qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Qemu-devel] [PATCH 6/7] cutils: Normalize qemu_strto[u]ll() signatu


From: Andreas Färber
Subject: Re: [Qemu-devel] [PATCH 6/7] cutils: Normalize qemu_strto[u]ll() signature
Date: Fri, 25 Sep 2015 15:27:00 +0200
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101 Thunderbird/38.2.0

Am 25.09.2015 um 14:56 schrieb Paolo Bonzini:
> On 25/09/2015 14:44, Andreas Färber wrote:
>>> Do we actually use long long and unsigned long long anywhere?
>>
>> The next patch does, because checkpatch asks for it. :)
>>
>> qemu_strtoull() has some special handling for Windows apparently.
> 
> No, I really mean the types. :) The qemu_strtoll/qemu_strtoull functions
> use {,u}int64_t because they are much more used than long long and
> unsigned long long.

Well, my answer still stands: The next patch has code using long long.

Problem is that uint64_t foo = strtoull(...) works, while
qemu_strtoull(..., &foo) causes a pointer mismatch warning treated as
error. I could've converted those to uint64_t (assuming the type is not
needed for something else), but I rather wanted to keep changes small.

If we want functions using [u]int64_t, we should name them
...strto[u]64, not mixing C and POSIX types.

But I assumed there may be some controversy, so I intentionally put this
after the actual bug fixes and test cases, they can easily be dropped. :)

Cheers,
Andreas

-- 
SUSE Linux GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany
GF: Felix Imendörffer, Jane Smithard, Graham Norton; HRB 21284 (AG Nürnberg)



reply via email to

[Prev in Thread] Current Thread [Next in Thread]