aboutsummaryrefslogtreecommitdiffstats
Commit message (Expand)AuthorAgeFilesLines
* 047: "the leaf is the commit" (feedback 009) set up before it runs: today's s...eunhhu24 hours5-3/+288
* 045: the app keyed and its range found by a seek, emitted as it says (no algo...eunhhu24 hours4-8/+373
* 029: two framings in the 009 section made plain: there is no disk leaf for a ...eunhhu25 hours1-2/+8
* 029: feedback 009's rule (one byte shape for the same data) applied to both p...eunhhu25 hours1-0/+44
* 045: feedback 009's zero copy measured: ranges read where its tree keeps it i...eunhhu25 hours3-14/+163
* 046, 029: the taken-over store sent a message too (opened() works on a copy: ...eunhhu25 hours7-20/+68
* 029: principle 2 passed its breaking experiment (046), written with what it s...eunhhu25 hours1-1/+37
* 046: 029's principle 2 put to its breaking experiment on one Pi, the tools an...eunhhu25 hours13-0/+2066
* 045, 029: a message end to end, the event's border written by hand too (app.c...eunhhu25 hours3-3/+174
* 029: the facts counted again: two new (N1 whole numbers, N4 unchanged after L...eunhhu26 hours1-7/+11
* 029: UAX #14's whole chain and the facts it needs: the handler 38,623 machine...eunhhu26 hours1-7/+45
* 045: UAX #14's whole app emitted by hand, the chain (app.c, app-way.h: on lin...eunhhu26 hours4-0/+703
* 029: range()'s chain written, and two of its own claims taken back: the execu...eunhhu27 hours1-14/+47
* 045: range() emitted by hand, Pure's meaning kept or dropped where the proof ...eunhhu27 hours3-0/+297
* 045: Pure's answers checked for real (cost.c and where.c compared only the C'...eunhhu27 hours4-24/+37
* 029: principle 1's break by mechanism (045's samples): dispatch 30.6%, lists ...eunhhu27 hours2-1/+52
* 045: where the machine instructions go, by mechanism (where.c samples the add...eunhhu27 hours6-0/+514
* 029: principle 1 broken on UAX #14 (045), written as its own rule asks: (a) 2...eunhhu27 hours3-2/+69
* 045: 029's principle 1 broken on UAX #14 before any emitter, the tools and th...eunhhu27 hours16-0/+607
* 029 second draft from a blank page (the first, five techniques listed, reject...eunhhu28 hours1-189/+95
* 029: spreading fns at load is a reason to try, not a result (an early return ...eunhhu29 hours1-1/+3
* CHECKPOINT: the evidence table rewritten on the app model as it is (aup, Pure...eunhhu29 hours1-0/+24
* 029 (draft, not a 1.0 freeze): past one node, outside feedback 007's five gap...eunhhu29 hours1-0/+194
* 115: the modules replaced are freed (0 bytes a change over 4,000, was 19,284)...eunhhu29 hours3-5/+37
* The modules an app ran are freed (research 115's next step): every app's modu...eunhhu29 hours10-60/+500
* CHECKPOINT: UAX #14 against Vega's C from one run (6,423 against 35.1 ns, 183...eunhhu31 hours1-1/+1
* CHECKPOINT: the gaps to 1.0 as they stand after 027 and 140 (the contract's r...eunhhu31 hours1-1/+41
* 140: determinism across builds at 027, 028 and 140's state: record_all.py's 4...eunhhu31 hours3-0/+986
* A save's arrays for many states asked with || (clang's -Wbitwise-instead-of-l...eunhhu31 hours1-1/+1
* 029's record_all.py: the pure-runs of the checks 027, 028 and 140 added, reco...eunhhu31 hours1-0/+67
* 140: LIST_VALUE's copy reviewed before building (measured by spec: UAX #14 ab...eunhhu31 hours4-0/+95
* A list value of numbers copied whole (140's LIST_VALUE review, no new opcode,...eunhhu31 hours4-3/+105
* 140: aup's frictions written away and the apps measured again (no core change...eunhhu32 hours5-7/+56
* 140: the decision section before 8 and 10 points to their sections; 8's Pytho...eunhhu33 hours1-1/+3
* 140's 8, box layout, written in aup with no new opcode (aup's examples/layout...eunhhu33 hours13-3/+735
* 140's 10, RFC 6298's retransmission timer, written in aup with no new opcode ...eunhhu33 hours7-2/+280
* 028 and 140: list values' copying measured a second time (UAX #14's Text reco...eunhhu33 hours2-0/+4
* 140's 5, line breaking (UAX #14, Unicode 16.0), written in aup with no new op...eunhhu33 hours10-3/+508
* pure-run MODULE --views: each snapshot without its states, for an app that ho...eunhhu33 hours1-3/+12
* 140 measured again without the workarounds: the three specs' checks pass on t...eunhhu34 hours9-13/+59
* CONCAT with an empty side gives the other as it is, uncopied (values do not c...eunhhu34 hours3-0/+24
* 140: small block fns inlined in aup, made and measured, not merged (the patch...eunhhu34 hours2-0/+292
* 028's bytes alone as a state, made: module state type 6 read as a text state ...eunhhu34 hours10-32/+261
* 028: bytes alone as a state (140's decision point: inflate and HTTP both met ...eunhhu35 hours1-0/+77
* Bytes compared (140's decision point: DNS and HTTP both met it): TEXT_COMPARE...eunhhu35 hours5-7/+70
* 140's 9, HTTP/1.1 requests a packet at a time (RFC 9112), written in aup with...eunhhu36 hours10-4/+714
* 140's DNS: the compiled path's 11% stated as measured (244 to 218 ns), which ...eunhhu36 hours1-1/+1
* 140's 1, DNS messages (RFC 1035), written in aup with no new opcode (aup's ex...eunhhu36 hours9-4/+639
* 140's inflate: the three workarounds' owners: a fn taking no bare list and a ...eunhhu37 hours1-0/+9
* 140's 3, inflate (RFC 1951), written in aup with no new opcode (aup's example...eunhhu37 hours11-12/+556