mirror of
https://github.com/golang/go
synced 2024-11-23 19:40:08 -07:00
5a90306344
Currently, this test allocates many objects and relies on heap-growth scavenging to happen unconditionally on heap-growth. However with the new pacing system for the scavenging, this is no longer true and the test is flaky. So, this change overhauls TestPhysicalMemoryUtilization to check the same aspect of the runtime, but in a much more robust way. Firstly, it sets up a much more constrained scenario: only 5 objects are allocated total with a maximum worst-case (i.e. the test fails) memory footprint of about 16 MiB. The test is now aware that scavenging will only happen if the heap growth causes us to push way past our scavenge goal, which is based on the heap goal. So, it makes the holes in the test much bigger and the actual retained allocations much smaller to keep the heap goal at the heap's minimum size. It does this twice to create exactly two unscavenged holes. Because the ratio between the size of the "saved" objects and the "condemned" object is so small, two holes are sufficient to create a consistent test. Then, the test allocates one enormous object (the size of the 4 other objects allocated, combined) with the intent that heap-growth scavenging should kick in and scavenge the holes. The heap goal will rise after this object is allocated, so it's very important we do all the scavenging in a single allocation that exceeds the heap goal because otherwise the rising heap goal could foil our test. Finally, we check memory use relative to HeapAlloc as before. Since the runtime should scavenge the entirety of the remaining holes, theoretically there should be no more free and unscavenged memory. However due to other allocations that may happen during the test we may still see unscavenged memory, so we need to have some threshold. We keep the current 10% threshold which, while arbitrary, is very conservative and should easily account for any other allocations the test makes. Before, we also had to ensure the allocations we were making looked large relative to the size of a heap arena since newly-mapped memory was considered unscavenged, and so that could significantly skew the test. However, thanks to the fix for #32012 we were able to reduce memory use to 16 MiB in the worst case. Fixes #32010. Change-Id: Ia38130481e292f581da7fa3289c98c99dc5394ed Reviewed-on: https://go-review.googlesource.com/c/go/+/177237 Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org> |
||
|---|---|---|
| .. | ||
| archive | ||
| bufio | ||
| builtin | ||
| bytes | ||
| cmd | ||
| compress | ||
| container | ||
| context | ||
| crypto | ||
| database/sql | ||
| debug | ||
| encoding | ||
| errors | ||
| expvar | ||
| flag | ||
| fmt | ||
| go | ||
| hash | ||
| html | ||
| image | ||
| index/suffixarray | ||
| internal | ||
| io | ||
| log | ||
| math | ||
| mime | ||
| net | ||
| os | ||
| path | ||
| plugin | ||
| reflect | ||
| regexp | ||
| runtime | ||
| sort | ||
| strconv | ||
| strings | ||
| sync | ||
| syscall | ||
| testdata | ||
| testing | ||
| text | ||
| time | ||
| unicode | ||
| unsafe | ||
| vendor | ||
| all.bash | ||
| all.bat | ||
| all.rc | ||
| bootstrap.bash | ||
| buildall.bash | ||
| clean.bash | ||
| clean.bat | ||
| clean.rc | ||
| cmp.bash | ||
| go.mod | ||
| go.sum | ||
| iostest.bash | ||
| make.bash | ||
| make.bat | ||
| Make.dist | ||
| make.rc | ||
| naclmake.bash | ||
| nacltest.bash | ||
| race.bash | ||
| race.bat | ||
| README.vendor | ||
| run.bash | ||
| run.bat | ||
| run.rc | ||
Vendoring in std and cmd
========================
The Go command maintains copies of external packages needed by the
standard library in the src/vendor and src/cmd/vendor directories.
In GOPATH mode, imports of vendored packages are resolved to these
directories following normal vendor directory logic
(see golang.org/s/go15vendor).
In module mode, std and cmd are modules (defined in src/go.mod and
src/cmd/go.mod). When a package outside std or cmd is imported
by a package inside std or cmd, the import path is interpreted
as if it had a "vendor/" prefix. For example, within "crypto/tls",
an import of "golang.org/x/crypto/cryptobyte" resolves to
"vendor/golang.org/x/crypto/cryptobyte". When a package with the
same path is imported from a package outside std or cmd, it will
be resolved normally. Consequently, a binary may be built with two
copies of a package at different versions if the package is
imported normally and vendored by the standard library.
Vendored packages are internally renamed with a "vendor/" prefix
to preserve the invariant that all packages have distinct paths.
This is necessary to avoid compiler and linker conflicts. Adding
a "vendor/" prefix also maintains the invariant that standard
library packages begin with a dotless path element.
The module requirements of std and cmd do not influence version
selection in other modules. They are only considered when running
module commands like 'go get' and 'go mod vendor' from a directory
in GOROOT/src.
Maintaining vendor directories
==============================
Before updating vendor directories, ensure that module mode is enabled.
Make sure GO111MODULE=off is not set ('on' or 'auto' should work).
Requirements may be added, updated, and removed with 'go get'.
The vendor directory may be updated with 'go mod vendor'.
A typical sequence might be:
cd src
go get -m golang.org/x/net@latest
go mod tidy
go mod vendor
Use caution when passing '-u' to 'go get'. The '-u' flag updates
modules providing all transitively imported packages, not just
the target module.
Note that 'go mod vendor' only copies packages that are transitively
imported by packages in the current module. If a new package is needed,
it should be imported before running 'go mod vendor'.