Re: linux-next: Tree for Mar 13

[Stephen Rothwell]

Hi Dominik,

On Tue, 13 Mar 2018 18:52:05 +0100 Dominik Brodowski 
 wrote:
>
> would it be possible to have
> 
>https://git.kernel.org/pub/scm/linux/kernel/git/brodo/linux.git 
> syscalls-next
> 
> added to -next? This will probably only be for one or two release cycles,
> but it might easen the development process of that patch series.

Added from today.  I put it near the start since Linus said he may take
(some of) it this release.

Thanks for adding your subsystem tree as a participant of linux-next.  As
you may know, this is not a judgement of your code.  The purpose of
linux-next is for integration testing and to lower the impact of
conflicts between subsystems in the next merge window. 

You will need to ensure that the patches/commits in your tree/series have
been:
 * submitted under GPL v2 (or later) and include the Contributor's
Signed-off-by,
 * posted to the relevant mailing list,
 * reviewed by you (or another maintainer of your subsystem tree),
 * successfully unit tested, and 
 * destined for the current or next Linux merge window.

Basically, this should be just what you would send to Linus (or ask him
to fetch).  It is allowed to be rebased if you deem it necessary.

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


pgpPZcLtBfF2O.pgp
Description: OpenPGP digital signature

Re: linux-next: Tree for Mar 13

[Stephen Rothwell]

Hi Dominik,

On Tue, 13 Mar 2018 18:52:05 +0100 Dominik Brodowski 
 wrote:
>
> would it be possible to have
> 
>https://git.kernel.org/pub/scm/linux/kernel/git/brodo/linux.git 
> syscalls-next
> 
> added to -next? This will probably only be for one or two release cycles,
> but it might easen the development process of that patch series.

Added from today.  I put it near the start since Linus said he may take
(some of) it this release.

Thanks for adding your subsystem tree as a participant of linux-next.  As
you may know, this is not a judgement of your code.  The purpose of
linux-next is for integration testing and to lower the impact of
conflicts between subsystems in the next merge window. 

You will need to ensure that the patches/commits in your tree/series have
been:
 * submitted under GPL v2 (or later) and include the Contributor's
Signed-off-by,
 * posted to the relevant mailing list,
 * reviewed by you (or another maintainer of your subsystem tree),
 * successfully unit tested, and 
 * destined for the current or next Linux merge window.

Basically, this should be just what you would send to Linus (or ask him
to fetch).  It is allowed to be rebased if you deem it necessary.

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


pgpPZcLtBfF2O.pgp
Description: OpenPGP digital signature

Re: linux-next: Tree for Mar 13

[Stephen Rothwell]

Hi Mike,

On Tue, 13 Mar 2018 11:57:40 -0400 Mike Snitzer  wrote:
>
> I had to revert the following commits to get this kernel to build
> (otherwise I got macro expansion errors, using RHEL7
> gcc-4.8.5-28.el7.x86_64):

Yeah, I reported that earlier (216 of my 244 builds failed - gcc 4.6.3)

> beb7eb2 kernelh-skip-single-eval-logic-on-literals-in-min-max-v3
> 19ff7e5 kernelh-skip-single-eval-logic-on-literals-in-min-max-v2
> c7c133f kernel.h: skip single-eval logic on literals in min()/max()
> 
> (it wasn't until I reverted commit c7c133f that the kernel build worked)

I will revert those from linux-next today unless someone sends a fix.

-- 
Cheers,
Stephen Rothwell


pgpwRHWBWLjIO.pgp
Description: OpenPGP digital signature

Re: linux-next: Tree for Mar 13

[Stephen Rothwell]

Hi Mike,

On Tue, 13 Mar 2018 11:57:40 -0400 Mike Snitzer  wrote:
>
> I had to revert the following commits to get this kernel to build
> (otherwise I got macro expansion errors, using RHEL7
> gcc-4.8.5-28.el7.x86_64):

Yeah, I reported that earlier (216 of my 244 builds failed - gcc 4.6.3)

> beb7eb2 kernelh-skip-single-eval-logic-on-literals-in-min-max-v3
> 19ff7e5 kernelh-skip-single-eval-logic-on-literals-in-min-max-v2
> c7c133f kernel.h: skip single-eval logic on literals in min()/max()
> 
> (it wasn't until I reverted commit c7c133f that the kernel build worked)

I will revert those from linux-next today unless someone sends a fix.

