Build failures in stable queues (4.4.y and later)

Build failures in stable queues (4.4.y and later)

[Guenter Roeck]

4.4.y, 4.9.y:

fs/cifs/cifsfs.c: In function 'cifs_statfs':
fs/cifs/cifsfs.c:198:27: error: 'struct cifs_tcon' has no member named 'vol_serial_number'
fs/cifs/cifsfs.c:200:45: error: 'struct cifs_tcon' has no member named 'vol_create_time'

4.14.y, 4.18.y:

kernel/printk/printk_safe.o: In function `vprintk_func':
kernel/printk/printk_safe.c:386: undefined reference to `vprintk_store'
kernel/printk/printk_safe.c:388: undefined reference to `defer_console_output'

Re: Build failures in stable queues (4.4.y and later)

[Guenter Roeck]

On 08/31/2018 06:47 AM, Guenter Roeck wrote:
> 4.4.y, 4.9.y:
> 
> fs/cifs/cifsfs.c: In function 'cifs_statfs':
> fs/cifs/cifsfs.c:198:27: error: 'struct cifs_tcon' has no member named 'vol_serial_number'
> fs/cifs/cifsfs.c:200:45: error: 'struct cifs_tcon' has no member named 'vol_create_time'
> 
> 4.14.y, 4.18.y:
> 
> kernel/printk/printk_safe.o: In function `vprintk_func':
> kernel/printk/printk_safe.c:386: undefined reference to `vprintk_store'
> kernel/printk/printk_safe.c:388: undefined reference to `defer_console_output'
> 
> 

I just had to share this one. I don't think I ever had _all_ boot tests fail,
much less for two releases. The good news is that some builds are still passing :-).

For 4.14.y:

Build results:
	total: 150 pass: 27 fail: 123
Qemu test results:
	total: 302 pass: 0 fail: 302

For 4.18.y:

Build results:
	total: 136 pass: 27 fail: 109
Qemu test results:
	total: 305 pass: 0 fail: 305

Guenter

Re: Build failures in stable queues (4.4.y and later)

[Guenter Roeck]

On 08/31/2018 07:57 AM, Guenter Roeck wrote:
> On 08/31/2018 06:47 AM, Guenter Roeck wrote:
>> 4.4.y, 4.9.y:
>>
>> fs/cifs/cifsfs.c: In function 'cifs_statfs':
>> fs/cifs/cifsfs.c:198:27: error: 'struct cifs_tcon' has no member named 'vol_serial_number'
>> fs/cifs/cifsfs.c:200:45: error: 'struct cifs_tcon' has no member named 'vol_create_time'
>>
>> 4.14.y, 4.18.y:
>>
>> kernel/printk/printk_safe.o: In function `vprintk_func':
>> kernel/printk/printk_safe.c:386: undefined reference to `vprintk_store'
>> kernel/printk/printk_safe.c:388: undefined reference to `defer_console_output'
>>
>>
> 

One more build failure in v4.9.y.queue:


Build reference: v4.9.124-64-g6d66e98
gcc version: sparc64-linux-gcc (GCC) 7.3.0

Building sparc32:allmodconfig ... passed
Building sparc32:defconfig ... failed
--------------
Error log:

arch/sparc/kernel/pcic.c: In function 'pcibios_fixup_bus':
arch/sparc/kernel/pcic.c:647:8: error: 'cmd' may be used uninitialized in this function

Re: Build failures in stable queues (4.4.y and later)

[Greg Kroah-Hartman]

On Fri, Aug 31, 2018 at 08:47:59AM -0700, Guenter Roeck wrote:
> One more build failure in v4.9.y.queue:
> 
> 
> Build reference: v4.9.124-64-g6d66e98
> gcc version: sparc64-linux-gcc (GCC) 7.3.0
> 
> Building sparc32:allmodconfig ... passed
> Building sparc32:defconfig ... failed
> --------------
> Error log:
> 
> arch/sparc/kernel/pcic.c: In function 'pcibios_fixup_bus':
> arch/sparc/kernel/pcic.c:647:8: error: 'cmd' may be used uninitialized in this function

That is really odd as there is no patch in the queue that touches this
file.  Could this have been caused by an older release?

thanks,

greg k-h

Re: Build failures in stable queues (4.4.y and later)

[Guenter Roeck]

On Fri, Aug 31, 2018 at 01:07:28PM -0700, Greg Kroah-Hartman wrote:
> On Fri, Aug 31, 2018 at 08:47:59AM -0700, Guenter Roeck wrote:
> > One more build failure in v4.9.y.queue:
> > 
> > 
> > Build reference: v4.9.124-64-g6d66e98
> > gcc version: sparc64-linux-gcc (GCC) 7.3.0
> > 
> > Building sparc32:allmodconfig ... passed
> > Building sparc32:defconfig ... failed
> > --------------
> > Error log:
> > 
> > arch/sparc/kernel/pcic.c: In function 'pcibios_fixup_bus':
> > arch/sparc/kernel/pcic.c:647:8: error: 'cmd' may be used uninitialized in this function
> 
> That is really odd as there is no patch in the queue that touches this
> file.  Could this have been caused by an older release?
> 

Ah, yes, sorry. I had added sparc32:defconfig recently since
sparc32:allmodconfig no longer builds in mainline. The fix is
commit 2dc77533f1e4 ("sparc: kernel/pcic: silence gcc 7.x
warning in pcibios_fixup_bus()"). The patch applies cleanly
to v4.9.y and fixes the problem.

