A few tests rely on finalizers running, but are doing tiny allocations.
These tests will break if, for example, the testing package does is own
tiny allocations before calling the test function (see CL 478955). The
tiny allocator will group these allocations together and the ones done
for the tests themselves will live longer than desired. Use types which
have/are pointers for these tests so they won't be allocated by the tiny
allocator.
While here, pick up a small refactor suggested by Michael Knyszek to use
the BlockUntilEmptyFinalizerQueue helper to wait for the finalizers to
run in TestFinalizerRegisterABI.
Change-Id: I39f477d61f81dc76c87fae215339f8a38979cf94
Reviewed-on: https://go-review.googlesource.com/c/go/+/529555
Auto-Submit: Michael Knyszek <mknyszek@google.com>
Reviewed-by: Carlos Amedee <carlos@golang.org>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Michael Knyszek <mknyszek@google.com>
Follow-up on CL 565518 which addressed issue #65790 but added
testdata/issue65970.go instead of testdata/issue65790.go.
Rename that file to match the issue.
For #65790.
Change-Id: I647c297286355137fa950fb6722e31ae4340393b
Reviewed-on: https://go-review.googlesource.com/c/go/+/567875
Reviewed-by: Robert Griesemer <gri@google.com>
Reviewed-by: Dmitri Shuralyov <dmitshur@golang.org>
Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
Auto-Submit: Robert Griesemer <gri@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
This CL refactors the interleaved fixpoint algorithm so that calls can
be inlined in any order. This has no immediate effect, but it will
allow a subsequent CL to prioritize calls by inlheur score.
Change-Id: I11a84d228e9c94732ee75f0d3c99bc90d83fea09
Reviewed-on: https://go-review.googlesource.com/c/go/+/567695
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Auto-Submit: Matthew Dempsky <mdempsky@google.com>
Reviewed-by: Than McIntosh <thanm@google.com>
Change-Id: I27ad7ce266edc255c93012b24b6f275dd1512dc2
GitHub-Last-Rev: d196b48875
GitHub-Pull-Request: golang/go#65981
Reviewed-on: https://go-review.googlesource.com/c/go/+/567538
Reviewed-by: Ian Lance Taylor <iant@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Cherry Mui <cherryyz@google.com>
Auto-Submit: Ian Lance Taylor <iant@google.com>
When returning an error which implements net.Error and reports
itself as a timeout, also report it as matching context.DeadlineExceeded.
This matches the behavior of timeout errors in the net package
and elsewhere.
Fixes#50856
Change-Id: I2ca911e3677a699af27ba89b1200401baa8b3b1b
Reviewed-on: https://go-review.googlesource.com/c/go/+/567537
Reviewed-by: Jonathan Amsterdam <jba@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
An upcoming CL will give this call more to do.
For now, separate out the compiler change that
stops inlining the computation.
Change-Id: I4c5cbd84a0694b306191bff38cc6ea2d69458d7d
Reviewed-on: https://go-review.googlesource.com/c/go/+/564556
Auto-Submit: Russ Cox <rsc@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Continuing conversion from C to Go, change timer API to use methods.
[This is one CL in a refactoring stack making very small changes
in each step, so that any subtle bugs that we miss can be more
easily pinpointed to a small change.]
Change-Id: I4cb88a366993a77aa4fad739793a7db7213cc38c
Reviewed-on: https://go-review.googlesource.com/c/go/+/564131
Reviewed-by: Ian Lance Taylor <iant@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
The status enumeration is simple enough now that we can
view it as a bit set instead. Switch to a bit set, freeing up
the remaining bits for use in followup work to allow
garbage-collecting timers.
[This is one CL in a refactoring stack making very small changes
in each step, so that any subtle bugs that we miss can be more
easily pinpointed to a small change.]
Change-Id: I5f331fe3db1b5cb52f8571091f97f8ba029f3ac9
Reviewed-on: https://go-review.googlesource.com/c/go/+/564130
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Ian Lance Taylor <iant@google.com>
Continue using timer.lock to simplify timer operations.
[This is one CL in a refactoring stack making very small changes
in each step, so that any subtle bugs that we miss can be more
easily pinpointed to a small change.]