-- 
Cheers,
Stephen Rothwell


pgpwRHWBWLjIO.pgp
Description: OpenPGP digital signature

Re: linux-next: Tree for Mar 13

[Kees Cook]

On Tue, Mar 13, 2018 at 8:57 AM, Mike Snitzer  wrote:
> I had to revert the following commits to get this kernel to build
> (otherwise I got macro expansion errors, using RHEL7
> gcc-4.8.5-28.el7.x86_64):
>
> beb7eb2 kernelh-skip-single-eval-logic-on-literals-in-min-max-v3
> 19ff7e5 kernelh-skip-single-eval-logic-on-literals-in-min-max-v2
> c7c133f kernel.h: skip single-eval logic on literals in min()/max()

Yeah, these need to be dropped from -mm. There isn't a sane solution
at the moment that works fully for all compiler versions. :(

-Kees

-- 
Kees Cook
Pixel Security

Re: linux-next: Tree for Mar 13

[Kees Cook]

On Tue, Mar 13, 2018 at 8:57 AM, Mike Snitzer  wrote:
> I had to revert the following commits to get this kernel to build
> (otherwise I got macro expansion errors, using RHEL7
> gcc-4.8.5-28.el7.x86_64):
>
> beb7eb2 kernelh-skip-single-eval-logic-on-literals-in-min-max-v3
> 19ff7e5 kernelh-skip-single-eval-logic-on-literals-in-min-max-v2
> c7c133f kernel.h: skip single-eval logic on literals in min()/max()

Yeah, these need to be dropped from -mm. There isn't a sane solution
at the moment that works fully for all compiler versions. :(

-Kees

-- 
Kees Cook
Pixel Security

Re: linux-next: Tree for Mar 13

[Randy Dunlap]

On 03/13/2018 08:57 AM, Mike Snitzer wrote:
> I had to revert the following commits to get this kernel to build
> (otherwise I got macro expansion errors, using RHEL7
> gcc-4.8.5-28.el7.x86_64):
> 
> beb7eb2 kernelh-skip-single-eval-logic-on-literals-in-min-max-v3
> 19ff7e5 kernelh-skip-single-eval-logic-on-literals-in-min-max-v2
> c7c133f kernel.h: skip single-eval logic on literals in min()/max()
> 
> (it wasn't until I reverted commit c7c133f that the kernel build worked)

Yep, all of my automated randconfig builds were useless...


> Here is a snippet of the errors if commit c7c133f was applied:
> 
> [root@rhel-storage-02 linux]# make
>   CHK include/config/kernel.release
>   CHK include/generated/uapi/linux/version.h
>   CHK include/generated/utsrelease.h
>   CC  kernel/bounds.s
>   CHK include/generated/bounds.h
>   CHK include/generated/timeconst.h
>   CC  arch/x86/kernel/asm-offsets.s
> In file included from ./include/linux/crypto.h:21:0,
>  from arch/x86/kernel/asm-offsets.c:9:
> ./include/linux/jiffies.h: In function ‘jiffies_delta_to_clock_t’:
> ./include/linux/kernel.h:856:2: error: first argument to
> ‘__builtin_choose_expr’ not a constant
>   __builtin_choose_expr(__builtin_constant_p(x) &&  \
>   ^
> ./include/linux/kernel.h:869:19: note: in expansion of macro ‘__max’
>  #define max(x, y) __max(typeof(x), typeof(y), x, y)
>^
> ./include/linux/jiffies.h:444:28: note: in expansion of macro ‘max’
>   return jiffies_to_clock_t(max(0L, delta));
> ^
> ./include/linux/mm.h: In function ‘get_mm_hiwater_rss’:
> ./include/linux/kernel.h:856:2: error: first argument to
> ‘__builtin_choose_expr’ not a constant
>   __builtin_choose_expr(__builtin_constant_p(x) &&  \
>   ^
> ./include/linux/kernel.h:869:19: note: in expansion of macro ‘__max’
>  #define max(x, y) __max(typeof(x), typeof(y), x, y)
>^
> ./include/linux/mm.h:1569:9: note: in expansion of macro ‘max’
>   return max(mm->hiwater_rss, get_mm_rss(mm));
>  ^
> ./include/linux/mm.h: In function ‘get_mm_hiwater_vm’:
> ./include/linux/kernel.h:856:2: error: first argument to
> ‘__builtin_choose_expr’ not a constant
>   __builtin_choose_expr(__builtin_constant_p(x) &&  \
>   ^
> ./include/linux/kernel.h:869:19: note: in expansion of macro ‘__max’
>  #define max(x, y) __max(typeof(x), typeof(y), x, y)
>^
> ./include/linux/mm.h:1574:9: note: in expansion of macro ‘max’
>   return max(mm->hiwater_vm, mm->total_vm);
>  ^
> ./include/linux/bvec.h: In function ‘bvec_iter_advance’:
> ./include/linux/kernel.h:833:2: error: first argument to
> ‘__builtin_choose_expr’ not a constant
>   __builtin_choose_expr(__builtin_constant_p(x) &&  \
>   ^
> ./include/linux/kernel.h:847:19: note: in expansion of macro ‘__min’
>  #define min(x, y) __min(typeof(x), typeof(y), x, y)
>^
> ./include/linux/bvec.h:59:2: note: in expansion of macro ‘min’
>   min((iter).bi_size, \
>   ^
> ...
> 


-- 
~Randy

Re: linux-next: Tree for Mar 13

[Randy Dunlap]

On 03/13/2018 08:57 AM, Mike Snitzer wrote:
> I had to revert the following commits to get this kernel to build
> (otherwise I got macro expansion errors, using RHEL7
> gcc-4.8.5-28.el7.x86_64):
> 
> beb7eb2 kernelh-skip-single-eval-logic-on-literals-in-min-max-v3
> 19ff7e5 kernelh-skip-single-eval-logic-on-literals-in-min-max-v2
> c7c133f kernel.h: skip single-eval logic on literals in min()/max()
> 
> (it wasn't until I reverted commit c7c133f that the kernel build worked)

Yep, all of my automated randconfig builds were useless...


> Here is a snippet of the errors if commit c7c133f was applied:
> 
> [root@rhel-storage-02 linux]# make
>   CHK include/config/kernel.release
>   CHK include/generated/uapi/linux/version.h
>   CHK include/generated/utsrelease.h
>   CC  kernel/bounds.s
>   CHK include/generated/bounds.h
>   CHK include/generated/timeconst.h
>   CC  arch/x86/kernel/asm-offsets.s
> In file included from ./include/linux/crypto.h:21:0,
>  from arch/x86/kernel/asm-offsets.c:9:
> ./include/linux/jiffies.h: In function ‘jiffies_delta_to_clock_t’:
> ./include/linux/kernel.h:856:2: error: first argument to
> ‘__builtin_choose_expr’ not a constant
>   __builtin_choose_expr(__builtin_constant_p(x) &&  \
>   ^
> ./include/linux/kernel.h:869:19: note: in expansion of macro ‘__max’
>  #define max(x, y) __max(typeof(x), typeof(y), x, y)
>^
> ./include/linux/jiffies.h:444:28: note: in expansion of macro ‘max’
>   return jiffies_to_clock_t(max(0L, delta));
> ^
> ./include/linux/mm.h: In function ‘get_mm_hiwater_rss’:
> ./include/linux/kernel.h:856:2: error: first argument to
> ‘__builtin_choose_expr’ not a constant
>   __builtin_choose_expr(__builtin_constant_p(x) &&  \
>   ^
> ./include/linux/kernel.h:869:19: note: in expansion of macro ‘__max’
>  #define max(x, y) __max(typeof(x), typeof(y), x, y)
>^
> ./include/linux/mm.h:1569:9: note: in expansion of macro ‘max’
>   return max(mm->hiwater_rss, get_mm_rss(mm));
>  ^
> ./include/linux/mm.h: In function ‘get_mm_hiwater_vm’:
> ./include/linux/kernel.h:856:2: error: first argument to
> ‘__builtin_choose_expr’ not a constant
>   __builtin_choose_expr(__builtin_constant_p(x) &&  \
>   ^
> ./include/linux/kernel.h:869:19: note: in expansion of macro ‘__max’
>  #define max(x, y) __max(typeof(x), typeof(y), x, y)
>^
> ./include/linux/mm.h:1574:9: note: in expansion of macro ‘max’
>   return max(mm->hiwater_vm, mm->total_vm);
>  ^
> ./include/linux/bvec.h: In function ‘bvec_iter_advance’:
> ./include/linux/kernel.h:833:2: error: first argument to
> ‘__builtin_choose_expr’ not a constant
>   __builtin_choose_expr(__builtin_constant_p(x) &&  \
>   ^
> ./include/linux/kernel.h:847:19: note: in expansion of macro ‘__min’
>  #define min(x, y) __min(typeof(x), typeof(y), x, y)
>^
> ./include/linux/bvec.h:59:2: note: in expansion of macro ‘min’
>   min((iter).bi_size, \
>   ^
> ...
> 


