build failure of next-20220906 due to 8bfc149ba24c ("scsi: qla2xxx: Enhance driver tracing with separate tunable and more")

build failure of next-20220906 due to 8bfc149ba24c ("scsi: qla2xxx: Enhance driver tracing with separate tunable and more")

[Sudip Mukherjee (Codethink)]

Hi All,

The builds of loongarch loongson3_defconfig have failed to build
next-20220906 with the error:

drivers/scsi/qla2xxx/qla_os.c: In function 'qla_trace_init':
drivers/scsi/qla2xxx/qla_os.c:2854:25: error: implicit declaration of function 'trace_array_get_by_name'; did you mean 'trace_array_set_clr_event'? [-Werror=implicit-function-declaration]
 2854 |         qla_trc_array = trace_array_get_by_name("qla2xxx");
      |                         ^~~~~~~~~~~~~~~~~~~~~~~
      |                         trace_array_set_clr_event

drivers/scsi/qla2xxx/qla_os.c: In function 'qla_trace_uninit':
drivers/scsi/qla2xxx/qla_os.c:2869:9: error: implicit declaration of function 'trace_array_put' [-Werror=implicit-function-declaration]
 2869 |         trace_array_put(qla_trc_array);
      |         ^~~~~~~~~~~~~~~


git bisect pointed to 8bfc149ba24c ("scsi: qla2xxx: Enhance driver tracing with separate tunable and more").

I will be happy to test any patch or provide any extra log if needed.


--
Regards
Sudip

Re: [EXT] build failure of next-20220906 due to 8bfc149ba24c ("scsi: qla2xxx: Enhance driver tracing with separate tunable and more")

[Arun Easi]

[-- Attachment #1: Type: text/plain, Size: 1472 bytes --]

Hi Sudip,

On Tue, 6 Sep 2022, 7:32am, Sudip Mukherjee (Codethink) wrote:

> External Email
> 
> ----------------------------------------------------------------------
> Hi All,
> 
> The builds of loongarch loongson3_defconfig have failed to build
> next-20220906 with the error:
> 
> drivers/scsi/qla2xxx/qla_os.c: In function 'qla_trace_init':
> drivers/scsi/qla2xxx/qla_os.c:2854:25: error: implicit declaration of function 'trace_array_get_by_name'; did you mean 'trace_array_set_clr_event'? [-Werror=implicit-function-declaration]
>  2854 |         qla_trc_array = trace_array_get_by_name("qla2xxx");
>       |                         ^~~~~~~~~~~~~~~~~~~~~~~
>       |                         trace_array_set_clr_event
> 
> drivers/scsi/qla2xxx/qla_os.c: In function 'qla_trace_uninit':
> drivers/scsi/qla2xxx/qla_os.c:2869:9: error: implicit declaration of function 'trace_array_put' [-Werror=implicit-function-declaration]
>  2869 |         trace_array_put(qla_trc_array);
>       |         ^~~~~~~~~~~~~~~
> 
> 
> git bisect pointed to 8bfc149ba24c ("scsi: qla2xxx: Enhance driver tracing with separate tunable and more").
> 
> I will be happy to test any patch or provide any extra log if needed.
> 
> 

This looks like is happening due to CONFIG_TRACING not being enabled (same 
as what kernel test bot reported). Could you try out the fix attached?

@Steven, please see the patch attached. Do you think the fix belongs 
better in linux/trace.h ?

Regards,
-Arun

[-- Attachment #2: Type: text/plain, Size: 1771 bytes --]

From ae4603712c9f3a49d67015f6386c520a77319ffa Mon Sep 17 00:00:00 2001
From: Arun Easi <aeasi@marvell.com>
Date: Fri, 2 Sep 2022 17:53:30 -0700
Subject: [PATCH] fixup! qla2xxx: Enhance driver tracing with separate tunable
 and more

