x86/intel_rdt: Re-enable pseudo-lock measurements

Commit 4a7a54a55e72 ("x86/intel_rdt: Disable PMU access") disabled
measurements of pseudo-locked regions because of incorrect usage
of the performance monitoring hardware.

Cache pseudo-locking measurements are now done correctly with the
in-kernel perf API and its use can be re-enabled at this time.

The adjustment to the in-kernel perf API also separated the L2 and L3
measurements that can be triggered separately from user space. The
re-enabling of the measurements is thus not a simple revert of the
original disable in order to accommodate the additional parameter
possible.

Signed-off-by: Reinette Chatre <reinette.chatre@intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: fenghua.yu@intel.com
Cc: tony.luck@intel.com
Cc: peterz@infradead.org
Cc: acme@kernel.org
Cc: gavin.hindman@intel.com
Cc: jithu.joseph@intel.com
Cc: dave.hansen@intel.com
Cc: hpa@zytor.com
Link: https://lkml.kernel.org/r/bfb9fc31692e0c62d9ca39062e55eceb6a0635b5.1537377064.git.reinette.chatre@intel.com

diff --git a/arch/x86/kernel/cpu/intel_rdt_pseudo_lock.c b/arch/x86/kernel/cpu/intel_rdt_pseudo_lock.c
index d68836139cf9d37556caf7a9528777ab5180c898..30e6c9f5a0ada75f9a0b7f7d85c9e05607b4368c 100644 (file)
--- a/arch/x86/kernel/cpu/intel_rdt_pseudo_lock.c
+++ b/arch/x86/kernel/cpu/intel_rdt_pseudo_lock.c
@@ -1236,7 +1236,7 @@ static ssize_t pseudo_lock_measure_trigger(struct file *file,
        buf[buf_size] = '\0';
        ret = kstrtoint(buf, 10, &sel);
        if (ret == 0) {
-               if (sel != 1)
+               if (sel != 1 && sel != 2 && sel != 3)
                        return -EINVAL;
                ret = debugfs_file_get(file->f_path.dentry);
                if (ret)