| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
| |
header's u16s alone (16 and 128 before); a stored list's record graph (PureLayout) holds a few records itself and more on the heap, 256 at most (an image numbers them in a byte), an image or a checkpoint not written then without room for it; a record's 16 fields kept, the list engine's width (a leaf's columns, a run's columns and narrow cells, the arrays' and packs' columns), to be widened with the lists' shapes sized to their fields; unit tests: 20 records of 10 fields (200 in all), an item of the last appended; a stored list whose six records each hold a list of the next saved and read back after a restart (its graph on the heap); counts: b41 against b40 with batches BFS +0.12%, without BFS -0.09%, the rest within 0.001%; the bank's queries (s40 and s41 in turns, six each) within 0.09% (the first three's in_credit +1.16% its spread: 2.75% between runs of one build); 008 and 027
|
| |
|
|
| |
in u16s), handlers or message fields: the header's u16s alone; a message's fields at most 64 (PURE_MESSAGE_FIELDS: a send pushes them all, as the stored sends' shapes hold them), outputs at most 255 (the store names them in a byte's count: the links and the outbox's image), the numbers of outputs a module dropped carried within it; an app's numbers of each input (last sent, saved, where the event began) in a block sized to its module's inputs (by_input), a replace laying the new one first and filling it by output name; image headers sized to the module (header_room); a view's arguments u32, a made column's field u16; pure-serve's timers grow to the module's inputs, its outlets and known senders its own count (OUTLETS, 32), a list view's JSON keys its columns on the stack for 64, else on the heap; unit test: 40 inputs and 70 views, the last of each used; counts: b40 against b39b within 0.01% with batches and without but BFS +0.09%; the bank's queries (s39b and s40 in turns, three each) in_credit -0.64%, per_branch -0.37%, field +0.34%, the rest within 0.22%; 008 and 027
|
| |
|
|
| |
(PURE_MODULE_STATES gone); a module's transient and counted states are flags of each, the lists views and functions change or name sets of states (state_words words, a bit each), the verifier's walked lists a list with its length, duplicate names found by sorting them; an app's eighteen arrays of each state (values, texts, lists, the stored lists' slots, the last save's and the burst's, a reader's, an event's values and marks) in one block sized to its module, a replace laying the new module's apart and taking it once nothing can fail; the runtime's lists placed past any state (PURE_LIST_HEARD 0x10000), their slots after the states'; a snapshot's arrays sized so too; pure_app_read false without room (pure-serve then shows none); a save's drafts, values and the tree's revisions on the stack for 64 states, else on the heap; the JIT compiles modules of 32 states at most (its sets a word), the app's lists read through a pointer; a view reads the app's values as they are (no state changes in one), no copy of 32 a evaluation; unit test: 300 states loaded and the last counted; a hole closed: pure_app_stop read the arrays a failed start never laid; PureApp 4,256 bytes, not 8,504; counts: b39b against b38 with batches within 0.06%, without batches knapsack +0.49%, join +0.23%, the rest within 0.17%; the bank's queries (s38 and s39b in turns, three each) field -1.99%, found -1.18%, in_credit -1.10%, large -0.99%, the rest -0.58% to +0.11% (the first build, s39, copied a view's values at run time's size: in_credit +2.4%); 008 and 027
|
| |
|
|
| |
of states, inputs, message fields, views (the columns the load makes among them, counted from the view and record entries first), handlers, records, record fields, functions and signature items point into one block the load lays out (lay_tables) and pure_module_release gives back, so PureModule is 2,752 bytes, not 24,320; a sanitized build poisons 16 bytes between tables, so reading past one is caught (none was); limits as they were; a test and five of research 044's timers gave the load a module never zeroed (it gives back what one holds), now zeroed; counts: b38 against b37 with batches within 0.25%, without batches matmul -0.45%, join +0.36%, knapsack +0.28%, BFS -0.25%; the bank's queries (s37 and s38 on one aup module in turns, three each) field -0.68%, found -0.51%, large -0.47%, the rest -0.11% to -0.20%; 027's 4f designed in four steps and 4f-1 as made
