linux-next: build failure after merge of the device-mapper tree

linux-next: build failure after merge of the device-mapper tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 431 bytes --]

Hi all,

After merging the device-mapper tree, today's linux-next build (powerpc
ppc64_defconfig) failed like this:

ERROR: modpost: "key_type_encrypted" [drivers/md/dm-crypt.ko] undefined!

Caused by commit

  5cacab0334b9 ("dm crypt: support using encrypted keys")

CONFIG_ENCRYPTED_KEYS is not set for this build.

I have used the device-mapper tree from next-20200421 fot today.

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

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

[Mike Snitzer]

On Tue, Feb 06 2024 at 10:40P -0500,
Stephen Rothwell <sfr@canb.auug.org.au> wrote:

> Hi all,
> 
> After merging the device-mapper tree, today's linux-next build (x86_64
> allmodconfig) failed like this:
> 
> drivers/md/dm-vdo/io-factory.c: In function 'read_ahead':
> drivers/md/dm-vdo/io-factory.c:118:17: error: too few arguments to function 'dm_bufio_prefetch'
>   118 |                 dm_bufio_prefetch(reader->client, block_number, read_ahead);
>       |                 ^~~~~~~~~~~~~~~~~
> In file included from drivers/md/dm-vdo/io-factory.h:9,
>                  from drivers/md/dm-vdo/io-factory.c:6:
> include/linux/dm-bufio.h:86:6: note: declared here
>    86 | void dm_bufio_prefetch(struct dm_bufio_client *c,
>       |      ^~~~~~~~~~~~~~~~~
> drivers/md/dm-vdo/io-factory.c: In function 'position_reader':
> drivers/md/dm-vdo/io-factory.c:182:24: error: too few arguments to function 'dm_bufio_read'
>   182 |                 data = dm_bufio_read(reader->client, block_number, &buffer);
>       |                        ^~~~~~~~~~~~~
> include/linux/dm-bufio.h:64:7: note: declared here
>    64 | void *dm_bufio_read(struct dm_bufio_client *c, sector_t block,
>       |       ^~~~~~~~~~~~~
> 
> Caused by commit
> 
>   82da73bac1ee ("dm vdo: add deduplication index storage interface")
> 
> interacting with commit
> 
>   3be93545346e ("dm bufio: Support IO priority")
> 
> I have used the device-mapper tree from next-20240206 for today.

I've just fixed the device-mapper tree. Sorry for the trouble,

Mike

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

[Matthew Sakai]

Acknowledged. We will get this sorted out as soon as possible.

Matt

On 2/6/24 22:40, Stephen Rothwell wrote:
> Hi all,
> 
> After merging the device-mapper tree, today's linux-next build (x86_64
> allmodconfig) failed like this:
> 
> drivers/md/dm-vdo/io-factory.c: In function 'read_ahead':
> drivers/md/dm-vdo/io-factory.c:118:17: error: too few arguments to function 'dm_bufio_prefetch'
>    118 |                 dm_bufio_prefetch(reader->client, block_number, read_ahead);
>        |                 ^~~~~~~~~~~~~~~~~
> In file included from drivers/md/dm-vdo/io-factory.h:9,
>                   from drivers/md/dm-vdo/io-factory.c:6:
> include/linux/dm-bufio.h:86:6: note: declared here
>     86 | void dm_bufio_prefetch(struct dm_bufio_client *c,
>        |      ^~~~~~~~~~~~~~~~~
> drivers/md/dm-vdo/io-factory.c: In function 'position_reader':
> drivers/md/dm-vdo/io-factory.c:182:24: error: too few arguments to function 'dm_bufio_read'
>    182 |                 data = dm_bufio_read(reader->client, block_number, &buffer);
>        |                        ^~~~~~~~~~~~~
> include/linux/dm-bufio.h:64:7: note: declared here
>     64 | void *dm_bufio_read(struct dm_bufio_client *c, sector_t block,
>        |       ^~~~~~~~~~~~~
> 
> Caused by commit
> 
>    82da73bac1ee ("dm vdo: add deduplication index storage interface")
> 
> interacting with commit
> 
>    3be93545346e ("dm bufio: Support IO priority")
> 
> I have used the device-mapper tree from next-20240206 for today.
> 

linux-next: build failure after merge of the device-mapper tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1375 bytes --]

Hi all,

After merging the device-mapper tree, today's linux-next build (x86_64
allmodconfig) failed like this:

drivers/md/dm-vdo/io-factory.c: In function 'read_ahead':
drivers/md/dm-vdo/io-factory.c:118:17: error: too few arguments to function 'dm_bufio_prefetch'
  118 |                 dm_bufio_prefetch(reader->client, block_number, read_ahead);
      |                 ^~~~~~~~~~~~~~~~~
In file included from drivers/md/dm-vdo/io-factory.h:9,
                 from drivers/md/dm-vdo/io-factory.c:6:
include/linux/dm-bufio.h:86:6: note: declared here
   86 | void dm_bufio_prefetch(struct dm_bufio_client *c,
      |      ^~~~~~~~~~~~~~~~~
drivers/md/dm-vdo/io-factory.c: In function 'position_reader':
drivers/md/dm-vdo/io-factory.c:182:24: error: too few arguments to function 'dm_bufio_read'
  182 |                 data = dm_bufio_read(reader->client, block_number, &buffer);
      |                        ^~~~~~~~~~~~~
include/linux/dm-bufio.h:64:7: note: declared here
   64 | void *dm_bufio_read(struct dm_bufio_client *c, sector_t block,
      |       ^~~~~~~~~~~~~

Caused by commit

  82da73bac1ee ("dm vdo: add deduplication index storage interface")

interacting with commit

  3be93545346e ("dm bufio: Support IO priority")

I have used the device-mapper tree from next-20240206 for today.

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

linux-next: build failure after merge of the device-mapper tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1025 bytes --]

Hi all,

After merging the device-mapper tree, today's linux-next build (powerpc
ppc64_defconfig) failed like this:

drivers/md/persistent-data/dm-extent-allocator.c: In function 'dm_alloc_context_get':
drivers/md/persistent-data/dm-extent-allocator.c:530:31: error: expected ';' before '}' token
  530 |         spin_unlock(&ea->lock)
      |                               ^
      |                               ;
  531 | }
      | ~                              
drivers/md/persistent-data/dm-extent-allocator.c: In function 'dm_alloc_context_put':
drivers/md/persistent-data/dm-extent-allocator.c:544:31: error: expected ';' before '}' token
  544 |         spin_unlock(&ea->lock)
      |                               ^
      |                               ;
  545 | }
      | ~                              

Caused by commit

  59d814674dd6 ("dm persistent data: Introduce extent allocator")

I have used the device-mapper tree from next-2023-914 for today.

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

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

[Demi Marie Obenour]

[-- Attachment #1: Type: text/plain, Size: 3422 bytes --]

On Fri, Jun 23, 2023 at 02:30:11PM +1000, Stephen Rothwell wrote:
> Hi all,
> 
> After merging the device-mapper tree, today's linux-next build (i386
> defconfig) failed like this:
> 
> In file included from include/linux/kernel.h:30,
>                  from arch/x86/include/asm/percpu.h:27,
>                  from arch/x86/include/asm/current.h:10,
>                  from include/linux/sched.h:12,
>                  from include/linux/kthread.h:6,
>                  from drivers/md/dm-core.h:13,
>                  from drivers/md/dm-ioctl.c:9:
> drivers/md/dm-ioctl.c: In function 'next_target':
> include/linux/kern_levels.h:5:25: error: format '%lu' expects argument of type 'long unsigned int', but argument 3 has type 'unsigned int' [-Werror=format=]
>     5 | #define KERN_SOH        "\001"          /* ASCII Start Of Header */
>       |                         ^~~~~~
> include/linux/printk.h:427:25: note: in definition of macro 'printk_index_wrap'
>   427 |                 _p_func(_fmt, ##__VA_ARGS__);                           \
>       |                         ^~~~
> include/linux/printk.h:498:9: note: in expansion of macro 'printk'
>   498 |         printk(KERN_ERR pr_fmt(fmt), ##__VA_ARGS__)
>       |         ^~~~~~
> include/linux/kern_levels.h:11:25: note: in expansion of macro 'KERN_SOH'
>    11 | #define KERN_ERR        KERN_SOH "3"    /* error conditions */
>       |                         ^~~~~~~~
> include/linux/printk.h:498:16: note: in expansion of macro 'KERN_ERR'
>   498 |         printk(KERN_ERR pr_fmt(fmt), ##__VA_ARGS__)
>       |                ^~~~~~~~
> include/linux/device-mapper.h:626:25: note: in expansion of macro 'pr_err'
>   626 | #define DMERR(fmt, ...) pr_err(DM_FMT(fmt), ##__VA_ARGS__)
>       |                         ^~~~~~
> drivers/md/dm-ioctl.c:1421:17: note: in expansion of macro 'DMERR'
>  1421 |                 DMERR("Next dm_target_spec (offset %u) is not %lu-byte aligned",
>       |                 ^~~~~
> cc1: all warnings being treated as errors
> 
> Caused by commit
> 
>   5df1daff2cc6 ("dm ioctl: Check dm_target_spec is sufficiently aligned")

Ugh, sorry about that.  I’m pretty sure the code would actually work
okay at runtime since unsigned long and unsigned int are passed the same
way, but this is still a bug.

> I have applied the following patch for today:
> 
> From: Stephen Rothwell <sfr@canb.auug.org.au>
> Date: Fri, 23 Jun 2023 14:24:29 +1000
> Subject: [PATCH] fix up for "dm ioctl: Check dm_target_spec is sufficiently aligned"
> 
> Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
> ---
>  drivers/md/dm-ioctl.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c
> index 5f0b641538d8..8b480d7825fe 100644
> --- a/drivers/md/dm-ioctl.c
> +++ b/drivers/md/dm-ioctl.c
> @@ -1418,7 +1418,7 @@ static int next_target(struct dm_target_spec *last, uint32_t next, const char *e
>  	}
>  
>  	if (next % __alignof__(struct dm_target_spec)) {
> -		DMERR("Next dm_target_spec (offset %u) is not %lu-byte aligned",
> +		DMERR("Next dm_target_spec (offset %u) is not %zu-byte aligned",
>  		      next, __alignof__(struct dm_target_spec));
>  		return -EINVAL;
>  	}
> -- 
> 2.39.2

That fix is correct, thanks!
-- 
Sincerely,
Demi Marie Obenour (she/her/hers)
Invisible Things Lab

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

linux-next: build failure after merge of the device-mapper tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 2974 bytes --]

Hi all,

After merging the device-mapper tree, today's linux-next build (i386
defconfig) failed like this:

In file included from include/linux/kernel.h:30,
                 from arch/x86/include/asm/percpu.h:27,
                 from arch/x86/include/asm/current.h:10,
                 from include/linux/sched.h:12,
                 from include/linux/kthread.h:6,
                 from drivers/md/dm-core.h:13,
                 from drivers/md/dm-ioctl.c:9:
drivers/md/dm-ioctl.c: In function 'next_target':
include/linux/kern_levels.h:5:25: error: format '%lu' expects argument of type 'long unsigned int', but argument 3 has type 'unsigned int' [-Werror=format=]
    5 | #define KERN_SOH        "\001"          /* ASCII Start Of Header */
      |                         ^~~~~~
include/linux/printk.h:427:25: note: in definition of macro 'printk_index_wrap'
  427 |                 _p_func(_fmt, ##__VA_ARGS__);                           \
      |                         ^~~~
include/linux/printk.h:498:9: note: in expansion of macro 'printk'
  498 |         printk(KERN_ERR pr_fmt(fmt), ##__VA_ARGS__)
      |         ^~~~~~
include/linux/kern_levels.h:11:25: note: in expansion of macro 'KERN_SOH'
   11 | #define KERN_ERR        KERN_SOH "3"    /* error conditions */
      |                         ^~~~~~~~
include/linux/printk.h:498:16: note: in expansion of macro 'KERN_ERR'
  498 |         printk(KERN_ERR pr_fmt(fmt), ##__VA_ARGS__)
      |                ^~~~~~~~
include/linux/device-mapper.h:626:25: note: in expansion of macro 'pr_err'
  626 | #define DMERR(fmt, ...) pr_err(DM_FMT(fmt), ##__VA_ARGS__)
      |                         ^~~~~~
drivers/md/dm-ioctl.c:1421:17: note: in expansion of macro 'DMERR'
 1421 |                 DMERR("Next dm_target_spec (offset %u) is not %lu-byte aligned",
      |                 ^~~~~
cc1: all warnings being treated as errors

Caused by commit

  5df1daff2cc6 ("dm ioctl: Check dm_target_spec is sufficiently aligned")

I have applied the following patch for today:

From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Fri, 23 Jun 2023 14:24:29 +1000
Subject: [PATCH] fix up for "dm ioctl: Check dm_target_spec is sufficiently aligned"

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 drivers/md/dm-ioctl.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c
index 5f0b641538d8..8b480d7825fe 100644
--- a/drivers/md/dm-ioctl.c
+++ b/drivers/md/dm-ioctl.c
@@ -1418,7 +1418,7 @@ static int next_target(struct dm_target_spec *last, uint32_t next, const char *e
 	}
 
 	if (next % __alignof__(struct dm_target_spec)) {
-		DMERR("Next dm_target_spec (offset %u) is not %lu-byte aligned",
+		DMERR("Next dm_target_spec (offset %u) is not %zu-byte aligned",
 		      next, __alignof__(struct dm_target_spec));
 		return -EINVAL;
 	}
-- 
2.39.2

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

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

[Mike Snitzer]

On Wed, Feb 10 2021 at 10:36pm -0500,
Stephen Rothwell <sfr@canb.auug.org.au> wrote:

> Hi all,
> 
> After merging the device-mapper tree, today's linux-next build (powerpc
> ppc64_defconfig) failed like this:
> 
> drivers/md/dm-linear.c:234:3: error: 'struct target_type' has no member named 'report_zones'
>   234 |  .report_zones = linear_report_zones,
>       |   ^~~~~~~~~~~~
> drivers/md/dm-crypt.c:3585:3: error: 'struct target_type' has no member named 'report_zones'
>  3585 |  .report_zones = crypt_report_zones,
>       |   ^~~~~~~~~~~~
> 
> Caused by commit
> 
>   7defd0da9dd2 ("dm: simplify target code conditional on CONFIG_BLK_DEV_ZONED")
> 
> The report_zones members existence is guarded by CONFIG_BLK_DEV_ZONED.
> 
> I have used the device-mapper tree from next-20210210 for today.

Sorry, I didn't test with CONFIG_BLK_DEV_ZONED disabled, now fixed.

Thanks,
Mike

linux-next: build failure after merge of the device-mapper tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 730 bytes --]

Hi all,

After merging the device-mapper tree, today's linux-next build (powerpc
ppc64_defconfig) failed like this:

drivers/md/dm-linear.c:234:3: error: 'struct target_type' has no member named 'report_zones'
  234 |  .report_zones = linear_report_zones,
      |   ^~~~~~~~~~~~
drivers/md/dm-crypt.c:3585:3: error: 'struct target_type' has no member named 'report_zones'
 3585 |  .report_zones = crypt_report_zones,
      |   ^~~~~~~~~~~~

Caused by commit

  7defd0da9dd2 ("dm: simplify target code conditional on CONFIG_BLK_DEV_ZONED")

The report_zones members existence is guarded by CONFIG_BLK_DEV_ZONED.

I have used the device-mapper tree from next-20210210 for today.

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

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

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 187 bytes --]

Hi Mike,

On Thu, 6 Dec 2018 08:41:06 -0500 Mike Snitzer <snitzer@redhat.com> wrote:
>
> Sorry for the trouble.  I'll get this fixed up.

Thanks.

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

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

[Mike Snitzer]

On Thu, Dec 06 2018 at  1:21am -0500,
Stephen Rothwell <sfr@canb.auug.org.au> wrote:

> Hi all,
> 
> After merging the device-mapper tree, today's linux-next build (powerpc
> ppc44x_defconfig) failed like this:
... 
> Caused by commit
> 
>   3a657cf59109 ("block: switch to per-cpu in-flight counters")
> 
> This build does not have CONFIG_SMP set and part->dkstats is not a
> pointer in this case.
> 
> I have reverted the device-mapper tree changes for today.

Sorry for the trouble.  I'll get this fixed up.

Thanks,
Mike

linux-next: build failure after merge of the device-mapper tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 14170 bytes --]

Hi all,

After merging the device-mapper tree, today's linux-next build (powerpc
ppc44x_defconfig) failed like this:

In file included from arch/powerpc/include/asm/local.h:144,
                 from include/linux/genhd.h:20,
                 from block/genhd.c:7:
block/genhd.c: In function 'part_inc_in_flight':
include/linux/percpu-defs.h:219:52: error: invalid operands to binary + (have 'struct disk_stats' and 'int')
  const void __percpu *__vpp_verify = (typeof((ptr) + 0))NULL; \
                                              ~~~~~ ^
include/asm-generic/local.h:31:40: note: in definition of macro 'local_inc'
 #define local_inc(l) atomic_long_inc(&(l)->a)
                                        ^
include/linux/percpu-defs.h:259:2: note: in expansion of macro '__verify_pcpu_ptr'
  __verify_pcpu_ptr(__p);      \
  ^~~~~~~~~~~~~~~~~
include/linux/percpu-defs.h:263:47: note: in expansion of macro 'VERIFY_PERCPU_PTR'
 #define per_cpu_ptr(ptr, cpu) ({ (void)(cpu); VERIFY_PERCPU_PTR(ptr); })
                                               ^~~~~~~~~~~~~~~~~
block/genhd.c:56:13: note: in expansion of macro 'per_cpu_ptr'
  local_inc(&per_cpu_ptr(part->dkstats, cpu)->in_flight[rw]);
             ^~~~~~~~~~~
include/linux/percpu-defs.h:260:10: error: invalid type argument of unary '*' (have 'struct disk_stats')
  (typeof(*(__p)) __kernel __force *)(__p);   \
          ^~~~~~
include/asm-generic/local.h:31:40: note: in definition of macro 'local_inc'
 #define local_inc(l) atomic_long_inc(&(l)->a)
                                        ^
include/linux/percpu-defs.h:263:47: note: in expansion of macro 'VERIFY_PERCPU_PTR'
 #define per_cpu_ptr(ptr, cpu) ({ (void)(cpu); VERIFY_PERCPU_PTR(ptr); })
                                               ^~~~~~~~~~~~~~~~~
block/genhd.c:56:13: note: in expansion of macro 'per_cpu_ptr'
  local_inc(&per_cpu_ptr(part->dkstats, cpu)->in_flight[rw]);
             ^~~~~~~~~~~
include/linux/percpu-defs.h:219:52: error: invalid operands to binary + (have 'struct disk_stats' and 'int')
  const void __percpu *__vpp_verify = (typeof((ptr) + 0))NULL; \
                                              ~~~~~ ^
include/asm-generic/local.h:31:40: note: in definition of macro 'local_inc'
 #define local_inc(l) atomic_long_inc(&(l)->a)
                                        ^
include/linux/percpu-defs.h:259:2: note: in expansion of macro '__verify_pcpu_ptr'
  __verify_pcpu_ptr(__p);      \
  ^~~~~~~~~~~~~~~~~
include/linux/percpu-defs.h:263:47: note: in expansion of macro 'VERIFY_PERCPU_PTR'
 #define per_cpu_ptr(ptr, cpu) ({ (void)(cpu); VERIFY_PERCPU_PTR(ptr); })
                                               ^~~~~~~~~~~~~~~~~
block/genhd.c:58:14: note: in expansion of macro 'per_cpu_ptr'
   local_inc(&per_cpu_ptr(part_to_disk(part)->part0.dkstats, cpu)->in_flight[rw]);
              ^~~~~~~~~~~
include/linux/percpu-defs.h:260:10: error: invalid type argument of unary '*' (have 'struct disk_stats')
  (typeof(*(__p)) __kernel __force *)(__p);   \
          ^~~~~~
include/asm-generic/local.h:31:40: note: in definition of macro 'local_inc'
 #define local_inc(l) atomic_long_inc(&(l)->a)
                                        ^
include/linux/percpu-defs.h:263:47: note: in expansion of macro 'VERIFY_PERCPU_PTR'
 #define per_cpu_ptr(ptr, cpu) ({ (void)(cpu); VERIFY_PERCPU_PTR(ptr); })
                                               ^~~~~~~~~~~~~~~~~
block/genhd.c:58:14: note: in expansion of macro 'per_cpu_ptr'
   local_inc(&per_cpu_ptr(part_to_disk(part)->part0.dkstats, cpu)->in_flight[rw]);
              ^~~~~~~~~~~
block/genhd.c: In function 'part_dec_in_flight':
include/linux/percpu-defs.h:219:52: error: invalid operands to binary + (have 'struct disk_stats' and 'int')
  const void __percpu *__vpp_verify = (typeof((ptr) + 0))NULL; \
                                              ~~~~~ ^
include/asm-generic/local.h:32:40: note: in definition of macro 'local_dec'
 #define local_dec(l) atomic_long_dec(&(l)->a)
                                        ^
include/linux/percpu-defs.h:259:2: note: in expansion of macro '__verify_pcpu_ptr'
  __verify_pcpu_ptr(__p);      \
  ^~~~~~~~~~~~~~~~~
include/linux/percpu-defs.h:263:47: note: in expansion of macro 'VERIFY_PERCPU_PTR'
 #define per_cpu_ptr(ptr, cpu) ({ (void)(cpu); VERIFY_PERCPU_PTR(ptr); })
                                               ^~~~~~~~~~~~~~~~~
block/genhd.c:69:13: note: in expansion of macro 'per_cpu_ptr'
  local_dec(&per_cpu_ptr(part->dkstats, cpu)->in_flight[rw]);
             ^~~~~~~~~~~
include/linux/percpu-defs.h:260:10: error: invalid type argument of unary '*' (have 'struct disk_stats')
  (typeof(*(__p)) __kernel __force *)(__p);   \
          ^~~~~~
include/asm-generic/local.h:32:40: note: in definition of macro 'local_dec'
 #define local_dec(l) atomic_long_dec(&(l)->a)
                                        ^
include/linux/percpu-defs.h:263:47: note: in expansion of macro 'VERIFY_PERCPU_PTR'
 #define per_cpu_ptr(ptr, cpu) ({ (void)(cpu); VERIFY_PERCPU_PTR(ptr); })
                                               ^~~~~~~~~~~~~~~~~
block/genhd.c:69:13: note: in expansion of macro 'per_cpu_ptr'
  local_dec(&per_cpu_ptr(part->dkstats, cpu)->in_flight[rw]);
             ^~~~~~~~~~~
include/linux/percpu-defs.h:219:52: error: invalid operands to binary + (have 'struct disk_stats' and 'int')
  const void __percpu *__vpp_verify = (typeof((ptr) + 0))NULL; \
                                              ~~~~~ ^
include/asm-generic/local.h:32:40: note: in definition of macro 'local_dec'
 #define local_dec(l) atomic_long_dec(&(l)->a)
                                        ^
include/linux/percpu-defs.h:259:2: note: in expansion of macro '__verify_pcpu_ptr'
  __verify_pcpu_ptr(__p);      \
  ^~~~~~~~~~~~~~~~~
include/linux/percpu-defs.h:263:47: note: in expansion of macro 'VERIFY_PERCPU_PTR'
 #define per_cpu_ptr(ptr, cpu) ({ (void)(cpu); VERIFY_PERCPU_PTR(ptr); })
                                               ^~~~~~~~~~~~~~~~~
block/genhd.c:71:14: note: in expansion of macro 'per_cpu_ptr'
   local_dec(&per_cpu_ptr(part_to_disk(part)->part0.dkstats, cpu)->in_flight[rw]);
              ^~~~~~~~~~~
include/linux/percpu-defs.h:260:10: error: invalid type argument of unary '*' (have 'struct disk_stats')
  (typeof(*(__p)) __kernel __force *)(__p);   \
          ^~~~~~
include/asm-generic/local.h:32:40: note: in definition of macro 'local_dec'
 #define local_dec(l) atomic_long_dec(&(l)->a)
                                        ^
include/linux/percpu-defs.h:263:47: note: in expansion of macro 'VERIFY_PERCPU_PTR'
 #define per_cpu_ptr(ptr, cpu) ({ (void)(cpu); VERIFY_PERCPU_PTR(ptr); })
                                               ^~~~~~~~~~~~~~~~~
block/genhd.c:71:14: note: in expansion of macro 'per_cpu_ptr'
   local_dec(&per_cpu_ptr(part_to_disk(part)->part0.dkstats, cpu)->in_flight[rw]);
              ^~~~~~~~~~~
block/genhd.c: In function 'part_in_flight':
include/linux/percpu-defs.h:219:52: error: invalid operands to binary + (have 'struct disk_stats' and 'int')
  const void __percpu *__vpp_verify = (typeof((ptr) + 0))NULL; \
                                              ~~~~~ ^
include/asm-generic/local.h:29:42: note: in definition of macro 'local_read'
 #define local_read(l) atomic_long_read(&(l)->a)
                                          ^
include/linux/percpu-defs.h:259:2: note: in expansion of macro '__verify_pcpu_ptr'
  __verify_pcpu_ptr(__p);      \
  ^~~~~~~~~~~~~~~~~
include/linux/percpu-defs.h:263:47: note: in expansion of macro 'VERIFY_PERCPU_PTR'
 #define per_cpu_ptr(ptr, cpu) ({ (void)(cpu); VERIFY_PERCPU_PTR(ptr); })
                                               ^~~~~~~~~~~~~~~~~
block/genhd.c:85:27: note: in expansion of macro 'per_cpu_ptr'
   inflight += local_read(&per_cpu_ptr(part->dkstats, cpu)->in_flight[0]) +
                           ^~~~~~~~~~~
include/linux/percpu-defs.h:260:10: error: invalid type argument of unary '*' (have 'struct disk_stats')
  (typeof(*(__p)) __kernel __force *)(__p);   \
          ^~~~~~
include/asm-generic/local.h:29:42: note: in definition of macro 'local_read'
 #define local_read(l) atomic_long_read(&(l)->a)
                                          ^
include/linux/percpu-defs.h:263:47: note: in expansion of macro 'VERIFY_PERCPU_PTR'
 #define per_cpu_ptr(ptr, cpu) ({ (void)(cpu); VERIFY_PERCPU_PTR(ptr); })
                                               ^~~~~~~~~~~~~~~~~
block/genhd.c:85:27: note: in expansion of macro 'per_cpu_ptr'
   inflight += local_read(&per_cpu_ptr(part->dkstats, cpu)->in_flight[0]) +
                           ^~~~~~~~~~~
include/linux/percpu-defs.h:219:52: error: invalid operands to binary + (have 'struct disk_stats' and 'int')
  const void __percpu *__vpp_verify = (typeof((ptr) + 0))NULL; \
                                              ~~~~~ ^
include/asm-generic/local.h:29:42: note: in definition of macro 'local_read'
 #define local_read(l) atomic_long_read(&(l)->a)
                                          ^
include/linux/percpu-defs.h:259:2: note: in expansion of macro '__verify_pcpu_ptr'
  __verify_pcpu_ptr(__p);      \
  ^~~~~~~~~~~~~~~~~
include/linux/percpu-defs.h:263:47: note: in expansion of macro 'VERIFY_PERCPU_PTR'
 #define per_cpu_ptr(ptr, cpu) ({ (void)(cpu); VERIFY_PERCPU_PTR(ptr); })
                                               ^~~~~~~~~~~~~~~~~
block/genhd.c:86:17: note: in expansion of macro 'per_cpu_ptr'
     local_read(&per_cpu_ptr(part->dkstats, cpu)->in_flight[1]);
                 ^~~~~~~~~~~
include/linux/percpu-defs.h:260:10: error: invalid type argument of unary '*' (have 'struct disk_stats')
  (typeof(*(__p)) __kernel __force *)(__p);   \
          ^~~~~~
include/asm-generic/local.h:29:42: note: in definition of macro 'local_read'
 #define local_read(l) atomic_long_read(&(l)->a)
                                          ^
include/linux/percpu-defs.h:263:47: note: in expansion of macro 'VERIFY_PERCPU_PTR'
 #define per_cpu_ptr(ptr, cpu) ({ (void)(cpu); VERIFY_PERCPU_PTR(ptr); })
                                               ^~~~~~~~~~~~~~~~~
block/genhd.c:86:17: note: in expansion of macro 'per_cpu_ptr'
     local_read(&per_cpu_ptr(part->dkstats, cpu)->in_flight[1]);
                 ^~~~~~~~~~~
block/genhd.c: In function 'part_in_flight_rw':
include/linux/percpu-defs.h:219:52: error: invalid operands to binary + (have 'struct disk_stats' and 'int')
  const void __percpu *__vpp_verify = (typeof((ptr) + 0))NULL; \
                                              ~~~~~ ^
include/asm-generic/local.h:29:42: note: in definition of macro 'local_read'
 #define local_read(l) atomic_long_read(&(l)->a)
                                          ^
include/linux/percpu-defs.h:259:2: note: in expansion of macro '__verify_pcpu_ptr'
  __verify_pcpu_ptr(__p);      \
  ^~~~~~~~~~~~~~~~~
include/linux/percpu-defs.h:263:47: note: in expansion of macro 'VERIFY_PERCPU_PTR'
 #define per_cpu_ptr(ptr, cpu) ({ (void)(cpu); VERIFY_PERCPU_PTR(ptr); })
                                               ^~~~~~~~~~~~~~~~~
block/genhd.c:107:30: note: in expansion of macro 'per_cpu_ptr'
   inflight[0] += local_read(&per_cpu_ptr(part->dkstats, cpu)->in_flight[0]);
                              ^~~~~~~~~~~
include/linux/percpu-defs.h:260:10: error: invalid type argument of unary '*' (have 'struct disk_stats')
  (typeof(*(__p)) __kernel __force *)(__p);   \
          ^~~~~~
include/asm-generic/local.h:29:42: note: in definition of macro 'local_read'
 #define local_read(l) atomic_long_read(&(l)->a)
                                          ^
include/linux/percpu-defs.h:263:47: note: in expansion of macro 'VERIFY_PERCPU_PTR'
 #define per_cpu_ptr(ptr, cpu) ({ (void)(cpu); VERIFY_PERCPU_PTR(ptr); })
                                               ^~~~~~~~~~~~~~~~~
block/genhd.c:107:30: note: in expansion of macro 'per_cpu_ptr'
   inflight[0] += local_read(&per_cpu_ptr(part->dkstats, cpu)->in_flight[0]);
                              ^~~~~~~~~~~
include/linux/percpu-defs.h:219:52: error: invalid operands to binary + (have 'struct disk_stats' and 'int')
  const void __percpu *__vpp_verify = (typeof((ptr) + 0))NULL; \
                                              ~~~~~ ^
include/asm-generic/local.h:29:42: note: in definition of macro 'local_read'
 #define local_read(l) atomic_long_read(&(l)->a)
                                          ^
include/linux/percpu-defs.h:259:2: note: in expansion of macro '__verify_pcpu_ptr'
  __verify_pcpu_ptr(__p);      \
  ^~~~~~~~~~~~~~~~~
include/linux/percpu-defs.h:263:47: note: in expansion of macro 'VERIFY_PERCPU_PTR'
 #define per_cpu_ptr(ptr, cpu) ({ (void)(cpu); VERIFY_PERCPU_PTR(ptr); })
                                               ^~~~~~~~~~~~~~~~~
block/genhd.c:108:30: note: in expansion of macro 'per_cpu_ptr'
   inflight[1] += local_read(&per_cpu_ptr(part->dkstats, cpu)->in_flight[1]);
                              ^~~~~~~~~~~
include/linux/percpu-defs.h:260:10: error: invalid type argument of unary '*' (have 'struct disk_stats')
  (typeof(*(__p)) __kernel __force *)(__p);   \
          ^~~~~~
include/asm-generic/local.h:29:42: note: in definition of macro 'local_read'
 #define local_read(l) atomic_long_read(&(l)->a)
                                          ^
include/linux/percpu-defs.h:263:47: note: in expansion of macro 'VERIFY_PERCPU_PTR'
 #define per_cpu_ptr(ptr, cpu) ({ (void)(cpu); VERIFY_PERCPU_PTR(ptr); })
                                               ^~~~~~~~~~~~~~~~~
block/genhd.c:108:30: note: in expansion of macro 'per_cpu_ptr'
   inflight[1] += local_read(&per_cpu_ptr(part->dkstats, cpu)->in_flight[1]);
                              ^~~~~~~~~~~

Caused by commit

  3a657cf59109 ("block: switch to per-cpu in-flight counters")

This build does not have CONFIG_SMP set and part->dkstats is not a
pointer in this case.

