Re: [HACKERS] [PATCH] SortSupport for macaddr type

2017-03-31 Thread Robert Haas
On Fri, Mar 31, 2017 at 3:03 PM, Brandur Leach wrote: > Thank you Peter for the assist here and great sleuthing in > the disassembly, and thanks Teodor for committing! > > Neha pointed out (thanks as well) a couple typos upthread > that I hadn't gotten around to fixing.

Re: [HACKERS] [PATCH] SortSupport for macaddr type

2017-03-31 Thread Brandur Leach
Hello, Thank you Peter for the assist here and great sleuthing in the disassembly, and thanks Teodor for committing! Neha pointed out (thanks as well) a couple typos upthread that I hadn't gotten around to fixing. I've attached a new three line patch to sort those out. Brandur On Wed, Mar 29,

Re: [HACKERS] [PATCH] SortSupport for macaddr type

2017-03-29 Thread Teodor Sigaev
Thank you, pushed Excellent! I've attached a new (and hopefully final) version of the patch. -- Teodor Sigaev E-mail: teo...@sigaev.ru WWW: http://www.sigaev.ru/ -- Sent via pgsql-hackers mailing list

Re: [HACKERS] [PATCH] SortSupport for macaddr type

2017-03-19 Thread Peter Geoghegan
On Sun, Mar 19, 2017 at 8:40 AM, Greg Stark wrote: >> Out of idle curiosity, I decided to generate disassembly of both >> macaddr_cmp_internal(), and the patch's abbreviated comparator. The >> former consists of 49 x86-64 instructions at -02 on my machine, >> totaling 135 bytes of

Re: [HACKERS] [PATCH] SortSupport for macaddr type

2017-03-19 Thread Greg Stark
On 18 March 2017 at 22:22, Peter Geoghegan wrote: > > Out of idle curiosity, I decided to generate disassembly of both > macaddr_cmp_internal(), and the patch's abbreviated comparator. The > former consists of 49 x86-64 instructions at -02 on my machine, > totaling 135 bytes of

Re: [HACKERS] [PATCH] SortSupport for macaddr type

2017-03-18 Thread Peter Geoghegan
On Sat, Mar 18, 2017 at 2:54 PM, Peter Geoghegan wrote: > This seems fine to me, especially > because it lets us compare macaddr using simple 3-way unsigned int > comparisons, which isn't otherwise the case. Out of idle curiosity, I decided to generate disassembly of both

Re: [HACKERS] [PATCH] SortSupport for macaddr type

2017-03-18 Thread Peter Geoghegan
On Tue, Mar 14, 2017 at 3:25 AM, Heikki Linnakangas wrote: > Looks straightforward at a quick read-through. I have just a couple of > questions. How much of the performance gain comes from avoiding the > FunctionCallInvoke overhead, by simply having SortSupport with a comparison

Re: [HACKERS] [PATCH] SortSupport for macaddr type

2017-03-14 Thread Heikki Linnakangas
Looks straightforward at a quick read-through. I have just a couple of questions. How much of the performance gain comes from avoiding the FunctionCallInvoke overhead, by simply having SortSupport with a comparison function, and how much comes from the "abbreviation"? Also, is it worth the

Re: [HACKERS] [PATCH] SortSupport for macaddr type

2017-03-02 Thread Neha Khatri
Hi Brandur, Couple of typo corrections required in patch: s/converstion/conversion s/go the heap/go to the heap The performance results you shared are for he Index Creation operation. Are there similar results for the sorting using ORDER BY queries too? Just curious. Regards, Neha

Re: [HACKERS] [PATCH] SortSupport for macaddr type

2017-03-02 Thread Julien Rouhaud
On Tue, Feb 28, 2017 at 08:58:24AM -0800, Brandur Leach wrote: > Hi Julien, > Hello Brandur, > Thanks for the expedient reply, even after I'd dropped the > ball for so long :) :) > Cool! I just re-read my own comment a few days later and I > think that it still mostly makes sense, but

Re: [HACKERS] [PATCH] SortSupport for macaddr type

2017-03-02 Thread Robert Haas
On Tue, Feb 28, 2017 at 10:28 PM, Brandur Leach wrote: > * I've been generating a new OID value with the > `unused_oids` script, but pretty much every time I rebase > I collide with someone else's addition and need to find a > new one. Is it better for me to pick an

Re: [HACKERS] [PATCH] SortSupport for macaddr type

2017-02-28 Thread Brandur Leach
Hi Julien, Thanks for the expedient reply, even after I'd dropped the ball for so long :) > Indeed, I should have checked more examples :/ There > isn't any clear pattern for this, so I guess any one > would be ok. Yeah, agreed. If it's alright with you, I ended up moving the naming back to

Re: [HACKERS] [PATCH] SortSupport for macaddr type

2017-02-25 Thread Julien Rouhaud
On Sun, Feb 05, 2017 at 01:56:41PM -0800, Brandur Leach wrote: Hello Brandur, thanks for the updated patch! > > > * you used macaddr_cmp_internal() function name, for uuid > > the same function is named uuid_internal_cmp(). Using > > the same naming pattern is probably better. > > I was a

Re: [HACKERS] [PATCH] SortSupport for macaddr type

2017-02-07 Thread Brandur Leach
And as a short follow up, I've moved the patch to the current commit fest: https://commitfest.postgresql.org/13/743/ On Sun, Feb 5, 2017 at 1:56 PM, Brandur Leach wrote: > Hi Julien, > > Thank you for taking the time to do this review, and my > apologies for the very

Re: [HACKERS] [PATCH] SortSupport for macaddr type

2017-02-05 Thread Brandur Leach
Hi Julien, Thank you for taking the time to do this review, and my apologies for the very delayed response. I lost track of this work and have only jumped back into it today. Please find attached a new version of the patch with your feedback integrated. I've also rebased the patch against the

Re: [HACKERS] [PATCH] SortSupport for macaddr type

2016-09-28 Thread Robert Haas
On Wed, Sep 14, 2016 at 6:14 AM, Julien Rouhaud wrote: > On 26/08/2016 19:44, Brandur wrote: >> Hello, > Hello, > >> I've attached a patch to add SortSupport for Postgres' macaddr which has the >> effect of improving the performance of sorting operations for the type.

Re: [HACKERS] [PATCH] SortSupport for macaddr type

2016-09-14 Thread Julien Rouhaud
On 26/08/2016 19:44, Brandur wrote: > Hello, > Hello, > I've attached a patch to add SortSupport for Postgres' macaddr which has the > effect of improving the performance of sorting operations for the type. The > strategy that I employ is very similar to that for UUID, which is to create >

Re: [HACKERS] [PATCH] SortSupport for macaddr type

2016-08-26 Thread Brandur
And here's a commitfest link: https://commitfest.postgresql.org/10/743/ Also, a correction to my original message: the unreferenced [1] footnote points back to the thread that included the patch for UUID SortSupport. [1]

[HACKERS] [PATCH] SortSupport for macaddr type

2016-08-26 Thread Brandur
Hello, I've attached a patch to add SortSupport for Postgres' macaddr which has the effect of improving the performance of sorting operations for the type. The strategy that I employ is very similar to that for UUID, which is to create abbreviated keys by packing as many bytes from the MAC