|
| |
|
|
| |
type a record field's (number, text, bool, i64, list, bytes; PURE_VIEW_TABLE now PURE_VIEW_LIST), a view entry 20 bytes (its table + 1, its items' record + 1, its list state + 1); a list view's rows a list state's items as they are or filtered (its record's fields its columns, made at load: each reads its item's field as it is), a list state's map (its columns following it, as tables were), or the items of a list its block leaves (copied into a kept list once a change, its rows and cells kept until the next); a list cell lent in its items' stored form (pure_app_cell_items, pure_items_next reading it), bytes as a text is; a pointing only at a list view over a state; pure-run writes bytes as hex and a list cell as an array of objects, pure-serve a bytes view as application/octet-stream and, in JSON, bytes as hex strings and list cells as arrays of objects; a hole closed: a view without a body never began its growth anew, so in a snapshot's reader (pure-serve's views) reading a kept list's list failed as memory (unreachable before 4e: no such view could be written), now marked a body at load; unit test list_views (as it is, filtered, computed, a map's list and bytes columns, a bytes view, six refused); check server_trees keeps its forests in a stored list shown by list views over HTTP against the model, a restart too; counts: the first build's bank tables rose (field +5.9%: view_of left inline at its seventh caller, serve's put_bytes too), so view_of is inlined, the made columns' and computed lists' paths and serve's new writers kept apart, a computed list's rows held as a filter's; b37 against b35 within 0.001% with batches and without; the bank's queries (s35 on the 4d aup's module, s37 on 4e's, six each in turns) field +0.67%, found +0.43%, large +0.31%, the rest within 0.20%; 008 and 027
|
| |
|
|
| |
messages, outputs and arguments (kinds 1 to 3), a message bound to what makes it (PureBinding: a key's press or its release, a text input, a timer, a frame, a pointing at a table), each binding's fields fixed (none; one text; one number, dt; one u32, the item's position, the table its view), one frame input at most; a handler's event its input's number; no DT or TEXT_INPUT (97 opcodes now 95, FIELD reads dt and the text), pure_app_input, frame, tick, text and point, pure_module_position, PureEdge or an app's ticks: every event is pure_app_message, saved after it unless deferred; pure-run sends each command's message to the input bound so (a release or a frame no input takes still an event, clearing the fault, so the workbench parity holds) and defers a stored app's saves past frames as before; pure-serve posts a press with its key's release, refuses timers, frames and pointings (no_input) and ticks timers as messages; holes closed: the tier found a handler's input by the old numbering, reading a message's number field as another input's (aup's numbers caught it), and serve's replies_borrow did too; module-probe presses as messages; unit tests: the test modules' builder (expand) writes the old shapes as bound messages, pointing over a pointing input and its table, stored frames deferred as a host defers them, a press's handler refused a FIELD; counts: b35 against b34 with batches BFS +0.03%, the rest within 0.001%, without batches BFS +0.12%, join +0.01%; the bank's queries (s34 on the 4c aup's module, s35 on 4d's, six each in turns) field -0.23%, found -0.14%, large -0.12%, in_credit +0.11%, the rest within 0.06%; 008 and 027
|
| |
|
|
| |