I have reverted the device-mapper tree changes for today.
-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

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

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 647 bytes --]

Hi Jens,

On Thu, 31 May 2018 09:02:01 -0600 Jens Axboe <axboe@kernel.dk> wrote:
>
> Talked to Mike offline, but for Stephen's sake I'll reiterate here
> too. Since I'm assuming that dm-writecache is the only problematic
> user here, and since Mike based dm-4.18 on the block tree, I considered
> it saner to leave things as-is. This means that the API removal is still
> in the tree.

Understood.

> Stephen, if you see further conflicts due to that, do let me know and
> I'll do a proper revert of the patch instead. Hopefully it'll be smooth
> sailing from here on out.

Yes, thanks, will do.

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

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

[Jens Axboe]

On 5/31/18 8:22 AM, Jens Axboe wrote:
> On 5/31/18 8:20 AM, Jens Axboe wrote:
>> On 5/31/18 6:10 AM, Mike Snitzer wrote:
>>> On Thu, May 31 2018 at 12:35am -0400,
>>> Jens Axboe <axboe@kernel.dk> wrote:
>>>
>>>> On May 30, 2018, at 10:23 PM, Stephen Rothwell <sfr@canb.auug.org.au> wrote:
>>>>>
>>>>> Hi all,
>>>>>
>>>>> After merging the device-mapper tree, today's linux-next build (x86_64
>>>>> allmodconfig) failed like this:
>>>>>
>>>>> drivers/md/dm-writecache.c: In function 'writecache_dtr':
>>>>> drivers/md/dm-writecache.c:1799:3: error: implicit declaration of function 'bioset_free'; did you mean 'bvec_free'? [-Werror=implicit-function-declaration]
>>>>>   bioset_free(wc->bio_set);
>>>>>   ^~~~~~~~~~~
>>>>>   bvec_free
>>>>> drivers/md/dm-writecache.c: In function 'writecache_ctr':
>>>>> drivers/md/dm-writecache.c:1929:17: error: implicit declaration of function 'bioset_create'; did you mean 'bioset_exit'? [-Werror=implicit-function-declaration]
>>>>>   wc->bio_set = bioset_create(BIO_POOL_SIZE,
>>>>>                 ^~~~~~~~~~~~~
>>>>>                 bioset_exit
>>>>> drivers/md/dm-writecache.c:1929:15: warning: assignment makes pointer from integer without a cast [-Wint-conversion]
>>>>>   wc->bio_set = bioset_create(BIO_POOL_SIZE,
>>>>>               ^
>>>>>
>>>>> Caused by commit
>>>>>
>>>>>  2105231db61b ("dm: add writecache target")
>>>>>
>>>>> interacting with commit
>>>>>
>>>>>  dad08527525f ("block: Drop bioset_create()")
>>>>>
>>>>> from the block tree.
>>>>>
>>>>> Can we please consider not immediately dropping APIs if at all
>>>>> possible. :-(
>>>>
>>>> I’ll drop the last patch, we can do that at the end of the merge window instead. 
>>>>
>>>
>>> FYI, I've since updated dm-writecache to use the new APIs
>>
>> Thanks Mike. I've rebased for-4.18/block to fold a bug fix for the
>> bounce code, and drop the patch that gets rid of the old API. When
>> we get to the end of the merge window, I'll push that last patch.
>> That's much more sensible, how it should have been done from the
>> get-go.
> 
> Actually, maybe I should roll them back, since you both fixed that
> up AND based on my tree - which is of course totally fine, but
> now I'm thinking the rebase will do more harm than good in this
> case.

Talked to Mike offline, but for Stephen's sake I'll reiterate here
too. Since I'm assuming that dm-writecache is the only problematic
user here, and since Mike based dm-4.18 on the block tree, I considered
it saner to leave things as-is. This means that the API removal is still
in the tree.

Stephen, if you see further conflicts due to that, do let me know and
I'll do a proper revert of the patch instead. Hopefully it'll be smooth
sailing from here on out.

-- 
Jens Axboe

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

[Jens Axboe]

On 5/31/18 8:20 AM, Jens Axboe wrote:
> On 5/31/18 6:10 AM, Mike Snitzer wrote:
>> On Thu, May 31 2018 at 12:35am -0400,
>> Jens Axboe <axboe@kernel.dk> wrote:
>>
>>> On May 30, 2018, at 10:23 PM, Stephen Rothwell <sfr@canb.auug.org.au> wrote:
>>>>
>>>> Hi all,
>>>>
>>>> After merging the device-mapper tree, today's linux-next build (x86_64
>>>> allmodconfig) failed like this:
>>>>
>>>> drivers/md/dm-writecache.c: In function 'writecache_dtr':
>>>> drivers/md/dm-writecache.c:1799:3: error: implicit declaration of function 'bioset_free'; did you mean 'bvec_free'? [-Werror=implicit-function-declaration]
>>>>   bioset_free(wc->bio_set);
>>>>   ^~~~~~~~~~~
>>>>   bvec_free
>>>> drivers/md/dm-writecache.c: In function 'writecache_ctr':
>>>> drivers/md/dm-writecache.c:1929:17: error: implicit declaration of function 'bioset_create'; did you mean 'bioset_exit'? [-Werror=implicit-function-declaration]
>>>>   wc->bio_set = bioset_create(BIO_POOL_SIZE,
>>>>                 ^~~~~~~~~~~~~
>>>>                 bioset_exit
>>>> drivers/md/dm-writecache.c:1929:15: warning: assignment makes pointer from integer without a cast [-Wint-conversion]
>>>>   wc->bio_set = bioset_create(BIO_POOL_SIZE,
>>>>               ^
>>>>
>>>> Caused by commit
>>>>
>>>>  2105231db61b ("dm: add writecache target")
>>>>
>>>> interacting with commit
>>>>
>>>>  dad08527525f ("block: Drop bioset_create()")
>>>>
>>>> from the block tree.
>>>>
>>>> Can we please consider not immediately dropping APIs if at all
>>>> possible. :-(
>>>
>>> I’ll drop the last patch, we can do that at the end of the merge window instead. 
>>>
>>
>> FYI, I've since updated dm-writecache to use the new APIs
> 
> Thanks Mike. I've rebased for-4.18/block to fold a bug fix for the
> bounce code, and drop the patch that gets rid of the old API. When
> we get to the end of the merge window, I'll push that last patch.
> That's much more sensible, how it should have been done from the
> get-go.

Actually, maybe I should roll them back, since you both fixed that
up AND based on my tree - which is of course totally fine, but
now I'm thinking the rebase will do more harm than good in this
case.

-- 
Jens Axboe

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

[Jens Axboe]

On 5/31/18 6:10 AM, Mike Snitzer wrote:
> On Thu, May 31 2018 at 12:35am -0400,
> Jens Axboe <axboe@kernel.dk> wrote:
> 
>> On May 30, 2018, at 10:23 PM, Stephen Rothwell <sfr@canb.auug.org.au> wrote:
>>>
>>> Hi all,
>>>
>>> After merging the device-mapper tree, today's linux-next build (x86_64
>>> allmodconfig) failed like this:
>>>
>>> drivers/md/dm-writecache.c: In function 'writecache_dtr':
>>> drivers/md/dm-writecache.c:1799:3: error: implicit declaration of function 'bioset_free'; did you mean 'bvec_free'? [-Werror=implicit-function-declaration]
>>>   bioset_free(wc->bio_set);
>>>   ^~~~~~~~~~~
>>>   bvec_free
>>> drivers/md/dm-writecache.c: In function 'writecache_ctr':
>>> drivers/md/dm-writecache.c:1929:17: error: implicit declaration of function 'bioset_create'; did you mean 'bioset_exit'? [-Werror=implicit-function-declaration]
>>>   wc->bio_set = bioset_create(BIO_POOL_SIZE,
>>>                 ^~~~~~~~~~~~~
>>>                 bioset_exit
>>> drivers/md/dm-writecache.c:1929:15: warning: assignment makes pointer from integer without a cast [-Wint-conversion]
>>>   wc->bio_set = bioset_create(BIO_POOL_SIZE,
>>>               ^
>>>
>>> Caused by commit
>>>
>>>  2105231db61b ("dm: add writecache target")
>>>
>>> interacting with commit
>>>
>>>  dad08527525f ("block: Drop bioset_create()")
>>>
>>> from the block tree.
>>>
>>> Can we please consider not immediately dropping APIs if at all
>>> possible. :-(
>>
>> I’ll drop the last patch, we can do that at the end of the merge window instead. 
>>
> 
> FYI, I've since updated dm-writecache to use the new APIs

Thanks Mike. I've rebased for-4.18/block to fold a bug fix for the
bounce code, and drop the patch that gets rid of the old API. When
we get to the end of the merge window, I'll push that last patch.
That's much more sensible, how it should have been done from the
get-go.

