13482 commits

This branch

This branch

All branches

Author	SHA1	Message	Date
Kovid Goyal	c19488f3be	Graphics protocol: Add a new delete mode for deleting images whose ids fall within a range ... Useful for bulk deletion. See #7080	2024-02-25 09:57:44 +05:30
Kovid Goyal	ad3ab877f8	Use a fast SIMD implementation to XOR data going into the disk cache	2024-02-25 09:57:43 +05:30
Kovid Goyal	88f3c8c5ee	Reduce max key size in disk cache ... We used only 12 byte keys no need to have a max key size more than 16	2024-02-25 09:57:43 +05:30
Kovid Goyal	91013a4e05	Faster image cache key generation	2024-02-25 09:57:43 +05:30
Kovid Goyal	de92470f0d	Improve performance of disk cache when there are thousands of small images ... Fixes #7080	2024-02-25 09:57:43 +05:30
Kovid Goyal	1c62b0f1ec	...	2024-02-25 09:57:43 +05:30
Kovid Goyal	b52af64ffe	Hide cursor during benchmark run	2024-02-25 09:57:43 +05:30
Kovid Goyal	1db7ac5f6b	Use our new shift by n functions to improve function to zero last N bytes ... Benchmark neutral but cleaner code using one less vector register and equal number of operations.	2024-02-25 09:57:43 +05:30
Kovid Goyal	e77a970ca1	Also implement arbitrary byte shift for 128 bit registers	2024-02-25 09:57:43 +05:30
Kovid Goyal	ac984d05f2	Fix gcc detection	2024-02-25 09:57:43 +05:30
Kovid Goyal	f16c2a0d67	Move checking for compiler brand into Env	2024-02-25 09:57:43 +05:30
Kovid Goyal	29a574a4bc	Prevent duplicate VZEROUPPER instructions	2024-02-25 09:57:43 +05:30
Kovid Goyal	a7c06b38e6	We dont actually need vzeroupper at start of function ... GCC emits vzeroupper automatically when compiling with native optimizations but we still need it otherwise	2024-02-25 09:57:43 +05:30
Kovid Goyal	16d36c46fe	Update to using math/rand/v2	2024-02-25 09:57:43 +05:30
Kovid Goyal	720618bc37	Use go 1.22 for building ... It supports PCALIGN on non ARM arches as well	2024-02-25 09:57:43 +05:30
Kovid Goyal	2f727e6561	...	2024-02-25 09:57:43 +05:30
Kovid Goyal	b65a5f78fd	Fix regression causing shells in darwin to not run in login mode	2024-02-25 09:57:43 +05:30
Kovid Goyal	0a1eb038a5	Implement functions for arbitrary byte shifts in vector registers	2024-02-25 09:57:42 +05:30
Kovid Goyal	e541c0534d	...	2024-02-25 09:57:42 +05:30
Kovid Goyal	eb1e3b33b4	Fix test failure on some systems ... Broken ass compilers strike again	2024-02-25 09:57:42 +05:30
Kovid Goyal	41ec46d5bb	...	2024-02-25 09:57:42 +05:30
Kovid Goyal	b021e9b648	Do the default func test last so we can see what the failure is on more explicitly	2024-02-25 09:57:42 +05:30
Kovid Goyal	ede4d7fbca	...	2024-02-25 09:57:42 +05:30
Kovid Goyal	d0797a025b	Add dedicated tests for find_either_of_two	2024-02-25 09:57:42 +05:30
Kovid Goyal	1acd223f45	...	2024-02-25 09:57:42 +05:30
Kovid Goyal	8639a2ac40	Update perf figures based on latest code	2024-02-25 09:57:42 +05:30
Kovid Goyal	f48e4ffd5e	Port aligned load based find algorithm to C	2024-02-25 09:57:42 +05:30
Kovid Goyal	c01b959723	Fix Go unaligned index implementation	2024-02-25 09:57:42 +05:30
Kovid Goyal	36773c09d3	Functions to get bytes to first match ignoring leading bytes	2024-02-25 09:57:42 +05:30
Kovid Goyal	687340003d	...	2024-02-25 09:57:42 +05:30
Kovid Goyal	7467307200	Add some alignment tests	2024-02-25 09:57:42 +05:30
Kovid Goyal	bbdb0b15f3	DRYer	2024-02-25 09:57:42 +05:30
Kovid Goyal	b5edd9ad57	Dont precalculate mask in loop body ... No need since we dont shift. Avoids the extra mask instructions for the not found case.	2024-02-25 09:57:42 +05:30
Kovid Goyal	a32e1aafa6	...	2024-02-25 09:57:41 +05:30
Kovid Goyal	f9fd6ffd46	Use only aligned loads for index funcs ... Also obviates the necessity for safe slice wrappers	2024-02-25 09:57:41 +05:30
Kovid Goyal	31a5fcf297	DRYer	2024-02-25 09:57:41 +05:30
Kovid Goyal	493fc900e9	Fix build on ARM	2024-02-25 09:57:41 +05:30
Kovid Goyal	3abdc54e4b	...	2024-02-25 09:57:41 +05:30
Kovid Goyal	618aeec709	Finally got gnome-terminal to run on my system ... Apparently it needed some kind of GTK desktop portal or the other 🙄 Interesting that its numbers are basically the same as alacritty's. Lot better than I remember, I guess the recent libvte performance work was good.	2024-02-25 09:57:41 +05:30
Kovid Goyal	4585361161	Micro optimization	2024-02-25 09:57:41 +05:30
Kovid Goyal	f64739c29b	Fix regression that broke handling of single byte control chars when cursor is on second cell of wide character	2024-02-25 09:57:41 +05:30
Kovid Goyal	f3830aa854	Avoid unnecessary if	2024-02-25 09:57:41 +05:30
Kovid Goyal	f1fe0bf40a	Code to easily compare SIMD and scalar decode in a live instance ... Also remove -mtune=intel as it fails with clang	2024-02-25 09:57:41 +05:30
Kovid Goyal	561712090d	Fix cmplt implementation	2024-02-25 09:57:41 +05:30
Kovid Goyal	d5f34c401d	Better vector registers to pre-calculate before the loop	2024-02-25 09:57:41 +05:30
Kovid Goyal	d9190ea675	DRYer	2024-02-25 09:57:41 +05:30
Kovid Goyal	57f4ea4d4a	Add some tests for broadcast from constant intrinsic	2024-02-25 09:57:41 +05:30
Kovid Goyal	9b0ae8d403	Dont use VEX encoded instructions for 128 bit ISA	2024-02-25 09:57:41 +05:30
Kovid Goyal	aed0611fb8	Avoid double trailing RET	2024-02-25 09:57:40 +05:30
Kovid Goyal	920b8a2496	Use VZEROUPPER in avx functions ... See https://www.intel.com/content/dam/develop/external/us/en/documents/11mc12-avoiding-2bavx-sse-2btransition-2bpenalties-2brh-2bfinal-809104.pdf	2024-02-25 09:57:40 +05:30