Change-Id: Iaf371315308425d132217eacb20b1e120a6833c5
Reviewed-on: https://go-review.googlesource.com/c/go/+/564127
Reviewed-by: Ian Lance Taylor <iant@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Continue using timer.lock to simplify timer operations.
[This is one CL in a refactoring stack making very small changes
in each step, so that any subtle bugs that we miss can be more
easily pinpointed to a small change.]
Change-Id: I2298cede902cbf0aea268c54d741190007a733c8
Reviewed-on: https://go-review.googlesource.com/c/go/+/564128
Reviewed-by: Ian Lance Taylor <iant@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Continue using timer.lock to simplify timer operations.
[This is one CL in a refactoring stack making very small changes
in each step, so that any subtle bugs that we miss can be more
easily pinpointed to a small change.]
Change-Id: I504335a010d6eb4d7d627145b64a896582158406
Reviewed-on: https://go-review.googlesource.com/c/go/+/564129
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Ian Lance Taylor <iant@google.com>
Continue using timer.lock to simplify timer operations.
[This is one CL in a refactoring stack making very small changes
in each step, so that any subtle bugs that we miss can be more
easily pinpointed to a small change.]
Change-Id: Ic12fd2630e8ac23cddd00fa7e3240a1ac19da596
Reviewed-on: https://go-review.googlesource.com/c/go/+/564126
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Ian Lance Taylor <iant@google.com>
Continue using timer.lock to simplify timer operations.
Note the removal of a previous potential deadlock.
(Explained at new line 325, there was a lock inversion
between individual timer locks and the 'timers' lock.)
[This is one CL in a refactoring stack making very small changes
in each step, so that any subtle bugs that we miss can be more
easily pinpointed to a small change.]
Change-Id: I8c9be00d13c6acd171a8aa2882a4fc844498f754
Reviewed-on: https://go-review.googlesource.com/c/go/+/564125
Reviewed-by: Ian Lance Taylor <iant@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
The state set is now simplified enough that all the CAS loops
are starting to look the same: they are just spin locks.
So introduce an actual timer.lock method and use it in deltimer.
[This is one CL in a refactoring stack making very small changes
in each step, so that any subtle bugs that we miss can be more
easily pinpointed to a small change.]
Change-Id: Ifd7f20eeede5c764ef10ecba64855c29a5ddbe39
Reviewed-on: https://go-review.googlesource.com/c/go/+/564124
Reviewed-by: Ian Lance Taylor <iant@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
When we make a change to a timer, we have to note the
desired change to t.when and then wait for the timer heap
owner to apply the change. There are two possible changes:
delete or set a new t.when. Most of the code for processing
these changes is the same, so we can simplify the code by
making both have the same state: timerDeleted is now
timerModified with t.nextwhen == 0.
This is part of a larger simplification of the state set.
[This is one CL in a refactoring stack making very small changes
in each step, so that any subtle bugs that we miss can be more
easily pinpointed to a small change.]
Change-Id: I1a2a12f8250bcd40f7b08b83f22c3a82b124eda6
Reviewed-on: https://go-review.googlesource.com/c/go/+/564123
Reviewed-by: Ian Lance Taylor <iant@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
For historical reasons, we have to treat a zero timer as
the same as an initialized timer that was stopped (removed).
The two states are already treated mostly identically.
Merge them.
This is part of a larger simplification of the state set.
[This is one CL in a refactoring stack making very small changes
in each step, so that any subtle bugs that we miss can be more
easily pinpointed to a small change.]
Change-Id: I9c3aeb8f92bafb18c47489c1ec20a7b87ac5cd9c
Reviewed-on: https://go-review.googlesource.com/c/go/+/564122
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Ian Lance Taylor <iant@google.com>
timerMoving is just a kind of "locked for modification",
so merge it into timerModifying.
This is part of a larger simplification of the state set.
[This is one CL in a refactoring stack making very small changes
in each step, so that any subtle bugs that we miss can be more
easily pinpointed to a small change.]
Change-Id: I5379122f96d9921ecda7a6a37cabd6c6b4d529a4
Reviewed-on: https://go-review.googlesource.com/c/go/+/564121
Reviewed-by: Ian Lance Taylor <iant@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
timerRemoving is just a kind of "locked for modification",
so merge it into timerModifying. This does potentially remove
a fast path from deltimer, in that deltimer of timerRemoving
is a fast-path exit while deltimer of timerModifying has to
wait for the timer to settle. Since all the timerModifying
critical paths are bounded and short, this should not matter.
