linux-next: build failure after merge of most of the trees

linux-next: build failure after merge of most of the trees

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 6006 bytes --]

Hi all,

After merging the all the trees, today's linux-next build (powerpc
allyesconfig) failed like this:

drivers/vhost/scsi.c: In function 'vhost_scsi_open':
drivers/vhost/scsi.c:1422:3: error: implicit declaration of function 'vzalloc' [-Werror=implicit-function-declaration]
   vs = vzalloc(sizeof(*vs));
   ^
drivers/vhost/scsi.c:1422:6: warning: assignment makes pointer from integer without a cast
   vs = vzalloc(sizeof(*vs));
      ^
drivers/target/target_core_pr.c: In function 'core_scsi3_update_and_write_aptpl':
drivers/target/target_core_pr.c:1985:2: error: implicit declaration of function 'vzalloc' [-Werror=implicit-function-declaration]
  buf = vzalloc(len);
  ^
drivers/target/target_core_pr.c:1985:6: warning: assignment makes pointer from integer without a cast
  buf = vzalloc(len);
      ^
drivers/target/target_core_pr.c:1991:3: error: implicit declaration of function 'vfree' [-Werror=implicit-function-declaration]
   vfree(buf);
   ^
drivers/target/target_core_transport.c: In function 'transport_alloc_session_tags':
drivers/target/target_core_transport.c:258:3: error: implicit declaration of function 'vzalloc' [-Werror=implicit-function-declaration]
   se_sess->sess_cmd_map = vzalloc(tag_num * tag_size);
   ^
drivers/target/target_core_transport.c:258:25: warning: assignment makes pointer from integer without a cast
   se_sess->sess_cmd_map = vzalloc(tag_num * tag_size);
                         ^
drivers/target/target_core_transport.c:270:4: error: implicit declaration of function 'vfree' [-Werror=implicit-function-declaration]
    vfree(se_sess->sess_cmd_map);
    ^
drivers/target/target_core_transport.c: In function 'transport_kmap_data_sg':
drivers/target/target_core_transport.c:2317:2: error: implicit declaration of function 'vmap' [-Werror=implicit-function-declaration]
  cmd->t_data_vmap = vmap(pages, cmd->t_data_nents,  VM_MAP, PAGE_KERNEL);
  ^
drivers/target/target_core_transport.c:2317:53: error: 'VM_MAP' undeclared (first use in this function)
  cmd->t_data_vmap = vmap(pages, cmd->t_data_nents,  VM_MAP, PAGE_KERNEL);
                                                     ^
drivers/target/target_core_transport.c: In function 'transport_kunmap_data_sg':
drivers/target/target_core_transport.c:2335:2: error: implicit declaration of function 'vunmap' [-Werror=implicit-function-declaration]
  vunmap(cmd->t_data_vmap);
  ^
drivers/target/target_core_file.c: In function 'fd_format_prot':
drivers/target/target_core_file.c:809:2: error: implicit declaration of function 'vzalloc' [-Werror=implicit-function-declaration]
  buf = vzalloc(unit_size);
  ^
drivers/target/target_core_file.c:809:6: warning: assignment makes pointer from integer without a cast
  buf = vzalloc(unit_size);
      ^
drivers/target/target_core_file.c:822:2: error: implicit declaration of function 'vfree' [-Werror=implicit-function-declaration]
  vfree(buf);
  ^
drivers/target/target_core_user.c: In function 'tcmu_configure_device':
drivers/target/target_core_user.c:895:2: error: implicit declaration of function 'vzalloc' [-Werror=implicit-function-declaration]
  udev->mb_addr = vzalloc(TCMU_RING_SIZE);
  ^
drivers/target/target_core_user.c:895:16: warning: assignment makes pointer from integer without a cast
  udev->mb_addr = vzalloc(TCMU_RING_SIZE);
                ^
drivers/target/target_core_user.c:947:2: error: implicit declaration of function 'vfree' [-Werror=implicit-function-declaration]
  vfree(udev->mb_addr);
  ^
drivers/target/iscsi/iscsi_target.c: In function 'iscsi_target_init_module':
drivers/target/iscsi/iscsi_target.c:557:2: error: implicit declaration of function 'vzalloc' [-Werror=implicit-function-declaration]
  iscsit_global->ts_bitmap = vzalloc(size);
  ^
drivers/target/iscsi/iscsi_target.c:557:27: warning: assignment makes pointer from integer without a cast
  iscsit_global->ts_bitmap = vzalloc(size);
                           ^
drivers/target/iscsi/iscsi_target.c:615:2: error: implicit declaration of function 'vfree' [-Werror=implicit-function-declaration]
  vfree(iscsit_global->ts_bitmap);
  ^
drivers/scsi/qla2xxx/tcm_qla2xxx.c: In function 'tcm_qla2xxx_init_lport':
drivers/scsi/qla2xxx/tcm_qla2xxx.c:1578:2: error: implicit declaration of function 'vmalloc' [-Werror=implicit-function-declaration]
  lport->lport_loopid_map = vmalloc(sizeof(struct tcm_qla2xxx_fc_loopid) *
  ^
drivers/scsi/qla2xxx/tcm_qla2xxx.c:1578:26: warning: assignment makes pointer from integer without a cast
  lport->lport_loopid_map = vmalloc(sizeof(struct tcm_qla2xxx_fc_loopid) *
                          ^
drivers/scsi/qla2xxx/tcm_qla2xxx.c: In function 'tcm_qla2xxx_make_lport':
drivers/scsi/qla2xxx/tcm_qla2xxx.c:1643:2: error: implicit declaration of function 'vfree' [-Werror=implicit-function-declaration]
  vfree(lport->lport_loopid_map);
  ^

