Re: [PATCH v7 2/2] tty: add rpmsg driver

[Jiri Slaby]

On 26. 03. 20, 11:59, Arnaud POULIQUEN wrote:
> 
> 
> On 3/26/20 1:01 AM, Joe Perches wrote:
>> On Wed, 2020-03-25 at 14:31 +0100, Jiri Slaby wrote:
>>> The question was exactly about that: can a compiler optimize it to a
>>> bare number or will strlen call remain there?
>>
>> $ cat str.c
>> #include <string.h>
>>
>> int foo(void)
>> {
>> 	return strlen("abc");
>> }
>>
>> $ gcc -c -O2 str.c
>> $ objdump -d str.o
>> str.o:     file format elf64-x86-64
>>
>>
>> Disassembly of section .text:
>>
>> 0000000000000000 <foo>:
>>    0:	f3 0f 1e fa          	endbr64 
>>    4:	b8 03 00 00 00       	mov    $0x3,%eax
>>    9:	c3                   	retq   
>>
>>
> same result with  arm gcc using  -O1 or -Og:
> 
> str.o:     file format elf32-littlearm
> 
> 
> Disassembly of section .text:
> 
> 00000000 <foo>:
>    0:	e3a00003 	mov	r0, #3
>    4:	e12fff1e 	bx	lr
> 
> So in conclusion replacing sizeof by srlen even if not optimized in -o0, right?

Right, gcc guys just confirmed, that it's constant-folded during parsing
already. I asked them as I tried to dump the tree.original and the
constant was already there.

So we are safe to use strlen, at least for gcc :P. Others should adapt
if they don't follow.

thanks,
-- 
js
suse labs