type 2 is a place in the text or the list it names (a bool 4, an i64 5), and seven anchor instructions take the place of the thirteen of cursors and indices (103 opcodes now 97): LOAD and STORE an anchor state, START and END of a text or a list, MOVE, COUNT and AT naming it and a unit (a character, a line, an item); the load holds an anchor in a text as a cursor and in a list as an index (a mark in a list refused) and each anchor instruction as the runtime's own one for it (PURE_OP_CUR_*, PURE_OP_INDEX_*, numbered past a module's in module-run.h with the kept lists' steps; a module holding one refused), so the executor's, the batches' and the JIT's steps are those they were; the verifier holds MOVE, COUNT and AT to the text or list they name; new: AT in a text (a cursor so many characters or lines from its start, not_found past the end); COUNT's lines from 0; unit test anchors (AT and COUNT in a text, AT in a list, four refused); the tests' modules written with the anchor instructions; counts: b34 against b33 with batches matmul -1.36%, the rest within 0.08%, without batches within 0.04%; the bank's queries (six each, s33 on the 4b-2 aup's module) in_credit +0.28%, per_branch +0.23%, ranks +0.10%, the rest 0.00%; 008 and 027
|
| |
|
|
| |
holding bytes doing their work: no collections table, DOC, DOC_SIZE or REPLACE (106 opcodes now 103, those from LEN on moved up), keyed messages or a base revision (pure_app_message takes an input and a payload), no PureData document callbacks (a data host opens states and saves them together), no burst by document key; the header 38 bytes; pure-edit-module gone; pure-serve without /objects, ETag, If-Match and If-None-Match, 412 or an app a thread for a module of documents alone (one without states or outputs runs so); a hole closed: an instruction in no handler's, view's or function's block was never verified yet the load's later passes read its arguments (a damaged module read out of bounds once the header moved what the mutation test damaged), now blanked to CONST 0; checks moved: server_documents (DOCS_SOURCE, keyed byte documents with the app's revision, against a Python model, one and three connection threads, a restart) in place of the parity with editor.pure, pipelining and bursts through kill -9 on its inputs and views, the lent budget on a text view (lending a document's read is server_lent_views' already), clocks and tokens on it; message tests without a document host; the HTTP benchmark's document suite runs the earlier document server and the baselines alone; the count harness builds against either message API (runs.sh build: -DPURE_KEYED_MESSAGES before 4b); counts: b33 against b32 with batches within 0.002% but BFS -0.33%, without batches BFS +0.63%, knapsack +0.52%, join +0.38%; the bank's queries (s32 on a v6 module, s33 on v7) field +0.68%, found +0.44%, large +0.42%, the rest within 0.13%; 008 as v7, 027's 4b-2 as made and 4c designed
|
| |
|
|
| |
5): held as a text is but for UTF-8, its default a constant; a function's bytes in a signature this type too (PURE_SIGNATURE_BYTES gone); no list keyed by bytes; replaced, a text field becomes a bytes field keeping its value (pure_cell_as, kinds_carry); images' headers take its kind; the JSON edge reads it from a string; pure-run shows it as hex; an append asks a record's fields held as bytes from a mask the load makes (record_held) instead of each field's type twice, so b32 runs fewer instructions than b30 (BFS without batches -1.05%, join -0.25%, BFS with batches -0.36%, the rest within 0.04%; the bank's queries s30 against s32 -0.01% to +0.30%); unit test bytes_fields (bytes not UTF-8, a text field replaced by a bytes field, three refused); server_trees with a bytes field; 008 and 027
|
| |
|
|
| |
rows whose records hold lists are read, as the message arrives, by stage 3's decoder into the event's lists (value_take into the rows' list), the event beginning with them; the JSON edge reads an object's array as its record's objects, 64 lists deep at most (items_of: an item's lists encoded first and the item moved down over them); the load no longer refuses them; a hole closed: the decoder read past the end where fewer than 4 bytes were left for a length (size - at - 4 wrapped around), unseen while it read only what was stored; unit test nested_rows (three events, four rows not whole changing nothing); check server_trees (300 random forests over HTTP against a Python model walking them in the same order, a require refusing some, a list field not an array, a field the record lacks and 65 lists deep refused, a restart); counts: b30 against b29 within 0.001% but BFS with batches +0.12%, the bank's queries -0.00% to +0.23%; 027's stage 4 planned in six steps (4a to 4f, the counts it starts from, 4b in two commits) and 4a as made; 008 on nested rows