Ouch :-(

Maybe commit 095dc8e0c368 ("tcp: fix/cleanup
inet_ehash_locks_alloc()")?  (that is the only commit I can find that
remove an include of vmalloc.h

Ah ha.  I applied this fix patch:

From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Thu, 28 May 2015 21:54:40 +1000
Subject: [PATCH] try putting the include of vmalloc.h back into inet_hashtables.h

Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
---
 include/net/inet_hashtables.h | 1 +
 1 file changed, 1 insertion(+)

diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h
index b73c88a19dd4..0046b9f91ed3 100644
--- a/include/net/inet_hashtables.h
+++ b/include/net/inet_hashtables.h
@@ -24,6 +24,7 @@
 #include <linux/spinlock.h>
 #include <linux/types.h>
 #include <linux/wait.h>
+#include <linux/vmalloc.h>
 
 #include <net/inet_connection_sock.h>
 #include <net/inet_sock.h>
-- 
2.1.4

So clearly there are people not reading Rule 1 in
Documentation/SubmitChecklist :-(
-- 
Cheers,
Stephen Rothwell                    sfr@canb.auug.org.au
http://www.canb.auug.org.au/~sfr/

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

Re: linux-next: build failure after merge of most of the trees

[Eric Dumazet]

On Thu, 2015-05-28 at 22:06 +1000, Stephen Rothwell wrote:
> Hi all,
> 
> After merging the all the trees, today's linux-next build (powerpc
> allyesconfig) failed like this:
> 
> drivers/vhost/scsi.c: In function 'vhost_scsi_open':
> drivers/vhost/scsi.c:1422:3: error: implicit declaration of function 'vzalloc' [-Werror=implicit-function-declaration]
>    vs = vzalloc(sizeof(*vs));
>    ^
> drivers/vhost/scsi.c:1422:6: warning: assignment makes pointer from integer without a cast
>    vs = vzalloc(sizeof(*vs));
>       ^
> drivers/target/target_core_pr.c: In function 'core_scsi3_update_and_write_aptpl':
> drivers/target/target_core_pr.c:1985:2: error: implicit declaration of function 'vzalloc' [-Werror=implicit-function-declaration]
>   buf = vzalloc(len);
>   ^
> drivers/target/target_core_pr.c:1985:6: warning: assignment makes pointer from integer without a cast
>   buf = vzalloc(len);
>       ^
> drivers/target/target_core_pr.c:1991:3: error: implicit declaration of function 'vfree' [-Werror=implicit-function-declaration]
>    vfree(buf);
>    ^
> drivers/target/target_core_transport.c: In function 'transport_alloc_session_tags':
> drivers/target/target_core_transport.c:258:3: error: implicit declaration of function 'vzalloc' [-Werror=implicit-function-declaration]
>    se_sess->sess_cmd_map = vzalloc(tag_num * tag_size);
>    ^
> drivers/target/target_core_transport.c:258:25: warning: assignment makes pointer from integer without a cast
>    se_sess->sess_cmd_map = vzalloc(tag_num * tag_size);
>                          ^
> drivers/target/target_core_transport.c:270:4: error: implicit declaration of function 'vfree' [-Werror=implicit-function-declaration]
>     vfree(se_sess->sess_cmd_map);
>     ^
> drivers/target/target_core_transport.c: In function 'transport_kmap_data_sg':
> drivers/target/target_core_transport.c:2317:2: error: implicit declaration of function 'vmap' [-Werror=implicit-function-declaration]
>   cmd->t_data_vmap = vmap(pages, cmd->t_data_nents,  VM_MAP, PAGE_KERNEL);
>   ^
> drivers/target/target_core_transport.c:2317:53: error: 'VM_MAP' undeclared (first use in this function)
>   cmd->t_data_vmap = vmap(pages, cmd->t_data_nents,  VM_MAP, PAGE_KERNEL);
>                                                      ^
> drivers/target/target_core_transport.c: In function 'transport_kunmap_data_sg':