-- 
Jens Axboe

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

[Mike Snitzer]

On Thu, May 31 2018 at 12:35am -0400,
Jens Axboe <axboe@kernel.dk> wrote:

> On May 30, 2018, at 10:23 PM, Stephen Rothwell <sfr@canb.auug.org.au> wrote:
> > 
> > Hi all,
> > 
> > After merging the device-mapper tree, today's linux-next build (x86_64
> > allmodconfig) failed like this:
> > 
> > drivers/md/dm-writecache.c: In function 'writecache_dtr':
> > drivers/md/dm-writecache.c:1799:3: error: implicit declaration of function 'bioset_free'; did you mean 'bvec_free'? [-Werror=implicit-function-declaration]
> >   bioset_free(wc->bio_set);
> >   ^~~~~~~~~~~
> >   bvec_free
> > drivers/md/dm-writecache.c: In function 'writecache_ctr':
> > drivers/md/dm-writecache.c:1929:17: error: implicit declaration of function 'bioset_create'; did you mean 'bioset_exit'? [-Werror=implicit-function-declaration]
> >   wc->bio_set = bioset_create(BIO_POOL_SIZE,
> >                 ^~~~~~~~~~~~~
> >                 bioset_exit
> > drivers/md/dm-writecache.c:1929:15: warning: assignment makes pointer from integer without a cast [-Wint-conversion]
> >   wc->bio_set = bioset_create(BIO_POOL_SIZE,
> >               ^
> > 
> > Caused by commit
> > 
> >  2105231db61b ("dm: add writecache target")
> > 
> > interacting with commit
> > 
> >  dad08527525f ("block: Drop bioset_create()")
> > 
> > from the block tree.
> > 
> > Can we please consider not immediately dropping APIs if at all
> > possible. :-(
> 
> I’ll drop the last patch, we can do that at the end of the merge window instead. 
> 

FYI, I've since updated dm-writecache to use the new APIs

Mike

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

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 219 bytes --]

Hi Jens,

On Wed, 30 May 2018 22:35:40 -0600 Jens Axboe <axboe@kernel.dk> wrote:
>
> I’ll drop the last patch, we can do that at the end of the merge window instead. 

Thanks.

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

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

[Jens Axboe]

On May 30, 2018, at 10:23 PM, Stephen Rothwell <sfr@canb.auug.org.au> wrote:
> 
> Hi all,
> 
> After merging the device-mapper tree, today's linux-next build (x86_64
> allmodconfig) failed like this:
> 
> drivers/md/dm-writecache.c: In function 'writecache_dtr':
> drivers/md/dm-writecache.c:1799:3: error: implicit declaration of function 'bioset_free'; did you mean 'bvec_free'? [-Werror=implicit-function-declaration]
>   bioset_free(wc->bio_set);
>   ^~~~~~~~~~~
>   bvec_free
> drivers/md/dm-writecache.c: In function 'writecache_ctr':
> drivers/md/dm-writecache.c:1929:17: error: implicit declaration of function 'bioset_create'; did you mean 'bioset_exit'? [-Werror=implicit-function-declaration]
>   wc->bio_set = bioset_create(BIO_POOL_SIZE,
>                 ^~~~~~~~~~~~~
>                 bioset_exit
> drivers/md/dm-writecache.c:1929:15: warning: assignment makes pointer from integer without a cast [-Wint-conversion]
>   wc->bio_set = bioset_create(BIO_POOL_SIZE,
>               ^
> 
> Caused by commit
> 
>  2105231db61b ("dm: add writecache target")
> 
> interacting with commit
> 
>  dad08527525f ("block: Drop bioset_create()")
> 
> from the block tree.
> 
> Can we please consider not immediately dropping APIs if at all
> possible. :-(

I’ll drop the last patch, we can do that at the end of the merge window instead. 

linux-next: build failure after merge of the device-mapper tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 3098 bytes --]

Hi all,

After merging the device-mapper tree, today's linux-next build (x86_64
allmodconfig) failed like this:

drivers/md/dm-writecache.c: In function 'writecache_dtr':
drivers/md/dm-writecache.c:1799:3: error: implicit declaration of function 'bioset_free'; did you mean 'bvec_free'? [-Werror=implicit-function-declaration]
   bioset_free(wc->bio_set);
   ^~~~~~~~~~~
   bvec_free
drivers/md/dm-writecache.c: In function 'writecache_ctr':
drivers/md/dm-writecache.c:1929:17: error: implicit declaration of function 'bioset_create'; did you mean 'bioset_exit'? [-Werror=implicit-function-declaration]
   wc->bio_set = bioset_create(BIO_POOL_SIZE,
                 ^~~~~~~~~~~~~
                 bioset_exit
drivers/md/dm-writecache.c:1929:15: warning: assignment makes pointer from integer without a cast [-Wint-conversion]
   wc->bio_set = bioset_create(BIO_POOL_SIZE,
               ^

Caused by commit

  2105231db61b ("dm: add writecache target")

interacting with commit

  dad08527525f ("block: Drop bioset_create()")

from the block tree.

Can we please consider not immediately dropping APIs if at all
possible. :-(

I have added the following merge fix patch (hopefully this is enough):

From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Thu, 31 May 2018 14:06:10 +1000
Subject: [PATCH] dm: fixup for "block: Drop bioset_create()"

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 drivers/md/dm-writecache.c | 12 +++++-------
 1 file changed, 5 insertions(+), 7 deletions(-)

diff --git a/drivers/md/dm-writecache.c b/drivers/md/dm-writecache.c
index 844c4fb2fcfc..b95f1f0e199d 100644
--- a/drivers/md/dm-writecache.c
+++ b/drivers/md/dm-writecache.c
@@ -211,7 +211,7 @@ struct dm_writecache {
 	struct completion flush_completion;
 	struct bio *flush_bio;
 
-	struct bio_set *bio_set;
+	struct bio_set bio_set;
 	mempool_t *copy_pool;
 
 	struct dm_kcopyd_client *dm_kcopyd;
@@ -1494,7 +1494,7 @@ static void __writecache_writeback_pmem(struct dm_writecache *wc, struct writeba
 
 		max_pages = e->wc_list_contiguous;
 
-		bio = bio_alloc_bioset(GFP_NOIO, max_pages, wc->bio_set);
+		bio = bio_alloc_bioset(GFP_NOIO, max_pages, &wc->bio_set);
 		wb = container_of(bio, struct writeback_struct, bio);
 		wb->wc = wc;
 		wb->bio.bi_end_io = writecache_writeback_endio;
@@ -1795,8 +1795,7 @@ static void writecache_dtr(struct dm_target *ti)
 	if (wc->flush_thread)
 		kthread_stop(wc->flush_thread);
 
-	if (wc->bio_set)
-		bioset_free(wc->bio_set);
+	bioset_exit(&wc->bio_set);
 
 	mempool_destroy(wc->copy_pool);
 
@@ -1926,11 +1925,10 @@ static int writecache_ctr(struct dm_target *ti, unsigned argc, char **argv)
 	}
 
 	if (WC_MODE_PMEM(wc)) {
-		wc->bio_set = bioset_create(BIO_POOL_SIZE,
+		r = bioset_init(&wc->bio_set, BIO_POOL_SIZE,
 					    offsetof(struct writeback_struct, bio),
 					    BIOSET_NEED_BVECS);
-		if (!wc->bio_set) {
-			r = -ENOMEM;
+		if (r) {
 			ti->error = "Could not allocate bio set";
 			goto bad;
 		}
-- 
2.17.0

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

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

[Bart Van Assche]

On 03/26/18 19:39, Stephen Rothwell wrote:
> diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
> index d5db9df7ea4b..0589a4da12bb 100644
> --- a/drivers/md/dm-table.c
> +++ b/drivers/md/dm-table.c
> @@ -1896,7 +1896,7 @@ void dm_table_set_restrictions(struct dm_table *t, struct request_queue *q,
>   		blk_queue_flag_set(QUEUE_FLAG_DISCARD, q);
>   
>   	if (dm_table_supports_secure_erase(t))
> -		queue_flag_set_unlocked(QUEUE_FLAG_SECERASE, q);
> +		blk_queue_flag_set(QUEUE_FLAG_SECERASE, q);
>   
>   	if (dm_table_supports_flush(t, (1UL << QUEUE_FLAG_WC))) {
>   		wc = true;
> 

The above patch looks fine to me. Thank you Stephen for having resolved 
this merge conflict.

Bart.

linux-next: build failure after merge of the device-mapper tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1678 bytes --]

Hi all,

After merging the device-mapper tree, today's linux-next build (powerpc
ppc64_defconfig) failed like this:

/home/sfr/next/next/drivers/md/dm-table.c: In function 'dm_table_set_restrictions':
/home/sfr/next/next/drivers/md/dm-table.c:1899:3: error: implicit declaration of function 'queue_flag_set_unlocked'; did you mean 'lookup_one_len_unlocked'? [-Werror=implicit-function-declaration]
   queue_flag_set_unlocked(QUEUE_FLAG_SECERASE, q);
   ^~~~~~~~~~~~~~~~~~~~~~~

Caused by commit

  9723d9db1a84 ("dm: add support for secure erase forwarding")

interacting with commit

  8a0ac14b8da9 ("block: Move the queue_flag_*() functions from a public into a private header file")

from the block tree.

I have added the following merge fix patch:

From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 27 Mar 2018 13:36:49 +1100
Subject: [PATCH] dm: merge fix for queue_flag_set_unlocked to
 blk_queue_flag_set API change

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 drivers/md/dm-table.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/md/dm-table.c b/drivers/md/dm-table.c
index d5db9df7ea4b..0589a4da12bb 100644
--- a/drivers/md/dm-table.c
+++ b/drivers/md/dm-table.c
@@ -1896,7 +1896,7 @@ void dm_table_set_restrictions(struct dm_table *t, struct request_queue *q,
 		blk_queue_flag_set(QUEUE_FLAG_DISCARD, q);
 
 	if (dm_table_supports_secure_erase(t))
-		queue_flag_set_unlocked(QUEUE_FLAG_SECERASE, q);
+		blk_queue_flag_set(QUEUE_FLAG_SECERASE, q);
 
 	if (dm_table_supports_flush(t, (1UL << QUEUE_FLAG_WC))) {
 		wc = true;
-- 
2.16.1

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

linux-next: build failure after merge of the device-mapper tree

[Stephen Rothwell]

Hi all,

After merging the device-mapper tree, today's linux-next build (x86_64
allmodconfig) failed like this:

drivers/md/dm-zoned-target.c: In function 'dmz_ctr':
drivers/md/dm-zoned-target.c:779:17: error: implicit declaration of function 'bioset_create_nobvec' [-Werror=implicit-function-declaration]
  dmz->bio_set = bioset_create_nobvec(DMZ_MIN_BIOS, 0);
                 ^
drivers/md/dm-zoned-target.c:779:15: warning: assignment makes pointer from integer without a cast [-Wint-conversion]
  dmz->bio_set = bioset_create_nobvec(DMZ_MIN_BIOS, 0);
               ^

Caused by commit

  1e6562930ed7 ("dm zoned: drive-managed zoned block device target")

interacting with commit

  011067b05668 ("blk: replace bioset_create_nobvec() with a flags arg to bioset_create()")

from the block tree.