This is part of a larger simplification of the state set.
[This is one CL in a refactoring stack making very small changes
in each step, so that any subtle bugs that we miss can be more
easily pinpointed to a small change.]
Change-Id: I039bf6a5a041a158dc3d1af8127f28eed50fc540
Reviewed-on: https://go-review.googlesource.com/c/go/+/564120
Reviewed-by: Ian Lance Taylor <iant@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Nothing actually needs to know the difference between these
two states, so merge them.
This is part of a larger simplification of the state set.
[This is one CL in a refactoring stack making very small changes
in each step, so that any subtle bugs that we miss can be more
easily pinpointed to a small change.]
Change-Id: Ia30699ac92e66467773942e7df1fb21470a6e51a
Reviewed-on: https://go-review.googlesource.com/c/go/+/564119
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Ian Lance Taylor <iant@google.com>
modtimer can always be used in place of addtimer.
Do that and delete addtimer, avoiding duplicated logic.
[This is one CL in a refactoring stack making very small changes
in each step, so that any subtle bugs that we miss can be more
easily pinpointed to a small change.]
Change-Id: I70291796bdac3bef5e0850f039f6f4a1da4498ae
Reviewed-on: https://go-review.googlesource.com/c/go/+/564118
Reviewed-by: Ian Lance Taylor <iant@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
No code changes, only code moves here.
Move all code that locks pp.timersLock into time.go
so that it is all in one place, for easier abstraction.
[This is one CL in a refactoring stack making very small changes
in each step, so that any subtle bugs that we miss can be more
easily pinpointed to a small change.]
Change-Id: I1b59af7780431ec6479440534579deb1a3d9d7a3
Reviewed-on: https://go-review.googlesource.com/c/go/+/564117
Reviewed-by: Ian Lance Taylor <iant@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
adjusttimers already contains the same logic. Use it instead.
This avoids having two copies of the code and is faster.
adjusttimers was formerly O(n log n) but is now O(n).
clearDeletedTimers was formerly O(n² log n) and is now gone!
[This is one CL in a refactoring stack making very small changes
in each step, so that any subtle bugs that we miss can be more
easily pinpointed to a small change.]
Change-Id: I32bf24817a589033dc304b359f8df10ea21f48fc
Reviewed-on: https://go-review.googlesource.com/c/go/+/564116
Reviewed-by: Ian Lance Taylor <iant@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
The current adjusttimers does an O(n) loop and then queues
a bunch of reinsertions, each of which is O(log n), for a worst
case of O(n log n) time plus an allocation of n elements.
Reestablishing the heap invariant from an arbitrarily ordered
slice can be done in O(n) time, so it is both simpler and faster
to avoid the allocated temporary queue and just re-init the
heap if we have damaged it. The cost of doing so is no worse
than the O(n) loop we already did.
This change also avoids holding multiple timers locked (status
set to timerMoving) at any given moment, as well as holding
individual timers locked for unbounded amounts of time,
as opposed to fixed-size critical sections.
[This is one CL in a refactoring stack making very small changes
in each step, so that any subtle bugs that we miss can be more
easily pinpointed to a small change.]
Change-Id: If966c1d1e66db797f4b19e7b1abbc06ab651764d
Reviewed-on: https://go-review.googlesource.com/c/go/+/564115
Reviewed-by: Ian Lance Taylor <iant@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Add a test that checks that all vendored packages in both src/ and
src/cmd are on an allowlist.
For #65678.
Change-Id: Ie9982f8b69c9c0fc6fef41c9dadeb464ecd7781a
Reviewed-on: https://go-review.googlesource.com/c/go/+/564355
Reviewed-by: Bryan Mills <bcmills@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Dmitri Shuralyov <dmitshur@golang.org>
Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
This CL adds new methods synonymous with the method of the same name
in reflect.Value to reflect.Type: OverflowComplex, OverflowFloat, OverflowInt, OverflowUint.
