proc: faster open/read/close with "permanent" files

Now that "struct proc_ops" exist we can start putting there stuff which
could not fly with VFS "struct file_operations"...

Most of fs/proc/inode.c file is dedicated to make open/read/.../close
reliable in the event of disappearing /proc entries which usually happens
if module is getting removed.  Files like /proc/cpuinfo which never
disappear simply do not need such protection.

Save 2 atomic ops, 1 allocation, 1 free per open/read/close sequence for such
"permanent" files.

Enable "permanent" flag for

	/proc/cpuinfo
	/proc/kmsg
	/proc/modules
	/proc/slabinfo
	/proc/stat
	/proc/sysvipc/*
	/proc/swaps

More will come once I figure out foolproof way to prevent out module
authors from marking their stuff "permanent" for performance reasons
when it is not.

This should help with scalability: benchmark is "read /proc/cpuinfo R times
by N threads scattered over the system".

	N	R	t, s (before)	t, s (after)
	-----------------------------------------------------
	64	4096	1.582458	1.530502	-3.2%
	256	4096	6.371926	6.125168	-3.9%
	1024	4096	25.64888	24.47528	-4.6%

Benchmark source:

#include <chrono>
#include <iostream>
#include <thread>
#include <vector>

#include <sys/types.h>
#include <sys/stat.h>
#include <fcntl.h>
#include <unistd.h>

const int NR_CPUS = sysconf(_SC_NPROCESSORS_ONLN);
int N;
const char *filename;
int R;

int xxx = 0;

int glue(int n)
{
	cpu_set_t m;
	CPU_ZERO(&m);
	CPU_SET(n, &m);
	return sched_setaffinity(0, sizeof(cpu_set_t), &m);
}

void f(int n)
{
	glue(n % NR_CPUS);

	while (*(volatile int *)&xxx == 0) {
	}

	for (int i = 0; i < R; i++) {
		int fd = open(filename, O_RDONLY);
		char buf[4096];
		ssize_t rv = read(fd, buf, sizeof(buf));
		asm volatile ("" :: "g" (rv));
		close(fd);
	}
}

int main(int argc, char *argv[])
{
	if (argc < 4) {
		std::cerr << "usage: " << argv[0] << ' ' << "N /proc/filename R
";
		return 1;
	}

	N = atoi(argv[1]);
	filename = argv[2];
	R = atoi(argv[3]);

	for (int i = 0; i < NR_CPUS; i++) {
		if (glue(i) == 0)
			break;
	}

	std::vector<std::thread> T;
	T.reserve(N);
	for (int i = 0; i < N; i++) {
		T.emplace_back(f, i);
	}

	auto t0 = std::chrono::system_clock::now();
	{
		*(volatile int *)&xxx = 1;
		for (auto& t: T) {
			t.join();
		}
	}
	auto t1 = std::chrono::system_clock::now();
	std::chrono::duration<double> dt = t1 - t0;
	std::cout << dt.count() << '
';

	return 0;
}

P.S.:
Explicit randomization marker is added because adding non-function pointer
will silently disable structure layout randomization.

[akpm@linux-foundation.org: coding style fixes]
Reported-by: kbuild test robot <lkp@intel.com>
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Alexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Joe Perches <joe@perches.com>
Link: http://lkml.kernel.org/r/20200222201539.GA22576@avx2
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

1 files changed, 137 insertions, 50 deletions

diff --git a/fs/proc/inode.c b/fs/proc/inode.c
index 05d31c464bee..fb4cace9ea41 100644
--- a/fs/proc/inode.c
+++ b/fs/proc/inode.c
@@ -259,135 +259,204 @@ void proc_entry_rundown(struct proc_dir_entry *de)
 	spin_unlock(&de->pde_unload_lock);
 }
 
+static loff_t pde_lseek(struct proc_dir_entry *pde, struct file *file, loff_t offset, int whence)
+{
+	typeof_member(struct proc_ops, proc_lseek) lseek;
+
+	lseek = pde->proc_ops->proc_lseek;
+	if (!lseek)
+		lseek = default_llseek;
+	return lseek(file, offset, whence);
+}
+
 static loff_t proc_reg_llseek(struct file *file, loff_t offset, int whence)
 {
 	struct proc_dir_entry *pde = PDE(file_inode(file));
 	loff_t rv = -EINVAL;
-	if (use_pde(pde)) {
-		typeof_member(struct proc_ops, proc_lseek) lseek;
 
-		lseek = pde->proc_ops->proc_lseek;
-		if (!lseek)
-			lseek = default_llseek;
-		rv = lseek(file, offset, whence);
+	if (pde_is_permanent(pde)) {
+		return pde_lseek(pde, file, offset, whence);
+	} else if (use_pde(pde)) {
+		rv = pde_lseek(pde, file, offset, whence);
 		unuse_pde(pde);
 	}
 	return rv;
 }
 
