linux-next: build failure after merge of the tty.current tree

linux-next: build failure after merge of the tty.current tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1233 bytes --]

Hi Greg,

After merging the tty.current tree, today's linux-next build (arm
multi_v7_defconfig) failed like this:

In file included from drivers/tty/serial/imx.c:42:0:
drivers/tty/serial/imx.c: In function 'imx_start_tx':
drivers/tty/serial/imx.c:570:27: error: request for member 'state' in something not a structure or union
  if (uart_circ_empty(&port.state->xmit))
                           ^
include/linux/serial_core.h:336:34: note: in definition of macro 'uart_circ_empty'
 #define uart_circ_empty(circ)  ((circ)->head == (circ)->tail)
                                  ^
drivers/tty/serial/imx.c:570:27: error: request for member 'state' in something not a structure or union
  if (uart_circ_empty(&port.state->xmit))
                           ^
include/linux/serial_core.h:336:50: note: in definition of macro 'uart_circ_empty'
 #define uart_circ_empty(circ)  ((circ)->head == (circ)->tail)
                                                  ^

Caused by commit c557d392fbf5 ("serial: Test for no tx data on tx
restart").  That can't have been build tested :-(

I have used the tty.current tree from next-20140710 for today.

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

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

[PATCH] serial: imx: Fix build breakage

[Peter Hurley]

Fix breakage introduced by
commit c557d392fbf5badd693ea1946a4317c87a26a716,
'serial: Test for no tx data on tx restart'.

Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
---
 drivers/tty/serial/imx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/tty/serial/imx.c b/drivers/tty/serial/imx.c
index 23c1dae..be1c842 100644
--- a/drivers/tty/serial/imx.c
+++ b/drivers/tty/serial/imx.c
@@ -563,7 +563,7 @@ static void imx_start_tx(struct uart_port *port)
 	struct imx_port *sport = (struct imx_port *)port;
 	unsigned long temp;
 
-	if (uart_circ_empty(&port.state->xmit))
+	if (uart_circ_empty(&port->state->xmit))
 		return;
 
 	if (USE_IRDA(sport)) {
-- 
2.0.0

Re: [PATCH] serial: imx: Fix build breakage

[Greg Kroah-Hartman]

On Thu, Jul 10, 2014 at 10:36:46PM -0400, Peter Hurley wrote:
> Fix breakage introduced by
> commit c557d392fbf5badd693ea1946a4317c87a26a716,
> 'serial: Test for no tx data on tx restart'.
> 
> Signed-off-by: Peter Hurley <peter@hurleysoftware.com>

You forgot a "Reported-by:" tag.

I'll go add it...

Re: [PATCH] serial: imx: Fix build breakage

[Randy Dunlap]

On 07/11/14 17:42, Greg Kroah-Hartman wrote:
> On Thu, Jul 10, 2014 at 10:36:46PM -0400, Peter Hurley wrote:
>> Fix breakage introduced by
>> commit c557d392fbf5badd693ea1946a4317c87a26a716,
>> 'serial: Test for no tx data on tx restart'.
>>
>> Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
> 
> You forgot a "Reported-by:" tag.
> 
> I'll go add it...

It's not clear to me that Peter forgot it.
Someone else brought this up -- it seems to me that
Documentation/SubmittingPatches is a little too strong about needing
permission to use Reported-by:

If this patch fixes a problem reported by somebody else, consider adding a
Reported-by: tag to credit the reporter for their contribution.  Please
note that this tag should not be added without the reporter's permission,
especially if the problem was not reported in a public forum.  That said,
if we diligently credit our bug reporters, they will, hopefully, be
inspired to help us again in the future.


Sure, if a problem was reported privately, then we need to have permission
to use Reported-by, but if it was reported in a public forum, that should be
implicit permission (IMHO).

eh?

