Cherry Zhang [Tue, 11 Feb 2020 15:27:15 +0000 (10:27 -0500)]
[dev.link] cmd/link: add defined package symbols first, before any non-package symbols
Currently, the loader adds defined package symbols and
non-package symbols to the global index space object by object.
This CL changes it to add all the defined package symbols first,
then all the non-package symbols. The advantage of doing this is
that when adding package symbols, by definition they cannot be
dup to each other, so we don't need to do a name lookup when
adding them.
We still add them to the lookup table (for now), since they may
still be referenced by name (e.g. through linkname). This CL is
also a prerequisite if we want to move to not adding package
symbols to the lookup table entirely (e.g. by using
pre-generated in-file lookup table).
Also update some comments to reflect the current state.
Cherry Zhang [Fri, 7 Feb 2020 15:26:06 +0000 (10:26 -0500)]
[dev.link] cmd/link: only read necessary fields in deadcode pass
Reading unnecessary fields from object file take time, even from
memory mapping. In deadcode, we do this a alot, so only read
necessary ones. In particular, for relocations, for non-type
symbols, we only need their target symbols and reloc types to
build the dependency graph, so don't read other fields. This
should speed up the deadcode pass.
Also cache an object's number of defined package symbols, as it
is accessed a lot in resolve.
This is a bit ugly from an API point of view. It would be nice if
the compiler could figure out some fields are not used so don't
fill them.
Change-Id: I5c4e9526b8a3b0aead9fa71901a51fb214f013a7
Reviewed-on: https://go-review.googlesource.com/c/go/+/218479
Run-TryBot: Cherry Zhang <cherryyz@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Jeremy Faller <jeremy@golang.org> Reviewed-by: Than McIntosh <thanm@google.com>
Cherry Zhang [Wed, 5 Feb 2020 21:52:12 +0000 (16:52 -0500)]
[dev.link] cmd/link: clean up some fields
With the new index mapping , we can clean up some fields.
Loader.max should always be equal to len(loader.objSyms) now. And
for external symbols we now give its "local index" as its index
in the payload array, so Relocs.extIdx is no longer useful.
Delete those fields.
Dmitri Shuralyov [Thu, 6 Feb 2020 00:24:55 +0000 (19:24 -0500)]
doc/go1.14: disable text/template processing in HTML page
HTML pages served by the website have the option to opt-in to template
processing, by including "Template: true" in the page metadata.
This functionality is documented at
https://github.com/golang/tools/blob/403f1254bdfd3da27c92a0e9e37dd180a9a82b3c/godoc/template.go#L5-L30.
Historically, the Go 1 release notes have used template processing
to a great extent, but release notes for all subsequent major Go
releases have not.
Since this feature is generally not used and not very well known,
it tends to do more harm than good by making it possible for errors
in the template to prevent the release notes from showing up at all.
Disable this feature for Go 1.14 release notes and onwards.
We can consider enabling it when there's a stronger need for it.
Fixes #37072
Updates #37070
Change-Id: If93553d52df12544b46c4edcf3aa5eddc2a155ad
Reviewed-on: https://go-review.googlesource.com/c/go/+/218058 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Documents io/ioutil.TempDir's new ability to create
predictable prefixes and suffixes, derived from the argument
'pattern', separated by the last '*' in it.
References: CL 198488
Updates #36878
Change-Id: I92c52fcc7d480ce74746e99e6e85a04efb87294f
Reviewed-on: https://go-review.googlesource.com/c/go/+/217780 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Cherry Zhang [Mon, 3 Feb 2020 17:37:35 +0000 (12:37 -0500)]
[dev.link] cmd/link: remove holes from global index space
In CL 217064, we made symbol's global index unique, but we still
reserve index space for each object file, which means we may
leave holes in the index space if the symbol is a dup or is
overwritten. In this CL, we stop reserving index spaces. Instead,
symbols are added one at a time, and only added if it does not
already exist. There is no more holes in the index space.
Change-Id: I3c4e67163c556ba1198e13065706510dac4692fb
Reviewed-on: https://go-review.googlesource.com/c/go/+/217519 Reviewed-by: Than McIntosh <thanm@google.com> Reviewed-by: Jeremy Faller <jeremy@golang.org>
Cherry Zhang [Sat, 1 Feb 2020 22:04:34 +0000 (17:04 -0500)]
[dev.link] cmd/link: remove sym.Symbol based host object support
We have converted all the host object readers to the new loader
based mechanism, and no longer do sym.Symbol host object loading.