> drivers/target/target_core_transport.c:2335:2: error: implicit declaration of function 'vunmap' [-Werror=implicit-function-declaration]
>   vunmap(cmd->t_data_vmap);
>   ^
> drivers/target/target_core_file.c: In function 'fd_format_prot':
> drivers/target/target_core_file.c:809:2: error: implicit declaration of function 'vzalloc' [-Werror=implicit-function-declaration]
>   buf = vzalloc(unit_size);
>   ^
> drivers/target/target_core_file.c:809:6: warning: assignment makes pointer from integer without a cast
>   buf = vzalloc(unit_size);
>       ^
> drivers/target/target_core_file.c:822:2: error: implicit declaration of function 'vfree' [-Werror=implicit-function-declaration]
>   vfree(buf);
>   ^
> drivers/target/target_core_user.c: In function 'tcmu_configure_device':
> drivers/target/target_core_user.c:895:2: error: implicit declaration of function 'vzalloc' [-Werror=implicit-function-declaration]
>   udev->mb_addr = vzalloc(TCMU_RING_SIZE);
>   ^
> drivers/target/target_core_user.c:895:16: warning: assignment makes pointer from integer without a cast
>   udev->mb_addr = vzalloc(TCMU_RING_SIZE);
>                 ^
> drivers/target/target_core_user.c:947:2: error: implicit declaration of function 'vfree' [-Werror=implicit-function-declaration]
>   vfree(udev->mb_addr);
>   ^
> drivers/target/iscsi/iscsi_target.c: In function 'iscsi_target_init_module':
> drivers/target/iscsi/iscsi_target.c:557:2: error: implicit declaration of function 'vzalloc' [-Werror=implicit-function-declaration]
>   iscsit_global->ts_bitmap = vzalloc(size);
>   ^
> drivers/target/iscsi/iscsi_target.c:557:27: warning: assignment makes pointer from integer without a cast
>   iscsit_global->ts_bitmap = vzalloc(size);
>                            ^
> drivers/target/iscsi/iscsi_target.c:615:2: error: implicit declaration of function 'vfree' [-Werror=implicit-function-declaration]
>   vfree(iscsit_global->ts_bitmap);
>   ^
> drivers/scsi/qla2xxx/tcm_qla2xxx.c: In function 'tcm_qla2xxx_init_lport':
> drivers/scsi/qla2xxx/tcm_qla2xxx.c:1578:2: error: implicit declaration of function 'vmalloc' [-Werror=implicit-function-declaration]
>   lport->lport_loopid_map = vmalloc(sizeof(struct tcm_qla2xxx_fc_loopid) *
>   ^
> drivers/scsi/qla2xxx/tcm_qla2xxx.c:1578:26: warning: assignment makes pointer from integer without a cast
>   lport->lport_loopid_map = vmalloc(sizeof(struct tcm_qla2xxx_fc_loopid) *
>                           ^
> drivers/scsi/qla2xxx/tcm_qla2xxx.c: In function 'tcm_qla2xxx_make_lport':
> drivers/scsi/qla2xxx/tcm_qla2xxx.c:1643:2: error: implicit declaration of function 'vfree' [-Werror=implicit-function-declaration]
>   vfree(lport->lport_loopid_map);
>   ^
> 
> Ouch :-(
> 
> Maybe commit 095dc8e0c368 ("tcp: fix/cleanup
> inet_ehash_locks_alloc()")?  (that is the only commit I can find that
> remove an include of vmalloc.h
> 
> Ah ha.  I applied this fix patch:
> 
> From: Stephen Rothwell <sfr@canb.auug.org.au>
> Date: Thu, 28 May 2015 21:54:40 +1000
> Subject: [PATCH] try putting the include of vmalloc.h back into inet_hashtables.h
> 
> Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
> ---
>  include/net/inet_hashtables.h | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/include/net/inet_hashtables.h b/include/net/inet_hashtables.h
> index b73c88a19dd4..0046b9f91ed3 100644
> --- a/include/net/inet_hashtables.h
> +++ b/include/net/inet_hashtables.h
> @@ -24,6 +24,7 @@
>  #include <linux/spinlock.h>
>  #include <linux/types.h>
>  #include <linux/wait.h>
> +#include <linux/vmalloc.h>
>  
>  #include <net/inet_connection_sock.h>
>  #include <net/inet_sock.h>
> -- 

