Use the Vector/HeapVector projection constructor. [chromium/src : main]
0 views
Skip to first unread message
Michael Lippautz (Gerrit)
Mar 23, 2026, 4:34:49 AM (8 days ago) Mar 23
to Steinar H Gunderson, Anton Bikineev, Anders Hartvoll Ruud, chrom...@appspot.gserviceaccount.com, Menard, Alexis, chromium...@chromium.org, Kentaro Hara, apavlo...@chromium.org, blink-re...@chromium.org, blink-...@chromium.org, kinuko...@chromium.org, kouhe...@chromium.org, oilpan-rev...@chromium.org
Attention needed from Anders Hartvoll Ruud and Steinar H Gunderson
Michael Lippautz voted and added 1 comment![]( "Open in Gerrit")
Votes added by Michael Lippautz
| Code-Review | +1 |
1 comment
Patchset-level comments
File-level comment, Patchset 2 (Latest):
Michael Lippautz . resolved
lgtm, constructing with the right size is the cheapest primitive for the GC
Related details
Attention is currently required from:
- Anders Hartvoll Ruud
- Steinar H Gunderson
Submit Requirements:
Code-Coverage
Code-Owners
Code-Review
Review-Enforcement
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
Anders Hartvoll Ruud (Gerrit)
Mar 23, 2026, 5:29:33 AM (8 days ago) Mar 23
to Steinar H Gunderson, Michael Lippautz, Anton Bikineev, chrom...@appspot.gserviceaccount.com, Menard, Alexis, chromium...@chromium.org, Kentaro Hara, apavlo...@chromium.org, blink-re...@chromium.org, blink-...@chromium.org, kinuko...@chromium.org, kouhe...@chromium.org, oilpan-rev...@chromium.org
Attention needed from Steinar H Gunderson
Anders Hartvoll Ruud voted and added 1 comment![]( "Open in Gerrit")
Votes added by Anders Hartvoll Ruud
| Code-Review | +1 |
1 comment
Patchset-level comments
Attention is currently required from:
- Steinar H Gunderson
Submit Requirements:
Code-Coverage
Code-Owners
Code-Review
Review-Enforcement
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
Steinar H Gunderson (Gerrit)
Mar 23, 2026, 5:33:34 AM (8 days ago) Mar 23
to Anders Hartvoll Ruud, Michael Lippautz, Anton Bikineev, chrom...@appspot.gserviceaccount.com, Menard, Alexis, chromium...@chromium.org, Kentaro Hara, apavlo...@chromium.org, blink-re...@chromium.org, blink-...@chromium.org, kinuko...@chromium.org, kouhe...@chromium.org, oilpan-rev...@chromium.org
Steinar H Gunderson voted Commit-Queue+2![]( "Open in Gerrit")
Submit Requirements:
Code-Coverage
Code-Owners
Code-Review
Review-Enforcement
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
Chromium LUCI CQ (Gerrit)
Mar 23, 2026, 6:24:23 AM (8 days ago) Mar 23
to Steinar H Gunderson, Anders Hartvoll Ruud, Michael Lippautz, Anton Bikineev, chrom...@appspot.gserviceaccount.com, Menard, Alexis, chromium...@chromium.org, Kentaro Hara, apavlo...@chromium.org, blink-re...@chromium.org, blink-...@chromium.org, kinuko...@chromium.org, kouhe...@chromium.org, oilpan-rev...@chromium.org
Chromium LUCI CQ submitted the change![]( "Open in Gerrit")
Change information
Commit message:
Use the Vector/HeapVector projection constructor.
Vector and HeapVector have grown a new constructor that take in a
range and a projection (a lambda), effectively creating the vector
in one go. (Actually HeapVector's was limited to other HeapVectors,
but we extend it to the same Range lambda as Vector.) This replaces
our previous pattern of ReserveInitialCapacity and repeated
push_back (or AppendUnchecked):
- It is safer, as there is no way to reserve a different number of
elements from the number you AppendUnchecked, and because it
naturally lends itself to spanification of the input data.
- It generates less code, since HeapVector does not need barriers
in this case (the constructed object is guaranteed to never
leave the stack while we are pushing the elements). It also does
not need any bounds-checking.
- It is also possibly marginally faster, again due to the above.
However, the effect is very slight; on Speedometer3, it is either
neutral or very weakly positive.
The aesthetics are debatable, but overall it seems mostly fine to me,
so we use it for all relevant CSS loops that I could find, by grepping
for ReserveInitialCapacity. (The primary missing cases are the ones
where we need a filter, so that we don't actually end up pushing
as many elements as we assume we will.)
Change-Id: I5b4cc0ab4398166d8c838c99a919be9cb3746090
Reviewed-by: Anders Hartvoll Ruud <and...@chromium.org>
Reviewed-by: Michael Lippautz <mlip...@chromium.org>
Commit-Queue: Steinar H Gunderson <se...@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1603351}
Files:
- M third_party/blink/renderer/core/css/basic_shape_functions.cc
- M third_party/blink/renderer/core/css/css_basic_shape_values.cc
- M third_party/blink/renderer/core/css/css_property_value_set.cc
- M third_party/blink/renderer/core/css/css_scope_rule.cc
- M third_party/blink/renderer/core/css/resolver/matched_properties_cache.cc
- M third_party/blink/renderer/core/css/resolver/style_builder_converter.cc
- M third_party/blink/renderer/platform/heap/collection_support/heap_vector.h
Change size: M
Delta: 7 files changed, 65 insertions(+), 83 deletions(-)
Branch: refs/heads/main
Submit Requirements:
Code-Review: +1 by Anders Hartvoll Ruud, +1 by Michael Lippautz
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
Reply all
Reply to author
Forward
0 new messages