|
| |
|
|
| |
(aup's stored_nested) alike in memory and on a store restarting eight times, every step's fault, states and a view (2,400 snapshots, 54 restarts); killed at the tree's four stages and the log's two, all or nothing; replacement moving nested lists by name at depth; appending into a stored item's list costs about 170 ns an item it already holds (append_cost.c: 42 µs an add on average over the first 500, 738 µs over 8,000), building one by appends quadratic; instruction counts: the first commit's BFS +1.45% with batches and +1.68% without, from branches in the hot steps, gone with the load's internal ops (b29 against b27: BFS 0.000%/+0.034%, the others within 0.008%; the bank's queries -0.25% to +0.22%); what remains (a message's rows holding lists, a list field's .count in a fold, for over xs[i].f); 027's stage 3 as made (the representation chosen against the plan, its cost); 008's wording (kind 5 in images' headers alone); CHECKPOINT's direction row through stage 3
|
| |
|
|
| |
descent (aup's examples/json_tree.aup: value(d, i) -> Json, number giving the value and where it ends, each call's own let's lists, string's escapes joined by char(u), a depth-first walk passing items as values) against Python's reading walked so: JSONTestSuite's 95 y_ taken alike, no n_ taken, 2b's 6 i_ taken (a lone surrogate refused by char as encoding); 100,000 and 50,000 deep refused as memory (the frames fill the event's room, 0.2 s), 30,000 deep taken (past Python's json); 300 random documents alike; in make check (module_specs' digest now depth first, three paths); 142 lines against the same design's 105 in Python (json_rec.py; the bar missed: 24 lines of declarations, 23 of braces), 46 times cJSON's time and 1.6 times plain Python's (600 ns a byte, 248 Pure instructions; 71% in the steps, about 10% in calls, 66,000 calls a document, most of them small block fns like skip); characters against Python's ord in aup's check; probes: p5-chars compiles, p4-rec (a view calling itself) refused; instruction counts: the workloads' modules the same, batches within 0.001% of 2b (BFS -0.21%), the executor alone -0.001% to +0.30%, the bank's queries -0.30% to +0.11% (four runs each; bar 1%); json_rec_time.c (cJSON, a depth-first digest compared first); CHECKPOINT's direction row through 2c
|
| |
|
|
| |
batches within 0.001% of 2a (BFS +0.36%, within its own runs' spread), the executor alone -1.3% to +0.4% (step's four cases more, mixed signs; -1.31% to +0.59% from before stage 1), the bank's queries -0.66% to +0.05% (four runs each; bar 1%); CHECKPOINT's direction row through 027's 2b
|
| |
|
|
| |
examples/json_tree.aup: a variant, lists in records, match, appends into a field's list, number(t)) and walked for a digest, against Python's reading: JSONTestSuite's 95 y_ taken alike, no n_ taken, 6 i_ taken; 300 random nested documents alike; in make check (module_specs, the suite and 100 random documents, three paths); 167 lines against the same design's 133 in Python (the bar missed: the type declarations), 40 times cJSON's time and 1.3 times plain Python's (556 ns a byte, 525 instructions); json_tree.py, json_tree_time.c (cJSON linked, digests compared first)
|
| |
|
|
| |
executor alone within 0.14% of stage 1, the bank's queries +0.28% to +0.80% where they hold still and +0.17% to +0.45% on average where they do not (four runs each; bar 1%): step's four cases more, not the held slots emptied each event (left to modules that hold some, the counts the same)
|
| |
|
|
| |
examples/json.aup, 80 lines against the same design's 90 in Python): JSONTestSuite's 95 y_ cases taken and 188 n_ refused, 21 of the 35 i_ taken, the three paths alike, in make check (module_specs, the 316 small cases in tests/specs/json under their MIT license, the two large ones made and hashed); 116 times C's time and 1.2 times plain Python's (299 ns a byte, 339 instructions); && and || made to decide alone (writing it, i < n && d[i] read past the end) with AND and OR kept where the right side cannot fail, every other module byte for byte the same; SHA-256 again with formats and joined bytes, 42 lines and 159 times C
|
| |
|
|
| |
