linux-kernel - Re: arch/x86/include/asm/preempt.h:75:9: sparse: context imbalance in 'vcpu_enter_guest'

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <ed5c8e6c-b9e7-b816-ad55-21a47c556926@redhat.com>
Date:   Tue, 9 May 2017 17:29:04 +0200
From:   Paolo Bonzini <pbonzini@...hat.com>
To:     kbuild test robot <fengguang.wu@...el.com>
Cc:     kbuild-all@...org, linux-kernel@...r.kernel.org,
        David Matlack <dmatlack@...gle.com>,
        Bandan Das <bsd@...hat.com>
Subject: Re: arch/x86/include/asm/preempt.h:75:9: sparse: context imbalance in
 'vcpu_enter_guest' - unexpected unlock



On 09/05/2017 14:32, kbuild test robot wrote:
>    ./arch/x86/kvm/trace.h:1153:1: sparse: odd constant _Bool cast (ffffffffffffffff becomes 1)
>    ./arch/x86/kvm/trace.h:1322:1: sparse: odd constant _Bool cast (ffffffffffffffff becomes 1)
>    ./arch/x86/kvm/trace.h:1322:1: sparse: odd constant _Bool cast (ffffffffffffffff becomes 1)
>>> arch/x86/include/asm/preempt.h:75:9: sparse: context imbalance in 'vcpu_enter_guest' - unexpected unlock

I can't really make much sense of this, but vcpu_enter_guest seems okay
to me.

Paolo

> vim +/vcpu_enter_guest +75 arch/x86/include/asm/preempt.h
> 
> c2daa3be Peter Zijlstra    2013-08-14  59  static __always_inline void clear_preempt_need_resched(void)
> c2daa3be Peter Zijlstra    2013-08-14  60  {
> b3ca1c10 Christoph Lameter 2014-04-07  61  	raw_cpu_or_4(__preempt_count, PREEMPT_NEED_RESCHED);
> c2daa3be Peter Zijlstra    2013-08-14  62  }
> c2daa3be Peter Zijlstra    2013-08-14  63  
> c2daa3be Peter Zijlstra    2013-08-14  64  static __always_inline bool test_preempt_need_resched(void)
> c2daa3be Peter Zijlstra    2013-08-14  65  {
> b3ca1c10 Christoph Lameter 2014-04-07  66  	return !(raw_cpu_read_4(__preempt_count) & PREEMPT_NEED_RESCHED);
> c2daa3be Peter Zijlstra    2013-08-14  67  }
> c2daa3be Peter Zijlstra    2013-08-14  68  
> c2daa3be Peter Zijlstra    2013-08-14  69  /*
> c2daa3be Peter Zijlstra    2013-08-14  70   * The various preempt_count add/sub methods
> c2daa3be Peter Zijlstra    2013-08-14  71   */
> c2daa3be Peter Zijlstra    2013-08-14  72  
> c2daa3be Peter Zijlstra    2013-08-14  73  static __always_inline void __preempt_count_add(int val)
> c2daa3be Peter Zijlstra    2013-08-14  74  {
> b3ca1c10 Christoph Lameter 2014-04-07 @75  	raw_cpu_add_4(__preempt_count, val);
> c2daa3be Peter Zijlstra    2013-08-14  76  }
> c2daa3be Peter Zijlstra    2013-08-14  77  
> c2daa3be Peter Zijlstra    2013-08-14  78  static __always_inline void __preempt_count_sub(int val)
> c2daa3be Peter Zijlstra    2013-08-14  79  {
> b3ca1c10 Christoph Lameter 2014-04-07  80  	raw_cpu_add_4(__preempt_count, -val);
> c2daa3be Peter Zijlstra    2013-08-14  81  }
> c2daa3be Peter Zijlstra    2013-08-14  82  
> ba1f14fb Peter Zijlstra    2013-11-28  83  /*
> 
> :::::: The code at line 75 was first introduced by commit
> :::::: b3ca1c10d7b32fdfdfaf5484eda486323f52d9be percpu: add raw_cpu_ops
> 
> :::::: TO: Christoph Lameter <cl@...ux.com>
> :::::: CC: Linus Torvalds <torvalds@...ux-foundation.org>
> 
> ---
> 0-DAY kernel test infrastructure                Open Source Technology Center
> https://lists.01.org/pipermail/kbuild-all                   Intel Corporation
>