public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Jens Axboe <axboe@kernel.dk>
To: "Jason A. Donenfeld" <Jason@zx2c4.com>
Cc: tytso@mit.edu, hch@lst.de, linux-kernel@vger.kernel.org
Subject: Re: [PATCHSET 0/2] Fix splice from random/urandom
Date: Thu, 19 May 2022 15:02:28 -0600	[thread overview]
Message-ID: <8e6c98d4-03e9-3eb5-3d4e-b9a9faeb677a@kernel.dk> (raw)
In-Reply-To: <a6c843ff-a3d7-ce6a-4e99-70968834a02a@kernel.dk>

[-- Attachment #1: Type: text/plain, Size: 1860 bytes --]

On 5/19/22 2:49 PM, Jens Axboe wrote:
> On 5/19/22 2:05 PM, Jason A. Donenfeld wrote:
>> Hi Jens,
>>
>> On Thu, May 19, 2022 at 01:31:31PM -0600, Jens Axboe wrote:
>>> Hi,
>>>
>>> We recently had a failure on a kernel upgrade because splice no longer
>>> works on random/urandom. This is due to:
>>>
>>> 6e2c7421f02 ("fs: don't allow splice read/write without explicit ops")
>>
>> Thanks for this. I'd noticed this a few months ago and assumed it has
>> just always been that way, and hadn't gotten to looking at what was up.
>>
>> I'll take a look at these patches in detail when I'm home in a few
>> hours, but one thing maybe you can answer more easily than my digging
>> is:
> 
> Sounds good, thanks!
> 
>> There's a lot of attention in random.c devoted to not leaving any output
>> around on the stack or in stray buffers. The explicit use of
>> copy_to_user() makes it clear that the output isn't being copied
>> anywhere other than what's the user's responsibility to cleanup. I'm
>> wondering if the switch to copy_to_iter() introduces any buffering or
>> gotchas that you might be aware of.
> 
> No, it's just a wrapper around copying to the user memory pointed to by
> the iov_iter. No extra buffering or anything like that. So I think it
> should be fine in that respect, and it actually cleans up the code a bit
> imho since the copy_to_iter() since the return value of "bytes copied"
> is easier to work with than the "bytes not copied".
> 
>> Also you may need to rebase this on the random.git tree at
>> https://git.kernel.org/pub/scm/linux/kernel/git/crng/random.git
> 
> OK, I will rebase it on that branch, not a problem.

Rebased patches attached, you can also find them here:

https://git.kernel.dk/cgit/linux-block/log/?h=random-splice

Did some basic sanity checking (and with splice too), and seems fine
rebased as well.

-- 
Jens Axboe

[-- Attachment #2: 0002-random-wire-up-fops-splice_read_iter.patch --]
[-- Type: text/x-patch, Size: 1128 bytes --]

From fd673dac9f7c905d95cbe334cc7519fb36ede678 Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@kernel.dk>
Date: Thu, 19 May 2022 13:26:55 -0600
Subject: [PATCH 2/2] random: wire up fops->splice_read_iter()

Now that random/urandom is using read_iter, we can wire it up to using
the generic splice read handler.

Fixes: 36e2c7421f02 ("fs: don't allow splice read/write without explicit ops")
Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 drivers/char/random.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/char/random.c b/drivers/char/random.c
index 147921d471d8..892513fb7479 100644
--- a/drivers/char/random.c
+++ b/drivers/char/random.c
@@ -1415,6 +1415,7 @@ const struct file_operations random_fops = {
 	.compat_ioctl = compat_ptr_ioctl,
 	.fasync = random_fasync,
 	.llseek = noop_llseek,
+	.splice_read = generic_file_splice_read,
 };
 
 const struct file_operations urandom_fops = {
@@ -1424,6 +1425,7 @@ const struct file_operations urandom_fops = {
 	.compat_ioctl = compat_ptr_ioctl,
 	.fasync = random_fasync,
 	.llseek = noop_llseek,
+	.splice_read = generic_file_splice_read,
 };
 
 
-- 
2.35.1


[-- Attachment #3: 0001-random-convert-to-using-fops-read_iter.patch --]
[-- Type: text/x-patch, Size: 4254 bytes --]

From c65f0221b72173c19398884f3f49ef775e2c668f Mon Sep 17 00:00:00 2001
From: Jens Axboe <axboe@kernel.dk>
Date: Thu, 19 May 2022 14:55:37 -0600
Subject: [PATCH 1/2] random: convert to using fops->read_iter()

This is a pre-requisite to writing up splice() again for the random
and urandom drivers.

Signed-off-by: Jens Axboe <axboe@kernel.dk>
---
 drivers/char/random.c | 39 +++++++++++++++++++--------------------
 1 file changed, 19 insertions(+), 20 deletions(-)

diff --git a/drivers/char/random.c b/drivers/char/random.c
index 0958fa91a964..147921d471d8 100644
--- a/drivers/char/random.c
+++ b/drivers/char/random.c
@@ -397,11 +397,12 @@ void get_random_bytes(void *buf, size_t len)
 }
 EXPORT_SYMBOL(get_random_bytes);
 
-static ssize_t get_random_bytes_user(void __user *ubuf, size_t len)
+static ssize_t get_random_bytes_user(struct iov_iter *to)
 {
-	size_t block_len, left, ret = 0;
+	size_t block_len, ret = 0;
 	u32 chacha_state[CHACHA_STATE_WORDS];
 	u8 output[CHACHA_BLOCK_SIZE];
+	size_t len = iov_iter_count(to);
 
 	if (!len)
 		return 0;
@@ -418,7 +419,7 @@ static ssize_t get_random_bytes_user(void __user *ubuf, size_t len)
 	 * the user directly.
 	 */
 	if (len <= CHACHA_KEY_SIZE) {
-		ret = len - copy_to_user(ubuf, &chacha_state[4], len);
+		ret = copy_to_iter(&chacha_state[4], len, to);
 		goto out_zero_chacha;
 	}
 
@@ -428,17 +429,12 @@ static ssize_t get_random_bytes_user(void __user *ubuf, size_t len)
 			++chacha_state[13];
 
 		block_len = min_t(size_t, len, CHACHA_BLOCK_SIZE);
-		left = copy_to_user(ubuf, output, block_len);
-		if (left) {
-			ret += block_len - left;
+		block_len = copy_to_iter(output, block_len, to);
+		if (!block_len)
 			break;
-		}
 
-		ubuf += block_len;
 		ret += block_len;
 		len -= block_len;
-		if (!len)
-			break;
 
 		BUILD_BUG_ON(PAGE_SIZE % CHACHA_BLOCK_SIZE != 0);
 		if (ret % PAGE_SIZE == 0) {
@@ -1248,6 +1244,9 @@ static void __cold try_to_generate_entropy(void)
 
 SYSCALL_DEFINE3(getrandom, char __user *, ubuf, size_t, len, unsigned int, flags)
 {
+	struct iovec iov = { .iov_base = ubuf };
+	struct iov_iter iter;
+
 	if (flags & ~(GRND_NONBLOCK | GRND_RANDOM | GRND_INSECURE))
 		return -EINVAL;
 
@@ -1270,7 +1269,9 @@ SYSCALL_DEFINE3(getrandom, char __user *, ubuf, size_t, len, unsigned int, flags
 		if (unlikely(ret))
 			return ret;
 	}
-	return get_random_bytes_user(ubuf, len);
+	iov.iov_len = len;
+	iov_iter_init(&iter, READ, &iov, 1, len);
+	return get_random_bytes_user(&iter);
 }
 
 static __poll_t random_poll(struct file *file, poll_table *wait)
@@ -1314,8 +1315,7 @@ static ssize_t random_write(struct file *file, const char __user *ubuf,
 	return (ssize_t)len;
 }
 
-static ssize_t urandom_read(struct file *file, char __user *ubuf,
-			    size_t len, loff_t *ppos)
+static ssize_t urandom_read_iter(struct kiocb *kiocb, struct iov_iter *to)
 {
 	static int maxwarn = 10;
 
@@ -1332,22 +1332,21 @@ static ssize_t urandom_read(struct file *file, char __user *ubuf,
 		else if (ratelimit_disable || __ratelimit(&urandom_warning)) {
 			--maxwarn;
 			pr_notice("%s: uninitialized urandom read (%zd bytes read)\n",
-				  current->comm, len);
+				  current->comm, iov_iter_count(to));
 		}
 	}
 
