David Laight 於 2022年6月1日 週三 下午4:25寫道：
>
> From: Yu-Jen Chang
> > Sent: 01 June 2022 06:59
> >
> > David Laight 於 2022年5月30日 週一 下午4:10寫道：
> > >
> > > From: Yu-Jen Chang
> > > > Sent: 28 May 2022 09:13
> > > >
> > > > The original assembly version of memchr() is implemented with
> > > > the byte-wi
From: Yu-Jen Chang
> Sent: 01 June 2022 06:59
>
> David Laight 於 2022年5月30日 週一 下午4:10寫道：
> >
> > From: Yu-Jen Chang
> > > Sent: 28 May 2022 09:13
> > >
> > > The original assembly version of memchr() is implemented with
> > > the byte-wise comparing technique, which does not fully
> > > use 64-bi
David Laight 於 2022年5月30日 週一 下午4:10寫道：
>
> From: Yu-Jen Chang
> > Sent: 28 May 2022 09:13
> >
> > The original assembly version of memchr() is implemented with
> > the byte-wise comparing technique, which does not fully
> > use 64-bits registers in x86_64 CPU. We use word-wide
> > comparing so tha
From: Yu-Jen Chang
> Sent: 28 May 2022 09:13
>
> The original assembly version of memchr() is implemented with
> the byte-wise comparing technique, which does not fully
> use 64-bits registers in x86_64 CPU. We use word-wide
> comparing so that 8 characters can be compared at the same time
> on x8
On Sun, May 29, 2022 at 12:40 AM Tao Zhou wrote:
>
> On Sat, May 28, 2022 at 04:12:35PM +0800, Yu-Jen Chang wrote:
>
> > The original assembly version of memchr() is implemented with
> > the byte-wise comparing technique, which does not fully
> > use 64-bits registers in x86_64 CPU. We use word-wi
On Sat, May 28, 2022 at 04:12:35PM +0800, Yu-Jen Chang wrote:
> The original assembly version of memchr() is implemented with
> the byte-wise comparing technique, which does not fully
> use 64-bits registers in x86_64 CPU. We use word-wide
> comparing so that 8 characters can be compared at the sa
