[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH] qemu-timer: remove unnecessary code
From: |
Paolo Bonzini |
Subject: |
Re: [Qemu-devel] [PATCH] qemu-timer: remove unnecessary code |
Date: |
Wed, 13 Jul 2016 13:40:31 +0200 |
User-agent: |
Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101 Thunderbird/45.1.1 |
On 13/07/2016 13:40, Cao jin wrote:
>
>
> On 07/13/2016 06:21 PM, Paolo Bonzini wrote:
>>
>>
>> On 13/07/2016 11:06, Cao jin wrote:
>>> When passed argument 'ns' is 0, macro DIV_ROUND_UP will return 0 also.
>>
>> It's potentially slower though.
>>
>
> Is it because the function in the i/o loop path, so the potentially
> extra arithmetical instructions matters?
It is quite common for ns to be zero, for example if a bottom half has
to be invoked.
However, qemu_timeout_ns_to_ms is not used in the really important path
(which is aio_poll in aio-posix.c) so I guess your patch is okay.
Thanks,
Paolo
>> Paolo
>>
>>> Cc: Paolo Bonzini <address@hidden>
>>> Signed-off-by: Cao jin <address@hidden>
>>> ---
>>> qemu-timer.c | 6 +-----
>>> 1 file changed, 1 insertion(+), 5 deletions(-)
>>>
>>> diff --git a/qemu-timer.c b/qemu-timer.c
>>> index eb22e92..cfe0893 100644
>>> --- a/qemu-timer.c
>>> +++ b/qemu-timer.c
>>> @@ -285,12 +285,8 @@ int qemu_timeout_ns_to_ms(int64_t ns)
>>> return -1;
>>> }
>>>
>>> - if (!ns) {
>>> - return 0;
>>> - }
>>> -
>>> /* Always round up, because it's better to wait too long than
>>> to wait too
>>> - * little and effectively busy-wait
>>> + * short and effectively busy-wait
>>> */
>>> ms = DIV_ROUND_UP(ns, SCALE_MS);
>>>
>>>
>>
>>
>>
>