Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <6a0cf3a8-2153-1bed-314e-f9ba2bc74400@linux.vnet.ibm.com>
Date:   Thu, 18 Aug 2022 16:16:47 +0530
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:91.0) Gecko/20100101
 Thunderbird/91.5.0
Subject: Re: [PATCH 02/16] powerpc: override __ALIGN() and __ALIGN_STR()
 macros
Content-Language: en-US
To:     Christophe Leroy <christophe.leroy@csgroup.eu>,
        "linuxppc-dev@lists.ozlabs.org" <linuxppc-dev@lists.ozlabs.org>
Cc:     "jpoimboe@redhat.com" <jpoimboe@redhat.com>,
        "peterz@infradead.org" <peterz@infradead.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        "aik@ozlabs.ru" <aik@ozlabs.ru>,
        "mpe@ellerman.id.au" <mpe@ellerman.id.au>,
        "mingo@redhat.com" <mingo@redhat.com>,
        "rostedt@goodmis.org" <rostedt@goodmis.org>,
        "mbenes@suse.cz" <mbenes@suse.cz>,
        "npiggin@gmail.com" <npiggin@gmail.com>,
        "chenzhongjin@huawei.com" <chenzhongjin@huawei.com>,
        "linux-arm-kernel@lists.infradead.org" 
        <linux-arm-kernel@lists.infradead.org>,
        "naveen.n.rao@linux.vnet.ibm.com" <naveen.n.rao@linux.vnet.ibm.com>,
        Sathvika Vasireddy <sv@linux.ibm.com>
References: <20220808114908.240813-1-sv@linux.ibm.com>
 <20220808114908.240813-3-sv@linux.ibm.com>
 <0d2438e6-9aaa-a9c9-9613-b833771c9362@csgroup.eu>
From:   Sathvika Vasireddy <sv@linux.vnet.ibm.com>
In-Reply-To: <0d2438e6-9aaa-a9c9-9613-b833771c9362@csgroup.eu>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk

Hi Christophe,

On 10/08/22 14:01, Christophe Leroy wrote:
>
> Le 08/08/2022 à 13:48, Sathvika Vasireddy a écrit :
>> Since we need an alignment of 4 bytes, override
>> __ALIGN() and __ALIGN_STR() accordingly.
> Why/When do we now need an alignment of 4 bytes ? Please explain the
> reason in the commit message.
Powerpc instructions must be word-aligned. Currently, there is an 
alignment of 16 bytes (by default),
and it is much more than what is required for powerpc (4 bytes).

The default expansion of these macros are:
#define __ALIGN        .align 4,0x90
#define __ALIGN_STR    ".align 4,0x90"

Since Powerpc Linux does not require a 16 byte alignment, override 
_ALIGN() and __ALIGN_STR() macros
to use required 4 byte alignment.

Sure, I'll explain the reason in commit message.
>
>> Signed-off-by: Sathvika Vasireddy <sv@linux.ibm.com>
>> ---
>>    arch/powerpc/include/asm/linkage.h | 4 ++++
>>    1 file changed, 4 insertions(+)
>>
>> diff --git a/arch/powerpc/include/asm/linkage.h b/arch/powerpc/include/asm/linkage.h
>> index b71b9582e754..8df88fe61438 100644
>> --- a/arch/powerpc/include/asm/linkage.h
>> +++ b/arch/powerpc/include/asm/linkage.h
>> @@ -2,8 +2,12 @@
>>    #ifndef _ASM_POWERPC_LINKAGE_H
>>    #define _ASM_POWERPC_LINKAGE_H
>>    
>> +#include <linux/stringify.h>
>>    #include <asm/types.h>
>>    
>> +#define __ALIGN			.align 2
>> +#define __ALIGN_STR		__stringify(__ALIGN)
> Why do you need to use __stringify ?
>
> ARM64 does:
>
> #define __ALIGN		.align 2
> #define __ALIGN_STR	".align 2"
We could use either __stringify() or quotes, both do the same thing.
While arm64 does ".align 2",

x86 does:
#define __ALIGN         .p2align 4, 0x90
#define __ALIGN_STR     __stringify(__ALIGN)
>
>
>
>
>> +
>>    #ifdef CONFIG_PPC64_ELF_ABI_V1
>>    #define cond_syscall(x) \
>>    	asm ("\t.weak " #x "\n\t.set " #x ", sys_ni_syscall\n"		\

Thanks for reviewing!

- Sathvika