I have applied this merge fix patch for today:

From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Mon, 19 Jun 2017 12:21:38 +1000
Subject: [PATCH] dm zoned: merge fix for "blk: replace bioset_create_nobvec()
 with a flags arg to bioset_create()"

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 drivers/md/dm-zoned-target.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/md/dm-zoned-target.c b/drivers/md/dm-zoned-target.c
index d25696caa7f3..2b538fa817f4 100644
--- a/drivers/md/dm-zoned-target.c
+++ b/drivers/md/dm-zoned-target.c
@@ -776,7 +776,7 @@ static int dmz_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 	ti->len = (sector_t)dmz_nr_chunks(dmz->metadata) << dev->zone_nr_sectors_shift;
 
 	/* Zone BIO */
-	dmz->bio_set = bioset_create_nobvec(DMZ_MIN_BIOS, 0);
+	dmz->bio_set = bioset_create(DMZ_MIN_BIOS, 0, 0);
 	if (!dmz->bio_set) {
 		ti->error = "Create BIO set failed";
 		ret = -ENOMEM;
-- 
2.11.0

-- 
Cheers,
Stephen Rothwell

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

[Mike Snitzer]

On Fri, Jun 09 2017 at 12:28am -0400,
Damien Le Moal <damien.lemoal@wdc.com> wrote:

> Stephen,
> 
> On 6/9/17 12:12, Stephen Rothwell wrote:
> > Hi all,
> > 
> > After merging the device-mapper tree, today's linux-next build (x86_64
> > allmodconfig) failed like this:
> > 
> > drivers/md/dm-zoned-target.c:947:2: error: unknown field 'suspend' specified in initializer
> >   .suspend  = dmz_suspend,
> >   ^
> > drivers/md/dm-zoned-target.c:947:14: error: initialization from incompatible pointer type [-Werror=incompatible-pointer-types]
> >   .suspend  = dmz_suspend,
> >               ^
> > 
> > Caused by commit
> > 
> >   02da2e15e81f ("dm zoned: drive-managed zoned block device target")
> > 
> > I have used the device-mapper tree from next-20170608 for today.
> 
> My apologies for that. My mistake.
> I just posted a patch to dm-devel to fix this.
> Everything should come in order after Mike's review.

I'm to blame for the above issue.  I've now fixed it up.

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

[Damien Le Moal]

Stephen,

On 6/9/17 12:12, Stephen Rothwell wrote:
> Hi all,
> 
> After merging the device-mapper tree, today's linux-next build (x86_64
> allmodconfig) failed like this:
> 
> drivers/md/dm-zoned-target.c:947:2: error: unknown field 'suspend' specified in initializer
>   .suspend  = dmz_suspend,
>   ^
> drivers/md/dm-zoned-target.c:947:14: error: initialization from incompatible pointer type [-Werror=incompatible-pointer-types]
>   .suspend  = dmz_suspend,
>               ^
> 
> Caused by commit
> 
>   02da2e15e81f ("dm zoned: drive-managed zoned block device target")
> 
> I have used the device-mapper tree from next-20170608 for today.

My apologies for that. My mistake.
I just posted a patch to dm-devel to fix this.
Everything should come in order after Mike's review.

Best regards.

-- 
Damien Le Moal, Ph.D.
Sr Manager, System Software Group,
Western Digital Research
Damien.LeMoal@wdc.com
Tel: (+81) 0466-98-3593 (Ext. 51-3593)
1 kirihara-cho, Fujisawa, Kanagawa, 252-0888 Japan
www.wdc.com, www.hgst.com

linux-next: build failure after merge of the device-mapper tree

[Stephen Rothwell]

Hi all,

After merging the device-mapper tree, today's linux-next build (x86_64
allmodconfig) failed like this:

drivers/md/dm-zoned-target.c:947:2: error: unknown field 'suspend' specified in initializer
  .suspend  = dmz_suspend,
  ^
drivers/md/dm-zoned-target.c:947:14: error: initialization from incompatible pointer type [-Werror=incompatible-pointer-types]
  .suspend  = dmz_suspend,
              ^

Caused by commit

  02da2e15e81f ("dm zoned: drive-managed zoned block device target")

I have used the device-mapper tree from next-20170608 for today.

-- 
Cheers,
Stephen Rothwell

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

[Mike Snitzer]

On Thu, Mar 03 2016 at  9:59pm -0500,
Stephen Rothwell <sfr@canb.auug.org.au> wrote:

> Hi all,
> 
> After merging the device-mapper tree, today's linux-next build (powerpc
> ppc64_defconfig) failed like this:
> 
> drivers/md/dm.c: In function '__split_and_process_non_flush':
> drivers/md/dm.c:1761:6: error: 'ret' undeclared (first use in this function)
>   if (ret < 0)
>       ^
> 
> Caused by commit
> 
>   f838997d1784 ("dm: return error if bio_integrity_clone() fails in clone_bio()")
> 
> That one is a bit annoying ... clearly not build tested :-(

Yeap, I'll just let that go.

> I have used the device-mapper tree from next-20160303 for today.

Not sure what happened.  I've fixed it.  next-20160303 is fine though
since the bad commit was pushed today.

linux-next: build failure after merge of the device-mapper tree

[Stephen Rothwell]

Hi all,

After merging the device-mapper tree, today's linux-next build (powerpc
ppc64_defconfig) failed like this:

drivers/md/dm.c: In function '__split_and_process_non_flush':
drivers/md/dm.c:1761:6: error: 'ret' undeclared (first use in this function)
  if (ret < 0)
      ^

Caused by commit

  f838997d1784 ("dm: return error if bio_integrity_clone() fails in clone_bio()")

That one is a bit annoying ... clearly not build tested :-(

I have used the device-mapper tree from next-20160303 for today.

-- 
Cheers,
Stephen Rothwell

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

[Mike Snitzer]

On Sun, Nov 01 2015 at  5:43am -0500,
Stephen Rothwell <sfr@canb.auug.org.au> wrote:

> Hi all,
> 
> After merging the device-mapper tree, today's linux-next build (powerpc
> ppc64_defconfig) failed like this:
> 
> drivers/md/dm-mpath.c: In function 'multipath_prepare_ioctl':
> drivers/md/dm-mpath.c:1570:41: error: 'cmd' undeclared (first use in this function)
>    int err = scsi_verify_blk_ioctl(NULL, cmd);
>                                          ^
> 
> Caused by commit
> 
>   21a2807bc3ff ("dm: refactor ioctl handling")
> 
> I have used the device-mapper tree from next-20151022 for today.

There was a ~3 hour window yesterday where this was broken.  I resolved
it almost 12 hours prior to you having sent this mail.

Do you batch update the trees you pull into linux-next at some point
much earlier in the day?

If/when you do hit a tree that is clearly broken at the time you
pulled: do you attempt to re-pull to get the latest _before_ reverting
to some prior linux-next version of the tree in question?

Anyway, sorry the device-mapper tree was broken for a small window of
time yesterday -- but of course linux-next pulled during that window ;)

linux-next: build failure after merge of the device-mapper tree

[Stephen Rothwell]

Hi all,

After merging the device-mapper tree, today's linux-next build (powerpc
ppc64_defconfig) failed like this:

drivers/md/dm-mpath.c: In function 'multipath_prepare_ioctl':
drivers/md/dm-mpath.c:1570:41: error: 'cmd' undeclared (first use in this function)
   int err = scsi_verify_blk_ioctl(NULL, cmd);
                                         ^

Caused by commit

  21a2807bc3ff ("dm: refactor ioctl handling")

I have used the device-mapper tree from next-20151022 for today.

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

linux-next: build failure after merge of the device-mapper tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1064 bytes --]

Hi all,

After merging the device-mapper tree, today's linux-next build (x86_64
allmodconfig) failed like this:

drivers/md/dm-thin.c: In function '__pool_destroy':
drivers/md/dm-thin.c:2780:2: error: implicit declaration of function 'vfree' [-Werror=implicit-function-declaration]
  vfree(pool->cell_sort_array);
  ^
drivers/md/dm-thin.c: In function 'pool_create':
drivers/md/dm-thin.c:2893:2: error: implicit declaration of function 'vmalloc' [-Werror=implicit-function-declaration]
  pool->cell_sort_array = vmalloc(sizeof(*pool->cell_sort_array) * CELL_SORT_ARRAY_SIZE);
  ^
drivers/md/dm-thin.c:2893:24: warning: assignment makes pointer from integer without a cast
  pool->cell_sort_array = vmalloc(sizeof(*pool->cell_sort_array) * CELL_SORT_ARRAY_SIZE);
                        ^

Caused by commit

  94212d3b6054 ("dm thin: allocate the cell_sort_array dynamically")

Forgot to include vmalloc.h

I have used the device-mapper from next-20150703 for today.

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

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

[Jens Axboe]

On 05/22/2015 07:26 AM, Mike Snitzer wrote:
> On Fri, May 22 2015 at  8:49am -0400,
> Mike Snitzer <snitzer@redhat.com> wrote:
>
>> On Thu, May 21 2015 at 11:27pm -0400,
>> Stephen Rothwell <sfr@canb.auug.org.au> wrote:
>>
>>> Hi all,
>>>
>>> After merging the device-mapper tree, today's linux-next build (powerpc
>>> ppc64_defconfig) failed like this:
>>>
>>> block/bio.c:310:20: error: redefinition of 'bio_inc_remaining'
>>>   static inline void bio_inc_remaining(struct bio *bio)
>>>                      ^
>>> In file included from block/bio.c:20:0:
>>> include/linux/bio.h:664:20: note: previous definition of 'bio_inc_remaining' was here
>>>   static inline void bio_inc_remaining(struct bio *bio)
>>>                      ^
>>>
>>> Caused by commit d61b82161c5a ("block: remove management of
>>> bi_remaining when restoring original bi_end_io").
>>>
>>> OK, this is a bad automatic merge caused by the fact that the block
>>> tree and the device-mapper tree both contain the same set fo patches as
>>> different commits.
>>>
>>> Commit c4cf5261f8bf from the block tree is the same as commit
>>> 8395711c4f90 from the device-mapper tree and both add
>>> bio_inc_remaining() to include/linux/bio.h.  The above commit in the
>>> device-mapper tree removes it from there, but the merge decided to keep
>>> it.
>>>
>>> That shared patch above is part of a series of 14 patches that are
>>> identical patches but different commits and are causing quite a few
>>> conflicts.  It looks like those patches were rebased onto v4.10rc3 as
>>> the base for the device-mapper tree.  A better idea would have been to
>>> start with v4.1rc3 (if that is necessary) and merge commit 5b3f341f098d
>>> ("blk-mq: make plug work for mutiple disks and queues") from the block
>>> tree - assuming that Jens will guarantee not to rebase that part of his
>>> tree.
>>>
>>> Please clean this mess up.
>>
>> Will do.
>
> I've fixed linux-dm's for-next to be properly based on linux-block's
> for-4.2/core -- I also resent Jens three block patches for 4.2 that
> motivated me to rebase ontop of block to begin with.

