Closed WillClinger closed 7 years ago
Tasks 1 and 3 completed by changeset 19ad5f5102e08659e1e90fdc462ccc695782ea40
Task 2 completed by changeset 98e16c25f50f87398428a58330f4de5ada98fb43
Benchmarks to add for R7RS Red Edition:
dderiv
chudnovsky
bibfreq
bibfreq2
normalization (?)
listsort
vecsort
hashtable
R7RS Red Edition libraries whose performance would remain largely untested by those benchmarks:
(scheme list) ; SRFI 1
(scheme vector) ; SRFI 133
(scheme set) ; SRFI 113
(scheme charset) ; SRFI 14
(scheme ilist) ; SRFI 116
(scheme rlist) ; SRFI 101 with systematic renamings
(scheme ideque) ; SRFI 134
(scheme text) ; SRFI 135
(scheme generator) ; SRFI 121
(scheme lseq) ; SRFI 127
(scheme stream) ; SRFI 41
(scheme box) ; SRFI 111
(scheme list-queue) ; SRFI 117
(scheme ephemeron) ; SRFI 124
Aside from the (scheme box)
and (scheme ephemeron)
libraries, those libraries probably deserve new tests. Benchmarks of reasonable size would probably test only only a few of the basic procedures from each, but that's okay.
The only R7RS Red Edition libraries that remain untested by any benchmarks are
(scheme ideque) ; SRFI 134
(scheme box) ; SRFI 111
(scheme list-queue) ; SRFI 117
(scheme ephemeron) ; SRFI 124
The queue and deque libraries deserve benchmarks, but I'm inclined to let that wait for the next version of Larceny, so I'm closing this issue.
R7RS Red Edition is an interim standard consisting of the R7RS Small standard plus 17 more standard libraries:
Upgrading Larceny to R7RS Red Edition involves four related tasks:
test/R7RS/Lib
tests