Guenter

Re: Build failures in stable queues (4.4.y and later)

[Greg Kroah-Hartman]

On Fri, Aug 31, 2018 at 01:19:05PM -0700, Guenter Roeck wrote:
> On Fri, Aug 31, 2018 at 01:07:28PM -0700, Greg Kroah-Hartman wrote:
> > On Fri, Aug 31, 2018 at 08:47:59AM -0700, Guenter Roeck wrote:
> > > One more build failure in v4.9.y.queue:
> > > 
> > > 
> > > Build reference: v4.9.124-64-g6d66e98
> > > gcc version: sparc64-linux-gcc (GCC) 7.3.0
> > > 
> > > Building sparc32:allmodconfig ... passed
> > > Building sparc32:defconfig ... failed
> > > --------------
> > > Error log:
> > > 
> > > arch/sparc/kernel/pcic.c: In function 'pcibios_fixup_bus':
> > > arch/sparc/kernel/pcic.c:647:8: error: 'cmd' may be used uninitialized in this function
> > 
> > That is really odd as there is no patch in the queue that touches this
> > file.  Could this have been caused by an older release?
> > 
> 
> Ah, yes, sorry. I had added sparc32:defconfig recently since
> sparc32:allmodconfig no longer builds in mainline. The fix is
> commit 2dc77533f1e4 ("sparc: kernel/pcic: silence gcc 7.x
> warning in pcibios_fixup_bus()"). The patch applies cleanly
> to v4.9.y and fixes the problem.

No problem, I'll go queue that up now.

greg k-h

Re: Build failures in stable queues (4.4.y and later)

[Greg Kroah-Hartman]

On Fri, Aug 31, 2018 at 07:57:23AM -0700, Guenter Roeck wrote:
> On 08/31/2018 06:47 AM, Guenter Roeck wrote:
> > 4.4.y, 4.9.y:
> > 
> > fs/cifs/cifsfs.c: In function 'cifs_statfs':
> > fs/cifs/cifsfs.c:198:27: error: 'struct cifs_tcon' has no member named 'vol_serial_number'
> > fs/cifs/cifsfs.c:200:45: error: 'struct cifs_tcon' has no member named 'vol_create_time'
> > 
> > 4.14.y, 4.18.y:
> > 
> > kernel/printk/printk_safe.o: In function `vprintk_func':
> > kernel/printk/printk_safe.c:386: undefined reference to `vprintk_store'
> > kernel/printk/printk_safe.c:388: undefined reference to `defer_console_output'
> > 
> > 
> 
> I just had to share this one. I don't think I ever had _all_ boot tests fail,
> much less for two releases. The good news is that some builds are still passing :-).
> 
> For 4.14.y:
> 
> Build results:
> 	total: 150 pass: 27 fail: 123
> Qemu test results:
> 	total: 302 pass: 0 fail: 302
> 
> For 4.18.y:
> 
> Build results:
> 	total: 136 pass: 27 fail: 109
> Qemu test results:
> 	total: 305 pass: 0 fail: 305

Ugh, things are really broken, sorry about that.  I'm at the OSS
conference and pushed out patches without being able to test them first
and then was too busy to read email for 24 hours :(

I'll fix this up later today, sorry.

greg k-h

Re: Build failures in stable queues (4.4.y and later)

[Greg Kroah-Hartman]

On Fri, Aug 31, 2018 at 06:47:02AM -0700, Guenter Roeck wrote:
> 4.14.y, 4.18.y:
> 
> kernel/printk/printk_safe.o: In function `vprintk_func':
> kernel/printk/printk_safe.c:386: undefined reference to `vprintk_store'
> kernel/printk/printk_safe.c:388: undefined reference to `defer_console_output'

This should now be fixed, thanks.

greg k-h

Re: Build failures in stable queues (4.4.y and later)

[Greg Kroah-Hartman]

On Fri, Aug 31, 2018 at 06:47:02AM -0700, Guenter Roeck wrote:
> 4.4.y, 4.9.y:
> 
> fs/cifs/cifsfs.c: In function 'cifs_statfs':
> fs/cifs/cifsfs.c:198:27: error: 'struct cifs_tcon' has no member named 'vol_serial_number'
> fs/cifs/cifsfs.c:200:45: error: 'struct cifs_tcon' has no member named 'vol_create_time'

Should now be fixed, thanks.

greg k-h