f2fs compile problem in next-20191220 on x86-32

f2fs compile problem in next-20191220 on x86-32

[Pavel Machek]

[-- Attachment #1: Type: text/plain, Size: 440 bytes --]

Hi!

I'm getting this:

  LD      .tmp_vmlinux1
  ld: fs/f2fs/file.o: in function `f2fs_truncate_blocks':
  file.c:(.text+0x2968): undefined reference to `__udivdi3'
  make: *** [Makefile:1079: vmlinux] Error 1

when attempting to compile kernel for x86-32.

Best regards,
									Pavel
-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 181 bytes --]

Re: f2fs compile problem in next-20191220 on x86-32

[Pavel Machek]

[-- Attachment #1: Type: text/plain, Size: 1535 bytes --]

On Sun 2019-12-22 16:49:17, Pavel Machek wrote:
> Hi!
> 
> I'm getting this:
> 
>   LD      .tmp_vmlinux1
>   ld: fs/f2fs/file.o: in function `f2fs_truncate_blocks':
>   file.c:(.text+0x2968): undefined reference to `__udivdi3'
>   make: *** [Makefile:1079: vmlinux] Error 1
> 
> when attempting to compile kernel for x86-32.

David bisected it:

https://bugzilla.kernel.org/show_bug.cgi?id=205967

And the bug is actually easy to see:

+int f2fs_truncate_blocks(struct inode *inode, u64 from, bool lock)
+{
+       u64 free_from = from;
+
+       /*
+        * for compressed file, only support cluster size
+        * aligned truncation.
+        */
+       if (f2fs_compressed_file(inode)) {
+               size_t cluster_size = PAGE_SIZE <<
+                                       F2FS_I(inode)->i_log_cluster_size;
+
+               free_from = roundup(from, cluster_size);

#define roundup(x, y) (                                 \
{                                                       \
        typeof(y) __y = y;                              \
        (((x) + (__y - 1)) / __y) * __y;                \
}                                                       \

div64 is needed instead of div in the roundup macro. Or actually... It
is quite stupid to use roundup like this on value that is power of
two, right?

Best regards,
									Pavel
-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 195 bytes --]

Re: f2fs compile problem in next-20191220 on x86-32

[Chao Yu]

On 2019/12/25 21:04, Pavel Machek wrote:
> On Sun 2019-12-22 16:49:17, Pavel Machek wrote:
>> Hi!
>>
>> I'm getting this:
>>
>>   LD      .tmp_vmlinux1
>>   ld: fs/f2fs/file.o: in function `f2fs_truncate_blocks':
>>   file.c:(.text+0x2968): undefined reference to `__udivdi3'
>>   make: *** [Makefile:1079: vmlinux] Error 1
>>
>> when attempting to compile kernel for x86-32.
> 
> David bisected it:
> 
> https://bugzilla.kernel.org/show_bug.cgi?id=205967
> 
> And the bug is actually easy to see:
> 
> +int f2fs_truncate_blocks(struct inode *inode, u64 from, bool lock)
> +{
> +       u64 free_from = from;
> +
> +       /*
> +        * for compressed file, only support cluster size
> +        * aligned truncation.
> +        */
> +       if (f2fs_compressed_file(inode)) {
> +               size_t cluster_size = PAGE_SIZE <<
> +                                       F2FS_I(inode)->i_log_cluster_size;
> +
> +               free_from = roundup(from, cluster_size);
> 
> #define roundup(x, y) (                                 \
> {                                                       \
>         typeof(y) __y = y;                              \
>         (((x) + (__y - 1)) / __y) * __y;                \
> }                                                       \
> 
> div64 is needed instead of div in the roundup macro. Or actually... It
> is quite stupid to use roundup like this on value that is power of
> two, right?

This has been fixed in dev branch, could you check that? People still saw this
issue because linux-next did not update after we fix this bug.

https://git.kernel.org/pub/scm/linux/kernel/git/jaegeuk/f2fs.git/commit/?h=dev

Thanks,

> 
> Best regards,
> 									Pavel
>