mm: add infrastructure for get_user_pages_fast() benchmarking

Performance of get_user_pages_fast() is critical for some workloads, but
it's tricky to test it directly.

This patch provides /sys/kernel/debug/gup_benchmark that helps with
testing performance of it.

See tools/testing/selftests/vm/gup_benchmark.c for userspace
counterpart.

Link: http://lkml.kernel.org/r/20170908215603.9189-2-kirill.shutemov@linux.intel.com
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Shuah Khan <shuah@kernel.org>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Thorsten Leemhuis <regressions@leemhuis.info>
Cc: Jonathan Corbet <corbet@lwn.net>
Cc: Huang Ying <ying.huang@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm/Kconfig

@@ -756,3 +756,12 @@ config PERCPU_STATS
 	  This feature collects and exposes statistics via debugfs. The
 	  information includes global and per chunk statistics, which can
 	  be used to help understand percpu memory usage.
+
+config GUP_BENCHMARK
+	bool "Enable infrastructure for get_user_pages_fast() benchmarking"
+	default n
+	help
+	  Provides /sys/kernel/debug/gup_benchmark that helps with testing
+	  performance of get_user_pages_fast().
+
+	  See tools/testing/selftests/vm/gup_benchmark.c

mm/Makefile

@@ -80,6 +80,7 @@ obj-$(CONFIG_PAGE_COUNTER) += page_counter.o
 obj-$(CONFIG_MEMCG) += memcontrol.o vmpressure.o
 obj-$(CONFIG_MEMCG_SWAP) += swap_cgroup.o
 obj-$(CONFIG_CGROUP_HUGETLB) += hugetlb_cgroup.o
+obj-$(CONFIG_GUP_BENCHMARK) += gup_benchmark.o
 obj-$(CONFIG_MEMORY_FAILURE) += memory-failure.o
 obj-$(CONFIG_HWPOISON_INJECT) += hwpoison-inject.o
 obj-$(CONFIG_DEBUG_KMEMLEAK) += kmemleak.o

mm/gup_benchmark.c

@@ -0,0 +1,100 @@
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/slab.h>
+#include <linux/uaccess.h>
+#include <linux/ktime.h>
+#include <linux/debugfs.h>
+
+#define GUP_FAST_BENCHMARK	_IOWR('g', 1, struct gup_benchmark)
+
+struct gup_benchmark {
+	__u64 delta_usec;
+	__u64 addr;
+	__u64 size;
+	__u32 nr_pages_per_call;
+	__u32 flags;
+};
+
+static int __gup_benchmark_ioctl(unsigned int cmd,
+		struct gup_benchmark *gup)
+{
+	ktime_t start_time, end_time;
+	unsigned long i, nr, nr_pages, addr, next;
+	struct page **pages;
+
+	nr_pages = gup->size / PAGE_SIZE;
+	pages = kvmalloc(sizeof(void *) * nr_pages, GFP_KERNEL);
+	if (!pages)
+		return -ENOMEM;
+
+	i = 0;
+	nr = gup->nr_pages_per_call;
+	start_time = ktime_get();
+	for (addr = gup->addr; addr < gup->addr + gup->size; addr = next) {
+		if (nr != gup->nr_pages_per_call)
+			break;
+
+		next = addr + nr * PAGE_SIZE;
+		if (next > gup->addr + gup->size) {
+			next = gup->addr + gup->size;
+			nr = (next - addr) / PAGE_SIZE;
+		}
+
+		nr = get_user_pages_fast(addr, nr, gup->flags & 1, pages + i);
+		i += nr;
+	}
+	end_time = ktime_get();
+
+	gup->delta_usec = ktime_us_delta(end_time, start_time);
+	gup->size = addr - gup->addr;
+
+	for (i = 0; i < nr_pages; i++) {
+		if (!pages[i])
+			break;
+		put_page(pages[i]);
+	}
+
+	kvfree(pages);
+	return 0;
+}
+
+static long gup_benchmark_ioctl(struct file *filep, unsigned int cmd,
+		unsigned long arg)
+{
+	struct gup_benchmark gup;
+	int ret;
+
+	if (cmd != GUP_FAST_BENCHMARK)
+		return -EINVAL;
+
+	if (copy_from_user(&gup, (void __user *)arg, sizeof(gup)))
+		return -EFAULT;
+
+	ret = __gup_benchmark_ioctl(cmd, &gup);
+	if (ret)
+		return ret;
+
+	if (copy_to_user((void __user *)arg, &gup, sizeof(gup)))
+		return -EFAULT;
+
+	return 0;
+}
+
+static const struct file_operations gup_benchmark_fops = {
+	.open = nonseekable_open,
+	.unlocked_ioctl = gup_benchmark_ioctl,
+};
+
+static int gup_benchmark_init(void)
+{
+	void *ret;
+
+	ret = debugfs_create_file_unsafe("gup_benchmark", 0600, NULL, NULL,
+			&gup_benchmark_fops);
+	if (!ret)
+		pr_warn("Failed to create gup_benchmark in debugfs");
+
+	return 0;
+}
+
+late_initcall(gup_benchmark_init);