3/3 is already in for-linus, it'll go in for 4.1. 1-2 will go to 4.2/core.


-- 
Jens Axboe

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

[Mike Snitzer]

On Fri, May 22 2015 at  8:49am -0400,
Mike Snitzer <snitzer@redhat.com> wrote:

> On Thu, May 21 2015 at 11:27pm -0400,
> Stephen Rothwell <sfr@canb.auug.org.au> wrote:
> 
> > Hi all,
> > 
> > After merging the device-mapper tree, today's linux-next build (powerpc
> > ppc64_defconfig) failed like this:
> > 
> > block/bio.c:310:20: error: redefinition of 'bio_inc_remaining'
> >  static inline void bio_inc_remaining(struct bio *bio)
> >                     ^
> > In file included from block/bio.c:20:0:
> > include/linux/bio.h:664:20: note: previous definition of 'bio_inc_remaining' was here
> >  static inline void bio_inc_remaining(struct bio *bio)
> >                     ^
> > 
> > Caused by commit d61b82161c5a ("block: remove management of
> > bi_remaining when restoring original bi_end_io").
> > 
> > OK, this is a bad automatic merge caused by the fact that the block
> > tree and the device-mapper tree both contain the same set fo patches as
> > different commits.
> > 
> > Commit c4cf5261f8bf from the block tree is the same as commit
> > 8395711c4f90 from the device-mapper tree and both add
> > bio_inc_remaining() to include/linux/bio.h.  The above commit in the
> > device-mapper tree removes it from there, but the merge decided to keep
> > it.
> > 
> > That shared patch above is part of a series of 14 patches that are
> > identical patches but different commits and are causing quite a few
> > conflicts.  It looks like those patches were rebased onto v4.10rc3 as
> > the base for the device-mapper tree.  A better idea would have been to
> > start with v4.1rc3 (if that is necessary) and merge commit 5b3f341f098d
> > ("blk-mq: make plug work for mutiple disks and queues") from the block
> > tree - assuming that Jens will guarantee not to rebase that part of his
> > tree.
> > 
> > Please clean this mess up.
> 
> Will do.

I've fixed linux-dm's for-next to be properly based on linux-block's
for-4.2/core -- I also resent Jens three block patches for 4.2 that
motivated me to rebase ontop of block to begin with.

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

[Mike Snitzer]

On Thu, May 21 2015 at 11:27pm -0400,
Stephen Rothwell <sfr@canb.auug.org.au> wrote:

> Hi all,
> 
> After merging the device-mapper tree, today's linux-next build (powerpc
> ppc64_defconfig) failed like this:
> 
> block/bio.c:310:20: error: redefinition of 'bio_inc_remaining'
>  static inline void bio_inc_remaining(struct bio *bio)
>                     ^
> In file included from block/bio.c:20:0:
> include/linux/bio.h:664:20: note: previous definition of 'bio_inc_remaining' was here
>  static inline void bio_inc_remaining(struct bio *bio)
>                     ^
> 
> Caused by commit d61b82161c5a ("block: remove management of
> bi_remaining when restoring original bi_end_io").
> 
> OK, this is a bad automatic merge caused by the fact that the block
> tree and the device-mapper tree both contain the same set fo patches as
> different commits.
> 
> Commit c4cf5261f8bf from the block tree is the same as commit
> 8395711c4f90 from the device-mapper tree and both add
> bio_inc_remaining() to include/linux/bio.h.  The above commit in the
> device-mapper tree removes it from there, but the merge decided to keep
> it.
> 
> That shared patch above is part of a series of 14 patches that are
> identical patches but different commits and are causing quite a few
> conflicts.  It looks like those patches were rebased onto v4.10rc3 as
> the base for the device-mapper tree.  A better idea would have been to
> start with v4.1rc3 (if that is necessary) and merge commit 5b3f341f098d
> ("blk-mq: make plug work for mutiple disks and queues") from the block
> tree - assuming that Jens will guarantee not to rebase that part of his
> tree.
> 
> Please clean this mess up.

Will do.
 
> I have used the version of the device-mapper tree from next-20150521
> for today.

Thanks,
Mike

linux-next: build failure after merge of the device-mapper tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1711 bytes --]

Hi all,

After merging the device-mapper tree, today's linux-next build (powerpc
ppc64_defconfig) failed like this:

block/bio.c:310:20: error: redefinition of 'bio_inc_remaining'
 static inline void bio_inc_remaining(struct bio *bio)
                    ^
In file included from block/bio.c:20:0:
include/linux/bio.h:664:20: note: previous definition of 'bio_inc_remaining' was here
 static inline void bio_inc_remaining(struct bio *bio)
                    ^

Caused by commit d61b82161c5a ("block: remove management of
bi_remaining when restoring original bi_end_io").

OK, this is a bad automatic merge caused by the fact that the block
tree and the device-mapper tree both contain the same set fo patches as
different commits.

Commit c4cf5261f8bf from the block tree is the same as commit
8395711c4f90 from the device-mapper tree and both add
bio_inc_remaining() to include/linux/bio.h.  The above commit in the
device-mapper tree removes it from there, but the merge decided to keep
it.

That shared patch above is part of a series of 14 patches that are
identical patches but different commits and are causing quite a few
conflicts.  It looks like those patches were rebased onto v4.10rc3 as
the base for the device-mapper tree.  A better idea would have been to
start with v4.1rc3 (if that is necessary) and merge commit 5b3f341f098d
("blk-mq: make plug work for mutiple disks and queues") from the block
tree - assuming that Jens will guarantee not to rebase that part of his
tree.

Please clean this mess up.

I have used the version of the device-mapper tree from next-20150521
for today.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]

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

[Mike Snitzer]

On Tue, Jan 14 2014 at 10:52pm -0500,
Stephen Rothwell <sfr@canb.auug.org.au> wrote:

> Hi all,
> 
> After merging the device-mapper tree, today's linux-next build (powerpc
> ppc64_defconfig) failed like this:
> 
> ERROR: ".dm_bufio_get_device_size" [drivers/md/dm-snapshot.ko] undefined!
> ERROR: ".dm_bufio_release" [drivers/md/dm-snapshot.ko] undefined!
> ERROR: ".dm_bufio_client_destroy" [drivers/md/dm-snapshot.ko] undefined!
> ERROR: ".dm_bufio_prefetch" [drivers/md/dm-snapshot.ko] undefined!
> ERROR: ".dm_bufio_set_minimum_buffers" [drivers/md/dm-snapshot.ko] undefined!
> ERROR: ".dm_bufio_forget" [drivers/md/dm-snapshot.ko] undefined!
> ERROR: ".dm_bufio_client_create" [drivers/md/dm-snapshot.ko] undefined!
> ERROR: ".dm_bufio_read" [drivers/md/dm-snapshot.ko] undefined!
> 
> Presumably caused by commit b41bf7440bcf ("dm snapshot: use dm-bufio").

Hi Stephen,

That commit was missing a Kconfig update to have DM_SNAPSHOT select
DM_BUFIO.  I've rebased the "dm snapshot: use dm-bufio" commit to include
the Kconfig change and pushed to 'for-next'.

Thanks,
Mike

linux-next: build failure after merge of the device-mapper tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 905 bytes --]

Hi all,

After merging the device-mapper tree, today's linux-next build (powerpc
ppc64_defconfig) failed like this:

ERROR: ".dm_bufio_get_device_size" [drivers/md/dm-snapshot.ko] undefined!
ERROR: ".dm_bufio_release" [drivers/md/dm-snapshot.ko] undefined!
ERROR: ".dm_bufio_client_destroy" [drivers/md/dm-snapshot.ko] undefined!
ERROR: ".dm_bufio_prefetch" [drivers/md/dm-snapshot.ko] undefined!
ERROR: ".dm_bufio_set_minimum_buffers" [drivers/md/dm-snapshot.ko] undefined!
ERROR: ".dm_bufio_forget" [drivers/md/dm-snapshot.ko] undefined!
ERROR: ".dm_bufio_client_create" [drivers/md/dm-snapshot.ko] undefined!
ERROR: ".dm_bufio_read" [drivers/md/dm-snapshot.ko] undefined!

Presumably caused by commit b41bf7440bcf ("dm snapshot: use dm-bufio").

I have used the device-mapper tree from next-20140114 for today.

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

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

[Jens Axboe]

On 11/05/2013 03:42 PM, Stephen Rothwell wrote:
> Hi Jens,
> 
> On Tue, 05 Nov 2013 09:22:11 -0700 Jens Axboe <axboe@kernel.dk> wrote:
>>
>> Just a heads up that I plan on sending the block bits shortly, hopefully
>> merging sooner rather than later. This round is huge, so I'd like to get
>> it in asap.
> 
> Given the problems we have seen since the inclusion into linux-next of
> some of it, don't you think it could do with a little more bedding in?

It'll either simmer without problems for the next day or two, or the
immutable bits will get pushed to the next merge window.

-- 
Jens Axboe

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

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 474 bytes --]

Hi Jens,

On Tue, 05 Nov 2013 09:22:11 -0700 Jens Axboe <axboe@kernel.dk> wrote:
>
> Just a heads up that I plan on sending the block bits shortly, hopefully
> merging sooner rather than later. This round is huge, so I'd like to get
> it in asap.

Given the problems we have seen since the inclusion into linux-next of
some of it, don't you think it could do with a little more bedding in?

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

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

[Jens Axboe]

On 11/05/2013 05:54 AM, Mike Snitzer wrote:
> On Mon, Nov 04 2013 at 10:32pm -0500,
> Stephen Rothwell <sfr@canb.auug.org.au> wrote:
> 
>> Hi Mike,
>>
>> On Mon, 4 Nov 2013 22:20:54 -0500 Mike Snitzer <snitzer@redhat.com> wrote:
>>>
>>> On Mon, Nov 04 2013 at  9:51pm -0500,
>>> Stephen Rothwell <sfr@canb.auug.org.au> wrote:
>>>
>>>> After merging the device-mapper tree, today's linux-next build (x86_64
>>>> allmodconfig) failed like this:
>>>>
>>>> drivers/md/dm-cache-target.c: In function 'bio_writes_complete_block':
>>>> drivers/md/dm-cache-target.c:1020:7: error: 'struct bio' has no member named 'bi_size'
>>>>    (bio->bi_size == (cache->sectors_per_block << SECTOR_SHIFT));
>>>>        ^
>>>>
>>>> Caused by commit 1bd270d800ec ("dm cache: promotion optimisation for
>>>> writes") interacting with commit ed2d2f9a8265 ("block: Abstract out bvec
>>>> iterator") from the block tree.
>>>>
>>>> I applied this fix up patch and can carry it as necessary:
>>>
>>> Yes that is needed, as is this (which begs the question: how should I be
>>
>> I will add that tomorrow.
>>
>>> handling the fact that post-merge changes are needed?  Especially since
>>> Linus will need the changes too):
>>
>> Who ever gets their tree in last has to tell Linus about these extra
>> merge fix ups as part of their pull request.
> 
> OK, that's what I thought, thanks for confirming.