Hi Stephen

We were alerted of this problem thanks to kbuild test robot.

This fix is not a definitive one I hope.

Golden rule is that vmalloc() users must include vmalloc.h themselves,
not by an indirect include.

include/net/inet_hashtables.h no longer uses vmalloc(), so this include
is not needed anymore.

I sent one fix, and prepared others, but I prefer that each offender is
fixed.

http://git.kernel.org/cgit/linux/kernel/git/davem/net-next.git/commit/?id=c6bfda8d72c485b2eee1642cbcbd92e3fbb93ed3

Thanks

Re: linux-next: build failure after merge of most of the trees

[David Miller]

From: Stephen Rothwell <sfr@canb.auug.org.au>
Date: Thu, 28 May 2015 22:06:07 +1000

> Ouch :-(

The only thing I will say on this matter is that the _only_ way this
problem will go away is if someone does the work necessary to get rid
of that implicit vmalloc.h include that happens on all x86 platform
builds.

So if you want this to stop happening, work on that.

I've applied the following to net-next, thanks for your report.

====================
[PATCH] treewide: Add missing vmalloc.h inclusion.

All of these files were only building on non-x86 because of
the indirect of inclusion of vmalloc.h by, of all things,
"net/inet_hashtables.h"

None of this got caught during build testing, because on x86
there is an implicit vmalloc.h include via on of the arch asm/
headers.

