sysctl: Convert to iter interfaces

Using the read_iter/write_iter interfaces allows for in-kernel users
to set sysctls without using set_fs().  Also, the buffer is a string,
so give it the real type of 'char *', not void *.

[AV: Christoph's fixup folded in]

Signed-off-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c
index 6c1166c..3178992 100644 (file)
--- a/fs/proc/proc_sysctl.c
+++ b/fs/proc/proc_sysctl.c
@@ -12,6 +12,7 @@
 #include <linux/cred.h>
 #include <linux/namei.h>
 #include <linux/mm.h>
+#include <linux/uio.h>
 #include <linux/module.h>
 #include <linux/bpf-cgroup.h>
 #include <linux/mount.h>
@@ -540,13 +541,14 @@ out:
        return err;
 }
 
-static ssize_t proc_sys_call_handler(struct file *filp, void __user *ubuf,
-               size_t count, loff_t *ppos, int write)
+static ssize_t proc_sys_call_handler(struct kiocb *iocb, struct iov_iter *iter,
+               int write)
 {
-       struct inode *inode = file_inode(filp);
+       struct inode *inode = file_inode(iocb->ki_filp);
        struct ctl_table_header *head = grab_header(inode);
        struct ctl_table *table = PROC_I(inode)->sysctl_entry;
-       void *kbuf;
+       size_t count = iov_iter_count(iter);
+       char *kbuf;
        ssize_t error;
 
        if (IS_ERR(head))
@@ -569,32 +571,30 @@ static ssize_t proc_sys_call_handler(struct file *filp, void __user *ubuf,
        error = -ENOMEM;
        if (count >= KMALLOC_MAX_SIZE)
                goto out;
+       kbuf = kzalloc(count + 1, GFP_KERNEL);
+       if (!kbuf)
+               goto out;
 
        if (write) {
-               kbuf = memdup_user_nul(ubuf, count);
-               if (IS_ERR(kbuf)) {
-                       error = PTR_ERR(kbuf);
-                       goto out;
-               }
-       } else {
-               kbuf = kzalloc(count, GFP_KERNEL);
-               if (!kbuf)
-                       goto out;
+               error = -EFAULT;
+               if (!copy_from_iter_full(kbuf, count, iter))
+                       goto out_free_buf;
+               kbuf[count] = '\0';
        }
 
        error = BPF_CGROUP_RUN_PROG_SYSCTL(head, table, write, &kbuf, &count,
-                                          ppos);
+                                          &iocb->ki_pos);
        if (error)
                goto out_free_buf;
 
        /* careful: calling conventions are nasty here */
-       error = table->proc_handler(table, write, kbuf, &count, ppos);
+       error = table->proc_handler(table, write, kbuf, &count, &iocb->ki_pos);
        if (error)
                goto out_free_buf;
 
        if (!write) {
                error = -EFAULT;
-               if (copy_to_user(ubuf, kbuf, count))
+               if (copy_to_iter(kbuf, count, iter) < count)
                        goto out_free_buf;
        }
 
@@ -607,16 +607,14 @@ out:
        return error;
 }
 
-static ssize_t proc_sys_read(struct file *filp, char __user *buf,
-                               size_t count, loff_t *ppos)
+static ssize_t proc_sys_read(struct kiocb *iocb, struct iov_iter *iter)
 {
-       return proc_sys_call_handler(filp, (void __user *)buf, count, ppos, 0);
+       return proc_sys_call_handler(iocb, iter, 0);
 }
 
-static ssize_t proc_sys_write(struct file *filp, const char __user *buf,
-                               size_t count, loff_t *ppos)
+static ssize_t proc_sys_write(struct kiocb *iocb, struct iov_iter *iter)
 {
-       return proc_sys_call_handler(filp, (void __user *)buf, count, ppos, 1);
+       return proc_sys_call_handler(iocb, iter, 1);
 }
 
 static int proc_sys_open(struct inode *inode, struct file *filp)
@@ -853,8 +851,10 @@ static int proc_sys_getattr(const struct path *path, struct kstat *stat,
 static const struct file_operations proc_sys_file_operations = {
        .open           = proc_sys_open,
        .poll           = proc_sys_poll,
-       .read           = proc_sys_read,
-       .write          = proc_sys_write,
+       .read_iter      = proc_sys_read,
+       .write_iter     = proc_sys_write,
+       .splice_read    = generic_file_splice_read,
+       .splice_write   = iter_file_splice_write,
        .llseek         = default_llseek,
 };
 

diff --git a/include/linux/bpf-cgroup.h b/include/linux/bpf-cgroup.h
index 64f3670..82b26a1 100644 (file)
--- a/include/linux/bpf-cgroup.h
+++ b/include/linux/bpf-cgroup.h
@@ -136,7 +136,7 @@ int __cgroup_bpf_check_dev_permission(short dev_type, u32 major, u32 minor,
 
 int __cgroup_bpf_run_filter_sysctl(struct ctl_table_header *head,
                                   struct ctl_table *table, int write,
-                                  void **buf, size_t *pcount, loff_t *ppos,
+                                  char **buf, size_t *pcount, loff_t *ppos,
                                   enum bpf_attach_type type);
 
 int __cgroup_bpf_run_filter_setsockopt(struct sock *sock, int *level,

diff --git a/kernel/bpf/cgroup.c b/kernel/bpf/cgroup.c
index 83ff127..2262993 100644 (file)
--- a/kernel/bpf/cgroup.c
+++ b/kernel/bpf/cgroup.c
@@ -1226,7 +1226,7 @@ const struct bpf_verifier_ops cg_dev_verifier_ops = {
  */
 int __cgroup_bpf_run_filter_sysctl(struct ctl_table_header *head,
                                   struct ctl_table *table, int write,
-                                  void **buf, size_t *pcount, loff_t *ppos,
+                                  char **buf, size_t *pcount, loff_t *ppos,
                                   enum bpf_attach_type type)
 {
        struct bpf_sysctl_kern ctx = {