Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Parallel gsum and gmean #3202

Merged
merged 13 commits into from
Dec 11, 2018
Merged

Parallel gsum and gmean #3202

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
13 commits
Select commit Hold shift + click to select a range
ca74947
interim
mattdowle Dec 7, 2018
60e6c40
interim
mattdowle Dec 8, 2018
daada95
interim
mattdowle Dec 8, 2018
3e1af9c
interim
mattdowle Dec 11, 2018
4c472a6
assigning g from each batch didn't work quite as well
mattdowle Dec 11, 2018
9b9f1aa
(o,g) pairs better
mattdowle Dec 11, 2018
4d63fba
tidy
mattdowle Dec 11, 2018
d6f8968
gsum real
mattdowle Dec 11, 2018
9bf300d
Merge branch 'master' into parallel_gforce
mattdowle Dec 11, 2018
670cce3
Merge branch 'master' into parallel_gforce
mattdowle Dec 11, 2018
b356dc7
Merge branch 'master' into parallel_gforce
mattdowle Dec 11, 2018
313e90d
coverage
mattdowle Dec 11, 2018
0a955e2
coverage
mattdowle Dec 11, 2018
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions CRAN_Release.cmd
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -236,6 +236,8 @@ print(Sys.time()); require(data.table); print(Sys.time()); started.at<-proc.time

# Investigated and ignore :
# Tests 648 and 1262 (see their comments) have single precision issues under valgrind that don't occur on CRAN, even Solaris.
# Old comment from gsumm.c ... // long double usage here used to result in test 648 failing when run under valgrind
// http://valgrind.org/docs/manual/manual-core.html#manual-core.limits"
# Ignore all "set address range perms" warnings :
# http://stackoverflow.com/questions/13558067/what-does-this-valgrind-warning-mean-warning-set-address-range-perms
# Ignore heap summaries around test 1705 and 1707/1708 due to the fork() test opening/closing, I guess.
Expand Down
15 changes: 15 additions & 0 deletions inst/tests/tests.Rraw
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13029,6 +13029,21 @@ test(1967.75, x[ , .(v = sum(v)), by = i1:i4], x[-10L])
test(1967.76, x[1:5, sum(v), by = list(i5 = 1:5 %% 2L), verbose = TRUE],
data.table(i5 = 1:0, V1 = c(0, 0)), output = 'i clause present but columns used in by not detected')

# gforce integer overflow coerce to double
DT = data.table(A=1:5, B=-3i, C=2147483647L)
test(1968.1, DT[, sum(B), by=A%%2L], error="Type 'complex' not supported by GForce sum (gsum). Either add the")
test(1968.2, storage.mode(DT$C), "integer")
test(1968.3, DT[, sum(C), by=A%%2L], data.table(A=c(1L,0L), V1=c(6442450941, 4294967294)),
warning="sum.*integer column.*more than type 'integer' can hold.*coerced to 'numeric'")
DT[3,C:=NA]
test(1968.4, DT[, sum(C), by=A%%2L], data.table(A=c(1L,0L), V1=c(NA, 4294967294)), warning="coerced to 'numeric'")
test(1968.5, DT[, sum(C,na.rm=TRUE), by=A%%2L], data.table(A=c(1L,0L), V1=c(4294967294, 4294967294)), warning="coerced to 'numeric'")
DT[4,C:=NA]
test(1968.6, DT[, sum(C,na.rm=TRUE), by=A%%2L], data.table(A=c(1L,0L), V1=c(4294967294, 2147483647)), warning="coerced to 'numeric'")
DT[2,C:=NA]
test(1968.7, DT[, sum(C,na.rm=TRUE), by=A%%2L], data.table(A=c(1L,0L), V1=c(4294967294, 0)), warning="coerced to 'numeric'")


###################################
# Add new tests above this line #
###################################
Expand Down
Loading