Fixes#60427
Change-Id: I7a0bb35629e59a7429820f13fcd3a6f120194bc6
GitHub-Last-Rev: 26c11bcffe
GitHub-Pull-Request: golang/go#65955
Reviewed-on: https://go-review.googlesource.com/c/go/+/567296
Auto-Submit: Ian Lance Taylor <iant@google.com>
Reviewed-by: Dmitri Shuralyov <dmitshur@google.com>
Reviewed-by: Ian Lance Taylor <iant@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
For #65164
Change-Id: Ied19cebd113ef91c34f613cafbeb92a335d6420d
GitHub-Last-Rev: 8118be6e30
GitHub-Pull-Request: golang/go#65444
Reviewed-on: https://go-review.googlesource.com/c/go/+/560597
Commit-Queue: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Keith Randall <khr@google.com>
Reviewed-by: Than McIntosh <thanm@google.com>
Auto-Submit: Ian Lance Taylor <iant@golang.org>
When the compiler writes PtrToThis field of noalg type, it generates
its pointer type. Mark them as noalg to prevent put them in typelinks.
Fixes#65957
Change-Id: Icbc3b18bc866f9138c7648e42dd500a80326f72b
Reviewed-on: https://go-review.googlesource.com/c/go/+/567335
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Auto-Submit: Keith Randall <khr@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Keith Randall <khr@google.com>
This makes version.go holding core version checking code only.
No functional changes.
Change-Id: Ia88a48166cad2698765697dd7a8625b56ecc2226
Reviewed-on: https://go-review.googlesource.com/c/go/+/567536
Reviewed-by: Robert Griesemer <gri@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Robert Findley <rfindley@google.com>
Auto-Submit: Robert Griesemer <gri@google.com>
Change-Id: I39af932b789cd18dc4bfc84f9667b1c32c9825f4
Reviewed-on: https://go-review.googlesource.com/c/go/+/567476
Reviewed-by: Robert Findley <rfindley@google.com>
Reviewed-by: Robert Griesemer <gri@google.com>
Auto-Submit: Robert Griesemer <gri@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Permit type parameters on type alias declarations depending on
Go language version.
Implement various version checks such that at most one version
error is reported per type alias declaration.
Add tparams field to Alias type node.
Missing:
- instantiation of alias types
- API additions (requires proposal)
For #46477.
Change-Id: Ica658292bd096d3bceb513027d3353501a6c58e4
Reviewed-on: https://go-review.googlesource.com/c/go/+/566856
Auto-Submit: Robert Griesemer <gri@google.com>
Reviewed-by: Robert Griesemer <gri@google.com>
Reviewed-by: Robert Findley <rfindley@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
This change moves formatting-specific functionality into a new file
format.go and rearranges the code in the errors.go files to be in
the same order for go/types and types2, making them more similar.
No functionality change. This is a pure code reordering.
Change-Id: Ibd818792397be146ad510a6c7308c85d3bb65956
Reviewed-on: https://go-review.googlesource.com/c/go/+/566555
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Auto-Submit: Robert Griesemer <gri@google.com>
Reviewed-by: Robert Findley <rfindley@google.com>
Reviewed-by: Robert Griesemer <gri@google.com>
This CL reduces the amount of code that needs to be maintained
manually by about 320 LOC.
Change-Id: Ib797e86dd4abd3a9fe614787103f0e5cebd5e4ae
Reviewed-on: https://go-review.googlesource.com/c/go/+/565838
Reviewed-by: Robert Griesemer <gri@google.com>
Reviewed-by: Robert Findley <rfindley@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Auto-Submit: Robert Griesemer <gri@google.com>
This CL reduces the amount of code that needs to be maintained
manually by about 340 LOC.
Change-Id: If7b96c30e5a2f1ff28ebf4ca2f3ac3f73d6a8865
Reviewed-on: https://go-review.googlesource.com/c/go/+/565839
Reviewed-by: Robert Griesemer <gri@google.com>
Auto-Submit: Robert Griesemer <gri@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Robert Findley <rfindley@google.com>
Factor out calling or typechecker error handler from error_.report.
In error_.report, decide if the typechecker error handler needs to
be called once or multiple times.
This change enables the use of sub-errors for types2 and go/types,
with the error handler taking care of deciding how many "separate"
errors are reported via the API.
Use new error reporting in go/types mono and initorder computation;
with the above adjustments, these changes should now pass gopls tests.