-- 
~Randy

Re: [PATCH] serial: imx: Fix build breakage

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 1190 bytes --]

Hi Randy,

On Fri, 11 Jul 2014 19:00:48 -0700 Randy Dunlap <rdunlap@infradead.org> wrote:
>
> It's not clear to me that Peter forgot it.
> Someone else brought this up -- it seems to me that
> Documentation/SubmittingPatches is a little too strong about needing
> permission to use Reported-by:
> 
> If this patch fixes a problem reported by somebody else, consider adding a
> Reported-by: tag to credit the reporter for their contribution.  Please
> note that this tag should not be added without the reporter's permission,
> especially if the problem was not reported in a public forum.  That said,
> if we diligently credit our bug reporters, they will, hopefully, be
> inspired to help us again in the future.
> 
> 
> Sure, if a problem was reported privately, then we need to have permission
> to use Reported-by, but if it was reported in a public forum, that should be
> implicit permission (IMHO).
> 
> eh?

Yes, indeed.

Just to be clear, I hereby give my permission for the addition of
Reported-by: tags with my email address for anything I report
publically or privately.  :-)

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

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

Re: [PATCH] serial: imx: Fix build breakage

[Greg Kroah-Hartman]

On Fri, Jul 11, 2014 at 07:00:48PM -0700, Randy Dunlap wrote:
> On 07/11/14 17:42, Greg Kroah-Hartman wrote:
> > On Thu, Jul 10, 2014 at 10:36:46PM -0400, Peter Hurley wrote:
> >> Fix breakage introduced by
> >> commit c557d392fbf5badd693ea1946a4317c87a26a716,
> >> 'serial: Test for no tx data on tx restart'.
> >>
> >> Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
> > 
> > You forgot a "Reported-by:" tag.
> > 
> > I'll go add it...
> 
> It's not clear to me that Peter forgot it.

Why?  Stephen pointed out the problem, in a public place, so I gave him
credit for it.

greg k-h

Re: [PATCH] serial: imx: Fix build breakage

[Randy Dunlap]

On 07/11/14 19:16, Greg Kroah-Hartman wrote:
> On Fri, Jul 11, 2014 at 07:00:48PM -0700, Randy Dunlap wrote:
>> On 07/11/14 17:42, Greg Kroah-Hartman wrote:
>>> On Thu, Jul 10, 2014 at 10:36:46PM -0400, Peter Hurley wrote:
>>>> Fix breakage introduced by
>>>> commit c557d392fbf5badd693ea1946a4317c87a26a716,
>>>> 'serial: Test for no tx data on tx restart'.
>>>>
>>>> Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
>>>
>>> You forgot a "Reported-by:" tag.
>>>
>>> I'll go add it...
>>
>> It's not clear to me that Peter forgot it.
> 
> Why?  Stephen pointed out the problem, in a public place, so I gave him
> credit for it.

SubmittingPatches does not give you permission to do that,
but Stephen just did so.


-- 
~Randy

Re: [PATCH] serial: imx: Fix build breakage

[Peter Hurley]

On 07/11/2014 10:00 PM, Randy Dunlap wrote:
> On 07/11/14 17:42, Greg Kroah-Hartman wrote:
>> On Thu, Jul 10, 2014 at 10:36:46PM -0400, Peter Hurley wrote:
>>> Fix breakage introduced by
>>> commit c557d392fbf5badd693ea1946a4317c87a26a716,
>>> 'serial: Test for no tx data on tx restart'.
>>>
>>> Signed-off-by: Peter Hurley <peter@hurleysoftware.com>
>>
>> You forgot a "Reported-by:" tag.
>>
>> I'll go add it...
>
> It's not clear to me that Peter forgot it.

Yeah, it was an oversight. I was rushing to get the patch done because
I was embarrassed about breaking the build.

The kbuild test robot actually notified me first though :)

> Someone else brought this up -- it seems to me that
> Documentation/SubmittingPatches is a little too strong about needing
> permission to use Reported-by:
>
> If this patch fixes a problem reported by somebody else, consider adding a
> Reported-by: tag to credit the reporter for their contribution.  Please
> note that this tag should not be added without the reporter's permission,
> especially if the problem was not reported in a public forum.  That said,
> if we diligently credit our bug reporters, they will, hopefully, be
> inspired to help us again in the future.