-	return get_random_bytes_user(ubuf, len);
+	return get_random_bytes_user(to);
 }
 
-static ssize_t random_read(struct file *file, char __user *ubuf,
-			   size_t len, loff_t *ppos)
+static ssize_t random_read_iter(struct kiocb *kiocb, struct iov_iter *to)
 {
 	int ret;
 
 	ret = wait_for_random_bytes();
 	if (ret != 0)
 		return ret;
-	return get_random_bytes_user(ubuf, len);
+	return get_random_bytes_user(to);
 }
 
 static long random_ioctl(struct file *f, unsigned int cmd, unsigned long arg)
@@ -1409,7 +1408,7 @@ static int random_fasync(int fd, struct file *filp, int on)
 }
 
 const struct file_operations random_fops = {
-	.read = random_read,
+	.read_iter = random_read_iter,
 	.write = random_write,
 	.poll = random_poll,
 	.unlocked_ioctl = random_ioctl,
@@ -1419,7 +1418,7 @@ const struct file_operations random_fops = {
 };
 
 const struct file_operations urandom_fops = {
-	.read = urandom_read,
+	.read_iter = urandom_read_iter,
 	.write = random_write,
 	.unlocked_ioctl = random_ioctl,
 	.compat_ioctl = compat_ptr_ioctl,
-- 
2.35.1


  reply	other threads:[~2022-05-19 21:02 UTC|newest]

Thread overview: 37+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-05-19 19:31 [PATCHSET 0/2] Fix splice from random/urandom Jens Axboe
2022-05-19 19:31 ` [PATCH 1/2] random: convert to using fops->read_iter() Jens Axboe
2022-05-19 23:12   ` Jason A. Donenfeld
2022-05-19 23:20     ` Jason A. Donenfeld
2022-05-19 23:21       ` Jens Axboe
2022-05-19 23:21         ` Jason A. Donenfeld
2022-05-19 23:21     ` Jens Axboe
2022-05-19 19:31 ` [PATCH 2/2] random: wire up fops->splice_read_iter() Jens Axboe
2022-05-19 19:48 ` [PATCHSET 0/2] Fix splice from random/urandom Christoph Hellwig
2022-05-19 19:55   ` Jens Axboe
2022-05-20  6:02     ` Christoph Hellwig
2022-05-20 12:51       ` Jens Axboe
2022-05-19 20:05 ` Jason A. Donenfeld
2022-05-19 20:49   ` Jens Axboe
2022-05-19 21:02     ` Jens Axboe [this message]
2022-05-19 23:15       ` Jason A. Donenfeld
2022-05-19 23:22         ` Jens Axboe
2022-05-19 23:25           ` Jason A. Donenfeld
2022-05-19 23:33             ` Jens Axboe
2022-05-19 23:39               ` Jason A. Donenfeld
2022-05-19 23:44                 ` Jens Axboe
2022-05-19 23:13     ` Jason A. Donenfeld
2022-05-19 23:19       ` Jens Axboe
2022-05-19 23:23         ` Jason A. Donenfeld
2022-05-19 23:25           ` Jens Axboe
2022-05-19 23:27             ` Jason A. Donenfeld
2022-05-19 23:57               ` Jens Axboe
2022-05-20  0:00                 ` Jason A. Donenfeld
2022-05-20  0:02                   ` Jens Axboe
2022-05-20  0:48                     ` Jason A. Donenfeld
2022-05-20  0:56                       ` Jens Axboe
2022-05-20  1:00                         ` Jason A. Donenfeld
2022-05-20  1:05                           ` Jens Axboe
2022-05-20  1:10                           ` Jens Axboe
2022-05-20 12:43                             ` Jason A. Donenfeld
2022-05-20 12:49                               ` Jens Axboe
2022-05-20 13:04                                 ` Jason A. Donenfeld

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=8e6c98d4-03e9-3eb5-3d4e-b9a9faeb677a@kernel.dk \
    --to=axboe@kernel.dk \
    --cc=Jason@zx2c4.com \
    --cc=hch@lst.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=tytso@mit.edu \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox