The original plan was to collect allocation stacks
for all memory blocks. But it was never implemented
and it's not in near plans and it's unclear how to do it at all.
R=golang-dev, dave, bradfitz
CC=golang-dev
https://golang.org/cl/12724044
Add new proginfo function that returns information about a
Prog*. The information includes various instruction
description bits as well as a list of required registers set
and used and indexing registers used.
Convert the large instruction switches to use proginfo.
This information was formerly duplicated in multiple
optimization passes, inconsistently. For example, the
information about which registers an instruction requires
appeared three times for most instructions.
Most of the switches were incomplete or incorrect in some way.
For example, the switch in copyu did not list cases for INCB,
JPS, MOVAPD, MOVBWSX, MOVBWZX, PCDATA, POPQ, PUSHQ, STD,
TESTB, TESTQ, and XCHGL. Those were all falling into the
"unknown instruction" default case and stopping the rewrite,
perhaps unnecessarily. Similarly, the switch in needc only
listed a handful of the instructions that use or set the carry bit.
We still need to decide whether to use proginfo to generalize
a few of the remaining smaller switches in peep.c.
If this goes well, we'll make similar changes in 8g and 5g.
R=ken2
CC=golang-dev
https://golang.org/cl/12637051
On entry to a function, zero the results and zero the pointer
section of the local variables.
This is an intermediate step on the way to precise collection
of Go frames.
This can incur a significant (up to 30%) slowdown, but it also ensures
that the garbage collector never looks at a word in a Go frame
and sees a stale pointer value that could cause a space leak.
(C frames and assembly frames are still possibly problematic.)
This CL is required to start making collection of interface values
as precise as collection of pointer values are today.
Since we have to dereference the interface type to understand
whether the value is a pointer, it is critical that the type field be
initialized.
A future CL by Carl will make the garbage collection pointer
bitmaps context-sensitive. At that point it will be possible to
remove most of the zeroing. The only values that will still need
zeroing are values whose addresses escape the block scoping
of the function but do not escape to the heap.
benchmark old ns/op new ns/op delta
BenchmarkBinaryTree17 4420289180 4331060459 -2.02%
BenchmarkFannkuch11 3442469663 3277706251 -4.79%
BenchmarkFmtFprintfEmpty 100 142 +42.00%
BenchmarkFmtFprintfString 262 310 +18.32%
BenchmarkFmtFprintfInt 213 281 +31.92%
BenchmarkFmtFprintfIntInt 355 431 +21.41%
BenchmarkFmtFprintfPrefixedInt 321 383 +19.31%
BenchmarkFmtFprintfFloat 444 533 +20.05%
BenchmarkFmtManyArgs 1380 1559 +12.97%
BenchmarkGobDecode 10240054 11794915 +15.18%
BenchmarkGobEncode 17350274 19970478 +15.10%
BenchmarkGzip 455179460 460699139 +1.21%
BenchmarkGunzip 114271814 119291574 +4.39%
BenchmarkHTTPClientServer 89051 89894 +0.95%
BenchmarkJSONEncode 40486799 52691558 +30.15%
BenchmarkJSONDecode 94193361 112428781 +19.36%
BenchmarkMandelbrot200 4747060 4748043 +0.02%
BenchmarkGoParse 6363798 6675098 +4.89%
BenchmarkRegexpMatchEasy0_32 129 171 +32.56%
BenchmarkRegexpMatchEasy0_1K 365 395 +8.22%
BenchmarkRegexpMatchEasy1_32 106 152 +43.40%
BenchmarkRegexpMatchEasy1_1K 952 1245 +30.78%
BenchmarkRegexpMatchMedium_32 198 283 +42.93%
BenchmarkRegexpMatchMedium_1K 79006 101097 +27.96%
BenchmarkRegexpMatchHard_32 3478 5115 +47.07%
BenchmarkRegexpMatchHard_1K 110245 163582 +48.38%
BenchmarkRevcomp 777384355 793270857 +2.04%
BenchmarkTemplate 136713089 157093609 +14.91%
BenchmarkTimeParse 1511 1761 +16.55%
BenchmarkTimeFormat 535 850 +58.88%
benchmark old MB/s new MB/s speedup
BenchmarkGobDecode 74.95 65.07 0.87x
BenchmarkGobEncode 44.24 38.43 0.87x
BenchmarkGzip 42.63 42.12 0.99x
BenchmarkGunzip 169.81 162.67 0.96x
BenchmarkJSONEncode 47.93 36.83 0.77x
BenchmarkJSONDecode 20.60 17.26 0.84x
BenchmarkGoParse 9.10 8.68 0.95x
BenchmarkRegexpMatchEasy0_32 247.24 186.31 0.75x
BenchmarkRegexpMatchEasy0_1K 2799.20 2591.93 0.93x
BenchmarkRegexpMatchEasy1_32 299.31 210.44 0.70x
BenchmarkRegexpMatchEasy1_1K 1074.71 822.45 0.77x
BenchmarkRegexpMatchMedium_32 5.04 3.53 0.70x
BenchmarkRegexpMatchMedium_1K 12.96 10.13 0.78x
BenchmarkRegexpMatchHard_32 9.20 6.26 0.68x
BenchmarkRegexpMatchHard_1K 9.29 6.26 0.67x
BenchmarkRevcomp 326.95 320.40 0.98x
BenchmarkTemplate 14.19 12.35 0.87x
R=cshapiro
CC=golang-dev
https://golang.org/cl/12616045
Probably we should remove this type before Go 1 contract has settled,
but too late. Instead, keep InvalidAddrError close to package generic
error types.
R=golang-dev, r
CC=golang-dev
https://golang.org/cl/12670044
Prior to this change, pointer maps encoded the disposition of
a word using a single bit. A zero signaled a non-pointer
value and a one signaled a pointer value. Interface values,
which are a effectively a union type, were conservatively
labeled as a pointer.
This change widens the logical element size of the pointer map
to two bits per word. As before, zero signals a non-pointer
value and one signals a pointer value. Additionally, a two
signals an iface pointer and a three signals an eface pointer.
Following other changes to the runtime, values two and three
will allow a type information to drive interpretation of the
subsequent word so only those interface values containing a
pointer value will be scanned.
R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/12689046
On my Mac, cuts the API checks from 15 seconds to 6 seconds.
Also clean up some tag confusion: go run list-of-files ignores tags.
R=bradfitz, gri
CC=golang-dev
https://golang.org/cl/12699048
Again, it still allocates but the code is simple.
benchmark old ns/op new ns/op delta
BenchmarkReadSlice1000Int32s 35580 11465 -67.78%
benchmark old MB/s new MB/s speedup
BenchmarkReadSlice1000Int32s 112.42 348.86 3.10x
R=golang-dev, bradfitz
CC=golang-dev
https://golang.org/cl/12694048
AllTags lists all the tags that can affect the decision
about which files to include. Tools scanning packages
can use this to decide how many variants there are
and what they are.
R=bradfitz
CC=golang-dev
https://golang.org/cl/12703044
There are a few different places in the code that escape
possibly-problematic characters like < > and &.
This one was the only one missing &, so add it.
This means that if you Marshal a string, you get the
same answer you do if you Marshal a string and
pass it through the compactor. (Ironically, the
compaction makes the string longer.)
Because html/template invokes json.Marshal to
prepare escaped strings for JavaScript, this changes
the form of some of the escaped strings, but not
their meaning.
R=golang-dev, bradfitz
CC=golang-dev
https://golang.org/cl/12708044
lookup_plan9.go's lookupSRV is using the wrong order for srv results. order should be weight, priority, port, following the response from /net/dns:
chi Aug 9 20:31:13 Rread tag 20 count 61 '_xmpp-client._tcp.offblast.org srv 5 0 5222 iota.offblast.org' 72
R=golang-dev, bradfitz
CC=ality, golang-dev, r, rsc
https://golang.org/cl/12708043
This change makes the way cc constructs pointer maps closer to
what gc does and is being done in preparation for changes to
the internal content of the pointer map such as a change to
distinguish interface pointers from ordinary pointers.
