Fix this build error when CONFIG_BLOCK is not set.
fs/compat_ioctl.c:1604: error: 'SG_SET_TRANSFORM' undeclared
Signed-off-by: Alexander Beregalov <[email protected]>
---
fs/compat_ioctl.c | 2 ++
1 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c
index 14cbc83..e4d4765 100644
--- a/fs/compat_ioctl.c
+++ b/fs/compat_ioctl.c
@@ -1600,8 +1600,10 @@ static long do_ioctl_trans(int fd, unsigned int cmd,
case KDSKBMETA:
case KDSKBLED:
case KDSETLED:
+#ifdef CONFIG_BLOCK
/* SG stuff */
case SG_SET_TRANSFORM:
+#endif
/* AUTOFS */
case AUTOFS_IOC_READY:
case AUTOFS_IOC_FAIL:
--
1.6.6.rc3
On Mon, Dec 21, 2009 at 12:21:32AM +0300, Alexander Beregalov wrote:
> Fix this build error when CONFIG_BLOCK is not set.
> fs/compat_ioctl.c:1604: error: 'SG_SET_TRANSFORM' undeclared
Do you want that through VFS tree? I'm going to send a bunch of assorted
fixes tomorrow morning, so I can pass that one at the same time...
2009/12/21 Al Viro <[email protected]>:
> On Mon, Dec 21, 2009 at 12:21:32AM +0300, Alexander Beregalov wrote:
>> Fix this build error when CONFIG_BLOCK is not set.
>> fs/compat_ioctl.c:1604: error: 'SG_SET_TRANSFORM' undeclared
>
> Do you want that through VFS tree? I'm going to send a bunch of assorted
> fixes tomorrow morning, so I can pass that one at the same time...
Yes, please.
No driver uses SG_SET_TRANSFORM any more in Linux, since the ide-scsi
driver was removed in 2.6.29. The compat-ioctl cleanup series moved
the handling for this around, which broke building without CONFIG_BLOCK.
Just remove the code handling it for compat mode.
---
On Sunday 20 December 2009, Alexander Beregalov wrote:
> 2009/12/21 Al Viro <[email protected]>:
> > On Mon, Dec 21, 2009 at 12:21:32AM +0300, Alexander Beregalov wrote:
> >> Fix this build error when CONFIG_BLOCK is not set.
> >> fs/compat_ioctl.c:1604: error: 'SG_SET_TRANSFORM' undeclared
> >
> > Do you want that through VFS tree? I'm going to send a bunch of assorted
> > fixes tomorrow morning, so I can pass that one at the same time...
>
> Yes, please.
Sorry for being slow on this, I already had the fix prepared when Randy
found it last week but had not gotten around to submitting it.
Al, the patch that Alexander has sent is obviously correct as well,
so if you have already prepared the series just keep that one, otherwise
please replace it with this one.
--- a/fs/compat_ioctl.c
+++ b/fs/compat_ioctl.c
@@ -1600,8 +1600,6 @@ static long do_ioctl_trans(int fd, unsigned int cmd,
case KDSKBMETA:
case KDSKBLED:
case KDSETLED:
- /* SG stuff */
- case SG_SET_TRANSFORM:
/* AUTOFS */
case AUTOFS_IOC_READY:
case AUTOFS_IOC_FAIL: