arm64: pointer_auth.h:40:3: error: implicit declaration of function ‘get_random_bytes’; did you mean ‘get_random_once’?

arm64: pointer_auth.h:40:3: error: implicit declaration of function ‘get_random_bytes’; did you mean ‘get_random_once’?

[Naresh Kamboju]

Linus master branch failed to build the arm64 architecture due to
below error and warnings.

kernel config:
https://builds.tuxbuild.com/Dp3s1p4uYnGpImQwCvhzEw/kernel.config

git_repo: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
target_arch: arm64
toolchain: gcc-9
git_short_log:
83bdc7275e62 (\random32: remove net_rand_state from the latent entropy
gcc plugin\)
git_describe: v5.8-rc7-40-g83bdc7275e62

make -sk KBUILD_BUILD_USER=TuxBuild -C/linux -j16 ARCH=arm64
CROSS_COMPILE=aarch64-linux-gnu- HOSTCC=gcc CC="sccache
aarch64-linux-gnu-gcc" O=build Image
#
In file included from ../arch/arm64/include/asm/smp.h:33,
                 from ../include/linux/smp.h:82,
                 from ../include/linux/percpu.h:7,
                 from ../include/linux/random.h:14,
                 from ../arch/arm64/kernel/pointer_auth.c:5:
../arch/arm64/include/asm/pointer_auth.h: In function ‘ptrauth_keys_init_user’:
../arch/arm64/include/asm/pointer_auth.h:40:3: error: implicit
declaration of function ‘get_random_bytes’; did you mean
‘get_random_once’? [-Werror=implicit-function-declaration]
   40 |   get_random_bytes(&keys->apia, sizeof(keys->apia));
      |   ^~~~~~~~~~~~~~~~
      |   get_random_once
In file included from ../arch/arm64/kernel/pointer_auth.c:5:
../include/linux/random.h: At top level:
../include/linux/random.h:41:13: warning: conflicting types for
‘get_random_bytes’
   41 | extern void get_random_bytes(void *buf, int nbytes);
      |             ^~~~~~~~~~~~~~~~
In file included from ../arch/arm64/include/asm/smp.h:33,
                 from ../include/linux/smp.h:82,
                 from ../include/linux/percpu.h:7,
                 from ../include/linux/random.h:14,
                 from ../arch/arm64/kernel/pointer_auth.c:5:
../arch/arm64/include/asm/pointer_auth.h:40:3: note: previous implicit
declaration of ‘get_random_bytes’ was here
   40 |   get_random_bytes(&keys->apia, sizeof(keys->apia));
      |   ^~~~~~~~~~~~~~~~
cc1: some warnings being treated as errors

build log link,
https://gitlab.com/Linaro/lkft/kernel-runs/-/jobs/662908458

-- 
Linaro LKFT
https://lkft.linaro.org

Re: arm64: pointer_auth.h:40:3: error: implicit declaration of function ‘get_random_bytes’; did you mean ‘get_random_once’?

[Stephen Rothwell]

[-- Attachment #1: Type: text/plain, Size: 337 bytes --]

Hi Naresh,

On Thu, 30 Jul 2020 16:25:49 +0530 Naresh Kamboju <naresh.kamboju@linaro.org> wrote:
>
> Linus master branch failed to build the arm64 architecture due to
> below error and warnings.

Thanks for the report.

Please see https://lore.kernel.org/lkml/202007292007.D87DBD34B@keescook/

-- 
Cheers,
Stephen Rothwell

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

Re: arm64: pointer_auth.h:40:3: error: implicit declaration of function ‘get_random_bytes’; did you mean ‘get_random_once’?

[Grygorii Strashko]

On 30/07/2020 14:15, Stephen Rothwell wrote:
> Hi Naresh,
> 
> On Thu, 30 Jul 2020 16:25:49 +0530 Naresh Kamboju <naresh.kamboju@linaro.org> wrote:
>>
>> Linus master branch failed to build the arm64 architecture due to
>> below error and warnings.
> 
> Thanks for the report.
> 
> Please see https://lore.kernel.org/lkml/202007292007.D87DBD34B@keescook/

I'd like to note that below set of patches:
1c9df907da83 random: fix circular include dependency on arm64 after addition of percpu.h
83bdc7275e62 random32: remove net_rand_state from the latent entropy gcc plugin
f227e3ec3b5c random32: update the net random state on interrupt and activity

also break ARM build for the case:
#if defined(CONFIG_SMP) && !defined(CONFIG_CPU_V6)

config: keystone_defconfig

   CC      arch/arm/kernel/signal.o
In file included from ../include/linux/random.h:14,
                  from ../arch/arm/kernel/signal.c:8:
../arch/arm/include/asm/percpu.h: In function ‘__my_cpu_offset’:
../arch/arm/include/asm/percpu.h:29:34: error: ‘current_stack_pointer’ undeclared (first use in this function); did you mean ‘user_stack_pointer’?
    : "Q" (*(const unsigned long *)current_stack_pointer));
                                   ^~~~~~~~~~~~~~~~~~~~~
                                   user_stack_pointer
../arch/arm/include/asm/percpu.h:29:34: note: each undeclared identifier is reported only once for each function it appears in
../scripts/Makefile.build:280: recipe for target 'arch/arm/kernel/signal.o' failed
make[2]: *** [arch/arm/kernel/signal.o] Error 1
/home/grygorii/kernel.org/linux-master/linux/Makefile:1756: recipe for target 'arch/arm/kernel' failed
make[1]: *** [arch/arm/kernel] Error 2



-- 
Best regards,
grygorii

Re: arm64: pointer_auth.h:40:3: error: implicit declaration of function ‘get_random_bytes’; did you mean ‘get_random_once’?

[Linus Torvalds]

On Thu, Jul 30, 2020 at 11:47 AM Grygorii Strashko
<grygorii.strashko@ti.com> wrote:
>
>
> I'd like to note that below set of patches:
> 1c9df907da83 random: fix circular include dependency on arm64 after addition of percpu.h
> 83bdc7275e62 random32: remove net_rand_state from the latent entropy gcc plugin
> f227e3ec3b5c random32: update the net random state on interrupt and activity
>
> also break ARM build for the case:
> #if defined(CONFIG_SMP) && !defined(CONFIG_CPU_V6)

Lovely.

I think I'll have to revert 1c9df907da83 after all, and  use Marc
Zyngier's patch to just remove that troublesome <asm/pointer_auth.h>
include in arm64 instead.

This turned out to be much messier than it should have been.

            Linus

Re: arm64: pointer_auth.h:40:3: error: implicit declaration of function ‘get_random_bytes’; did you mean ‘get_random_once’?

[Grygorii Strashko]

On 30/07/2020 21:56, Linus Torvalds wrote:
> On Thu, Jul 30, 2020 at 11:47 AM Grygorii Strashko
> <grygorii.strashko@ti.com> wrote:
>>
>>
>> I'd like to note that below set of patches:
>> 1c9df907da83 random: fix circular include dependency on arm64 after addition of percpu.h
>> 83bdc7275e62 random32: remove net_rand_state from the latent entropy gcc plugin
>> f227e3ec3b5c random32: update the net random state on interrupt and activity
>>
>> also break ARM build for the case:
>> #if defined(CONFIG_SMP) && !defined(CONFIG_CPU_V6)
> 
> Lovely.
> 
> I think I'll have to revert 1c9df907da83 after all, and  use Marc
> Zyngier's patch to just remove that troublesome <asm/pointer_auth.h>
> include in arm64 instead.
> 
> This turned out to be much messier than it should have been.

FYI. Below diff seems fixes build:

diff --git a/arch/arm/include/asm/percpu.h b/arch/arm/include/asm/percpu.h
index f44f448537f2..1a3eedbac4a2 100644
--- a/arch/arm/include/asm/percpu.h
+++ b/arch/arm/include/asm/percpu.h
@@ -5,6 +5,8 @@
  #ifndef _ASM_ARM_PERCPU_H_
  #define _ASM_ARM_PERCPU_H_
  
+#include <asm/thread_info.h>
+
  /*
   * Same as asm-generic/percpu.h, except that we store the per cpu offset
   * in the TPIDRPRW. TPIDRPRW only exists on V6K and V7


-- 
Best regards,
grygorii

Re: arm64: pointer_auth.h:40:3: error: implicit declaration of function ‘get_random_bytes’; did you mean ‘get_random_once’?

[Linus Torvalds]

On Thu, Jul 30, 2020 at 12:05 PM Grygorii Strashko
<grygorii.strashko@ti.com> wrote:
>
> FYI. Below diff seems fixes build:

Thanks, I'll do that instead, it looks like the right thing to do regardless.

Mind sending me a sign-off for it (and commit message would be lovely
too, but I can make something up)? Yeah, it's just a one-liner, but
let's do this right.

Thanks,
            Linus

Re: arm64: pointer_auth.h:40:3: error: implicit declaration of function ‘get_random_bytes’; did you mean ‘get_random_once’?

[Grygorii Strashko]

On 30/07/2020 22:40, Linus Torvalds wrote:
> On Thu, Jul 30, 2020 at 12:05 PM Grygorii Strashko
> <grygorii.strashko@ti.com> wrote:
>>
>> FYI. Below diff seems fixes build:
> 
> Thanks, I'll do that instead, it looks like the right thing to do regardless.
> 
> Mind sending me a sign-off for it (and commit message would be lovely
> too, but I can make something up)? Yeah, it's just a one-liner, but
> let's do this right.

I've also boot-tested it on Keystone 2 platforms.

Best regards,
grygorii

---
 From fa1238a8e4b7c603fcc0fb5f6169c212035aef4f Mon Sep 17 00:00:00 2001
From: Grygorii Strashko <grygorii.strashko@ti.com>
Date: Thu, 30 Jul 2020 22:05:01 +0300
Subject: [PATCH] ARM: percpu.h: fix build error
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Fix build error for the case:
  defined(CONFIG_SMP) && !defined(CONFIG_CPU_V6)

config: keystone_defconfig

CC      arch/arm/kernel/signal.o
In file included from ../include/linux/random.h:14,
                  from ../arch/arm/kernel/signal.c:8:
../arch/arm/include/asm/percpu.h: In function ‘__my_cpu_offset’:
../arch/arm/include/asm/percpu.h:29:34: error: ‘current_stack_pointer’ undeclared (first use in this function); did you mean ‘user_stack_pointer’?
    : "Q" (*(const unsigned long *)current_stack_pointer));
                                   ^~~~~~~~~~~~~~~~~~~~~
                                   user_stack_pointer

Fixes: f227e3ec3b5c ("random32: update the net random state on interrupt and activity")
Signed-off-by: Grygorii Strashko <grygorii.strashko@ti.com>
---
  arch/arm/include/asm/percpu.h | 2 ++
  1 file changed, 2 insertions(+)

diff --git a/arch/arm/include/asm/percpu.h b/arch/arm/include/asm/percpu.h
index f44f448537f2..1a3eedbac4a2 100644
--- a/arch/arm/include/asm/percpu.h
+++ b/arch/arm/include/asm/percpu.h
@@ -5,6 +5,8 @@
  #ifndef _ASM_ARM_PERCPU_H_
  #define _ASM_ARM_PERCPU_H_
  
+#include <asm/thread_info.h>
+
  /*
   * Same as asm-generic/percpu.h, except that we store the per cpu offset
   * in the TPIDRPRW. TPIDRPRW only exists on V6K and V7
-- 
2.17.1

Re: arm64: pointer_auth.h:40:3: error: implicit declaration of function 'get_random_bytes'; did you mean 'get_random_once'?

[Willy Tarreau]

On Thu, Jul 30, 2020 at 12:40:01PM -0700, Linus Torvalds wrote:
> On Thu, Jul 30, 2020 at 12:05 PM Grygorii Strashko
> <grygorii.strashko@ti.com> wrote:
> >
> > FYI. Below diff seems fixes build:
> 
> Thanks, I'll do that instead, it looks like the right thing to do regardless.
> 
> Mind sending me a sign-off for it (and commit message would be lovely
> too, but I can make something up)? Yeah, it's just a one-liner, but
> let's do this right.

Don't you want to take Mark's patch anyway in addition to all this ? In
case anyone meets yet another build issue, they'd have more luck trying
to revert any other patch. Right now if they revert one, it breaks the
build in a different way and doesn't help much.

Willy

Re: arm64: pointer_auth.h:40:3: error: implicit declaration of function 'get_random_bytes'; did you mean 'get_random_once'?

[Linus Torvalds]

On Thu, Jul 30, 2020 at 7:16 PM Willy Tarreau <w@1wt.eu> wrote:
>
> Don't you want to take Mark's patch anyway in addition to all this ? In
> case anyone meets yet another build issue, they'd have more luck trying
> to revert any other patch. Right now if they revert one, it breaks the
> build in a different way and doesn't help much.

I think Will picked up Mark's patch into the arm64 tree, so I felt I
should let it come in eventually that way.

The immediate problems seem to have been resolved.

Knock wood.

             Linus

Re: arm64: pointer_auth.h:40:3: error: implicit declaration of function 'get_random_bytes'; did you mean 'get_random_once'?

[Willy Tarreau]

On Thu, Jul 30, 2020 at 10:25:41PM -0700, Linus Torvalds wrote:
> On Thu, Jul 30, 2020 at 7:16 PM Willy Tarreau <w@1wt.eu> wrote:
> >
> > Don't you want to take Mark's patch anyway in addition to all this ? In
> > case anyone meets yet another build issue, they'd have more luck trying
> > to revert any other patch. Right now if they revert one, it breaks the
> > build in a different way and doesn't help much.
> 
> I think Will picked up Mark's patch into the arm64 tree, so I felt I
> should let it come in eventually that way.

OK then!

> The immediate problems seem to have been resolved.
> 
> Knock wood.

Yes, that's my feeling as well. I'm going to resend you your patch
moving prandom32 definitions to prandom.h rebased on these latest
changes, add add a commit message. It would be nice for 5.9, but
maybe it could put an end to this mess once for all :-/

Willy

Re: arm64: pointer_auth.h:40:3: error: implicit declaration of function 'get_random_bytes'; did you mean 'get_random_once'?

[Will Deacon]

On Thu, Jul 30, 2020 at 10:25:41PM -0700, Linus Torvalds wrote:
> On Thu, Jul 30, 2020 at 7:16 PM Willy Tarreau <w@1wt.eu> wrote:
> >
> > Don't you want to take Mark's patch anyway in addition to all this ? In
> > case anyone meets yet another build issue, they'd have more luck trying
> > to revert any other patch. Right now if they revert one, it breaks the
> > build in a different way and doesn't help much.
> 
> I think Will picked up Mark's patch into the arm64 tree, so I felt I
> should let it come in eventually that way.

Yup, you'll get it later today with a few other fixes that have trickled
in.

Cheers,

Will