Just a heads up that I plan on sending the block bits shortly, hopefully
merging sooner rather than later. This round is huge, so I'd like to get
it in asap.

-- 
Jens Axboe

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

[Mike Snitzer]

On Mon, Nov 04 2013 at 10:32pm -0500,
Stephen Rothwell <sfr@canb.auug.org.au> wrote:

> Hi Mike,
> 
> On Mon, 4 Nov 2013 22:20:54 -0500 Mike Snitzer <snitzer@redhat.com> wrote:
> >
> > On Mon, Nov 04 2013 at  9:51pm -0500,
> > Stephen Rothwell <sfr@canb.auug.org.au> wrote:
> > 
> > > After merging the device-mapper tree, today's linux-next build (x86_64
> > > allmodconfig) failed like this:
> > > 
> > > drivers/md/dm-cache-target.c: In function 'bio_writes_complete_block':
> > > drivers/md/dm-cache-target.c:1020:7: error: 'struct bio' has no member named 'bi_size'
> > >    (bio->bi_size == (cache->sectors_per_block << SECTOR_SHIFT));
> > >        ^
> > > 
> > > Caused by commit 1bd270d800ec ("dm cache: promotion optimisation for
> > > writes") interacting with commit ed2d2f9a8265 ("block: Abstract out bvec
> > > iterator") from the block tree.
> > > 
> > > I applied this fix up patch and can carry it as necessary:
> > 
> > Yes that is needed, as is this (which begs the question: how should I be
> 
> I will add that tomorrow.
> 
> > handling the fact that post-merge changes are needed?  Especially since
> > Linus will need the changes too):
> 
> Who ever gets their tree in last has to tell Linus about these extra
> merge fix ups as part of their pull request.

OK, that's what I thought, thanks for confirming.

Mike

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

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1237 bytes --]

Hi Mike,

On Mon, 4 Nov 2013 22:20:54 -0500 Mike Snitzer <snitzer@redhat.com> wrote:
>
> On Mon, Nov 04 2013 at  9:51pm -0500,
> Stephen Rothwell <sfr@canb.auug.org.au> wrote:
> 
> > After merging the device-mapper tree, today's linux-next build (x86_64
> > allmodconfig) failed like this:
> > 
> > drivers/md/dm-cache-target.c: In function 'bio_writes_complete_block':
> > drivers/md/dm-cache-target.c:1020:7: error: 'struct bio' has no member named 'bi_size'
> >    (bio->bi_size == (cache->sectors_per_block << SECTOR_SHIFT));
> >        ^
> > 
> > Caused by commit 1bd270d800ec ("dm cache: promotion optimisation for
> > writes") interacting with commit ed2d2f9a8265 ("block: Abstract out bvec
> > iterator") from the block tree.
> > 
> > I applied this fix up patch and can carry it as necessary:
> 
> Yes that is needed, as is this (which begs the question: how should I be

I will add that tomorrow.

> handling the fact that post-merge changes are needed?  Especially since
> Linus will need the changes too):

Who ever gets their tree in last has to tell Linus about these extra
merge fix ups as part of their pull request.

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

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

[Mike Snitzer]

On Mon, Nov 04 2013 at  9:51pm -0500,
Stephen Rothwell <sfr@canb.auug.org.au> wrote:

> Hi all,
> 
> After merging the device-mapper tree, today's linux-next build (x86_64
> allmodconfig) failed like this:
> 
> drivers/md/dm-cache-target.c: In function 'bio_writes_complete_block':
> drivers/md/dm-cache-target.c:1020:7: error: 'struct bio' has no member named 'bi_size'
>    (bio->bi_size == (cache->sectors_per_block << SECTOR_SHIFT));
>        ^
> 
> Caused by commit 1bd270d800ec ("dm cache: promotion optimisation for
> writes") interacting with commit ed2d2f9a8265 ("block: Abstract out bvec
> iterator") from the block tree.
> 
> I applied this fix up patch and can carry it as necessary:

Hi Stephen,

Yes that is needed, as is this (which begs the question: how should I be
handling the fact that post-merge changes are needed?  Especially since
Linus will need the changes too):

 drivers/md/dm-cache-target.c |   12 ++++++------
 1 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/drivers/md/dm-cache-target.c b/drivers/md/dm-cache-target.c
index a977d5c..52d7d75 100644
--- a/drivers/md/dm-cache-target.c
+++ b/drivers/md/dm-cache-target.c
@@ -85,6 +85,12 @@ static void dm_unhook_bio(struct dm_hook_info *h, struct bio *bio)
 {
 	bio->bi_end_io = h->bi_end_io;
 	bio->bi_private = h->bi_private;
+
+	/*
+	 * Must bump bi_remaining to allow bio to complete with
+	 * restored bi_end_io.
+	 */
+	atomic_inc(&bio->bi_remaining);
 }
 
 /*----------------------------------------------------------------*/
@@ -746,12 +752,6 @@ static void writethrough_endio(struct bio *bio, int err)
 
 	dm_unhook_bio(&pb->hook_info, bio);
 
-	/*
-	 * Must bump bi_remaining to allow bio to complete with
-	 * restored bi_end_io.
-	 */
-	atomic_inc(&bio->bi_remaining);
-
 	if (err) {
 		bio_endio(bio, err);
 		return;

linux-next: build failure after merge of the device-mapper tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1601 bytes --]

Hi all,

After merging the device-mapper tree, today's linux-next build (x86_64
allmodconfig) failed like this:

drivers/md/dm-cache-target.c: In function 'bio_writes_complete_block':
drivers/md/dm-cache-target.c:1020:7: error: 'struct bio' has no member named 'bi_size'
   (bio->bi_size == (cache->sectors_per_block << SECTOR_SHIFT));
       ^

Caused by commit 1bd270d800ec ("dm cache: promotion optimisation for
writes") interacting with commit ed2d2f9a8265 ("block: Abstract out bvec
iterator") from the block tree.

I applied this fix up patch and can carry it as necessary:

From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Tue, 5 Nov 2013 13:49:25 +1100
Subject: [PATCH] dm_cache: fix up for bvec iterator abstraction

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 drivers/md/dm-cache-target.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/md/dm-cache-target.c b/drivers/md/dm-cache-target.c
index a977d5c6d544..5ce0f5ec6d74 100644
--- a/drivers/md/dm-cache-target.c
+++ b/drivers/md/dm-cache-target.c
@@ -1017,7 +1017,7 @@ static void issue_overwrite(struct dm_cache_migration *mg, struct bio *bio)
 static bool bio_writes_complete_block(struct cache *cache, struct bio *bio)
 {
 	return (bio_data_dir(bio) == WRITE) &&
-		(bio->bi_size == (cache->sectors_per_block << SECTOR_SHIFT));
+		(bio->bi_iter.bi_size == (cache->sectors_per_block << SECTOR_SHIFT));
 }
 
 static void avoid_copy(struct dm_cache_migration *mg)
-- 
1.8.4.rc3

-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au

[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]

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

[Alasdair G Kergon]

On Mon, Aug 02, 2010 at 11:31:48AM +1000, Stephen Rothwell wrote:
> I will use the device-mapper tree from next-20100730 for today.
 
Gah - I forgot to repush it, sorry.

Milan - I made it so the inner function always sets an error message, instead
of making the caller have a special case to detect whether or not it had been
set.

> P.S. the patch summaries could be more informative ... i.e. you should
> put "Subject:" lines in the patches.

They are the quilt filenames and the subject line gets generated automatically
from that when I import into git to submit to Linus.

Alasdair

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

[Milan Broz]

On 08/02/2010 03:31 AM, Stephen Rothwell wrote:
> After merging the device-mapper tree, today's linux-next build (powerpc
> ppc64_defconfig) failed like this:
> 
> drivers/md/dm-crypt.c: In function 'crypt_ctr_cipher':
> drivers/md/dm-crypt.c:1095: error: 'bad_mem' undeclared (first use in this function)

hmmmm, Alasdair you edited that patch?
The version I sent compiled cleanly.
(moreover it is just part of patchset which is not yet fully there, seems)

Milan

linux-next: build failure after merge of the device-mapper tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 721 bytes --]

Hi Alasdair,

After merging the device-mapper tree, today's linux-next build (powerpc
ppc64_defconfig) failed like this:

drivers/md/dm-crypt.c: In function 'crypt_ctr_cipher':
drivers/md/dm-crypt.c:1095: error: 'bad_mem' undeclared (first use in this function)

Caused by commit 460328b294ddd132945bdf536a9130f2d6cd6ace
("dm-crypt-simplify-crypt_ctr").

Please do test builds ... "No functional change in this patch." :-(

I will use the device-mapper tree from next-20100730 for today.

P.S. the patch summaries could be more informative ... i.e. you should
put "Subject:" lines in the patches.
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

[-- Attachment #2: Type: application/pgp-signature, Size: 490 bytes --]

linux-next: build failure after merge of the device-mapper tree

[Stephen Rothwell]

Hi Alasdair,

After merging the device-mapper tree, today's linux-next build (powerpc
ppc64_defconfig) failed like this:

drivers/md/dm.c: In function '__clone_and_map':
/data/sfr/next/drivers/md/dm.c:1245: error: implicit declaration of function 'bio_rw_flagged'
/data/sfr/next/drivers/md/dm.c:1245: error: 'BIO_RW_DISCARD' undeclared (first use in this function)

Caused by commit d134095c1564b101987d8f2920d0928cae669be2
("dm-linear-support-discard") interacting with commit
74450be123b6f3cb480c358a056be398cce6aa6e ("block: unify flags for struct
bio and struct request") from the block tree.

I hate API changes :-)

I applied the following merge fix up patch:

From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Wed, 30 Jun 2010 12:54:22 +1000
Subject: [PATCH] device-mapper: fix for bio API change

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 drivers/md/dm.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index 518bfe0..20b4248 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -1242,7 +1242,7 @@ static int __clone_and_map(struct clone_info *ci)
 	if (unlikely(bio_empty_barrier(bio)))
 		return __clone_and_map_empty_barrier(ci);
 
-	if (unlikely(bio_rw_flagged(bio, BIO_RW_DISCARD)))
+	if (unlikely(bio->bi_rw & REQ_DISCARD))
 		return __clone_and_map_discard(ci);
 
 	ti = dm_table_find_target(ci->map, ci->sector);
-- 
1.7.1


-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/