-- 
~Randy

Re: linux-next: Tree for Mar 13

[Mike Snitzer]

I had to revert the following commits to get this kernel to build
(otherwise I got macro expansion errors, using RHEL7
gcc-4.8.5-28.el7.x86_64):

beb7eb2 kernelh-skip-single-eval-logic-on-literals-in-min-max-v3
19ff7e5 kernelh-skip-single-eval-logic-on-literals-in-min-max-v2
c7c133f kernel.h: skip single-eval logic on literals in min()/max()

(it wasn't until I reverted commit c7c133f that the kernel build worked)

Here is a snippet of the errors if commit c7c133f was applied:

[root@rhel-storage-02 linux]# make
  CHK include/config/kernel.release
  CHK include/generated/uapi/linux/version.h
  CHK include/generated/utsrelease.h
  CC  kernel/bounds.s
  CHK include/generated/bounds.h
  CHK include/generated/timeconst.h
  CC  arch/x86/kernel/asm-offsets.s
In file included from ./include/linux/crypto.h:21:0,
 from arch/x86/kernel/asm-offsets.c:9:
./include/linux/jiffies.h: In function ‘jiffies_delta_to_clock_t’:
./include/linux/kernel.h:856:2: error: first argument to
‘__builtin_choose_expr’ not a constant
  __builtin_choose_expr(__builtin_constant_p(x) &&  \
  ^
./include/linux/kernel.h:869:19: note: in expansion of macro ‘__max’
 #define max(x, y) __max(typeof(x), typeof(y), x, y)
   ^
./include/linux/jiffies.h:444:28: note: in expansion of macro ‘max’
  return jiffies_to_clock_t(max(0L, delta));
^
./include/linux/mm.h: In function ‘get_mm_hiwater_rss’:
./include/linux/kernel.h:856:2: error: first argument to
‘__builtin_choose_expr’ not a constant
  __builtin_choose_expr(__builtin_constant_p(x) &&  \
  ^
./include/linux/kernel.h:869:19: note: in expansion of macro ‘__max’
 #define max(x, y) __max(typeof(x), typeof(y), x, y)
   ^
./include/linux/mm.h:1569:9: note: in expansion of macro ‘max’
  return max(mm->hiwater_rss, get_mm_rss(mm));
 ^
./include/linux/mm.h: In function ‘get_mm_hiwater_vm’:
./include/linux/kernel.h:856:2: error: first argument to
‘__builtin_choose_expr’ not a constant
  __builtin_choose_expr(__builtin_constant_p(x) &&  \
  ^
./include/linux/kernel.h:869:19: note: in expansion of macro ‘__max’
 #define max(x, y) __max(typeof(x), typeof(y), x, y)
   ^
./include/linux/mm.h:1574:9: note: in expansion of macro ‘max’
  return max(mm->hiwater_vm, mm->total_vm);
 ^
./include/linux/bvec.h: In function ‘bvec_iter_advance’:
./include/linux/kernel.h:833:2: error: first argument to
‘__builtin_choose_expr’ not a constant
  __builtin_choose_expr(__builtin_constant_p(x) &&  \
  ^
./include/linux/kernel.h:847:19: note: in expansion of macro ‘__min’
 #define min(x, y) __min(typeof(x), typeof(y), x, y)
   ^
./include/linux/bvec.h:59:2: note: in expansion of macro ‘min’
  min((iter).bi_size, \
  ^
...

Re: linux-next: Tree for Mar 13

[Mike Snitzer]

I had to revert the following commits to get this kernel to build
(otherwise I got macro expansion errors, using RHEL7
gcc-4.8.5-28.el7.x86_64):

beb7eb2 kernelh-skip-single-eval-logic-on-literals-in-min-max-v3
19ff7e5 kernelh-skip-single-eval-logic-on-literals-in-min-max-v2
c7c133f kernel.h: skip single-eval logic on literals in min()/max()

(it wasn't until I reverted commit c7c133f that the kernel build worked)

Here is a snippet of the errors if commit c7c133f was applied:

[root@rhel-storage-02 linux]# make
  CHK include/config/kernel.release
  CHK include/generated/uapi/linux/version.h
  CHK include/generated/utsrelease.h
  CC  kernel/bounds.s
  CHK include/generated/bounds.h
  CHK include/generated/timeconst.h
  CC  arch/x86/kernel/asm-offsets.s
In file included from ./include/linux/crypto.h:21:0,
 from arch/x86/kernel/asm-offsets.c:9:
./include/linux/jiffies.h: In function ‘jiffies_delta_to_clock_t’:
./include/linux/kernel.h:856:2: error: first argument to
‘__builtin_choose_expr’ not a constant
  __builtin_choose_expr(__builtin_constant_p(x) &&  \
  ^
./include/linux/kernel.h:869:19: note: in expansion of macro ‘__max’
 #define max(x, y) __max(typeof(x), typeof(y), x, y)
   ^
./include/linux/jiffies.h:444:28: note: in expansion of macro ‘max’
  return jiffies_to_clock_t(max(0L, delta));
^
./include/linux/mm.h: In function ‘get_mm_hiwater_rss’:
./include/linux/kernel.h:856:2: error: first argument to
‘__builtin_choose_expr’ not a constant
  __builtin_choose_expr(__builtin_constant_p(x) &&  \
  ^
./include/linux/kernel.h:869:19: note: in expansion of macro ‘__max’
 #define max(x, y) __max(typeof(x), typeof(y), x, y)
   ^
./include/linux/mm.h:1569:9: note: in expansion of macro ‘max’
  return max(mm->hiwater_rss, get_mm_rss(mm));
 ^
./include/linux/mm.h: In function ‘get_mm_hiwater_vm’:
./include/linux/kernel.h:856:2: error: first argument to
‘__builtin_choose_expr’ not a constant
  __builtin_choose_expr(__builtin_constant_p(x) &&  \
  ^
./include/linux/kernel.h:869:19: note: in expansion of macro ‘__max’
 #define max(x, y) __max(typeof(x), typeof(y), x, y)
   ^
./include/linux/mm.h:1574:9: note: in expansion of macro ‘max’
  return max(mm->hiwater_vm, mm->total_vm);
 ^
./include/linux/bvec.h: In function ‘bvec_iter_advance’:
./include/linux/kernel.h:833:2: error: first argument to
‘__builtin_choose_expr’ not a constant
  __builtin_choose_expr(__builtin_constant_p(x) &&  \
  ^
./include/linux/kernel.h:847:19: note: in expansion of macro ‘__min’
 #define min(x, y) __min(typeof(x), typeof(y), x, y)
   ^
./include/linux/bvec.h:59:2: note: in expansion of macro ‘min’
  min((iter).bi_size, \
  ^
...