capability: just use a 'u64' instead of a 'u32[2]' array

Back in 2008 we extended the capability bits from 32 to 64, and we did
it by extending the single 32-bit capability word from one word to an
array of two words.  It was then obfuscated by hiding the "2" behind two
macro expansions, with the reasoning being that maybe it gets extended
further some day.

That reasoning may have been valid at the time, but the last thing we
want to do is to extend the capability set any more.  And the array of
values not only causes source code oddities (with loops to deal with
it), but also results in worse code generation.  It's a lose-lose
situation.

So just change the 'u32[2]' into a 'u64' and be done with it.

We still have to deal with the fact that the user space interface is
designed around an array of these 32-bit values, but that was the case
before too, since the array layouts were different (ie user space
doesn't use an array of 32-bit values for individual capability masks,
but an array of 32-bit slices of multiple masks).

So that marshalling of data is actually simplified too, even if it does
remain somewhat obscure and odd.

This was all triggered by my reaction to the new "cap_isidentical()"
introduced recently.  By just using a saner data structure, it went from

	unsigned __capi;
	CAP_FOR_EACH_U32(__capi) {
		if (a.cap[__capi] != b.cap[__capi])
			return false;
	}
	return true;

to just being

	return a.val == b.val;

instead.  Which is rather more obvious both to humans and to compilers.

Cc: Mateusz Guzik <mjguzik@gmail.com>
Cc: Casey Schaufler <casey@schaufler-ca.com>
Cc: Serge Hallyn <serge@hallyn.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Paul Moore <paul@paul-moore.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

1 files changed, 45 insertions, 59 deletions

diff --git a/kernel/capability.c b/kernel/capability.c
index 339a44dfe2f4..3e058f41df32 100644
--- a/kernel/capability.c
+++ b/kernel/capability.c
@@ -20,13 +20,6 @@
 #include <linux/user_namespace.h>
 #include <linux/uaccess.h>
 
-/*
- * Leveraged for setting/resetting capabilities
- */
-
-const kernel_cap_t __cap_empty_set = CAP_EMPTY_SET;
-EXPORT_SYMBOL(__cap_empty_set);
-
 int file_caps_enabled = 1;
 
 static int __init file_caps_disable(char *str)
@@ -151,6 +144,7 @@ SYSCALL_DEFINE2(capget, cap_user_header_t, header, cap_user_data_t, dataptr)
 	pid_t pid;
 	unsigned tocopy;
 	kernel_cap_t pE, pI, pP;
+	struct __user_cap_data_struct kdata[2];
 
 	ret = cap_validate_magic(header, &tocopy);
 	if ((dataptr == NULL) || (ret != 0))
@@ -163,42 +157,46 @@ SYSCALL_DEFINE2(capget, cap_user_header_t, header, cap_user_data_t, dataptr)
 		return -EINVAL;
 
 	ret = cap_get_target_pid(pid, &pE, &pI, &pP);
-	if (!ret) {
-		struct __user_cap_data_struct kdata[_KERNEL_CAPABILITY_U32S];
-		unsigned i;
-
-		for (i = 0; i < tocopy; i++) {
-			kdata[i].effective = pE.cap[i];
-			kdata[i].permitted = pP.cap[i];
-			kdata[i].inheritable = pI.cap[i];
-		}
-
-		/*
-		 * Note, in the case, tocopy < _KERNEL_CAPABILITY_U32S,
-		 * we silently drop the upper capabilities here. This
-		 * has the effect of making older libcap
-		 * implementations implicitly drop upper capability
-		 * bits when they perform a: capget/modify/capset
-		 * sequence.
-		 *
-		 * This behavior is considered fail-safe
-		 * behavior. Upgrading the application to a newer
-		 * version of libcap will enable access to the newer
-		 * capabilities.
-		 *
-		 * An alternative would be to return an error here
-		 * (-ERANGE), but that causes legacy applications to
-		 * unexpectedly fail; the capget/modify/capset aborts
-		 * before modification is attempted and the application
-		 * fails.
-		 */
-		if (copy_to_user(dataptr, kdata, tocopy
-				 * sizeof(struct __user_cap_data_struct))) {
-			return -EFAULT;
-		}
-	}
+	if (ret)
+		return ret;
 