This fixes all of these

Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
 crypto/algif_skcipher.c                | 1 +
 drivers/scsi/qla2xxx/tcm_qla2xxx.c     | 1 +
 drivers/target/iscsi/iscsi_target.c    | 1 +
 drivers/target/target_core_file.c      | 1 +
 drivers/target/target_core_pr.c        | 1 +
 drivers/target/target_core_transport.c | 1 +
 drivers/target/target_core_user.c      | 1 +
 drivers/vhost/scsi.c                   | 1 +
 8 files changed, 8 insertions(+)

diff --git a/crypto/algif_skcipher.c b/crypto/algif_skcipher.c
index 37110fd..4d1c315 100644
--- a/crypto/algif_skcipher.c
+++ b/crypto/algif_skcipher.c
@@ -444,6 +444,7 @@ static int skcipher_recvmsg(struct kiocb *unused, struct socket *sock,
 			err = skcipher_wait_for_data(sk, flags);
 			if (err)
 				goto unlock;
+			used = ctx->used;
 		}
 
 		used = min_t(unsigned long, used, iov_iter_count(&msg->msg_iter));
diff --git a/drivers/scsi/qla2xxx/tcm_qla2xxx.c b/drivers/scsi/qla2xxx/tcm_qla2xxx.c
index 73f9fee..54c986a 100644
--- a/drivers/scsi/qla2xxx/tcm_qla2xxx.c
+++ b/drivers/scsi/qla2xxx/tcm_qla2xxx.c
@@ -27,6 +27,7 @@
 #include <linux/moduleparam.h>
 #include <generated/utsrelease.h>
 #include <linux/utsname.h>
+#include <linux/vmalloc.h>
 #include <linux/init.h>
 #include <linux/list.h>
 #include <linux/slab.h>
diff --git a/drivers/target/iscsi/iscsi_target.c b/drivers/target/iscsi/iscsi_target.c
index aebde32..f2ce95c 100644
--- a/drivers/target/iscsi/iscsi_target.c
+++ b/drivers/target/iscsi/iscsi_target.c
@@ -21,6 +21,7 @@
 #include <linux/crypto.h>
 #include <linux/completion.h>
 #include <linux/module.h>
+#include <linux/vmalloc.h>
 #include <linux/idr.h>
 #include <asm/unaligned.h>
 #include <scsi/scsi_device.h>
diff --git a/drivers/target/target_core_file.c b/drivers/target/target_core_file.c
index d836de2..5f8b119 100644
--- a/drivers/target/target_core_file.c
+++ b/drivers/target/target_core_file.c
@@ -30,6 +30,7 @@
 #include <linux/slab.h>
 #include <linux/spinlock.h>
 #include <linux/module.h>
+#include <linux/vmalloc.h>
 #include <linux/falloc.h>
 #include <scsi/scsi.h>
 #include <scsi/scsi_host.h>
diff --git a/drivers/target/target_core_pr.c b/drivers/target/target_core_pr.c
index 283cf78..06cd53e 100644
--- a/drivers/target/target_core_pr.c
+++ b/drivers/target/target_core_pr.c
@@ -27,6 +27,7 @@
 #include <linux/slab.h>
 #include <linux/spinlock.h>
 #include <linux/list.h>
+#include <linux/vmalloc.h>
 #include <linux/file.h>
 #include <scsi/scsi.h>
 #include <scsi/scsi_cmnd.h>
diff --git a/drivers/target/target_core_transport.c b/drivers/target/target_core_transport.c
index 0adc0f6..c99d2ea 100644
--- a/drivers/target/target_core_transport.c
+++ b/drivers/target/target_core_transport.c
@@ -34,6 +34,7 @@
 #include <linux/cdrom.h>
 #include <linux/module.h>
 #include <linux/ratelimit.h>
+#include <linux/vmalloc.h>
 #include <asm/unaligned.h>
 #include <net/sock.h>
 #include <net/tcp.h>
diff --git a/drivers/target/target_core_user.c b/drivers/target/target_core_user.c
index 1a1bcf7..ca43a10 100644
--- a/drivers/target/target_core_user.c
+++ b/drivers/target/target_core_user.c
@@ -21,6 +21,7 @@
 #include <linux/idr.h>
 #include <linux/timer.h>
 #include <linux/parser.h>
+#include <linux/vmalloc.h>
 #include <scsi/scsi.h>
 #include <scsi/scsi_host.h>
 #include <linux/uio_driver.h>
diff --git a/drivers/vhost/scsi.c b/drivers/vhost/scsi.c
index dc78d87..16b45ca 100644
--- a/drivers/vhost/scsi.c
+++ b/drivers/vhost/scsi.c
@@ -35,6 +35,7 @@
 #include <linux/compat.h>
 #include <linux/eventfd.h>
 #include <linux/fs.h>
+#include <linux/vmalloc.h>
 #include <linux/miscdevice.h>
 #include <asm/unaligned.h>
 #include <scsi/scsi.h>
-- 
2.1.0

