Re: [QUESTION] Usage of '0b' as a prefix for numerical constants?

[Eric Blake]

On 11/22/19 8:56 AM, Aleksandar Markovic wrote:
> Hello, all.
>
> I am currently reviewing some code, and I see it uses '0b' as a prefix
> of numerical constants, similar to these examples:
>
> switch (((inst >> 22) & 0b1111111000) | ((inst >> 12) & 0b0000000111)) {
>
> or
>
> ARRAY_FIELD_DP32(s->regs, CRB_INTF_ID, RID, 0b0000);

Binary constants introduced by 0b are a gcc'ism, copied by clang, and 
thus usable in qemu if we want to (similar to our use of the ?: 
operator, the {} initializer, the ranged case 0 ... 7, 
__attribute__((cleanup)), ...).  But it is not standard C.

> I remember a while ago, something stopped me from using '0b' as a
> prefix in my own code (was it checkpatch.pl, or perhaps some statement
> on coding style, or a compiler, or something else - I don't really
> remember), so I didn't use it, and used '0x' (hexadecimal constant).
>
> What is really the view of the community on usage of '0b'?

For small constants, 0b111 is just about as readable as 0x7.  But for 
large constants, I much prefer 0x7f over 0b1111111.

> Please C language standard and compiler experts, and also regular
> participants like me, speak up.

If you want to provide a patch for coding standards (either admitting 
that yes we use the extension and here are some guidelines on using it, 
or declaring no new uses of it and maybe patching existing uses to 
switch to hex constants), then go for it.  Maybe wait for more opinions 
to come in to see which color more of the developers prefer for their 
bikeshed.

--
Eric Blake, Principal Software Engineer
Red Hat, Inc.           +1-919-301-3226
Virtualization:  qemu.org | libvirt.org