH 5/5] arm-cci: Fix CCI PMU event validation' \
    /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

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).