Remove that support. So we don't have to take care of that in
future changes to the loader.
TODO: there are still code outside the loader than can be
removed. This CL only focuses on the loader.
Change-Id: I67dd75f631964921a7c7a6576ed95b071241484a
Reviewed-on: https://go-review.googlesource.com/c/go/+/217518 Reviewed-by: Than McIntosh <thanm@google.com> Reviewed-by: Jeremy Faller <jeremy@golang.org>
Emmanuel T Odeke [Wed, 5 Feb 2020 19:09:34 +0000 (11:09 -0800)]
doc/go1.14: document http.ServeFile large file fix for Windows
Document that for Windows, net/http.ServeFile can now
correctly serve files >2GB after we fixed internal/poll.SendFile
to transmit files larger than the default Windows limit of 2GB.
References: CL 192518, CL 194218
Updates #36878
Change-Id: Ibefc8b2841bc0cee3a89884a680085f99d2b6928
Reviewed-on: https://go-review.googlesource.com/c/go/+/217779 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Bryan C. Mills [Wed, 5 Feb 2020 15:22:07 +0000 (10:22 -0500)]
doc/install.html: streamline the “Test your installation” step and make it module-agnostic
In CL 199417, we updated “How to Write Go Code” to give a basic
introduction to modules and to include module-mode commands.
However, most new users will end up reading “Getting Started”
(doc/install.html) before “How to Write Go Code”, and we forgot to
update the handful of commands there for module mode.
Before this change, the “Test your installation” section also covered
quite a few operations beoyond merely testing the installation: it
included setting up a GOPATH, building a binary, and installing and
cleaning binaries. Those are valuable operations to learn, but they
arguably belong in “How to Write Go Code”, not “Test your
installation” — and having all that extra detail in the install
instructions may well discourage folks from further essential reading.
Rather than updating all of those operations here, I've removed them.
A companion CL will update “How to Write Go Code” to ensure that it
mentions GOPATH (as the location of the module cache and the default
install location for binaries) and 'go clean -i'.
Updates #37042
Change-Id: I157f21ccbe3896575fa1115dc821abf6c71ed15e
Reviewed-on: https://go-review.googlesource.com/c/go/+/217840
Run-TryBot: Bryan C. Mills <bcmills@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org> Reviewed-by: Tyler Bui-Palsulich <tbp@google.com>
Dmitri Shuralyov [Wed, 5 Feb 2020 10:23:42 +0000 (05:23 -0500)]
doc: rename HTML element IDs to avoid duplicates
These 3 release notes have had an element ID collision because both
the runtime changes and changes to the package "runtime" used the
same ID. Fix it by using a "pkg-" prefix for the runtime package.
Move the "runtime-again" ID from CL 129635 to a nearby <dt> element
so that existing links to https://golang.org/doc/go1.11#runtime-again
don't break.
Fixes #37036
Updates #36878
Change-Id: Ib68d93acfac802fd84c0a57485937e45dea2064a
Reviewed-on: https://go-review.googlesource.com/c/go/+/217797 Reviewed-by: Toshihiro Shiino <shiino.toshihiro@gmail.com> Reviewed-by: Agniva De Sarker <agniva.quicksilver@gmail.com> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Ian Lance Taylor [Mon, 3 Feb 2020 23:53:53 +0000 (15:53 -0800)]
runtime: don't treat SIGURG as a bad signal
It's possible for the scheduler to try to preempt a goroutine running
on a thread created by C code just as the goroutine returns from Go code
to C code. If that happens, the goroutine will have a nil g,
which would normally cause us to enter the badsignal code.
The badsignal code will allocate an M, reset the signal handler,
and raise the signal. This is all wasted work for SIGURG,
as the default behavior is for the kernel to ignore the signal.
It also means that there is a period of time when preemption requests
are ignored, because the signal handler is reset to the default.
And, finally, it triggers a bug on 386 OpenBSD 6.2. So stop doing it.
