-
-
Notifications
You must be signed in to change notification settings - Fork 32.1k
bpo-38328: Speed up the creation time of constant list literals. #16498
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.
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
Conversation
@@ -1222,7 +1222,7 @@ def get_gen(): yield 1 | |||
# list | |||
samples = [[], [1,2,3], ['1', '2', '3']] | |||
for sample in samples: | |||
check(sample, vsize('Pn') + len(sample)*self.P) | |||
check(list(sample), vsize('Pn') + len(sample)*self.P) |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This is a sys.getsizeof
test. The 3-element lists are overallocated, so we resize them here.
c41475a
to
d53e48d
Compare
@serhiy-storchaka Since you proposed something similar a while back, do you mind taking a look at this when you have the chance? |
Python/peephole.c
Outdated
h, i+1, consts, j); | ||
h, i+1-islist, consts, j); | ||
if (h >= 0 && islist) { | ||
h = copy_op_arg(codestr, i, BUILD_LIST_UNPACK, 1, i+1); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
h
is not used. Remove h =
.
It can also be replaced with
codestr[i] = PACKOPARG(BUILD_LIST_UNPACK, 1);
Thanks for the feedback @serhiy-storchaka. Tests are passing, ready for another review. |
Thank @serhiy-storchaka. Is anything else needed here? |
d0f4746
to
091e9cd
Compare
Nothing particular. I just want to fix the issue with list overallocation first. Your PR is good, thank you. |
This PR contains a small change to the peephole optimizer that converts sequences of:
LOAD_CONST(a)
,LOAD_CONST(b)
, ...,BUILD_LIST(n)
to
LOAD_CONST((a, b, ...))
,BUILD_LIST_UNPACK(1)
The improvement quickly becomes significant for lists larger than a few items:
This can be tested on any version of Python by comparing the performance of
[0, 1, 2, ...]
vs[*(0, 1, 2, ...)]
. The common cases of empty and single-element lists are not affected by this change.This is related to bpo-33325, but that was an invasive change for all collection literals that had an unknown affect on performance. I've limited this one to lists and kept it to a few lines in the peephole optimizer.
https://bugs.python.org/issue38328