__hw_perf_counter_init() on x86
It seems that for most of the error returns this function can give, we'll leak the active_counters count grabbed at the beginning of the function and therefore we'll never re-enable the LAPIC NMI watchdog even once all the perf counter users go away.
Just FYI...
Quoted 7 lines · David Miller
It seems that for most of the error returns this function can give, we'll leak the
active_counterscount grabbed at the beginning of the function and therefore we'll never re-enable the LAPIC NMI watchdog even once all the perf counter users go away.Just FYI...
Thanks!
---
Subject: perf_counter: x86: Fix PMU resource leak
Dave noticed that we leak the PMU resource reservations when we fail the hardware counter init.
--- arch/x86/kernel/cpu/perf_counter.c | 9 ++++++--- 1 files changed, 6 insertions(+), 3 deletions(-)
arch/x86/kernel/cpu/perf_counter.c +6 -3
@@ -924,6 +924,8 @@ static int __hw_perf_counter_init(struct perf_counter *counter)if (err)return err; + counter->destroy = hw_perf_counter_destroy; +/* * Generate PMC IRQs: * (keep 'enabled' bit clear for now)
@@ -953,8 +955,6 @@ static int __hw_perf_counter_init(struct perf_counter *counter)return -EOPNOTSUPP ; } - counter->destroy = hw_perf_counter_destroy; -/* * Raw event type provide the config in the event structure */
@@ -2093,8 +2093,11 @@ const struct pmu *hw_perf_counter_init(struct perf_counter *counter)int err; err =__hw_perf_counter_init (counter); -if (err) +if (err) { +if (counter->destroy) + counter->destroy (counter);return ERR_PTR (err); + }return &pmu; }
David Miller
Acked-byFrom: Peter Zijlstra <[email protected]>
Date: Wed, 09 Sep 2009 10:04:47 +0200
Quoted 7 lines
Subject:
perf_counter: x86: Fix PMU resource leakDave noticed that we leak the PMU resource reservations when we fail the hardware counter init.
Reported-by David Miller <[email protected]>Signed-off-by Peter Zijlstra <[email protected]>
Looks great.
[tip:perfcounters/urgent] perf_counter: x86: Fix PMU resource leak
In reply to Peter ZijlstraCommit-ID: a1792cdacaf5180e04e07811e220c4a3b4a9c33e Gitweb: http://git.kernel.org/tip/a1792cdacaf5180e04e07811e220c4a3b4a9c33e Author: Peter Zijlstra <[email protected]> AuthorDate: Wed, 9 Sep 2009 10:04:47 +0200 Committer: Ingo Molnar <[email protected]> CommitDate: Sun, 20 Sep 2009 19:53:57 +0200
perf_counter: x86: Fix PMU resource leak
Dave noticed that we leak the PMU resource reservations when we fail the hardware counter init.
LKML-Reference: <1252483487.7746.164.camel@twins>
--- arch/x86/kernel/cpu/perf_counter.c | 9 ++++++--- 1 files changed, 6 insertions(+), 3 deletions(-)
arch/x86/kernel/cpu/perf_counter.c +6 -3
@@ -924,6 +924,8 @@ static int __hw_perf_counter_init(struct perf_counter *counter)if (err)return err; + counter->destroy = hw_perf_counter_destroy; +/* * Generate PMC IRQs: * (keep 'enabled' bit clear for now)
@@ -953,8 +955,6 @@ static int __hw_perf_counter_init(struct perf_counter *counter)return -EOPNOTSUPP ; } - counter->destroy = hw_perf_counter_destroy; -/* * Raw event type provide the config in the event structure */
@@ -2107,8 +2107,11 @@ const struct pmu *hw_perf_counter_init(struct perf_counter *counter)int err; err =__hw_perf_counter_init (counter); -if (err) +if (err) { +if (counter->destroy) + counter->destroy (counter);return ERR_PTR (err); + }return &pmu; }