Also: adjust some format strings to avoid vet errors.
Change-Id: If05a7044399b4783c596c69a8158619f83c21c70
Reviewed-on: https://go-review.googlesource.com/c/go/+/566537
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Robert Findley <rfindley@google.com>
Reviewed-by: Robert Griesemer <gri@google.com>
Auto-Submit: Robert Griesemer <gri@google.com>
ReverseProxy uses a httptrace.ClientTrace.Got1xxResponse trace hook
to capture 1xx response headers for proxying. This hook can be called
asynchrnously after RoundTrip returns. (This should only happen when
RoundTrip has failed for some reason.) Add synchronization so we don't
attempt to modifying the ResponseWriter headers map from the hook
after another goroutine has begun making use of it.
Fixes#65123
Change-Id: I8b7ecb1a140f7ba7e37b9d27b8a20bca41a118b1
Reviewed-on: https://go-review.googlesource.com/c/go/+/567216
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Jonathan Amsterdam <jba@google.com>
Auto-Submit: Damien Neil <dneil@google.com>
Fixes#65669
Change-Id: Ifd2011dd604fef399e4352b804fc2f6a9e74096e
Reviewed-on: https://go-review.googlesource.com/c/go/+/566237
Reviewed-by: Ian Lance Taylor <iant@google.com>
Reviewed-by: Keith Randall <khr@google.com>
Auto-Submit: Keith Randall <khr@golang.org>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Keith Randall <khr@golang.org>
CL 517775 moved early deadcode into unified writer. with new way to
handle dead code with label statement involved: any statements after
terminating statement will be considered dead until next label
statement.
However, this is not safe, because code after label statement may still
refer to dead statements between terminating and label statement.
It's only safe to remove statements after terminating *and* label one.
Fixes#65593
Change-Id: Idb630165240931fad50789304a9e4535f51f56e2
Reviewed-on: https://go-review.googlesource.com/c/go/+/565596
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Keith Randall <khr@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Run-TryBot: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Auto-Submit: Cuong Manh Le <cuong.manhle.vn@gmail.com>
During calls to the race detector on arm64, we switch to the g0 stack if
we aren't already on it. If we are already on the g0 stack, the race
detector library code can then create a stack frame using the stack
pointer coming from Go code. The race detector library can go on to
write values to the top of its stack frame. But the Go ABI for arm64
saves the caller's frame pointer in the word below the current stack
frame. So, the saved frame pointer on the stack can be clobbered by the
race detector. Decrement the stack pointer to account for where the
frame pointer is saved, like we do for asmcgocall.
Change-Id: I66e5e4a671c3befc10776bac6869810ecf71790d
Reviewed-on: https://go-review.googlesource.com/c/go/+/561515
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Keith Randall <khr@golang.org>
Auto-Submit: Keith Randall <khr@google.com>
Reviewed-by: Keith Randall <khr@google.com>
Reviewed-by: Michael Knyszek <mknyszek@google.com>
When readying a goroutine, the scheduler typically places the readied
goroutine in pp.runnext, which will typically be the next goroutine to
run in the schedule.
In order to prevent a set of ping-pong goroutines from simply switching
back and forth via runnext and starving the rest of the run queue, a
goroutine scheduled via runnext shares a time slice (pp.schedtick) with
the previous goroutine.
sysmon detects "long-running goroutines", which really means Ps using
the same pp.schedtick for too long, and preempts them to allow the rest
of the run queue to run. Thus this avoids starvation via runnext.
However, wasm has no threads, and thus no sysmon. Without sysmon to
preempt, the possibility for starvation returns. Avoid this by disabling
runnext entirely on wasm. This means that readied goroutines always go
on the end of the run queue and thus cannot starve via runnext.
Note that this CL doesn't do anything about single long-running
goroutines. Without sysmon to preempt them, a single goroutine that
fails to yield will starve the run queue indefinitely.
For #65178.
Change-Id: I10859d088776125a2af8c9cd862b6e071da628b5
Cq-Include-Trybots: luci.golang.try:gotip-js-wasm,gotip-wasip1-wasm_wasmtime,gotip-wasip1-wasm_wazero
Reviewed-on: https://go-review.googlesource.com/c/go/+/559798
Auto-Submit: Bryan Mills <bcmills@google.com>
Reviewed-by: Michael Pratt <mpratt@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Add commentary explaining why and how we create method wrappers.
