Re: include/linux/compiler_types.h:315:38: error: call to '__compiletime_assert_511' declared with attribute error: BUILD_BUG_ON failed: offsetof(struct can_frame, len) != offsetof(struct canfd_frame, len) || offsetof(struct can_frame, data) != offsetof(struc...

[Oliver Hartkopp <socketcan@hartkopp.net>]

On 19.03.21 09:06, kernel test robot wrote:
> Hi Oliver,
> 
> FYI, the error/warning still remains.
> 

Hm - I have no clue either.

> tree:   https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
> head:   8b12a62a4e3ed4ae99c715034f557eb391d6b196
> commit: c7b74967799b1af52b3045d69d4c26836b2d41de can: replace can_dlc as variable/element for payload length

The patch which introduced the union I suspected to be the problem is 
some commits earlier ...

>     net/can/af_can.c:891:2: note: in expansion of macro 'BUILD_BUG_ON'
>       891 |  BUILD_BUG_ON(offsetof(struct can_frame, len) !=
>           |  ^~~~~~~~~~~~

The only idea which does not change the functionality but may help the 
macro expansion is to revert the change from "can_dlc" -> "len" in 
af_can.c :

https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/diff/net/can/af_can.c?id=c7b74967799b1af52b3045d69d4c26836b2d41de

Is it possible for you to revert that single line for a test?

diff --git a/net/can/af_can.c b/net/can/af_can.c
index cce2af10eb3e..1c95ede2c9a6 100644
--- a/net/can/af_can.c
+++ b/net/can/af_can.c
@@ -865,11 +865,11 @@ static struct pernet_operations can_pernet_ops 
__read_mostly = {
  static __init int can_init(void)
  {
         int err;

         /* check for correct padding to be able to use the structs 
similarly */
-       BUILD_BUG_ON(offsetof(struct can_frame, len) !=
+       BUILD_BUG_ON(offsetof(struct can_frame, can_dlc) !=
                      offsetof(struct canfd_frame, len) ||
                      offsetof(struct can_frame, data) !=
                      offsetof(struct canfd_frame, data));

         pr_info("can: controller area network core\n");

Unfortunately I was not able to reproduce the issue here.

Best regards,
Oliver



> 
> 
> vim +/__compiletime_assert_511 +315 include/linux/compiler_types.h
> 
> eb5c2d4b45e3d2 Will Deacon 2020-07-21  301
> eb5c2d4b45e3d2 Will Deacon 2020-07-21  302  #define _compiletime_assert(condition, msg, prefix, suffix) \
> eb5c2d4b45e3d2 Will Deacon 2020-07-21  303  	__compiletime_assert(condition, msg, prefix, suffix)
> eb5c2d4b45e3d2 Will Deacon 2020-07-21  304
> eb5c2d4b45e3d2 Will Deacon 2020-07-21  305  /**
> eb5c2d4b45e3d2 Will Deacon 2020-07-21  306   * compiletime_assert - break build and emit msg if condition is false
> eb5c2d4b45e3d2 Will Deacon 2020-07-21  307   * @condition: a compile-time constant condition to check
> eb5c2d4b45e3d2 Will Deacon 2020-07-21  308   * @msg:       a message to emit if condition is false
> eb5c2d4b45e3d2 Will Deacon 2020-07-21  309   *
> eb5c2d4b45e3d2 Will Deacon 2020-07-21  310   * In tradition of POSIX assert, this macro will break the build if the
> eb5c2d4b45e3d2 Will Deacon 2020-07-21  311   * supplied condition is *false*, emitting the supplied error message if the
> eb5c2d4b45e3d2 Will Deacon 2020-07-21  312   * compiler has support to do so.
> eb5c2d4b45e3d2 Will Deacon 2020-07-21  313   */
> eb5c2d4b45e3d2 Will Deacon 2020-07-21  314  #define compiletime_assert(condition, msg) \
> eb5c2d4b45e3d2 Will Deacon 2020-07-21 @315  	_compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
> eb5c2d4b45e3d2 Will Deacon 2020-07-21  316
> 
> :::::: The code at line 315 was first introduced by commit
> :::::: eb5c2d4b45e3d2d5d052ea6b8f1463976b1020d5 compiler.h: Move compiletime_assert() macros into compiler_types.h
> 
> :::::: TO: Will Deacon <will@kernel.org>
> :::::: CC: Will Deacon <will@kernel.org>
> 
> ---
> 0-DAY CI Kernel Test Service, Intel Corporation
> https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org
>