> On Fri, Nov 30, 2018 at 10:31 PM Юрий Соколов wrote:
>
>> Due to lack of response and taking into account this commentary, I'm marking
>> this patch as "Returned with feedback", but hopefully I can pick it up later
>> to
>> improve.
>
> Good luck.
Unexpected for me, but promising reaction - re
пт, 30 нояб. 2018 г., 19:21 Dmitry Dolgov 9erthali...@gmail.com:
> > On Sat, Nov 10, 2018 at 8:37 PM Andres Freund
> wrote:
> >
> > On 2018-11-10 20:18:33 +0100, Dmitry Dolgov wrote:
> > > > On Mon, 2 Jul 2018 at 15:54, Jesper Pedersen <
> jesper.peder...@redhat.com> wrote:
> > > >
> > > > The pa
> On Sat, Nov 10, 2018 at 8:37 PM Andres Freund wrote:
>
> On 2018-11-10 20:18:33 +0100, Dmitry Dolgov wrote:
> > > On Mon, 2 Jul 2018 at 15:54, Jesper Pedersen
> > > wrote:
> > >
> > > The patch from November 27, 2017 still applies (with hunks),
> > >
> > > https://commitfest.postgresql.org/1
On 2018-11-10 20:18:33 +0100, Dmitry Dolgov wrote:
> > On Mon, 2 Jul 2018 at 15:54, Jesper Pedersen
> > wrote:
> >
> > The patch from November 27, 2017 still applies (with hunks),
> >
> > https://commitfest.postgresql.org/18/1166/
> >
> > passes "make check-world" and shows performance improvem
> On Mon, 2 Jul 2018 at 15:54, Jesper Pedersen
> wrote:
>
> The patch from November 27, 2017 still applies (with hunks),
>
> https://commitfest.postgresql.org/18/1166/
>
> passes "make check-world" and shows performance improvements.
>
> Keeping it in "Ready for Committer".
Looks like for some
Hi,
On 03/01/2018 10:50 AM, Jesper Pedersen wrote:
As the patch still applies, make check-world passes and I believe that
Yura has provided feedback for Andres' comments I'll leave this entry
in "Ready for Committer".
The patch from November 27, 2017 still applies (with hunks), passes
"mak
Hi,
On 01/02/2018 11:09 AM, Jesper Pedersen wrote:
Oh... there were stupid error in previos file.
Attached fixed version.
As the patch still applies, make check-world passes and I believe that
Yura has provided feedback for Andres' comments I'll leave this entry in
"Ready for Committer".
Hi,
On 11/27/2017 07:41 AM, Юрий Соколов wrote:
I looked at assembly, and remembered, that last commit simplifies
`init_local_spin_delay` to just two-three writes of zeroes (looks
like compiler combines 2*4byte write into 1*8 write). Compared to
code around (especially in LWLockAcquire itself),
On Tue, Nov 28, 2017 at 4:10 AM, Jesper Pedersen
wrote:
> On 11/27/2017 07:41 AM, Юрий Соколов wrote:
>> Oh... there were stupid error in previos file.
>> Attached fixed version.
>
> I can reconfirm my performance findings with this patch; system same as
> up-thread.
I have moved this patch to ne
Hi Yura,
On 11/27/2017 07:41 AM, Юрий Соколов wrote:
I looked at assembly, and remembered, that last commit simplifies
`init_local_spin_delay` to just two-three writes of zeroes (looks
like compiler combines 2*4byte write into 1*8 write). Compared to
code around (especially in LWLockAcquire itse
2017-11-26 19:51 GMT+03:00 Юрий Соколов :
>
> 2017-11-06 18:07 GMT+03:00 Sokolov Yura :
> >
> > On 2017-10-20 11:54, Sokolov Yura wrote:
> >>
> >> Hello,
> >>
> >> On 2017-10-19 19:46, Andres Freund wrote:
> >>>
> >>> On 2017-10-19 14:36:56 +0300, Sokolov Yura wrote:
>
> > > + init_loca
2017-11-06 18:07 GMT+03:00 Sokolov Yura :
>
> On 2017-10-20 11:54, Sokolov Yura wrote:
>>
>> Hello,
>>
>> On 2017-10-19 19:46, Andres Freund wrote:
>>>
>>> On 2017-10-19 14:36:56 +0300, Sokolov Yura wrote:
> > + init_local_spin_delay(&delayStatus);
>
> The way you moved this a
12 matches
Mail list logo