I wish the Reported-by tag (and some others) was picked up by git send-email
automatically.

> Sure, if a problem was reported privately, then we need to have permission
> to use Reported-by, but if it was reported in a public forum, that should be
> implicit permission (IMHO).
>
> eh?

I agree.

Regards,
Peter Hurley

Re: linux-next: build failure after merge of the tty.current tree

[Peter Hurley]

On 07/10/2014 09:21 PM, Stephen Rothwell wrote:
> Hi Greg,
>
> After merging the tty.current tree, today's linux-next build (arm
> multi_v7_defconfig) failed like this:
[...]
> Caused by commit c557d392fbf5 ("serial: Test for no tx data on tx
> restart").  That can't have been build tested :-(

Sorry about that. I don't have a full cross-compiler setup for
build testing. Looks like that's something I'm going to have to add.

Regards,
Peter Hurley

Re: linux-next: build failure after merge of the tty.current tree

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 582 bytes --]

Hi Peter,

On Fri, 11 Jul 2014 11:06:56 -0400 Peter Hurley <peter@hurleysoftware.com> wrote:
>
> Sorry about that. I don't have a full cross-compiler setup for
> build testing. Looks like that's something I'm going to have to add.

see https://www.kernel.org/pub/tools/crosstool/

These are only good for building the kernel (no libc).

You don't necessarily need to build the whole kernel but often just
building the particular drivers is enough.  Especially if they are
directly modified ...

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

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

Re: linux-next: build failure after merge of the tty.current tree

[Peter Hurley]

On 07/11/2014 09:20 PM, Stephen Rothwell wrote:
> Hi Peter,
>
> On Fri, 11 Jul 2014 11:06:56 -0400 Peter Hurley <peter@hurleysoftware.com> wrote:
>>
>> Sorry about that. I don't have a full cross-compiler setup for
>> build testing. Looks like that's something I'm going to have to add.
>
> see https://www.kernel.org/pub/tools/crosstool/
>
> These are only good for building the kernel (no libc).
>
> You don't necessarily need to build the whole kernel but often just
> building the particular drivers is enough.  Especially if they are
> directly modified ...

Hi Stephen,

Thanks for the link. I had found Shuah Khan's slideshow pdf on cross-compiler
packages earlier and so have already started down that path.

Regards,
Peter Hurley

Re: linux-next: build failure after merge of the tty.current tree

[Peter Hurley]

On 07/12/2014 09:02 AM, Peter Hurley wrote:
> On 07/11/2014 09:20 PM, Stephen Rothwell wrote:
>> Hi Peter,
>>
>> On Fri, 11 Jul 2014 11:06:56 -0400 Peter Hurley <peter@hurleysoftware.com> wrote:
>>>
>>> Sorry about that. I don't have a full cross-compiler setup for
>>> build testing. Looks like that's something I'm going to have to add.
>>
>> see https://www.kernel.org/pub/tools/crosstool/
>>
>> These are only good for building the kernel (no libc).
>>
>> You don't necessarily need to build the whole kernel but often just
>> building the particular drivers is enough.  Especially if they are
>> directly modified ...
>
> Hi Stephen,
>
> Thanks for the link. I had found Shuah Khan's slideshow pdf on cross-compiler
> packages earlier and so have already started down that path.

Just an fyi,

The cross-compiler packages of the emdebian project in the unstable (aka, "sid")
distribution have broken dependencies.

So the instructions here
http://events.linuxfoundation.org/sites/events/files/slides/Shuah_Khan_cross_compile_linux.pdf
and here
http://linuxdriverproject.org/mediawiki/index.php/Cross-compiling_Linux_kernel_on_x86_64#Installation_Steps
are no longer applicable.

Regards,
Peter Hurley