R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/12692043
I've placed net.runtime_Semacquire into netpoll.goc,
but netbsd does not yet use netpoll.goc.
R=golang-dev, bradfitz, iant
CC=golang-dev
https://golang.org/cl/12699045
The mutex, fdMutex, handles locking and lifetime of sysfd,
and serializes Read and Write methods.
This allows to strip 2 sync.Mutex.Lock calls,
2 sync.Mutex.Unlock calls, 1 defer and some amount
of misc overhead from every network operation.
On linux/amd64, Intel E5-2690:
benchmark old ns/op new ns/op delta
BenchmarkTCP4Persistent 9595 9454 -1.47%
BenchmarkTCP4Persistent-2 8978 8772 -2.29%
BenchmarkTCP4ConcurrentReadWrite 4900 4625 -5.61%
BenchmarkTCP4ConcurrentReadWrite-2 2603 2500 -3.96%
In general it strips 70-500 ns from every network operation depending
on processor model. On my relatively new E5-2690 it accounts to ~5%
of network op cost.
Fixes#6074.
R=golang-dev, bradfitz, alex.brainman, iant, mikioh.mikioh
CC=golang-dev
https://golang.org/cl/12418043
The old code was caching per-type struct field info. Instead,
cache type-specific encoding funcs, tailored for that
particular type to avoid unnecessary reflection at runtime.
Once the machine is built once, future encodings of that type
just run the func.
benchmark old ns/op new ns/op delta
BenchmarkCodeEncoder 48424939 36975320 -23.64%
benchmark old MB/s new MB/s speedup
BenchmarkCodeEncoder 40.07 52.48 1.31x
Additionally, the numbers seem stable now at ~52 MB/s, whereas
the numbers for the old code were all over the place: 11 MB/s,
40 MB/s, 13 MB/s, 39 MB/s, etc. In the benchmark above I compared
against the best I saw the old code do.
R=rsc, adg
CC=gobot, golang-dev, r
https://golang.org/cl/9129044
Simple approach. Still generates garbage, but not as much.
benchmark old ns/op new ns/op delta
BenchmarkWriteSlice1000Int32s 40260 18791 -53.33%
benchmark old MB/s new MB/s speedup
BenchmarkWriteSlice1000Int32s 99.35 212.87 2.14x
Fixes#2634.
R=golang-dev, crawshaw
CC=golang-dev
https://golang.org/cl/12680046
Introduce freezetheworld function that is a best-effort attempt to stop any concurrently running goroutines. Call it during crash.
Fixes#5873.
R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/12054044
Original CL by rsc (11916045):
The motivation for disallowing them was RFC 4180 saying
"The last field in the record must not be followed by a comma."
I believe this is an admonition to CSV generators, not readers.
When reading, anything followed by a comma is not the last field.
Fixes#5892.
R=golang-dev, rsc, r
CC=golang-dev
https://golang.org/cl/12294043
By separating finding the end of the comment from the end of the action,
we can diagnose malformed comments better.
Also tweak the documentation to make the comment syntax clearer.
Fixes#6022.
R=golang-dev, dsymonds
CC=golang-dev
https://golang.org/cl/12570044
g% 6c ~/x.c
/Users/rsc/x.c:1 duplicate types given: STRUCT s and VOID
/Users/rsc/x.c:1 no return at end of function: f
g%
Fixes#6083.
R=ken2
CC=golang-dev
https://golang.org/cl/12691043
See issue 4949 for a full explanation.
Allocs go from 1 to zero in the non-addressable case.
Fixes#4949.
BenchmarkInterfaceBig 90 14 -84.01%
BenchmarkInterfaceSmall 14 14 +0.00%
R=golang-dev, bradfitz
CC=golang-dev
https://golang.org/cl/12646043
Unlike the existing net package own pollster, runtime-integrated
network pollster on BSD variants, actually kqueue, requires a socket
that has beed passed to syscall.Listen previously for a stream
listener.
This CL separates pollDesc.Init of Unix network pollster from newFD
to avoid any breakages in the transition from Unix network pollster
to runtime-integrated pollster. Upcoming CLs will rearrange the call
order of pollster and syscall functions like the following;
- For dialers that open active connections, pollDesc.Init will be
called in between syscall.Bind and syscall.Connect.
- For stream listeners that open passive stream connections,
pollDesc.Init will be called just after syscall.Listen.
- For datagram listeners that open datagram connections,
pollDesc.Init will be called just after syscall.Bind.
This is in preparation for runtime-integrated network pollster for BSD
variants.
Update #5199
R=dvyukov, bradfitz
CC=golang-dev
https://golang.org/cl/12663043
Having a trailing dot in the string doesn't really simplify
the checking loop in isDomainName. Avoid this unnecessary allocation.
Also make the valid domain names more explicit by adding some more
test cases.
benchmark old ns/op new ns/op delta
BenchmarkDNSNames 2420.0 983.0 -59.38%
benchmark old allocs new allocs delta
BenchmarkDNSNames 12 0 -100.00%
benchmark old bytes new bytes delta
BenchmarkDNSNames 336 0 -100.00%
R=golang-dev, bradfitz
CC=golang-dev
https://golang.org/cl/12662043
MOVBS and MOVHS are defined as duplicates of MOVB and MOVH,
and perform sign-extension moving.
No change is made to code generation.
Update #1837
R=rsc, bradfitz
CC=golang-dev
https://golang.org/cl/12682043
- adjusted test files so that they actually type-check
- adjusted go1.txt, go1.1.txt, next.txt
- to run, provide build tag: api_tool
Fixes#4538.
R=bradfitz
CC=golang-dev
https://golang.org/cl/12300043
The ResponseWriter's ReadFrom method was causing side effects on
the output before any data was read.
Now, bail out early and do a normal copy (which does a read
before writing) when our input and output are known to not to
be the pair of types we need for sendfile.
Fixes#5660
R=golang-dev, rsc, nightlyone
CC=golang-dev
https://golang.org/cl/12632043
I moved the pointer block from one end of the frame
to the other toward the end of working on the last CL,
and of course that made the optimization no longer work.
Now it works again:
0030 (bug361.go:12) DATA gclocals·0+0(SB)/4,$4
0030 (bug361.go:12) DATA gclocals·0+4(SB)/4,$3
0030 (bug361.go:12) GLOBL gclocals·0+0(SB),8,$8
Fixes arm build (this time for sure!).
TBR=golang-dev
CC=cshapiro, golang-dev, iant
https://golang.org/cl/12627044
Sort non-pointer-containing data to the low end of the
stack frame, and make the bitmaps only cover the
pointer-containing top end.
Generates significantly less garbage collection bitmap
for programs with large byte buffers on the stack.
Only 2% shorter for godoc, but 99.99998% shorter
in some test cases.
Fixes arm build.
TBR=golang-dev
CC=cshapiro, golang-dev, iant
https://golang.org/cl/12541047
Individual variables bigger than 10 MB are now
moved to the heap, as if they had escaped on
their own.
This avoids ridiculous stacks for programs that
do things like
x := [1<<30]byte{}
... use x ...
If 10 MB is too small, we can raise the limit.
Fixes#6077.
R=ken2
CC=golang-dev
https://golang.org/cl/12650045
GetQueuedCompletionStatusEx allows to dequeue a batch of completion
notifications, which is more efficient than dequeueing one by one.
benchmark old ns/op new ns/op delta
BenchmarkClientServerParallel4 100605 90945 -9.60%
BenchmarkClientServerParallel4-2 90225 74504 -17.42%
R=golang-dev, alex.brainman
CC=golang-dev
https://golang.org/cl/12436044
In prep for Robert's forthcoming cmd/api rewrite which
depends on the go.tools subrepo, we'll need to be more
careful about how and when we run cmd/api.
Rather than implement this policy in both run.bash and
run.bat, this change moves the policy and mechanism into
cmd/api/run.go, which will then evolve.
The plan is in a TODO in run.go.
R=golang-dev, gri
CC=golang-dev
https://golang.org/cl/12482044
The test takes up to 64 seconds on windows builders.
I've tried to reduce number of iterations in the test,
but it does not affect run time.
Fixes#6054.
R=golang-dev, alex.brainman
CC=golang-dev
https://golang.org/cl/12531043
Previously, all word aligned locations in the local variables
area were scanned as conservative roots. With this change, a
bitmap is generated describing the locations of pointer values
in local variables.
With this change the argument bitmap information has been
changed to only store information about arguments. The locals
member, has been removed. In its place, the bitmap data for
local variables is now used to store the size of locals. If
the size is negative, the magnitude indicates the size of the
local variables area.
R=rsc
CC=golang-dev
https://golang.org/cl/12328044
Remove NOPROF/DUPOK from everything.
Edits done with a script, except pclinetest.asm which depended
on the DUPOK flag on main().
R=golang-dev, bradfitz
CC=golang-dev
https://golang.org/cl/12613044
There were some issues with the code sometimes using base64.StdEncoding,
and sometimes base64.URLEncoding.
Encoding basic authentication is now always done by the same code.
Fixes#5970.
R=golang-dev, bradfitz
CC=golang-dev
https://golang.org/cl/12397043
We can then include this file in assembly to replace
cryptic constants like "7" with meaningful constants
like "(NOPROF|DUPOK|NOSPLIT)".
Converting just pkg/runtime/asm*.s for now. Dropping NOPROF
and DUPOK from lots of places where they aren't needed.
More .s files to come in a subsequent changelist.
A nonzero number in the textflag field now means
"has not been converted yet".
R=golang-dev, daniel.morsing, rsc, khr
CC=golang-dev
https://golang.org/cl/12568043
HTTP/1.0 connections are closed implicitly, unless otherwise specified.
Note that this change does not test or fix "request too large" responses.
Reasoning: (a) it complicates tests and fixes, (b) they should be rare,
and (c) this is just a minor wire optimization, and thus not really worth worrying
about in this context.
Fixes#5955.
R=golang-dev, bradfitz
CC=golang-dev
https://golang.org/cl/12435043
A response to a HEAD request is supposed to look the same as a
response to a GET request, just without a body.
HEAD requests are incredibly rare in the wild.
The Go net/http package has so far treated HEAD requests
specially: a Write on our default ResponseWriter returned
ErrBodyNotAllowed, telling handlers that something was wrong.
This was to optimize the fast path for HEAD requests, but:
1) because HEAD requests are incredibly rare, they're not
worth having a fast path for.
2) Letting the http.Handler handle but do nop Writes is still
very fast.
3) this forces ugly error handling into the application.
e.g. https://code.google.com/p/go/source/detail?r=6f596be7a31e
and related.
4) The net/http package nowadays does Content-Type sniffing,
but you don't get that for HEAD.
5) The net/http package nowadays does Content-Length counting
for small (few KB) responses, but not for HEAD.
6) ErrBodyNotAllowed was useless. By the time you received it,
you had probably already done all your heavy computation
and I/O to calculate what to write.
So, this change makes HEAD requests like GET requests.
We now count content-length and sniff content-type for HEAD
requests. If you Write, it doesn't return an error.
If you want a fast-path in your code for HEAD, you have to do
it early and set all the response headers yourself. Just like
before. If you choose not to Write in HEAD requests, be sure
to set Content-Length if you know it. We won't write
"Content-Length: 0" because you might've just chosen to not
write (or you don't know your Content-Length in advance).
Fixes#5454
R=golang-dev, dsymonds
CC=golang-dev
https://golang.org/cl/12583043
If the padding is huge, we crashed by blowing the buffer. That's easy: make sure
we have a big enough buffer by allocating in problematic cases.
Zero padding floats was just wrong in general: the space would appear in the
middle.
Fixes#6044.
R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/12498043
This CL refactors the existing listenerSockaddr function into several
methods on netFD.
This is in preparation for runtime-integrated network pollster for BSD
variants.
Update #5199
R=golang-dev, dave, alex.brainman, dvyukov, remyoudompheng
CC=golang-dev
https://golang.org/cl/12023043
Updates #6046.
This CL just does maxstring and concatstring. There are other functions
to fix but doing them a few at a time will help isolate any (unlikely)
breakages these changes bring up in architectures I can't test
myself.
R=golang-dev, dave, iant
CC=golang-dev
https://golang.org/cl/12519044
I broke it with the darwin getwd attrlist stuff (0583e9d36dd).
plan9 doesn't have syscall.ENOTSUP.
It's in api/go1.txt as a symbol always available (not context-specific):
pkg syscall, const ENOTSUP Errno
... but plan9 isn't considered by cmd/api, so it only looks
universally available. Alternatively, we could add a fake ENOTSUP
to plan9, but they were making efforts earlier to clean their
syscall package, so I'd prefer not to dump more in it.
R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/12509044
This change replaces the hard-coded switch on compression method
in zipfile reader and writer with a map into which users can
register compressors and decompressors in their init()s.
R=golang-dev, bradfitz, rsc
CC=golang-dev
https://golang.org/cl/12421043
NetBSD and OpenBSD are broken like OS X is. Good to know.
Drop required count from avg/2 to avg/3, because the
Plan 9 builder just barely missed avg/2 in one of its runs.
R=golang-dev, dvyukov
CC=golang-dev
https://golang.org/cl/12548043
Looks like latest FreeBSD doesn't set address family identifer
for RTAX_NETMASK stuff; probably RTAX_GENMASK too, not confirmed.
This CL tries to identify address families by using the length of
each socket address if possible.
The issue is confirmed on FreeBSD 9.1.
R=golang-dev, bradfitz
CC=golang-dev
https://golang.org/cl/12332043
Unlike the existing net package own pollster, runtime-integrated
network pollster on BSD variants, actually kqueue, requires a socket
that has beed passed to syscall.Listen previously for a stream
listener.
This CL separates pollDesc.Init (actually runtime_pollOpen) from newFD
to allow control of each state of sockets and adds init method to netFD
instead. Upcoming CLs will rearrange the call order of runtime-integrated
pollster and syscall functions like the following;
- For dialers that open active connections, runtime_pollOpen will be
called in between syscall.Bind and syscall.Connect.
- For stream listeners that open passive stream connections,
runtime_pollOpen will be called just after syscall.Listen.
- For datagram listeners that open datagram connections,
runtime_pollOpen will be called just after syscall.Bind.
This is in preparation for runtime-integrated network pollster for BSD
variants.
Update #5199
R=dvyukov, alex.brainman, minux.ma
CC=golang-dev
https://golang.org/cl/8608044
Update #6046.
This CL just does findnull and findnullw. There are other functions
to fix but doing them a few at a time will help isolate any (unlikely)
breakages these changes bring up in architectures I can't test
myself.
R=golang-dev, dsymonds
CC=golang-dev
https://golang.org/cl/12520043
Embed all data necessary for read/write operations directly into netFD.
benchmark old ns/op new ns/op delta
BenchmarkTCP4Persistent 27669 23341 -15.64%
BenchmarkTCP4Persistent-2 18173 12558 -30.90%
BenchmarkTCP4Persistent-4 10390 7319 -29.56%
This change will intentionally break all builders to see
how many allocations they do per read/write.
This will be fixed soon afterwards.
R=golang-dev, alex.brainman
CC=golang-dev
https://golang.org/cl/12413043
gcpc/gcsp are used by GC in similar situation.
gcpc/gcsp are also more stable than gp->sched,
because gp->sched is mutated by entersyscall/exitsyscall
in morestack and mcall. So it has higher chances of being inconsistent.
Also, rename gcpc/gcsp to syscallpc/syscallsp.
This is the same as reverted change 12250043
with save marked as textflag 7.
The problem was that if save calls morestack,
then subsequent lessstack spoils g->sched.pc/sp.
And that bad values were remembered in g->syscallpc/sp.
Entersyscallblock had the same problem,
but it was never triggered to date.
R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/12478043
Basically a partial rollback of 12053043 until I can
figure out what is really going on.
Fixes bug 6051.
R=golang-dev
CC=golang-dev
https://golang.org/cl/12496043
This means that pprof will no longer report profiles on OS X.
That's unfortunate, but the profiles were often wrong and, worse,
it was difficult to tell whether the profile was wrong or not.
The workarounds were making the scheduler more complex,
possibly caused a deadlock (see issue 5519), and did not actually
deliver reliable results.
It may be possible for adventurous users to apply a patch to
their kernels to get working results, or perhaps having no results
will encourage someone to do the work of creating a profiling
thread like on Windows. Issue 6047 has details.
Fixes#5519.
Fixes#6047.
R=golang-dev, bradfitz, r
CC=golang-dev
https://golang.org/cl/12429045
This means that in the common case (modern kernel), we only
make 1 system call to dup instead of two, and we also avoid
grabbing the syscall.ForkLock.
R=golang-dev, iant
CC=golang-dev
https://golang.org/cl/12476043
you do reflect.call with too big an argument list.
Not worth the hassle.
Fixes#6023Fixes#6033
R=golang-dev, bradfitz, dave
CC=golang-dev
https://golang.org/cl/12485043
For normal slices a[i:j] we're generating 3 bounds
checks: j<={len(string),cap(slice)}, j<=j (!), and i<=j.
Somehow snuck in as part of the [i:j:k] implementation
where the second check does something.
Remove the second check when we don't need it.
R=rsc, r
CC=golang-dev
https://golang.org/cl/12311046
While we're here, add a test for the same functionality in gzip,
which was already implemented, and add bzip2 CRC checks.
Fixes#5772.
R=golang-dev, r
CC=golang-dev
https://golang.org/cl/12387044
Break all 386 builders.
««« original CL description
runtime: use gcpc/gcsp during traceback of goroutines in syscalls
gcpc/gcsp are used by GC in similar situation.
gcpc/gcsp are also more stable than gp->sched,
because gp->sched is mutated by entersyscall/exitsyscall
in morestack and mcall. So it has higher chances of being inconsistent.
Also, rename gcpc/gcsp to syscallpc/syscallsp.
R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/12250043
»»»
R=rsc
CC=golang-dev
https://golang.org/cl/12424045
It was needed for the old scheduler,
because there temporary could be more threads than gomaxprocs.
In the new scheduler gomaxprocs is always respected.
R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/12438043
gcpc/gcsp are used by GC in similar situation.
gcpc/gcsp are also more stable than gp->sched,
because gp->sched is mutated by entersyscall/exitsyscall
in morestack and mcall. So it has higher chances of being inconsistent.
Also, rename gcpc/gcsp to syscallpc/syscallsp.
R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/12250043
In the event that code tries to use a hash function that isn't compiled
in and panics, give the developer a fighting chance of figuring out
which hash function it needed.
R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/12420045
Runtime netpoll supports at most one read waiter
and at most one write waiter. It's responsibility
of net package to ensure that. Currently windows
implementation allows more than one waiter in Accept.
It leads to "fatal error: netpollblock: double wait".
R=golang-dev, bradfitz
CC=golang-dev
https://golang.org/cl/12400045
Whether the keys are concatenated or separate (or a mixture) depends on the server.
Fixes#5979.
R=golang-dev, bradfitz
CC=golang-dev
https://golang.org/cl/12433043
Windows dynamic priority boosting assumes that a process has different types
of dedicated threads -- GUI, IO, computational, etc. Go processes use
equivalent threads that all do a mix of GUI, IO, computations, etc.
In such context dynamic priority boosting does nothing but harm, so turn it off.
In particular, if 2 goroutines do heavy IO on a server uniprocessor machine,
windows rejects to schedule timer thread for 2+ seconds when priority boosting is enabled.
Fixes#5971.
R=alex.brainman
CC=golang-dev
https://golang.org/cl/12406043
The test isn't checking deliberate panics so catching them just makes the code longer.
R=golang-dev, dsymonds
CC=golang-dev
https://golang.org/cl/12420043
This CL makes IPAddr, UDPAddr and TCPAddr implement sockaddr
interface, UnixAddr is already sockaddr interface compliant, and
reduces unnecessary conversions between net.Addr, net.sockaddr and
syscall.Sockaddr.
This is in preparation for runtime-integrated network pollster for BSD
variants.
Update #5199
R=golang-dev, dave, bradfitz
CC=golang-dev
https://golang.org/cl/12010043
Also, add a meaningful error message when an encoding which
can't be parsed is found.
Fixes#5801.
R=golang-dev, bradfitz, rsc
CC=golang-dev
https://golang.org/cl/12343043
Fixes#5822.
Will no doubt cause other problems, but Apple has forced our hand.
R=golang-dev, bradfitz, khr
CC=golang-dev
https://golang.org/cl/12350044
cmd/api is a tool to prevent the Go developers from breaking
the Go 1 API promise. It has no utility to end users and
doesn't run on arbitrary packages (it's always been full of
hacks for its bespoke type checker to work on the standard
library)
Robert's in-progress rewrite depends on the go.tools repo for
go/types, so we won't be able to ship this tool later
anyway. Just remove it from binary distributions.
A future change to run.bash can conditionally build & run
cmd/api, perhaps automatically fetching go/types if
necessary. I assume people don't want to vendor go/types into
a private gopath just for cmd/api.
I will need help with run.bat.
R=golang-dev, adg, dsymonds, rsc
CC=golang-dev
https://golang.org/cl/12316043
It's okay to preempt at ordinary function calls because
compilers arrange that there are no live registers to save
on entry to the function call.
The software floating point routines are function calls
masquerading as individual machine instructions. They are
expected to keep all the registers intact. In particular,
they are expected not to clobber all the floating point
registers.
The floating point registers are kept per-M, because they
are not live at non-preemptive goroutine scheduling events,
and so keeping them per-M reduces the number of 132-byte
register blocks we are keeping in memory.
Because they are per-M, allowing the goroutine to be
rescheduled during software floating point simulation
would mean some other goroutine could overwrite the registers
or perhaps the goroutine would continue running on a different
M entirely.
Disallow preemption during the software floating point
routines to make sure that a function full of floating point
instructions has the same floating point registers throughout
its execution.
R=golang-dev, dave
CC=golang-dev
https://golang.org/cl/12298043
Per suggestion from Russ in February. Then strings.IndexByte
can be implemented in terms of the shared code in pkg runtime.
Update #3751
R=golang-dev, r
CC=golang-dev
https://golang.org/cl/12289043
I used just enough of the data provided by Matt in Issue 5915 to trigger
issue 5915. As luck would have it, using slightly less of it triggered
issue 5962.
Fixes#5915.
Fixes#5962.
R=golang-dev, bradfitz
CC=golang-dev
https://golang.org/cl/12288043
Also move chatty recent additions to -v -v.
For what it's worth:
$ go build -o /dev/null -ldflags -v cmd/go
...
0.87 pclntab=1110836 bytes, funcdata total 69700 bytes
...
$
This broke the ELF builds last time because I tried to dedup
the funcdata in case the same funcdata was pointed at by
multiple functions. That doesn't currently happen, so I've
removed that test.
If we start doing bitmap coalescing we'll need to figure out
how to measure the size more carefully, but I think at that
point the bitmaps will be an extra indirection away from the
funcdata anyway, so the dedup I used before wouldn't help.
R=ken2
CC=golang-dev
https://golang.org/cl/12269043
This allows to at least determine goroutine "identity".
Now it looks like:
goroutine 12 [running]:
goroutine running on other thread; stack unavailable
created by testing.RunTests
src/pkg/testing/testing.go:440 +0x88e
R=golang-dev, r, rsc
CC=golang-dev
https://golang.org/cl/12248043
We see timeouts in these tests on some platforms,
but not on the others. The hypothesis is that
the problematic platforms are slow uniprocessors.
Stack traces do not suggest that the process
is completely hang, and it is able to schedule
the alarm goroutine. And if it actually hangs,
we still will be able to detect that.
R=golang-dev, r
CC=golang-dev
https://golang.org/cl/12253043
With dup3, we can avoid an extra system call on some machines
while holding syscall.ForkLock. Currently we have to
syscall.Dup + syscall.CloseOnExec.
On machines with Linux and a new enough kernel, this can just
be dup3.
R=golang-dev, r
CC=golang-dev
https://golang.org/cl/12170045
Preemption during the software floating point code
could cause m (R9) to change, so that when the
original registers were restored at the end of the
floating point handler, the changed and correct m
would be replaced by the old and incorrect m.
TBR=dvyukov
CC=golang-dev
https://golang.org/cl/11883045
Operations on int64 are very stack consuming with 5c.
Fixes netbsd/arm build.
Before: TEXT runtime.timediv+0(SB),7,$52-16
After: TEXT runtime.timediv+0(SB),7,$44-16
The stack usage is unchanged on 386:
TEXT runtime.timediv+0(SB),7,$8-16
R=golang-dev, dvyukov, bradfitz
CC=golang-dev
https://golang.org/cl/12182044
This patch introduces specialized functions for initial
and final permutations, and precomputes the output of the
third permutation on the S-box elements.
benchmark old ns/op new ns/op delta
BenchmarkEncrypt 3581 1226 -65.76%
BenchmarkDecrypt 3590 1224 -65.91%
benchmark old MB/s new MB/s speedup
BenchmarkEncrypt 2.23 6.52 2.92x
BenchmarkDecrypt 2.23 6.53 2.93x
R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/12072045
Update #5139.
Double wakeup on Note was reported several times,
but no reliable reproducer.
There also was a strange report about weird value of epoll fd.
Maybe it's corruption of global data...
R=golang-dev, bradfitz
CC=golang-dev
https://golang.org/cl/12182043
Sysmon thread parks if no goroutines are running (runtime.sched.npidle ==
runtime.gomaxprocs).
Currently it's unparked when a goroutine enters syscall, it was enough
to retake P's from blocking syscalls.
But it's not enough for reliable goroutine preemption. We need to ensure that
sysmon runs if any goroutines are running.
R=rsc
CC=golang-dev
https://golang.org/cl/12176043
Submitted with some unrelated changes that were not intended to go in.
««« original CL description
runtime: do not park sysmon thread if any goroutines are running
Sysmon thread parks if no goroutines are running (runtime.sched.npidle == runtime.gomaxprocs).
Currently it's unparked when a goroutine enters syscall, it was enough
to retake P's from blocking syscalls.
But it's not enough for reliable goroutine preemption. We need to ensure that
sysmon runs if any goroutines are running.
R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/12167043
»»»
R=rsc
CC=golang-dev
https://golang.org/cl/12171044
This is required to properly unwind reflect.methodValueCall/makeFuncStub.
Fixes#5954.
Stats for 'go install std':
61849 total INSTCALL
24655 currently have ArgSize metadata
27278 have ArgSize metadata with this change
godoc size before: 11351888, after: 11364288
R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/12163043
Sysmon thread parks if no goroutines are running (runtime.sched.npidle == runtime.gomaxprocs).
Currently it's unparked when a goroutine enters syscall, it was enough
to retake P's from blocking syscalls.
But it's not enough for reliable goroutine preemption. We need to ensure that
sysmon runs if any goroutines are running.
R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/12167043
Adds layout cases with seconds for stdISO8601 and stdNumTZ with and without colons. Update time.Format to append seconds for those cases.
Fixes#4934.
R=golang-dev, r, bradfitz
CC=golang-dev
https://golang.org/cl/8132044
This means that printing a Node will produce output that can be used as valid input.
It won't be exactly the same - some spacing may be different - but it will mean the same.
Fixes#4593.
R=golang-dev, dsymonds
CC=golang-dev
https://golang.org/cl/12006047
When scanning input and "white space" is permitted, a carriage return
followed immediately by a newline (\r\n) is treated exactly the same
as a plain newline (\n). I hope this makes it work better on Windows.
We do it everywhere, not just on Windows, since why not?
Fixes#5391.
R=golang-dev, dsymonds
CC=golang-dev
https://golang.org/cl/12142043
When comparing strings, check these (in order):
- length mismatch => not equal
- string pointer equal => equal
- if length is short:
- memeq on body
- if length is long:
- compare first&last few bytes, if different => not equal
- save entry as a possible match
- after checking every entry, if there is only one possible
match, use memeq on that entry. Otherwise, fallback to hash.
benchmark old ns/op new ns/op delta
BenchmarkSameLengthMap 43 4 -89.77%
Fixes#5194.
Update #3885.
R=golang-dev, bradfitz, khr, rsc
CC=golang-dev
https://golang.org/cl/12128044
The prefix was not uniformly applied and is probably better left off anyway.
Fixes#4944.
R=golang-dev, dsymonds
CC=golang-dev
https://golang.org/cl/12140043
There is a chance that the SIGQUIT will make the test process
dump its stacks as part of exiting, which would be nice for
finding out what it is doing.
Right now the builders are occasionally timing out running
the runtime test. I hope this will give us some information
about the state of the runtime.
R=golang-dev, dave
CC=golang-dev
https://golang.org/cl/12041051
struct Hmap is the header for a map value.
CL 8377046 made flags a uint32 so that it could be updated atomically,
but that bumped the struct to 56 bytes, which allocates as 64 bytes (on amd64).
hash0 is initialized from runtime.fastrand1, which returns a uint32,
so the top 32 bits were always zero anyway. Declare it as a uint32
to reclaim 4 bytes and bring the Hmap size back down to a 48-byte allocation.
Fixes#5237.
R=golang-dev, khr, khr
CC=bradfitz, dvyukov, golang-dev
https://golang.org/cl/12034047
If netFD is closed by finalizer, runtime netpoll descriptor is not freed.
R=golang-dev, dave, alex.brainman
CC=golang-dev
https://golang.org/cl/12037043
EscapeText now escapes 0xFFFD returned from DecodeRune as 0xFFFD, rather than passing through the original byte.
Fixes#5880.
R=golang-dev, r, bradfitz, adg
CC=golang-dev
https://golang.org/cl/11975043
I want to see the timing information in build logs,
and we can't see the logs for "ok" builds.
So make the build fail everywhere.
Will roll back immediately.
TBR=dvyukov
CC=golang-dev
https://golang.org/cl/12058046
notetsleep: nosplit stack overflow
120 assumed on entry to notetsleep
96 after notetsleep uses 24
88 on entry to runtime.semasleep
32 after runtime.semasleep uses 56
24 on entry to runtime.nanotime
-8 after runtime.nanotime uses 32
Nanotime seems to be using only 24 bytes of stack space.
Unless I am missing something.
R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/12041044
notetsleep: nosplit stack overflow
120 assumed on entry to notetsleep
80 after notetsleep uses 40
72 on entry to runtime.futexsleep
16 after runtime.futexsleep uses 56
8 on entry to runtime.printf
-16 after runtime.printf uses 24
R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/12047043
Split stack checks (morestack) corrupt g->sched,
but g->sched must be preserved consistent for GC/traceback.
The change implements runtime.notetsleepg function,
which does entersyscall/exitsyscall and is carefully arranged
to not call any split functions in between.
R=rsc
CC=golang-dev
https://golang.org/cl/11575044
Close netpoll descriptor along with socket.
Ensure that error paths close the descriptor as well.
R=golang-dev, mikioh.mikioh, alex.brainman
CC=golang-dev
https://golang.org/cl/11987043
This is in preparation for runtime-integrated network pollster for BSD
variants.
Update #5199
R=golang-dev, fvbommel, dave
CC=golang-dev
https://golang.org/cl/11984043
This is in preparation for runtime-integrated network pollster for BSD
variants.
Update #5199
R=golang-dev, dave
CC=golang-dev
https://golang.org/cl/11932044
This CL extends existing sockaddr interface to accommodate not only
internet protocol family endpoint addressess but unix network family
endpoint addresses.
This is in preparation for runtime-integrated network pollster for BSD
variants.
Update #5199
R=golang-dev, dave
CC=golang-dev
https://golang.org/cl/11979043
If netpoll has been told to block, it must not return with nil,
otherwise scheduler assumes that netpoll is disabled.
R=golang-dev, alex.brainman
CC=golang-dev
https://golang.org/cl/11920044
Make it accept type, combine flags.
Several reasons for the change:
1. mallocgc and settype must be atomic wrt GC
2. settype is called from only one place now
3. it will help performance (eventually settype
functionality must be combined with markallocated)
4. flags are easier to read now (no mallocgc(sz, 0, 1, 0) anymore)
R=golang-dev, iant, nightlyone, rsc, dave, khr, bradfitz, r
CC=golang-dev
https://golang.org/cl/10136043
The rlimit arguments for prlimit are reversed for linux 32-bit (386 and arm).
Getrlimit becomes Setrlimit and vice versa.
Fixes#5949.
R=iant, mikioh.mikioh, rsc
CC=golang-dev
https://golang.org/cl/11803043
Don't require a full-scale callback for calls to the special
prologue functions.
Always use a simple wrapper function for C functions, so that
we can handle static functions defined in the import "C"
comment.
Disable a test that relies on gc-specific function names.
Fixes#5905.
R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/11406047
Backends do not exactly expect receiving binary operators with
constant operands or use workarounds to move them to
register/stack in order to handle them.
Fixes#5841.
R=golang-dev, daniel.morsing, rsc
CC=golang-dev
https://golang.org/cl/11107044
An ARM version of md5block.go with a big improvement in
throughput (up to 2.5x) and a reduction in object size (21%).
Code size
Before 3100 bytes
After 2424 bytes
21% smaller
Benchmarks on Rasperry Pi
benchmark old ns/op new ns/op delta
BenchmarkHash8Bytes 11703 6636 -43.30%
BenchmarkHash1K 38057 21881 -42.50%
BenchmarkHash8K 208131 142735 -31.42%
BenchmarkHash8BytesUnaligned 11457 6570 -42.66%
BenchmarkHash1KUnaligned 69334 26841 -61.29%
BenchmarkHash8KUnaligned 455120 182223 -59.96%
benchmark old MB/s new MB/s speedup
BenchmarkHash8Bytes 0.68 1.21 1.78x
BenchmarkHash1K 26.91 46.80 1.74x
BenchmarkHash8K 39.36 57.39 1.46x
BenchmarkHash8BytesUnaligned 0.70 1.22 1.74x
BenchmarkHash1KUnaligned 14.77 38.15 2.58x
BenchmarkHash8KUnaligned 18.00 44.96 2.50x
benchmark old allocs new allocs delta
BenchmarkHash8Bytes 1 0 -100.00%
BenchmarkHash1K 2 0 -100.00%
BenchmarkHash8K 2 0 -100.00%
BenchmarkHash8BytesUnaligned 1 0 -100.00%
BenchmarkHash1KUnaligned 2 0 -100.00%
BenchmarkHash8KUnaligned 2 0 -100.00%
benchmark old bytes new bytes delta
BenchmarkHash8Bytes 64 0 -100.00%
BenchmarkHash1K 128 0 -100.00%
BenchmarkHash8K 128 0 -100.00%
BenchmarkHash8BytesUnaligned 64 0 -100.00%
BenchmarkHash1KUnaligned 128 0 -100.00%
BenchmarkHash8KUnaligned 128 0 -100.00%
This also adds another test which makes sure that the sums
over larger blocks work properly. I wrote this test when I was
worried about memory corruption.
R=golang-dev, dave, bradfitz, rsc, ajstarks
CC=golang-dev, minux.ma, remyoudompheng
https://golang.org/cl/11648043
The problem is that the cdecl() function in cmd/cgo/godefs.go isn't
properly translating the Go array type to a C array type when an
asterisk follows the [] in the array type declaration (it is perfectly
legal to put the asterisk on either side of the [] in go syntax,
depending on how you set up your pointers).
That said, the cdefs tool is only designed to translate from Go types
generated using the cgo *godefs* tool -- where the godefs tool is
designed to translate gcc-style C types into Go types. In essence, the
cdefs tool translates from gcc-style C types to Go types (via the godefs
tool), then back to kenc-style C types. Because of this, cdefs does not
need to know how to translate arbitraty Go types into C, just the ones
produced by godefs.
The problem is that during this translation process, the logic is
slightly wrong when going from (e.g.):
char *array[10];
to:
array [10]*int8;
back to:
int8 *array[10];
In the current implementation of cdecl(), the translation from the Go
type declaration back to the kenc-style declaration looks for Go
types of the form:
name *[]type;
rather than the actual generated Go type declaration of:
name []*type;
Both are valid Go syntax, with slightly different semantics, but the
latter is the only one that can ever be generated by the godefs tools.
(The semantics of the former are not directly expressible in a
single C statement -- you would have to have to first typedef the array
type, then declare a pointer to that typedef'd type in a separate
statement).
This commit changes the logic of cdecl() to look properly for, and
translate, Go type declarations of the form:
name []*type;
Additionally, the original implementation only allowed for a single
asterisk and a single sized aray (i.e. only a single level of pointer
indirection, and only one set of []) on the type, whereas the patched
version allows for an arbitrary number of both.
Tests are included in misc/cgo/testcdefs and the all.bash script has been
updated to account for these.
R=golang-dev, bradfitz, dave, iant
CC=golang-dev
https://golang.org/cl/11377043
Revision 15629 (8d71734a0cb0) removed the serverConn interface
that was introduce in revision 7718 (ee5e80c62862). The
serverConn interface was there for use by gccgo on Solaris,
and it is still needed there. Solaris does not support
connecting to the syslog daemon over TCP, and gccgo simply
calls the C library function. This CL restores the
interface.
R=golang-dev, bradfitz, rsc
CC=golang-dev
https://golang.org/cl/11737043
Currently Darwin and FreeBSD support and NetBSD and OpenBSD do not
support EV_RECEIPT flag. We will drop use of EV_RECEIPT for now.
Also enables to build runtime-integrated network pollster on
freebsd/amd64,386 and openbsd/amd64,386. It just does build but never
runs pollster stuff.
This is in preparation for runtime-integrated network pollster for BSD
variants.
Update #5199
R=dvyukov, minux.ma
CC=golang-dev
https://golang.org/cl/11759044
Debugging the Windows breakage I noticed that SEH
only exists on 386, so we can balance the two stacks
a little more on amd64 and reclaim another word.
Now we're down to just one word consumed by
cgocallback_gofunc, having reclaimed 25% of the
overall budget (4 words out of 16).
Separately, fix windows/386 - the SEH must be on the
m0 stack, as must the saved SP, so we are forced to have
a three-word frame for 386. It matters much less for
386, because there 128 bytes gives 32 words to use.
R=dvyukov, alex.brainman
CC=golang-dev
https://golang.org/cl/11551044
The existing function, IsOneOf, is hard to use. Since the slice comes
before the rune, in parallelism with the other Is functions, the slice
is clumsy to build. This CL adds a nicer-signatured In function of
equivalent functionality (its implementation is identical) that's much
easier to use. Compare:
unicode.IsOneOf([]*unicode.RangeTable{unicode.Letter, unicode.Number}, r)
unicode.In(r, unicode.Letter, unicode.Number)
R=golang-dev, adg
CC=golang-dev
https://golang.org/cl/11672044
Tying preemption to stack splits means that we have to able to
complete the call to exitsyscall (inside cgocallbackg at least for now)
without any stack split checks, meaning that the whole sequence
has to work within 128 bytes of stack, unless we increase the size
of the red zone. This CL frees up 24 bytes along that critical path
on amd64. (The 32-bit systems have plenty of space because all
their words are smaller.)
R=dvyukov
CC=golang-dev
https://golang.org/cl/11676043
When making an HTTPS client request, respect the
ServerName field in the tls.Config.
Fixes#5829
R=golang-dev, agl, adg
CC=golang-dev
https://golang.org/cl/11691043
Phrases like "returns whether or not the image is opaque" could be
describing what the function does (it always returns, regardless of
the opacity) or what it returns (a boolean indicating the opacity).
Even when the "or not" is missing, the phrasing is bizarre.
Go with "reports whether", which is still clunky but at least makes
it clear we're talking about the return value.
These were edited by hand. A few were cleaned up in other ways.
R=golang-dev, dsymonds
CC=golang-dev
https://golang.org/cl/11699043
Change use of x+(SP) to access the stack frame into x-(SP)
Fixes#5925.
R=golang-dev, bradfitz, dave, remyoudompheng, nick, rsc
CC=dave cheney <dave, golang-dev
https://golang.org/cl/11647043
src/cmd/ld/lib.c:1379 set and not used: p
src/cmd/ld/lib.c:1426 format mismatch 6llux INT, arg 3
src/cmd/ld/lib.c:1437 format mismatch 6llux INT, arg 3
src/cmd/ld/lib.c:1456 format mismatch 6llux INT, arg 3
src/cmd/ld/lib.c:1477 format mismatch 6llux INT, arg 3
src/cmd/ld/lib.c:1459 set and not used: started
R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/11615044
notetsleepg is the same as notetsleep, but is called on user g.
It includes entersyscall/exitsyscall and will help to avoid
split stack functions in syscall status.
R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/11681043
Calls into math/rand are very slow, especially under race
detector because of heap accesses.
go test -bench . -run none -benchtime .1s
Before: 23.0s
After: 17.4s
Fixes#5837.
R=golang-dev, dave, r
CC=golang-dev
https://golang.org/cl/11564044
This does not change the default compiler on OS X to clang.
It appears that for now we can keep using gcc as long as we
enable a few more warning settings that are on-by-default
elsewhere.
R=golang-dev, bradfitz, dave
CC=golang-dev
https://golang.org/cl/11610044
This CL introduces a FUNCDATA number for runtime-specific
garbage collection metadata, changes the C and Go compilers
to emit that metadata, and changes the runtime to expect it.
The old pseudo-instructions that carried this information
are gone, as is the linker code to process them.
R=golang-dev, dvyukov, cshapiro
CC=golang-dev
https://golang.org/cl/11406044
So far no checked-in assembly needs these,
but it matches having them for M and G.
I needed these for some manual testing.
R=golang-dev, dvyukov
CC=golang-dev
https://golang.org/cl/11595043
whose argument size is unknown (C vararg functions, and
assembly code without an explicit specification).
We used to use 0 to mean "unknown" and 1 to mean "zero".
Now we use ArgsSizeUnknown (0x80000000) to mean "unknown".
R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/11590043
If the network is not polled for 10ms, sysmon starts polling network
on every iteration (every 20us) until another thread blocks in netpoll.
Fixes#5922.
R=golang-dev, iant
CC=golang-dev
https://golang.org/cl/11569043
It is an expensive test to run, and even more so with -race,
and causes timeouts on builders. It is doubtful that it would
find a race that other tests in this package wouldn't, so there
is little loss in excluding it.
Update #5837.
R=golang-dev, dave
CC=golang-dev
https://golang.org/cl/11568043
It assumes that the m will not change, and the m may
change if the goroutine is preempted.
R=golang-dev, r
CC=golang-dev
https://golang.org/cl/11560043
If we start a garbage collection on g0 during a
stack split or unsplit, we'll see morestack or lessstack
at the top of the stack. Record an argument frame size
for those, and record that they terminate the stack.
R=golang-dev, dvyukov
CC=golang-dev
https://golang.org/cl/11533043
Deferreturn is synthesizing a new call frame.
It must not be interrupted between copying the args there
and fixing up the program counter, or else the stack will
be in an inconsistent state, one that will confuse the
garbage collector.
R=golang-dev, dvyukov
CC=golang-dev
https://golang.org/cl/11522043
With preemption, _sfloat2 can show up in stack traces.
Write the function prototype in a way that accurately
shows the frame size and the fact that it might contain
pointers.
R=golang-dev, dvyukov
CC=golang-dev
https://golang.org/cl/11523043
Windows was the only one seeing this bug reliably in the builder,
but it was easy to reproduce using 'GOGC=1 go test strconv'.
concatstring looked like it took only one string, but in fact it
takes a long list of strings. Add an explicit ... so that the traceback
will not use the "fixed" frame size and instead look at the
frame size metadata recorded by the caller.
R=golang-dev
TBR=golang-dev
CC=golang-dev
https://golang.org/cl/11531043
This should fix the Windows build, or at least
what's breaking it at the moment.
Fixes#5904.
TBR=golang-dev
CC=golang-dev
https://golang.org/cl/11519044
Be consistent with os.File, strings.Reader, bytes.Reader, etc,
which all allow seeks past the end.
R=golang-dev, r
CC=golang-dev
https://golang.org/cl/11403043
Otherwise the tests in pkg/runtime fail:
runtime: unknown argument frame size for runtime.deferreturn called from 0x48657b [runtime_test.func·022]
fatal error: invalid stack
...
R=golang-dev, dave
CC=golang-dev
https://golang.org/cl/11483043
Update #543
I believe the runtime is strong enough now to reenable
preemption during the function prologue.
Assuming this is or can be made stable, it will be in Go 1.2.
More aggressive preemption is not planned for Go 1.2.
R=golang-dev, iant
CC=golang-dev
https://golang.org/cl/11433045
Currently preemption signal g->stackguard0==StackPreempt
can be lost if it is received when preemption is disabled
(e.g. m->lock!=0). This change duplicates the preemption
signal in g->preempt and restores g->stackguard0
when preemption is enabled.
Update #543.
R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/10792043
With this CL, I believe the runtime always knows
the frame size during the gc walk. There is no fallback
to "assume entire stack frame of caller" anymore.
R=golang-dev, khr, cshapiro, dvyukov
CC=golang-dev
https://golang.org/cl/11374044
Add support for ECDHE-ECDSA (RFC4492), which uses an ephemeral server
key pair to perform ECDH with ECDSA signatures. Like ECDHE-RSA,
ECDHE-ECDSA also provides PFS.
R=agl
CC=golang-dev
https://golang.org/cl/7006047
clearfat (used to zero initialize structures) will use AX for x86 block ops. If we write to AX while calculating the dest pointer, we will fill the structure with incorrect values.
Since 64-bit arithmetic uses AX to synthesize a 64-bit register, getting an adress by indexing with 64-bit ops can clobber the register.
Fixes#5820.
R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/11383043
If calling a function in package runtime, emit argument size
information around the call in case the call is to a variadic C function.
R=ken2
CC=golang-dev
https://golang.org/cl/11371043
I have not done the system call stubs in sys_*.s.
I hope to avoid that, because those do not block, so those
frames will not appear in stack traces during garbage
collection.
R=golang-dev, dvyukov, khr
CC=golang-dev
https://golang.org/cl/11360043
While we're here, fix Syscall9 on NetBSD and OpenBSD:
it was storing the results into the wrong memory locations.
I guess no one uses that function's results on those systems.
Part of cleaning up stack traces and argument frame information.
R=golang-dev, dvyukov
CC=golang-dev
https://golang.org/cl/11355044
The portable code in cmd/ld already knows how to process it,
we just have to ignore it during code generation.
R=ken2
CC=golang-dev
https://golang.org/cl/11363043
Design at http://golang.org/s/go12symtab.
This enables some cleanup of the garbage collector metadata
that will be done in future CLs.
This CL does not move the old symtab and pclntab back into
an unmapped section of the file. That's a bit tricky and will be
done separately.
Fixes#4020.
R=golang-dev, dave, cshapiro, iant, r
CC=golang-dev, nigeltao
https://golang.org/cl/11085043
Race instrumentation can allocate, switch stacks, preempt, etc.
All that is not allowed in between fork and exec.
Fixes#4840.
R=golang-dev, daniel.morsing, dave
CC=golang-dev
https://golang.org/cl/11324044
A type switch on a value with map index expressions,
could get a spurious instrumentation from a OTYPESW node.
These nodes do not need instrumentation because after
walk the type switch has been turned into a sequence
of ifs.
Fixes#5890.
R=golang-dev, dvyukov
CC=golang-dev
https://golang.org/cl/11308043
"M requires pointer receiver" can be misinterpreted to
mean that method M should have a pointer receiver but
does not. In fact the message means "M has a pointer
receiver" (and you don't have a pointer).
Fixes#5891.
R=ken2
CC=golang-dev
https://golang.org/cl/11313043
Sets both the duration from the last data packet to the first
keep alive packet and the duration between keep alive packets to be
the passed duration.
I've tested the function on both Darwin (10.8.4) and 4.2 Linux.
I've compiled (make.bash) for all the OS's and tested (all.bash)
on Darwin and Linux.
R=golang-dev, dave, rsc, dvyukov, presotto+facebook, nick
CC=golang-dev, veyron-team
https://golang.org/cl/11130044
Before:
$ go test -c -cover fmt
$ ./fmt.test -test.covermode=set
PASS
coverage: 65.1% of statements in strconv
$
After:
$ go test -c -cover fmt
$ ./fmt.test
PASS
coverage: 65.1% of statements in strconv
$
In addition to being cumbersome, the old flag didn't make sense:
the cover mode cannot be changed after the binary has been built.
Another useful effect of this CL is that if you happen to do
$ go test -c -covermode=atomic fmt
and then forget you did that and run benchmarks,
the final line of the output (the coverage summary) reminds you
that you are benchmarking with coverage enabled, which might
not be what you want.
$ ./fmt.test -test.bench .
PASS
BenchmarkSprintfEmpty 10000000 217 ns/op
BenchmarkSprintfString 2000000 755 ns/op
BenchmarkSprintfInt 2000000 774 ns/op
BenchmarkSprintfIntInt 1000000 1363 ns/op
BenchmarkSprintfPrefixedInt 1000000 1501 ns/op
BenchmarkSprintfFloat 1000000 1257 ns/op
BenchmarkManyArgs 500000 5346 ns/op
BenchmarkScanInts 1000 2562402 ns/op
BenchmarkScanRecursiveInt 500 3189457 ns/op
coverage: 91.4% of statements
$
As part of passing the new mode setting in via _testmain.go, merge
the two registration mechanisms into one extensible mechanism
(a struct).
R=r
CC=golang-dev
https://golang.org/cl/11219043
I want to think more carefully about this.
We put this in because Marshal encoded named []byte but Unmarshal rejected them.
And we noticed that Marshal's behavior was undocumented so we documented it.
But I am starting to think the docs and Unmarshal were correct and Marshal's
behavior was the problem.
Rolling back to give us more time to think.
««« original CL description
json: unmarshal types that are byte slices.
The json package cheerfully would marshal
type S struct {
IP net.IP
}
but would give an error when unmarshalling. This change allows any
type whose concrete type is a byte slice to be unmarshalled from a
string.
Fixes#5086.
R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/11161044
»»»
R=golang-dev, r
CC=golang-dev
https://golang.org/cl/11042046
In practice, rejecting an entire structure due to a single invalid byte
in a string is just too picky, and too hard to track down.
Be consistent with the bulk of the standard library by converting
invalid UTF-8 into UTF-8 with replacement runes.
R=golang-dev, crawshaw
CC=golang-dev
https://golang.org/cl/11211045
Recently addition to runtime test makes it take very close to 720s
of timeout limit on the netbsd-arm-qemu builder.
R=golang-dev, go.peter.90, rsc
CC=golang-dev
https://golang.org/cl/10935043
Deferred functions are not run by a call instruction. They are run by
the runtime editing registers to make the call start with a caller PC
returning to a
CALL deferreturn
instruction.
That instruction has always had the line number of the function's
closing brace, but that instruction's line number is irrelevant.
Stack traces show the line number of the instruction before the
return PC, because normally that's what started the call. Not so here.
The instruction before the CALL deferreturn could be almost anywhere
in the function; it's unrelated and its line number is incorrect to show.
Fix the line number by inserting a true hardware no-op with the right
line number before the returned-to CALL instruction. That is, the deferred
calls now appear to start with a caller PC returning to the second instruction
in this sequence:
NOP
CALL deferreturn
The traceback will show the line number of the NOP, which we've set
to be the line number of the function's closing brace.
The NOP here is not the usual pseudo-instruction, which would be
elided by the linker. Instead it is the real hardware instruction:
XCHG AX, AX on 386 and amd64, and AND.EQ R0, R0, R0 on ARM.
Fixes#5856.
R=ken2, ken
CC=golang-dev
https://golang.org/cl/11223043
If the stack frame size is larger than the known-unmapped region at the
bottom of the address space, then the stack split prologue cannot use the usual
condition:
SP - size >= stackguard
because SP - size may wrap around to a very large number.
Instead, if the stack frame is large, the prologue tests:
SP - stackguard >= size
(This ends up being a few instructions more expensive, so we don't do it always.)
Preemption requests register by setting stackguard to a very large value, so
that the first test (SP - size >= stackguard) cannot possibly succeed.
Unfortunately, that same very large value causes a wraparound in the
second test (SP - stackguard >= size), making it succeed incorrectly.
To avoid *that* wraparound, we have to amend the test:
stackguard != StackPreempt && SP - stackguard >= size
This test is only used for functions with large frames, which essentially
always split the stack, so the cost of the few instructions is noise.
This CL and CL 11085043 together fix the known issues with preemption,
at the beginning of a function, so we will be able to try turning it on again.
R=ken2
CC=golang-dev
https://golang.org/cl/11205043
This is a transcript before this change. I've capitalized the text being removed.
Note that it is always near another line that already says fmt, marked with <<<
$ cd $GOROOT/src/pkg/fmt
$ go test -cover
PASS
coverage FOR FMT: 91.3% of statements
ok fmt 0.040s <<<
$ go test -coverpkg strconv
PASS
coverage FOR FMT: 64.9% of statements in strconv
ok fmt 0.039s <<<
$ go test -cover -c
$ ./fmt.test -test.covermode=set <<<
PASS
coverage FOR FMT: 91.3% of statements
$ go test -coverpkg strconv -c
$ ./fmt.test -test.covermode=set <<<
PASS
coverage FOR FMT: 64.9% of statements in strconv
That the summary printed by 'go test [options] fmt' is unchanged:
$ go test -cover fmt
ok fmt 0.040s coverage: 91.3% of statements
$ go test -coverpkg strconv fmt
ok fmt 0.038s coverage: 64.9% of statements in strconv
R=r
CC=gobot, golang-dev
https://golang.org/cl/10932045
The current cas64 definition hard-codes the x86 behavior
of updating *old with the new value when the cas fails.
This is inconsistent with cas32 and casp.
Make it consistent.
This means that the cas64 uses will be epsilon less efficient
than they might be, because they have to do an unnecessary
memory load on x86. But so be it. Code clarity and consistency
is more important.
R=golang-dev, bradfitz
CC=golang-dev
https://golang.org/cl/10909045