byte the same, the executor alone moved matmul +0.43% and BFS -0.34% (knapsack, roots, join within 0.03%; the bar of 0.1% missed, mixed signs: four cases more in step), the bank's queries -0.22% to 0.00% (bar 1%); the four cases out of line in a noinline function (b23) cost 6-15% of the executor's instructions (a call taking frame from step: the register cliff) and were taken back; CHECKPOINT's direction row through 027's stage 1
|
| |
|
|
| |
examples/sha256.aup) right for FIPS 180-2's examples and 138 more messages on every path, in make check against hashlib; 52 lines against Python's 28 (the bar missed: no formatting, no bytes joined or sliced, no lists as values, moved into stage 2), 176 times C's time and 0.43 of plain Python's (1,591 ns a byte; a block's 19,129 instructions, 39% of them the inlined fn's literals not folded); 044's check, time.c against the same algorithm in C (digests compared first) and the Python reference
|
| |
|
|
| |
inflate, JSON, UTF-8 and line breaking, relational queries and graphs, text editing, box layout, HTTP/1.1 by packets, retries and timeouts, constant-time crypto) with five criteria (written in aup and Pure alone, no workaround, lines against a Python reference, time against C, the spec's own vectors passing in Pure's checks on three machines); ten probes (044) show today's aup stops at a byte by position, bit operations, nesting, recursion, variants and reading a character; the core's minimal candidates and the rule against piling up (only what two specs need); SHA-256 and JSON first
|
| |
|
|
| |
spare_count they moved its large arrays and knapsack ran 7% slower at the same instructions (8.79 against 8.29 ms), at its end about 2% (8.45); research 139's afterword: 026 not taken heavily (the user's word), the one light cut made (a nested count loop's rounds sweeping only when few lanes are left: matmul 25.85 to 24.3 ms), the walk loops kept sweeping each round (held, BFS's batches ran out), and the stop; 041 runs of b21
|
| |
|
|
| |
a load would make, interpreted a step at a time (scalar steps by the executor's rules, kernel steps over four-byte lane registers fixed at load with two buffers each, a list's cells once a batch, a deferred set after its distance check), against the same kernels called straight and natural.h's C: matmul 10.36 ms (straight 7.17; the executor's 25.8 with 3.6 of making), knapsack 4.16 (straight 3.76; the executor's 8.19); 139's question, bars and b20's split (matmul's 11.1 ms of interpretation, knapsack's 1.9)
|
| |
|
|
| |
at once in integers (join 116.0 to 110.5 ms, short of 108: what is left is the same cells' memory, read, add and write again), decide's bit found once (BFS 23.2, short of 21.5: the tail's time was its loads more than its instructions), and the batch executor's loops aligned to 64-byte lines (the hot loops' placement had swung knapsack 6% at the same instructions; the lists' loops aligned too cost the bank's queries 1-2% more instructions, so not they); against b9 matmul -36%, knapsack -28%, roots -7.6%, join -6.2%, BFS -3.9%, the executor alone, the bank queries and TPC-C (+0.7%) as they were; 025's bars and what each stage measured; 024 and CHECKPOINT say so
|
| |
|
|
| |
comparisons counting their lanes): knapsack 10.07 to 8.25 ms (bar 8.6; b9 11.46, the compiled code 7.9), its instructions 12.2% below b9's; what is left of knapsack's gap is the per-batch interpretation (about 3 ms), not the kernels (ADD 1.7, GT 1.7, the masked set 0.55 ms); the others the same by instructions; make check's server_view_workers timed out once (the known flake), passed 3/3 alone and in a full rerun; bank queries' instructions within 0.1% of b9's, TPC-C -0.3%; the two large gaps left (matmul's nested rounds, BFS's candidate tail) the user's to choose
|
| |
|
|
| |
no lanes while every lane's e is the same, a list's cells once a batch): against b9, interleaved, matmul 40.01 to 25.97 ms, knapsack 11.47 to 10.07, roots 8.97 to 8.29, join 117.2 to 115.8, BFS and sort the same; sum0 8.69 to 1.01 (bar 1.5); matmul short of its bar (21): the nested rounds' interpretation (about 14 ms) and the matrices' making (4-5) were left out of the arithmetic; the executor alone and the bank queries' instructions as b9 (0.08%), TPC-C +0.7%; stage 4 next (knapsack's six passes to three)
|
