[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <a3ed9194-ff36-19ff-8cdd-2cfba12c17a9@loongson.cn>
Date: Wed, 7 Dec 2022 17:13:54 +0800
From: Tiezhu Yang <yangtiezhu@...ngson.cn>
To: Huacai Chen <chenhuacai@...nel.org>
Cc: WANG Xuerui <kernel@...0n.name>,
Masami Hiramatsu <mhiramat@...nel.org>,
loongarch@...ts.linux.dev, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v7 1/4] LoongArch: Simulate branch and PC instructions
On 12/07/2022 11:06 AM, Huacai Chen wrote:
> Hi, Tiezhu,
>
> On Fri, Dec 2, 2022 at 9:08 PM Tiezhu Yang <yangtiezhu@...ngson.cn> wrote:
>>
>> According to LoongArch Reference Manual, simulate branch and
>> PC instructions, this is preparation for later patch.
>>
...
>> +static inline unsigned long sign_extended(unsigned long val, unsigned int idx)
>> +{
>> + if (val & (1UL << idx))
>> + return ~((1UL << (idx + 1)) - 1) | val;
>> + else
>> + return ((1UL << (idx + 1)) - 1) & val;
>> +}
> You can use existing __SIGNEX and its friends rather than reinvent them.
Thanks for your reminder.
In my opinion, this static inline function sign_extended()
is much more clear and readable than the macro __SIGNEX()
defined in alternative.c, the helper function bs_dest_*()
seems redundant too, use "pc + sign_extended()" is a more
straightforward way to simulate instruction according to
the ISA Manual, so here I prefer to keep it as is.
Additionally, we can use sign_extended() instead of __SIGNEX()
in alternative.c, the __SIGNEX() related code can be removed
in a seperate patch in some day.
Thanks,
Tiezhu
Powered by blists - more mailing lists