Fix this compilation error seen when CONFIG_TRACING is not enabled:

drivers/scsi/qla2xxx/qla_os.c: In function 'qla_trace_init':
drivers/scsi/qla2xxx/qla_os.c:2854:25: error: implicit declaration of function
'trace_array_get_by_name'; did you mean 'trace_array_set_clr_event'?
[-Werror=implicit-function-declaration]
 2854 |         qla_trc_array = trace_array_get_by_name("qla2xxx");
      |                         ^~~~~~~~~~~~~~~~~~~~~~~
      |                         trace_array_set_clr_event

drivers/scsi/qla2xxx/qla_os.c: In function 'qla_trace_uninit':
drivers/scsi/qla2xxx/qla_os.c:2869:9: error: implicit declaration of function
'trace_array_put' [-Werror=implicit-function-declaration]
 2869 |         trace_array_put(qla_trc_array);
      |         ^~~~~~~~~~~~~~~

Reported-by: kernel test robot <lkp@intel.com>
---
 drivers/scsi/qla2xxx/qla_def.h | 7 +++++++
 1 file changed, 7 insertions(+)

diff --git a/drivers/scsi/qla2xxx/qla_def.h b/drivers/scsi/qla2xxx/qla_def.h
index 3ec6a200942e..d0da737e51e2 100644
--- a/drivers/scsi/qla2xxx/qla_def.h
+++ b/drivers/scsi/qla2xxx/qla_def.h
@@ -35,6 +35,13 @@
 
 #include <uapi/scsi/fc/fc_els.h>
 
+#ifndef CONFIG_TRACING
+#ifndef trace_array_get_by_name
+#define trace_array_get_by_name(_trc_arr)	NULL
+#define trace_array_put(_trc_arr)
+#endif /* trace_array_get_by_name */
+#endif /* CONFIG_TRACING */
+
 /* Big endian Fibre Channel S_ID (source ID) or D_ID (destination ID). */
 typedef struct {
 	uint8_t domain;
-- 
2.27.0

Re: [EXT] build failure of next-20220906 due to 8bfc149ba24c ("scsi: qla2xxx: Enhance driver tracing with separate tunable and more")

[Steven Rostedt]

On Tue, 6 Sep 2022 09:05:38 -0700
Arun Easi <aeasi@marvell.com> wrote:

> This looks like is happening due to CONFIG_TRACING not being enabled (same 
> as what kernel test bot reported). Could you try out the fix attached?
> 
> @Steven, please see the patch attached. Do you think the fix belongs 
> better in linux/trace.h ?

Yes, of course. Feel free to send me a patch.

Thanks!

-- Steve

Re: [EXT] build failure of next-20220906 due to 8bfc149ba24c ("scsi: qla2xxx: Enhance driver tracing with separate tunable and more")

[Sudip Mukherjee]

On Tue, Sep 6, 2022 at 5:05 PM Arun Easi <aeasi@marvell.com> wrote:
>
> Hi Sudip,
>
> On Tue, 6 Sep 2022, 7:32am, Sudip Mukherjee (Codethink) wrote:
>
> > External Email
> >
> > ----------------------------------------------------------------------
> > Hi All,
> >
> > The builds of loongarch loongson3_defconfig have failed to build
> > next-20220906 with the error:
> >
> > drivers/scsi/qla2xxx/qla_os.c: In function 'qla_trace_init':
> > drivers/scsi/qla2xxx/qla_os.c:2854:25: error: implicit declaration of function 'trace_array_get_by_name'; did you mean 'trace_array_set_clr_event'? [-Werror=implicit-function-declaration]
> >  2854 |         qla_trc_array = trace_array_get_by_name("qla2xxx");
> >       |                         ^~~~~~~~~~~~~~~~~~~~~~~
> >       |                         trace_array_set_clr_event
> >
> > drivers/scsi/qla2xxx/qla_os.c: In function 'qla_trace_uninit':
> > drivers/scsi/qla2xxx/qla_os.c:2869:9: error: implicit declaration of function 'trace_array_put' [-Werror=implicit-function-declaration]
> >  2869 |         trace_array_put(qla_trc_array);
> >       |         ^~~~~~~~~~~~~~~
> >
> >
> > git bisect pointed to 8bfc149ba24c ("scsi: qla2xxx: Enhance driver tracing with separate tunable and more").
> >
> > I will be happy to test any patch or provide any extra log if needed.
> >
> >
>
> This looks like is happening due to CONFIG_TRACING not being enabled (same
> as what kernel test bot reported). Could you try out the fix attached?