+static ssize_t pde_read(struct proc_dir_entry *pde, struct file *file, char __user *buf, size_t count, loff_t *ppos)
+{
+	typeof_member(struct proc_ops, proc_read) read;
+
+	read = pde->proc_ops->proc_read;
+	if (read)
+		return read(file, buf, count, ppos);
+	return -EIO;
+}
+
 static ssize_t proc_reg_read(struct file *file, char __user *buf, size_t count, loff_t *ppos)
 {
 	struct proc_dir_entry *pde = PDE(file_inode(file));
 	ssize_t rv = -EIO;
-	if (use_pde(pde)) {
-		typeof_member(struct proc_ops, proc_read) read;
 
-		read = pde->proc_ops->proc_read;
-		if (read)
-			rv = read(file, buf, count, ppos);
+	if (pde_is_permanent(pde)) {
+		return pde_read(pde, file, buf, count, ppos);
+	} else if (use_pde(pde)) {
+		rv = pde_read(pde, file, buf, count, ppos);
 		unuse_pde(pde);
 	}
 	return rv;
 }
 
+static ssize_t pde_write(struct proc_dir_entry *pde, struct file *file, const char __user *buf, size_t count, loff_t *ppos)
+{
+	typeof_member(struct proc_ops, proc_write) write;
+
+	write = pde->proc_ops->proc_write;
+	if (write)
+		return write(file, buf, count, ppos);
+	return -EIO;
+}
+
 static ssize_t proc_reg_write(struct file *file, const char __user *buf, size_t count, loff_t *ppos)
 {
 	struct proc_dir_entry *pde = PDE(file_inode(file));
 	ssize_t rv = -EIO;
-	if (use_pde(pde)) {
-		typeof_member(struct proc_ops, proc_write) write;
 
-		write = pde->proc_ops->proc_write;
-		if (write)
-			rv = write(file, buf, count, ppos);
+	if (pde_is_permanent(pde)) {
+		return pde_write(pde, file, buf, count, ppos);
+	} else if (use_pde(pde)) {
+		rv = pde_write(pde, file, buf, count, ppos);
 		unuse_pde(pde);
 	}
 	return rv;
 }
 
+static __poll_t pde_poll(struct proc_dir_entry *pde, struct file *file, struct poll_table_struct *pts)
+{
+	typeof_member(struct proc_ops, proc_poll) poll;
+
+	poll = pde->proc_ops->proc_poll;
+	if (poll)
+		return poll(file, pts);
+	return DEFAULT_POLLMASK;
+}
+
 static __poll_t proc_reg_poll(struct file *file, struct poll_table_struct *pts)
 {
 	struct proc_dir_entry *pde = PDE(file_inode(file));
 	__poll_t rv = DEFAULT_POLLMASK;
-	if (use_pde(pde)) {
-		typeof_member(struct proc_ops, proc_poll) poll;
 
-		poll = pde->proc_ops->proc_poll;
-		if (poll)
-			rv = poll(file, pts);
+	if (pde_is_permanent(pde)) {
+		return pde_poll(pde, file, pts);
+	} else if (use_pde(pde)) {
+		rv = pde_poll(pde, file, pts);
 		unuse_pde(pde);
 	}
 	return rv;
 }
 
+static long pde_ioctl(struct proc_dir_entry *pde, struct file *file, unsigned int cmd, unsigned long arg)
+{
+	typeof_member(struct proc_ops, proc_ioctl) ioctl;
+
+	ioctl = pde->proc_ops->proc_ioctl;
+	if (ioctl)
+		return ioctl(file, cmd, arg);
+	return -ENOTTY;
+}
+
 static long proc_reg_unlocked_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 {
 	struct proc_dir_entry *pde = PDE(file_inode(file));
 	long rv = -ENOTTY;
-	if (use_pde(pde)) {
-		typeof_member(struct proc_ops, proc_ioctl) ioctl;
 
-		ioctl = pde->proc_ops->proc_ioctl;
-		if (ioctl)
-			rv = ioctl(file, cmd, arg);
+	if (pde_is_permanent(pde)) {
+		return pde_ioctl(pde, file, cmd, arg);
+	} else if (use_pde(pde)) {
+		rv = pde_ioctl(pde, file, cmd, arg);
 		unuse_pde(pde);
 	}
 	return rv;
 }
 
