Hojun-Cho/ww
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
1,080 Commits 1 Branch 0 Tags
a367bf984ee92fd1709fd39e7a2a98514feec00f
Commit Graph

6 Commits

Author SHA1 Message Date
Hojun-Cho
88f3d67b28 lib/math: re-fold F64_EXPBIAS to const f64info/f32info struct (#40) 
Restore the Hare-faithful floatinfo struct defs (ref/hare/math/floats.ha
:117,126), removing the Drew-flattened-(b) #129-era bypass (F64_EXPBIAS:int
scalar). #149 (db7523e) now lowers cross-module &def to LEAQ, so the struct
exports are addressable via the stof consumer pattern (&math.f64info ->
f: *floatinfo -> f.expbias). First real consumer of A.2 struct-composite
static-init (DATA byte-validated: f64info/f32info 40B each).

ADD export def f64info/f32info: floatinfo (hex masks value-identical to
Hare's (1<<52)-1 etc.; A.2 helper folds bare literals only, documented
at-site). DELETE export def F64_EXPBIAS (zero consumers). KEEP NAN_BITS/
INF_BITS u64 sentinels (ruling b: def NAN=0.0/0.0 blocked by #147) +
F64_EXPONENT_BIAS:u64 (bit-ops alias, Hare keeps both).

Test 952: pointer-param row reads &math.f64info + &math.f32info through
*math.floatinfo, all 5 fields each (pointer-param not direct field-read,
dodges #150). Make test 184/184 incl 990-997 byte-id + combined_ww_fresh.
lib/math not compiler-imported -> no regen.

Unblocks fold-4 stof.ha.
 2026-05-27 10:55:31 +09:00
Hojun-Cho
e28d6b2e9d lib/math: NAN/INF bits + flattened F64_EXPBIAS (#106 fold-1b) 
Flattened from Hare's const struct instances per #129 cgen module-let-
init gap; NAN/INF expressed as `def NAN_BITS:u64 = 0x...;` and
`def INF_BITS:u64 = 0x...;` materialized via f64frombits(NAN_BITS) at
use-site. F64_EXPBIAS:int flattened from f64info.expbias (cite
ref/hare/math/floats.ha:117); floatinfo struct definition preserved for
γ-cleanup re-fold when #129 closes. F32_EXPBIAS deferred (option-ii:
stof.ha consumers reach f.expbias via struct-field only).
 2026-05-26 19:19:48 +09:00
Hojun-Cho
e9620953a0 lib/math: F32 family + f32bits/f32frombits + floatinfo struct (#106 fold-1a) 
F32 width consts + bit converters mirror existing F64 family; floatinfo
struct definition (instances deferred to fold-1b per #129 cgen module-
let-init gap). Cite ref/hare/math/floats.ha; expbias: int per Drew.
 2026-05-26 19:10:02 +09:00
Hojun-Cho
f12676b880 test: 952 add frexpf64 runtime rows 2026-05-25 16:48:14 +09:00
Hojun-Cho
b4752aad21 test: 952 add issubnormalf64 + normalizef64 runtime rows 
Four cstage build+run rows for the fold-2a decompose step: issubnormalf64
across normals/zero/a constructed subnormal (f64frombits(1u64)), and
normalizef64 on both the normal path (n, 0) and the subnormal multiply
path (exp -52, result no longer subnormal). normalizef64's (f64, i64)
return rides the #102 16B-tuple-from-call receive; tuple-field f64s are
spilled to a let before any literal comparison to dodge the same
gate-blind XMM mis-load that holds back frexpf64 (see 952 header +
lib/math/floats.ww). Byte-id of a normalize-calling program is covered by
954's tuple-receive gate.
 2026-05-25 16:43:25 +09:00
Hojun-Cho
389f314473 test: math::floats fold-1 runtime regression 2026-05-25 13:17:47 +09:00