Re: linux-next: build failure after merge of most of the trees

[Joe Perches]

On Thu, 2015-05-28 at 11:42 -0700, David Miller wrote:
> I've applied the following to net-next, thanks for your report.
> 
> ====================
> [PATCH] treewide: Add missing vmalloc.h inclusion.
> 
> All of these files were only building on non-x86 because of
> the indirect of inclusion of vmalloc.h by, of all things,
> "net/inet_hashtables.h"
[]
> diff --git a/crypto/algif_skcipher.c b/crypto/algif_skcipher.c
[]
> @@ -444,6 +444,7 @@ static int skcipher_recvmsg(struct kiocb *unused, struct socket *sock,
>  			err = skcipher_wait_for_data(sk, flags);
>  			if (err)
>  				goto unlock;
> +			used = ctx->used;

huh?

Re: linux-next: build failure after merge of most of the trees

[David Miller]

From: Joe Perches <joe@perches.com>
Date: Thu, 28 May 2015 11:51:15 -0700

> On Thu, 2015-05-28 at 11:42 -0700, David Miller wrote:
>> I've applied the following to net-next, thanks for your report.
>> 
>> ====================
>> [PATCH] treewide: Add missing vmalloc.h inclusion.
>> 
>> All of these files were only building on non-x86 because of
>> the indirect of inclusion of vmalloc.h by, of all things,
>> "net/inet_hashtables.h"
> []
>> diff --git a/crypto/algif_skcipher.c b/crypto/algif_skcipher.c
> []
>> @@ -444,6 +444,7 @@ static int skcipher_recvmsg(struct kiocb *unused, struct socket *sock,
>>  			err = skcipher_wait_for_data(sk, flags);
>>  			if (err)
>>  				goto unlock;
>> +			used = ctx->used;
> 
> huh?

Bogus chunk in my local tree, didn't make it into the final commit I
pushed out.

But thanks for noticing.

Re: linux-next: build failure after merge of most of the trees

[Eric Dumazet]

On Thu, 2015-05-28 at 14:35 -0700, David Miller wrote:

> Bogus chunk in my local tree, didn't make it into the final commit I
> pushed out.

Thanks for taking care of this before me !

Re: linux-next: build failure after merge of most of the trees

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 636 bytes --]

Hi Eric,

On Thu, 28 May 2015 08:26:51 -0700 Eric Dumazet <eric.dumazet@gmail.com> wrote:
>
> We were alerted of this problem thanks to kbuild test robot.
> 
> This fix is not a definitive one I hope.

No, just something to allow me to get my tree to build so I could go to
bed :-)

> Golden rule is that vmalloc() users must include vmalloc.h themselves,
> not by an indirect include.

Yep, that is a special case of Rule 1.

> I sent one fix, and prepared others, but I prefer that each offender is
> fixed.

Yeah, thanks to Dave for that.

-- 
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 most of the trees

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 600 bytes --]

Hi Dave,

On Thu, 28 May 2015 11:42:03 -0700 (PDT) David Miller <davem@davemloft.net> wrote:
>
> The only thing I will say on this matter is that the _only_ way this
> problem will go away is if someone does the work necessary to get rid
> of that implicit vmalloc.h include that happens on all x86 platform
> builds.
> 
> So if you want this to stop happening, work on that.

I may have a look in my (copious) free time :-)

> I've applied the following to net-next, thanks for your report.

Thanks for that.

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

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