[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1a06a70d-5dae-93a7-c200-c5b58f1241d5@loongson.cn>
Date: Mon, 23 Oct 2023 17:21:38 +0800
From: Tiezhu Yang <yangtiezhu@...ngson.cn>
To: Xi Ruoyao <xry111@...111.site>,
Josh Poimboeuf <jpoimboe@...nel.org>,
Peter Zijlstra <peterz@...radead.org>,
Huacai Chen <chenhuacai@...nel.org>
Cc: loongarch@...ts.linux.dev, linux-kernel@...r.kernel.org,
loongson-kernel@...ts.loongnix.cn
Subject: Re: [PATCH v4 8/8] LoongArch: Add ORC stack unwinder support
On 10/23/2023 04:43 PM, Xi Ruoyao wrote:
> On Mon, 2023-10-23 at 16:11 +0800, Tiezhu Yang wrote:
>> +ifdef CONFIG_OBJTOOL
>> +KBUILD_CFLAGS += -fno-optimize-sibling-calls -fno-jump-tables -falign-functions=4
>> +endif
>
> Why do we need to regress the code generation so much for objtool?
Add the above options is to avoid generating some intractable
instructions for objtool by now, I hope all of the extra options
can be removed if possible.
For example, without -fno-jump-tables, there exist many objtool
warnings like this:
"sibling call from callable instruction with modified stack frame"
If the enhanced special function about jump table is implemented
in objtool, -fno-jump-tables will be removed in the future.
Thanks,
Tiezhu
Powered by blists - more mailing lists