| |
|
|
| |
no lanes while every lane's e is the same), interleaved with b11 and b12, and their counts: matmul 36.10 to 26.56 ms (instructions -20%), roots 8.82 to 8.31 (-6.1%), the executor alone as b9 on all five
|
| |
|
|
| |
interleaved): matmul 38.18 to 36.10 ms with 4.5% fewer instructions (segments in its nested rounds), knapsack's per-batch cost 0.68 to 0.63 us but knapsack 10.35 to 10.51 ms with 1% fewer instructions (its per-element loops placed elsewhere), sum0 1,743 instructions a batch: short of 025's stage-2 bars (9.2 ms, 0.45 us, 33 ms): 025 priced a segment's instruction at 50 cycles, the counts show about 25 instructions, and a batch's cost sits in finds, bounds, read records, lane setup, masks and the deferred store's check; a list's cells once a batch measured at 3% and not made; next, per element: stage 3's four-byte MUL (matmul 21 ms)
|
| |
|
|
| |
interleaved): knapsack's per-batch cost 0.90 to 0.68 us and knapsack 11.48 to 10.34 ms, short of 025's bars (0.60 us, 10.3 ms): 025 counted the pass's per-instruction head as lookups; sum0 2,330 to 1,822 instructions a batch (bar 1,550); matmul 40.0 to 38.2 ms with the same instructions and BFS 24.04 to 24.31 with 0.2% fewer (placement); sum0's 8.69 to 6.13 ms mostly the compiler now vectorizing the reductions' first loop; the executor alone the same instructions; what is left a batch is the pass's Val handling (stage 2); 041 runs.sh pairs, pairs and counts data
|
| |
|
|
| |
only values each batch: knapsack's per-batch cost measured at 0.90 us, 38% of its 11.5 ms (BATCH 1024 against 256 under a sampled profile: re-reading the block about 39%, value kinds 26%, list facts and dependence 22%, dispatch 14%), sum0's 0.33 us, matmul's inner-round dispatch about 13 ms and float-checked arithmetic about 22; five stages (facts at load, static kinds and list facts once a batch, typed kernels, fewer passes, the tails) with bars from those shares; 041 batch-costs data
|
| |
|
|
| |
executor's way (memory's order, ascending): clang -O2 1.87 ms, gcc -O3 2.11, gcc -O2 9.02 (C 6.2, the compiled code 7.9)
|
| |
|
|
| |
goto in one function (make.py: GCC spills every slot across the indirect branch) and threaded tail calls with the slots as vector arguments (make2.py: an operation its vector instructions, a load and a branch; clang's musttail, GCC's sibling calls); knapsack's inner loop, BFS's frontier and roots' Newton steps against tight passes and C: roots 5.8 ms (C 6.9, the compiled code 9.57), knapsack 29.4 (passes 2.9 with clang, 9.6 gcc; C 6.2; compiled 7.9), BFS 18.1 (passes 9.8; C 8.6; compiled 7.7); NEON gathers lane by lane cost the tiles what they gain in registers
|
| |
|
|
| |
passes through memory; a batch's fixed cost about 6,500 instructions, 1 us, measured by 1024 against 256 lanes: fewer lanes are slower, so 024's L1 sizing was the wrong lever; a candidate's tail about 250), what the database literature measured of the same shape (X100, Kersten 2018's cycles and instructions a tuple, Sompolski 2011, Gubner 2017's in-register aggregation, Photon, Ertl and Gregg), the ways judged by the philosophy (a batch program made once at load; register tiles with threaded dispatch as the one general way past the floor; a table of fused operations and AOT C out), and the deciding experiment with its bar first (three bodies in tiles of 4, 8, 16: within 1.3x the compiled code's cycles an element, or it is not the answer); 041 lanes data
|
| |
|
|
| |
batches BFS 24.0 ms, the compiled code 7.7, alone 92.7; knapsack 11.45 and matmul 40.0, up from b7's 10.3 and 36.5; the executor alone as before) and the search counts on b9 (BFS 112 instructions an edge, 57 cycles; the frontier sum 56 and 20; the compiled code 38.5 and 19, 35 and 15)
|
| |
|
|
| |
ms, the compiled code 7.66, alone 93.4; the others and the executor alone as b6's) and search/: the suite's graph loaded and one handler run again and again (frontier-run: each node's edges' to summed, 70 instructions an edge batched, 35 compiled; search-run: the BFS, 211 and 38.5), counted with icount (2026-10-08-search-counts.jsonl)
|
| |
|
|
| |