-	return ret;
+	/*
+	 * Annoying legacy format with 64-bit capabilities exposed
+	 * as two sets of 32-bit fields, so we need to split the
+	 * capability values up.
+	 */
+	kdata[0].effective   = pE.val; kdata[1].effective   = pE.val >> 32;
+	kdata[0].permitted   = pP.val; kdata[1].permitted   = pP.val >> 32;
+	kdata[0].inheritable = pI.val; kdata[1].inheritable = pI.val >> 32;
+
+	/*
+	 * Note, in the case, tocopy < _KERNEL_CAPABILITY_U32S,
+	 * we silently drop the upper capabilities here. This
+	 * has the effect of making older libcap
+	 * implementations implicitly drop upper capability
+	 * bits when they perform a: capget/modify/capset
+	 * sequence.
+	 *
+	 * This behavior is considered fail-safe
+	 * behavior. Upgrading the application to a newer
+	 * version of libcap will enable access to the newer
+	 * capabilities.
+	 *
+	 * An alternative would be to return an error here
+	 * (-ERANGE), but that causes legacy applications to
+	 * unexpectedly fail; the capget/modify/capset aborts
+	 * before modification is attempted and the application
+	 * fails.
+	 */
+	if (copy_to_user(dataptr, kdata, tocopy * sizeof(kdata[0])))
+		return -EFAULT;
+
+	return 0;
+}
+
+static kernel_cap_t mk_kernel_cap(u32 low, u32 high)
+{
+	return (kernel_cap_t) { (low | ((u64)high << 32)) & CAP_VALID_MASK };
 }
 
 /**
@@ -221,8 +219,8 @@ SYSCALL_DEFINE2(capget, cap_user_header_t, header, cap_user_data_t, dataptr)
  */
 SYSCALL_DEFINE2(capset, cap_user_header_t, header, const cap_user_data_t, data)
 {
-	struct __user_cap_data_struct kdata[_KERNEL_CAPABILITY_U32S];
-	unsigned i, tocopy, copybytes;
+	struct __user_cap_data_struct kdata[2] = { { 0, }, };
+	unsigned tocopy, copybytes;
 	kernel_cap_t inheritable, permitted, effective;
 	struct cred *new;
 	int ret;
@@ -246,21 +244,9 @@ SYSCALL_DEFINE2(capset, cap_user_header_t, header, const cap_user_data_t, data)
 	if (copy_from_user(&kdata, data, copybytes))
 		return -EFAULT;
 
-	for (i = 0; i < tocopy; i++) {
-		effective.cap[i] = kdata[i].effective;
-		permitted.cap[i] = kdata[i].permitted;
-		inheritable.cap[i] = kdata[i].inheritable;
-	}
-	while (i < _KERNEL_CAPABILITY_U32S) {
-		effective.cap[i] = 0;
-		permitted.cap[i] = 0;
-		inheritable.cap[i] = 0;
-		i++;
-	}
-
-	effective.cap[CAP_LAST_U32] &= CAP_LAST_U32_VALID_MASK;
-	permitted.cap[CAP_LAST_U32] &= CAP_LAST_U32_VALID_MASK;
-	inheritable.cap[CAP_LAST_U32] &= CAP_LAST_U32_VALID_MASK;
+	effective   = mk_kernel_cap(kdata[0].effective,   kdata[1].effective);
+	permitted   = mk_kernel_cap(kdata[0].permitted,   kdata[1].permitted);
+	inheritable = mk_kernel_cap(kdata[0].inheritable, kdata[1].inheritable);
 
 	new = prepare_creds();
 	if (!new)