perf/core: Add PERF_SAMPLE_DATA_PAGE_SIZE

Current perf can report both virtual addresses and physical addresses,
but not the MMU page size. Without the MMU page size information of the
utilized page, users cannot decide whether to promote/demote large pages
to optimize memory usage.

Add a new sample type for the data MMU page size.

Current perf already has a facility to collect data virtual addresses.
A page walker is required to walk the pages tables and calculate the
MMU page size from a given virtual address.

On some platforms, e.g., X86, the page walker is invoked in an NMI
handler. So the page walker must be NMI-safe and low overhead. Besides,
the page walker should work for both user and kernel virtual address.
The existing generic page walker, e.g., walk_page_range_novma(), is a
little bit complex and doesn't guarantee the NMI-safe. The follow_page()
is only for user-virtual address.

Add a new function perf_get_page_size() to walk the page tables and
calculate the MMU page size. In the function:
- Interrupts have to be disabled to prevent any teardown of the page
  tables.
- For user space threads, the current->mm is used for the page walker.
  For kernel threads and the like, the current->mm is NULL. The init_mm
  is used for the page walker. The active_mm is not used here, because
  it can be NULL.
  Quote from Peter Zijlstra,
  "context_switch() can set prev->active_mm to NULL when it transfers it
   to @next. It does this before @current is updated. So an NMI that
   comes in between this active_mm swizzling and updating @current will
   see !active_mm."
- The MMU page size is calculated from the page table level.

The method should work for all architectures, but it has only been
verified on X86. Should there be some architectures, which support perf,
where the method doesn't work, it can be fixed later separately.
Reporting the wrong page size would not be fatal for the architecture.

Some under discussion features may impact the method in the future.
Quote from Dave Hansen,
  "There are lots of weird things folks are trying to do with the page
   tables, like Address Space Isolation.  For instance, if you get a
   perf NMI when running userspace, current->mm->pgd is *different* than
   the PGD that was in use when userspace was running. It's close enough
   today, but it might not stay that way."
If the case happens later, lots of consecutive page walk errors will
happen. The worst case is that lots of page-size '0' are returned, which
would not be fatal.
In the perf tool, a check is implemented to detect this case. Once it
happens, a kernel patch could be implemented accordingly then.

Suggested-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20201001135749.2804-2-kan.liang@linux.intel.com

1 files changed, 103 insertions, 0 deletions

diff --git a/kernel/events/core.c b/kernel/events/core.c
index fb662eb4fb69..a796db2f3b57 100644
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -51,6 +51,7 @@
 #include <linux/proc_ns.h>
 #include <linux/mount.h>
 #include <linux/min_heap.h>
+#include <linux/highmem.h>
 
 #include "internal.h"
 
@@ -1894,6 +1895,9 @@ static void __perf_event_header_size(struct perf_event *event, u64 sample_type)
 	if (sample_type & PERF_SAMPLE_CGROUP)
 		size += sizeof(data->cgroup);
 
+	if (sample_type & PERF_SAMPLE_DATA_PAGE_SIZE)
+		size += sizeof(data->data_page_size);
+
 	event->header_size = size;
 }
 
@@ -6938,6 +6942,9 @@ void perf_output_sample(struct perf_output_handle *handle,
 	if (sample_type & PERF_SAMPLE_CGROUP)
 		perf_output_put(handle, data->cgroup);
 
+	if (sample_type & PERF_SAMPLE_DATA_PAGE_SIZE)
+		perf_output_put(handle, data->data_page_size);
+
 	if (sample_type & PERF_SAMPLE_AUX) {
 		perf_output_put(handle, data->aux_size);
 
@@ -6995,6 +7002,94 @@ static u64 perf_virt_to_phys(u64 virt)
 	return phys_addr;
 }
 
+#ifdef CONFIG_MMU
+
+/*
+ * Return the MMU page size of a given virtual address
+ */
+static u64 __perf_get_page_size(struct mm_struct *mm, unsigned long addr)
+{
+	pgd_t *pgd;
+	p4d_t *p4d;
+	pud_t *pud;
+	pmd_t *pmd;
+	pte_t *pte;
+
+	pgd = pgd_offset(mm, addr);
+	if (pgd_none(*pgd))
+		return 0;
+
+	p4d = p4d_offset(pgd, addr);
+	if (!p4d_present(*p4d))
+		return 0;
+
+	if (p4d_leaf(*p4d))
+		return 1ULL << P4D_SHIFT;
+
+	pud = pud_offset(p4d, addr);
+	if (!pud_present(*pud))
+		return 0;
+
+	if (pud_leaf(*pud))
+		return 1ULL << PUD_SHIFT;
+
+	pmd = pmd_offset(pud, addr);
+	if (!pmd_present(*pmd))
+		return 0;
+
+	if (pmd_leaf(*pmd))
+		return 1ULL << PMD_SHIFT;
+
+	pte = pte_offset_map(pmd, addr);
+	if (!pte_present(*pte)) {
+		pte_unmap(pte);
+		return 0;
+	}
+
+	pte_unmap(pte);
+	return PAGE_SIZE;
+}
+
+#else
+
+static u64 __perf_get_page_size(struct mm_struct *mm, unsigned long addr)
+{
+	return 0;
+}
+
+#endif
+
+static u64 perf_get_page_size(unsigned long addr)
+{
+	struct mm_struct *mm;
+	unsigned long flags;
+	u64 size;
+
+	if (!addr)
+		return 0;
+
+	/*
+	 * Software page-table walkers must disable IRQs,
+	 * which prevents any tear down of the page tables.
+	 */
+	local_irq_save(flags);
+
+	mm = current->mm;
+	if (!mm) {
+		/*
+		 * For kernel threads and the like, use init_mm so that
+		 * we can find kernel memory.
+		 */
+		mm = &init_mm;
+	}
+
+	size = __perf_get_page_size(mm, addr);
+
+	local_irq_restore(flags);
+
+	return size;
+}
+
 static struct perf_callchain_entry __empty_callchain = { .nr = 0, };
 
 struct perf_callchain_entry *
@@ -7150,6 +7245,14 @@ void perf_prepare_sample(struct perf_event_header *header,
 	}
 #endif
 
+	/*
+	 * PERF_DATA_PAGE_SIZE requires PERF_SAMPLE_ADDR. If the user doesn't
+	 * require PERF_SAMPLE_ADDR, kernel implicitly retrieve the data->addr,
+	 * but the value will not dump to the userspace.
+	 */
+	if (sample_type & PERF_SAMPLE_DATA_PAGE_SIZE)
+		data->data_page_size = perf_get_page_size(data->addr);
+
 	if (sample_type & PERF_SAMPLE_AUX) {
 		u64 size;