| From: | cca5507 <cca5507(at)qq(dot)com> |
|---|---|
| To: | David Rowley <dgrowleyml(at)gmail(dot)com> |
| Cc: | pgsql-hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org> |
| Subject: | Re: [Patch] Build the heap more efficient in tuplesort.c |
| Date: | 2025-12-01 16:05:30 |
| Message-ID: | tencent_1D73A1CE850DB5F1294932EA7844071D7608@qq.com |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
Hi,
Generally speaking, build the heap by tuplesort_heap_build() is O(n) while tuplesort_heap_insert() is
O(n log n), so the former is better.
I'm not sure why tuplesort_heap_build() is worse than tuplesort_heap_insert() when handling random
data sometimes, maybe my test method is wrong? Any ideas?
The v2-0001 refactor some code, mark some function as always inline and reduce some useless copy
of SortTuple.
--
Regards,
ChangAo Chen
| Attachment | Content-Type | Size |
|---|---|---|
| v2-0001-Build-the-heap-more-efficient-in-tuplesort.c.patch | application/octet-stream | 8.0 KB |
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Peter Eisentraut | 2025-12-01 16:08:15 | Re: warning: dereferencing type-punned pointer |
| Previous Message | Robert Haas | 2025-12-01 15:58:55 | Re: show size of DSAs and dshash tables in pg_dsm_registry_allocations |