Change-Id: Idf35a77d0483f24c2163e11f5e001fd5536cca63
Reviewed-on: https://go-review.googlesource.com/c/go/+/558395
Auto-Submit: Matthew Dempsky <mdempsky@google.com>
Reviewed-by: Carlos Amedee <carlos@golang.org>
Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
In CL https://golang.org/cl/550435, a segment that had already been reviewed and modified was rolled back due to an erroneous operation. Here, submit a correct code snippet.
Change-Id: I46adfad110d532aec5f557c350fe6dccbc583968
Reviewed-on: https://go-review.googlesource.com/c/go/+/565628
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Damien Neil <dneil@google.com>
Reviewed-by: qiulaidongfeng <2645477756@qq.com>
Reviewed-by: Carlos Amedee <carlos@golang.org>
Reviewed-by: Mateusz Poliwczak <mpoliwczak34@gmail.com>
Auto-Submit: Ian Lance Taylor <iant@golang.org>
Commit-Queue: Ian Lance Taylor <iant@golang.org>
When FileServer(Dir("file")) is used where "file" is a normal file and
not a directory, the server enters a redirect loop.
The usage of a file inplace of a directory path is not documented in
http.Dir and it could be considered undefined behavior.
This CL updates serveFile to check if we are trying to traverse a normal
file instead of a directory and return an error, preventing the redirect
loop.
Fixes#63769
Change-Id: I81e289444e7d0bd72189c2e7b763f5540333e2d0
GitHub-Last-Rev: 754c9a1167
GitHub-Pull-Request: golang/go#63860
Reviewed-on: https://go-review.googlesource.com/c/go/+/538719
Reviewed-by: Damien Neil <dneil@google.com>
Reviewed-by: Emmanuel Odeke <emmanuel@orijtech.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Bryan Mills <bcmills@google.com>
Auto-Submit: Bryan Mills <bcmills@google.com>
Commit-Queue: Bryan Mills <bcmills@google.com>
With the new routing style in go 1.22, declaring
http.Handle("GET /", h)
generates a conflict with route "/debug/pprof/" and the others declared in
the net/http/pprof package. You get an error such as:
panic: pattern "GET /" (registered at .../pprof.go:94): GET / matches
fewer methods than /debug/pprof/, but has a more general path pattern
This patch prevents that error. Adding GET is correct because no other
method makes sense with the /debug/pprof routes. However, a tool using any
method other than GET will break.
We preserve the traditional behaviour when GODEBUG=httpmuxgo121=1 is
specified.
Updates #65723
Change-Id: I49c21f5f3e802ad7538062d824354b2e4d8a800e
GitHub-Last-Rev: 35e4012663
GitHub-Pull-Request: golang/go#65791
Reviewed-on: https://go-review.googlesource.com/c/go/+/565176
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Jonathan Amsterdam <jba@google.com>
Reviewed-by: Carlos Amedee <carlos@golang.org>
For #65898.
Change-Id: I495e53060ac56b88a551ccd9901f25bbce97c714
Reviewed-on: https://go-review.googlesource.com/c/go/+/567215
Reviewed-by: Alan Donovan <adonovan@google.com>
Reviewed-by: Robert Findley <rfindley@google.com>
Reviewed-by: Robert Griesemer <gri@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Auto-Submit: Robert Griesemer <gri@google.com>
os.Stat and os.Lstat on Windows use GetFileInformationByHandleEx to
retrieve file information for reparse points and files that
GetFileAttributesEx does not handle.
However, GetFileInformationByHandleEx is only necessary for
reparse points, so we can avoid the call for regular files.
With this change we can drop the FAT hack that was added in CL 154377,
as files won't have the FILE_ATTRIBUTE_REPARSE_POINT attribute set
on that file system.
Change-Id: Id18639067a6c3fa1bb2c6706d5b79358c224fe37
Reviewed-on: https://go-review.googlesource.com/c/go/+/566397
Reviewed-by: Carlos Amedee <carlos@golang.org>
Reviewed-by: Alex Brainman <alex.brainman@gmail.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Reviewed-by: Bryan Mills <bcmills@google.com>
