linux-next: build failure after merge of the block tree

linux-next: build failure after merge of the block tree

[Stephen Rothwell]

Hi Jens,

After merging the block tree, today's linux-next build (powerpc
allnoconfig) failed like this:

fs/fcntl.o: In function `do_fcntl': 
fcntl.c:(.text+0x6d4): undefined reference to `__get_user_bad'
fcntl.c:(.text+0x730): undefined reference to `__get_user_bad'

Probably caused by commit

  c75b1d9421f8 ("fs: add fcntl() interface for setting/getting write life time hints")

On powerpc (at least) you cannot use get_user() to fetch anything larger
than "unsigned long" i.e. 32 bits on 32 bit powerpc.

This has been discussed before (and, I think, a fix attempted).

For now I have applied the following patch:

From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Wed, 28 Jun 2017 17:50:31 +1000
Subject: [PATCH] fs: don't read an enum directly from a u64 __user *

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 fs/fcntl.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/fs/fcntl.c b/fs/fcntl.c
index 24e233c75a33..19825eb7c40d 100644
--- a/fs/fcntl.c
+++ b/fs/fcntl.c
@@ -277,6 +277,7 @@ static long fcntl_rw_hint(struct file *file, unsigned int cmd,
 {
 	struct inode *inode = file_inode(file);
 	u64 *argp = (u64 __user *)arg;
+	u64 h;
 	enum rw_hint hint;
 
 	switch (cmd) {
@@ -285,8 +286,9 @@ static long fcntl_rw_hint(struct file *file, unsigned int cmd,
 			return -EFAULT;
 		return 0;
 	case F_SET_FILE_RW_HINT:
-		if (get_user(hint, argp))
+		if (copy_from_user(&h, argp, sizeof(h)))
 			return -EFAULT;
+		hint = (enum rw_hint)h;
 		if (!rw_hint_valid(hint))
 			return -EINVAL;
 
@@ -299,8 +301,9 @@ static long fcntl_rw_hint(struct file *file, unsigned int cmd,
 			return -EFAULT;
 		return 0;
 	case F_SET_RW_HINT:
-		if (get_user(hint, argp))
+		if (copy_from_user(&h, argp, sizeof(h)))
 			return -EFAULT;
+		hint = (enum rw_hint)h;
 		if (!rw_hint_valid(hint))
 			return -EINVAL;
 
-- 
2.11.0

-- 
Cheers,
Stephen Rothwell

Re: linux-next: build failure after merge of the block tree

[Jens Axboe]

On 06/28/2017 02:04 AM, Stephen Rothwell wrote:
> Hi Jens,
> 
> After merging the block tree, today's linux-next build (powerpc
> allnoconfig) failed like this:
> 
> fs/fcntl.o: In function `do_fcntl': 
> fcntl.c:(.text+0x6d4): undefined reference to `__get_user_bad'
> fcntl.c:(.text+0x730): undefined reference to `__get_user_bad'
> 
> Probably caused by commit
> 
>   c75b1d9421f8 ("fs: add fcntl() interface for setting/getting write life time hints")
> 
> On powerpc (at least) you cannot use get_user() to fetch anything larger
> than "unsigned long" i.e. 32 bits on 32 bit powerpc.
> 
> This has been discussed before (and, I think, a fix attempted).

Gah, thanks for letting me know. I'll test your patch and queue it
up to fix this issue.

-- 
Jens Axboe

Re: linux-next: build failure after merge of the block tree

[Jens Axboe]

On 06/28/2017 06:43 AM, Jens Axboe wrote:
> On 06/28/2017 02:04 AM, Stephen Rothwell wrote:
>> Hi Jens,
>>
>> After merging the block tree, today's linux-next build (powerpc
>> allnoconfig) failed like this:
>>
>> fs/fcntl.o: In function `do_fcntl': 
>> fcntl.c:(.text+0x6d4): undefined reference to `__get_user_bad'
>> fcntl.c:(.text+0x730): undefined reference to `__get_user_bad'
>>
>> Probably caused by commit
>>
>>   c75b1d9421f8 ("fs: add fcntl() interface for setting/getting write life time hints")
>>
>> On powerpc (at least) you cannot use get_user() to fetch anything larger
>> than "unsigned long" i.e. 32 bits on 32 bit powerpc.
>>
>> This has been discussed before (and, I think, a fix attempted).
> 
> Gah, thanks for letting me know. I'll test your patch and queue it
> up to fix this issue.

But put_user() is fine? Just checking here, since the change adds
both a u64 put and get user.

-- 
Jens Axboe

Re: linux-next: build failure after merge of the block tree

[Jens Axboe]

On 06/28/2017 08:01 AM, Jens Axboe wrote:
> On 06/28/2017 06:43 AM, Jens Axboe wrote:
>> On 06/28/2017 02:04 AM, Stephen Rothwell wrote:
>>> Hi Jens,
>>>
>>> After merging the block tree, today's linux-next build (powerpc
>>> allnoconfig) failed like this:
>>>
>>> fs/fcntl.o: In function `do_fcntl': 
>>> fcntl.c:(.text+0x6d4): undefined reference to `__get_user_bad'
>>> fcntl.c:(.text+0x730): undefined reference to `__get_user_bad'
>>>
>>> Probably caused by commit
>>>
>>>   c75b1d9421f8 ("fs: add fcntl() interface for setting/getting write life time hints")
>>>
>>> On powerpc (at least) you cannot use get_user() to fetch anything larger
>>> than "unsigned long" i.e. 32 bits on 32 bit powerpc.
>>>
>>> This has been discussed before (and, I think, a fix attempted).
>>
>> Gah, thanks for letting me know. I'll test your patch and queue it
>> up to fix this issue.
> 
> But put_user() is fine? Just checking here, since the change adds
> both a u64 put and get user.

I just changed all 4, at least that provides some symmetry in how
we copy things in and out for that set of fcntls.

-- 
Jens Axboe

Re: linux-next: build failure after merge of the block tree

[Stephen Rothwell]

Hi Jens,

On Wed, 28 Jun 2017 09:11:32 -0600 Jens Axboe <axboe@kernel.dk> wrote:
>
> On 06/28/2017 08:01 AM, Jens Axboe wrote:
> > But put_user() is fine? Just checking here, since the change adds
> > both a u64 put and get user.  

Yes, put_user is fine (it does 2 4 byte moves.  The asm is there to do
the 8 byte get_user, but the surrounding C code uses an unsigned long
for the destination in all cases (some other arches do the same).  I
don't remember why it is like that.

> I just changed all 4, at least that provides some symmetry in how
> we copy things in and out for that set of fcntls.

OK, thanks.

-- 
Cheers,
Stephen Rothwell