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: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <af37c513-6232-c35c-33e3-f6d8d82c8175@huawei.com>
Date:   Mon, 7 Sep 2020 09:51:28 +0800
From:   Yang Yingliang <yangyingliang@...wei.com>
To:     Christophe Leroy <christophe.leroy@...roup.eu>,
        <linuxppc-dev@...ts.ozlabs.org>, <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH -next] powerpc/book3s64: fix link error with
 CONFIG_PPC_RADIX_MMU=n


On 2020/9/6 14:50, Christophe Leroy wrote:
>
>
> Le 05/09/2020 à 13:25, Yang Yingliang a écrit :
>> Fix link error when CONFIG_PPC_RADIX_MMU is disabled:
>> powerpc64-linux-gnu-ld: 
>> arch/powerpc/platforms/pseries/lpar.o:(.toc+0x0): undefined reference 
>> to `mmu_pid_bits'
>>
>> Reported-by: Hulk Robot <hulkci@...wei.com>
>> Signed-off-by: Yang Yingliang <yangyingliang@...wei.com>
>> ---
>>   arch/powerpc/mm/book3s64/mmu_context.c | 4 ++++
>
> In your commit log, you are just mentionning 
> arch/powerpc/platforms/pseries/lpar.o, which is right.
>
> You shouldn't need to modify arch/powerpc/mm/book3s64/mmu_context.c at 
> all, see below.
>
>>   arch/powerpc/platforms/pseries/lpar.c  | 2 ++
>>   2 files changed, 6 insertions(+)
>>
>> diff --git a/arch/powerpc/mm/book3s64/mmu_context.c 
>> b/arch/powerpc/mm/book3s64/mmu_context.c
>> index 0ba30b8b935b..a8e292cd88f0 100644
>> --- a/arch/powerpc/mm/book3s64/mmu_context.c
>> +++ b/arch/powerpc/mm/book3s64/mmu_context.c
>> @@ -152,6 +152,7 @@ void hash__setup_new_exec(void)
>>     static int radix__init_new_context(struct mm_struct *mm)
>>   {
>> +#ifdef CONFIG_PPC_RADIX_MMU
>
> This shouldn't be required. radix__init_new_context() is only called 
> when radix_enabled() returns true.
> As it is a static function, when it is not called it gets optimised 
> away, so you will never get an undefined reference to `mmu_pid_bits` 
> there.
powerpc64-linux-gnu-ld: 
arch/powerpc/mm/book3s64/mmu_context.o:(.toc+0x0): undefined reference 
to `mmu_pid_bits'
powerpc64-linux-gnu-ld: 
arch/powerpc/mm/book3s64/mmu_context.o:(.toc+0x8): undefined reference 
to `mmu_base_pid'


mmu_context.c is always compiled, it uses mmu_pid_bits and mmu_base_pid.

>
>>       unsigned long rts_field;
>>       int index, max_id;
>>   @@ -177,6 +178,9 @@ static int radix__init_new_context(struct 
>> mm_struct *mm)
>>       mm->context.hash_context = NULL;
>>         return index;
>> +#else
>> +    return -ENOTSUPP;
>> +#endif
>>   }
>>     int init_new_context(struct task_struct *tsk, struct mm_struct *mm)
>> diff --git a/arch/powerpc/platforms/pseries/lpar.c 
>> b/arch/powerpc/platforms/pseries/lpar.c
>> index baf24eacd268..e454e218dbba 100644
>> --- a/arch/powerpc/platforms/pseries/lpar.c
>> +++ b/arch/powerpc/platforms/pseries/lpar.c
>> @@ -1726,10 +1726,12 @@ void __init hpte_init_pseries(void)
>>     void radix_init_pseries(void)
>>   {
>> +#ifdef CONFIG_PPC_RADIX_MMU
>
> This function is only called from 
> /arch/powerpc/mm/book3s64/radix_pgtable.c which is only built when 
> CONFIG_PPC_RADIX_MMU is selected.
>
> So the entire function should be encloded in the #ifdef.
OK, I will send a v2 later.
>
>>       pr_info("Using radix MMU under hypervisor\n");
>>         pseries_lpar_register_process_table(__pa(process_tb),
>>                           0, PRTB_SIZE_SHIFT - 12);
>> +#endif
>>   }
>>     #ifdef CONFIG_PPC_SMLPAR
>>
>
> Christophe
> .

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