compiled code 7.93), matmul 36.41 (14.20), join 118.3 (78.7), sort 55.5 (35.6), roots 8.87 (9.57); the executor alone as many instructions as b4's; costs three ways on b5 and b6 (tagged by version, b4's too: adds 0.65, ops 4.00, recur 21.8, recur4 36.0); frontier.c: the suite's BFS as frontier batches (105,232 of 999,913 edges asked again in order, 99,997 added; walking a batch's edges together 2.17x the edges)
|
| |
|
|
| |
(a step's result, a recurrence's value, the loop's value, a number that stays) rather than a stack's pushes and pops, an operation computed twice in its statement once (x the same until it is done): 1,000,000 rounds of MINSTD 44.4 to 25.7 ms (the compiled code 18.5, the executor alone 73.0); costs/recur.aup
|
| |
|
|
| |
c67f988): batches join 161.2 ms (the compiled code 74.4, alone 423.4), matmul 46.1, knapsack 15.1, roots 9.22; costs/: small modules timed three ways (an instruction about 20 ns a batch, a sum's end about 4 ns a round, an append about 100 ns as the executor's, the nested rounds 10.2 ms against the compiled code's 23.1)
|
| |
|
|
| |
9.57, alone 130.6), knapsack 13.80, matmul 50.03 (14.27, 914.9); join and sort as many instructions with batches as without
|
| |
|
|
| |
batches, the executor alone) and icount.c's instructions and cycles; stage 1 at 10a50dd: knapsack 13.57 ms with batches (1,040.9 alone, 7.52 compiled, C 6.23), the others as they were; the executor's instructions as e2e56dd's or fewer (roots 2.756 to 2.760 billion, cycles alike), knapsack's 21.7 to 0.46 billion with batches
|
| |
|
|
| |
operations apart, interpreted, 57.8 ms (C 32.6; fused by hand 48.4, the tier 2.46x); bfs: a node with more edges than the lanes runs its round alone (tested with lanes of 16, LANES_TEST); runs again
|
| |
|
|
| |
four-byte lanes ascending in memory, 4.54 ms (C 6.38, the tier 7.5; fused 2.70): narrow lanes in memory's order are most of the gain, fusing the rest; runs again
|
| |
|
|
| |
made at build time (no code made at run time), on Pure's forms and checks, against C natural and C given the same restructuring: roots 6.79 ms (C 6.86), knapsack fused four-byte 2.70 (C 6.47, C in the batch's blocks 1.94), matmul lanes over j 6.67 (C 7.10, C swapped and vector 4.88), join 37 / 46 with its recurrence interpreted (C 30), BFS 10.3 (C 10.0, C read ahead 7.0); unfused interpreted operations 11.2 and 17.7-22.1
|
| |
|
|
| |
generic code's (steps 0.1%), what the tier needs for it (a list's count as a changing value); 023 cites words at e2e56dd
|
| |
|
|
| |
knapsack 1.19, sort 1.37, join 2.54, words 5.58, matmul 1.85, BFS 1.61; geometric mean 1.92, from 3-300x at 6b45806), what each change did, the forms measured in assembly, the apps unchanged, what is left (text in the executor, BFS's appends, spills); CHECKPOINT says so
|
| |
|
|
| |
off the number pipes the divisions and sums wait on: roots 10.9 to 9.6 ms (1.39x C). 040 newton.c: roots' loop in the forms the tier may make it (the check as numbers 9.7 ms, in integers 8.7, none 7.7; next in r's register as gcc keeps it 7.0)
|
| |
|
|
| |
20.7 ms, pointer with multiply-add 13.1, apart 10.0, with checks 10.2, a bit tested before each load 13.4): the position's multiply and the pre-load branch cost, the checks hardly; 023 section 5 says so and what comes first (pointer induction, then a full list's copy)
|
| |
|
|
| |
knapsack 1.18, sort 1.37, join 2.62, words 5.46, matmul 3.07, BFS without reading ahead 1.58 (geometric mean 2.09; 3-300x at 6b45806)
|
| |
|
|
| |
apps there are): six pairs in turns, TPC-C's median -1.2% (the pairs -5.2 to +1.4%), the queries' medians within 1% but in_credit's (0.84 to 0.815 ms); the queries' instructions +0.3 to +1.5% on 49b6196 (2026-10-05-counts-49b6196.jsonl), -0.1 to +0.5% after the list's tree paths left HASHED's checks (2026-10-05-counts.jsonl). runs.sh serve, pairs and counts
|
| |
|
|
| |
lists' representation, its meaning) bare, with the checks a compiler could hoist, and with every check where the meaning puts it: hoisted 0.64-1.02x C (0.90 by the geometric mean), naive up to 12x
|
| |
|
|
| |
(knapsack), 4.8x (sort), 22x (join), 12x (words), 304x (matmul), BFS 1.57x without reading ahead; only roots and BFS run as code
|
| |
|
|
| |
knapsack, sort, join, words, matmul, BFS), each against the C one would write, in one process with the same inputs and answers; PURE_AHEAD=0 to measure the code without reading ahead
|