Thanks, that fixed the build for me. The patch did not apply directly
on next-20220906, so I had to modify it a little bit.

Tested-by: Sudip Mukherjee <sudipm.mukherjee@gmail.com>

-- 
Regards
Sudip

Re: [EXT] build failure of next-20220906 due to 8bfc149ba24c ("scsi: qla2xxx: Enhance driver tracing with separate tunable and more")

[Steven Rostedt]

On Tue, 6 Sep 2022 21:57:10 +0100
Sudip Mukherjee <sudipm.mukherjee@gmail.com> wrote:

> Thanks, that fixed the build for me. The patch did not apply directly
> on next-20220906, so I had to modify it a little bit.
> 
> Tested-by: Sudip Mukherjee <sudipm.mukherjee@gmail.com>

Well, the change needs to go into include/linux/trace.h though.

-- Steve

Re: [EXT] build failure of next-20220906 due to 8bfc149ba24c ("scsi: qla2xxx: Enhance driver tracing with separate tunable and more")

[Arun Easi]

On Tue, 6 Sep 2022, 2:03pm, Steven Rostedt wrote:

> On Tue, 6 Sep 2022 21:57:10 +0100
> Sudip Mukherjee <sudipm.mukherjee@gmail.com> wrote:
> 
> > Thanks, that fixed the build for me. The patch did not apply directly
> > on next-20220906, so I had to modify it a little bit.
> > 
> > Tested-by: Sudip Mukherjee <sudipm.mukherjee@gmail.com>
> 
> Well, the change needs to go into include/linux/trace.h though.
> 

Steve, I was thinking both the fixes stay at least for a short 
term the one in qla2xxx to avoid tree dependencies.

Since, the qla_def.h change goes as:

+#ifndef CONFIG_TRACING
+#ifndef trace_array_get_by_name

..it should co-exist when the "#define trace_array_get_by_name" gets 
merged in include/linux/trace.h. BTW, I will send out the changes to 
trace.h today.

Other alternatives/suggestions welcome.

Regards,
-Arun

Re: [EXT] build failure of next-20220906 due to 8bfc149ba24c ("scsi: qla2xxx: Enhance driver tracing with separate tunable and more")

[Steven Rostedt]

On Tue, 6 Sep 2022 14:26:31 -0700
Arun Easi <aeasi@marvell.com> wrote:

> Steve, I was thinking both the fixes stay at least for a short 
> term the one in qla2xxx to avoid tree dependencies.
> 
> Since, the qla_def.h change goes as:
> 
> +#ifndef CONFIG_TRACING
> +#ifndef trace_array_get_by_name
> 
> ..it should co-exist when the "#define trace_array_get_by_name" gets 
> merged in include/linux/trace.h. BTW, I will send out the changes to 
> trace.h today.
> 
> Other alternatives/suggestions welcome.

I doubt I'll have anything that conflicts with an update to
include/linux/trace.h, as it is seldom modified.

Just sent a patch out that updates that file and I'll review it, and then
you can push it through your tree.

-- Steve

Re: [EXT] build failure of next-20220906 due to 8bfc149ba24c ("scsi: qla2xxx: Enhance driver tracing with separate tunable and more")

[Arun Easi]

On Tue, 6 Sep 2022, 2:41pm, Steven Rostedt wrote:

> On Tue, 6 Sep 2022 14:26:31 -0700
> Arun Easi <aeasi@marvell.com> wrote:
> 
> > Steve, I was thinking both the fixes stay at least for a short 
> > term the one in qla2xxx to avoid tree dependencies.
> > 
> > Since, the qla_def.h change goes as:
> > 
> > +#ifndef CONFIG_TRACING
> > +#ifndef trace_array_get_by_name
> > 
> > ..it should co-exist when the "#define trace_array_get_by_name" gets 
> > merged in include/linux/trace.h. BTW, I will send out the changes to 
> > trace.h today.
> > 
> > Other alternatives/suggestions welcome.
> 
> I doubt I'll have anything that conflicts with an update to
> include/linux/trace.h, as it is seldom modified.
> 
> Just sent a patch out that updates that file and I'll review it, and then
> you can push it through your tree.
> 

Sounds good. Thanks Steve.

Regards,
-Arun