 #ifdef CONFIG_COMPAT
+static long pde_compat_ioctl(struct proc_dir_entry *pde, struct file *file, unsigned int cmd, unsigned long arg)
+{
+	typeof_member(struct proc_ops, proc_compat_ioctl) compat_ioctl;
+
+	compat_ioctl = pde->proc_ops->proc_compat_ioctl;
+	if (compat_ioctl)
+		return compat_ioctl(file, cmd, arg);
+	return -ENOTTY;
+}
+
 static long proc_reg_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 {
 	struct proc_dir_entry *pde = PDE(file_inode(file));
 	long rv = -ENOTTY;
-	if (use_pde(pde)) {
-		typeof_member(struct proc_ops, proc_compat_ioctl) compat_ioctl;
-
-		compat_ioctl = pde->proc_ops->proc_compat_ioctl;
-		if (compat_ioctl)
-			rv = compat_ioctl(file, cmd, arg);
+	if (pde_is_permanent(pde)) {
+		return pde_compat_ioctl(pde, file, cmd, arg);
+	} else if (use_pde(pde)) {
+		rv = pde_compat_ioctl(pde, file, cmd, arg);
 		unuse_pde(pde);
 	}
 	return rv;
 }
 #endif
 
+static int pde_mmap(struct proc_dir_entry *pde, struct file *file, struct vm_area_struct *vma)
+{
+	typeof_member(struct proc_ops, proc_mmap) mmap;
+
+	mmap = pde->proc_ops->proc_mmap;
+	if (mmap)
+		return mmap(file, vma);
+	return -EIO;
+}
+
 static int proc_reg_mmap(struct file *file, struct vm_area_struct *vma)
 {
 	struct proc_dir_entry *pde = PDE(file_inode(file));
 	int rv = -EIO;
-	if (use_pde(pde)) {
-		typeof_member(struct proc_ops, proc_mmap) mmap;
 
-		mmap = pde->proc_ops->proc_mmap;
-		if (mmap)
-			rv = mmap(file, vma);
+	if (pde_is_permanent(pde)) {
+		return pde_mmap(pde, file, vma);
+	} else if (use_pde(pde)) {
+		rv = pde_mmap(pde, file, vma);
 		unuse_pde(pde);
 	}
 	return rv;
 }
 
 static unsigned long
-proc_reg_get_unmapped_area(struct file *file, unsigned long orig_addr,
+pde_get_unmapped_area(struct proc_dir_entry *pde, struct file *file, unsigned long orig_addr,
 			   unsigned long len, unsigned long pgoff,
 			   unsigned long flags)
 {
-	struct proc_dir_entry *pde = PDE(file_inode(file));
-	unsigned long rv = -EIO;
-
-	if (use_pde(pde)) {
-		typeof_member(struct proc_ops, proc_get_unmapped_area) get_area;
+	typeof_member(struct proc_ops, proc_get_unmapped_area) get_area;
 
-		get_area = pde->proc_ops->proc_get_unmapped_area;
+	get_area = pde->proc_ops->proc_get_unmapped_area;
 #ifdef CONFIG_MMU
-		if (!get_area)
-			get_area = current->mm->get_unmapped_area;
+	if (!get_area)
+		get_area = current->mm->get_unmapped_area;
 #endif
+	if (get_area)
+		return get_area(file, orig_addr, len, pgoff, flags);
+	return orig_addr;
+}
+
+static unsigned long
+proc_reg_get_unmapped_area(struct file *file, unsigned long orig_addr,
+			   unsigned long len, unsigned long pgoff,
+			   unsigned long flags)
+{
+	struct proc_dir_entry *pde = PDE(file_inode(file));
+	unsigned long rv = -EIO;
 
-		if (get_area)
-			rv = get_area(file, orig_addr, len, pgoff, flags);
-		else
-			rv = orig_addr;
+	if (pde_is_permanent(pde)) {
+		return pde_get_unmapped_area(pde, file, orig_addr, len, pgoff, flags);
+	} else if (use_pde(pde)) {
+		rv = pde_get_unmapped_area(pde, file, orig_addr, len, pgoff, flags);
 		unuse_pde(pde);
 	}
 	return rv;
@@ -401,6 +470,13 @@ static int proc_reg_open(struct inode *inode, struct file *file)
 	typeof_member(struct proc_ops, proc_release) release;
 	struct pde_opener *pdeo;
 
+	if (pde_is_permanent(pde)) {
+		open = pde->proc_ops->proc_open;
+		if (open)
+			rv = open(inode, file);
+		return rv;
+	}
+
 	/*
 	 * Ensure that
 	 * 1) PDE's ->release hook will be called no matter what
@@ -450,6 +526,17 @@ static int proc_reg_release(struct inode *inode, struct file *file)
 {
 	struct proc_dir_entry *pde = PDE(inode);
 	struct pde_opener *pdeo;
+
+	if (pde_is_permanent(pde)) {
+		typeof_member(struct proc_ops, proc_release) release;
+
+		release = pde->proc_ops->proc_release;
+		if (release) {
+			return release(inode, file);
+		}
+		return 0;
+	}
+
 	spin_lock(&pde->pde_unload_lock);
 	list_for_each_entry(pdeo, &pde->pde_openers, lh) {
 		if (pdeo->file == file) {