No test because there is no real change in behavior (other than on OpenBSD),
the new code is just more efficient
Katie Hockman [Tue, 4 Feb 2020 17:48:20 +0000 (12:48 -0500)]
doc: remove paragraph break for upgrading to modules
Previously, the release notes broke up the sentences that modules
is now ready for production use and where to file issues for
migration problems into separate paragraphs. This made it look like
the migration paragraph was about upgrading to 1.14, not to modules,
and made the reading a bit confusing. Now the entire idea is in one
paragraph.
Dmitri Shuralyov [Tue, 4 Feb 2020 15:24:20 +0000 (15:24 +0000)]
doc/go1.14: remove TODO about Solaris port
The solaris-amd64-oraclerel builder is passing for the main Go repo
(on tip and release branches for 1.13 and 1.12), and golang.org/x repos
(also on tip and release branches for 1.13 and 1.12).
The builder is still maintained as described at
https://golang.org/issue/15581#issuecomment-550368581.
Updates #36878
Updates #15581
Change-Id: Icc6f7529ca2e05bb34f09ce4363d9582e80829c6
Reviewed-on: https://go-review.googlesource.com/c/go/+/217738 Reviewed-by: Ian Lance Taylor <iant@golang.org> Reviewed-by: Alexander Rakoczy <alex@golang.org>
Dmitri Shuralyov [Tue, 4 Feb 2020 15:15:39 +0000 (15:15 +0000)]
doc/go1.14: remove TODO about Illumos port
There is an active builder that was added in CL 201597,
and it is passing on Go tip and release-branch.go1.13
(with one failure that appears to be flaky due to being
out of memory). It's also passing on all golang.org/x repos
on tip and release-branch.go1.13. It's not configured to
run on Go 1.12 release branches.
Updates #36878
Updates #15581
Change-Id: I4ed7fc62c11a09743832fca39bd61fa0cf6e7ded
Reviewed-on: https://go-review.googlesource.com/c/go/+/217737 Reviewed-by: Ian Lance Taylor <iant@golang.org> Reviewed-by: Alexander Rakoczy <alex@golang.org>
Tobias Klauser [Sun, 2 Feb 2020 13:43:54 +0000 (14:43 +0100)]
doc/go1.14: remove TODO about Dragonfly passing
Both the Dragonfly release and tip builder have been passing for a
while. The net package's interface API is working on both builders since
CL 202317 which has been re-vendored in CL 202438.
Updates #34368
Updates #36878
Change-Id: I187178b3a59f2604187af453207fb4e24a56105c
Reviewed-on: https://go-review.googlesource.com/c/go/+/217358 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Toshihiro Shiino [Tue, 4 Feb 2020 12:32:58 +0000 (12:32 +0000)]
doc/go1.14: add missing slashes
This saves a redirect and makes the document more consistent.
Change-Id: I67840f5dc05dffd8893a055618eb202b682a0ebc
Reviewed-on: https://go-review.googlesource.com/c/go/+/217698 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Filippo Valsorda [Fri, 31 Jan 2020 23:41:03 +0000 (15:41 -0800)]
std,cmd: sync go.mod with new release branches
cd src
go get golang.org/x/sys@release-branch.go1.14-std
go mod tidy && go mod vendor
cd cmd
go get golang.org/x/sys@release-branch.go1.14-cmd
go get golang.org/x/tools@release-branch.go1.14
go mod tidy && go mod vendor
Fixes #36851
Change-Id: Ib0a78eec90c92515b6f421972102cc8e8d0c08a7
Reviewed-on: https://go-review.googlesource.com/c/go/+/217305
Run-TryBot: Filippo Valsorda <filippo@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Filippo Valsorda [Fri, 31 Jan 2020 22:49:11 +0000 (14:49 -0800)]
math/big: simplify GCD docs
We don't usually document past behavior (like "As of Go 1.14 ...") and
in isolation the current docs made it sound like a and b could only be
negative or zero.
Change-Id: I0d3c2b8579a9c01159ce528a3128b1478e99042a
Reviewed-on: https://go-review.googlesource.com/c/go/+/217302 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Filippo Valsorda [Fri, 31 Jan 2020 22:54:21 +0000 (14:54 -0800)]
doc/go1.14: fix math/big.(*Int).GCD notes
GCD is a method, not a function, so the link was broken.
Change-Id: Icbb09d39959e7c71a48987b15bb82febe12c3a19
Reviewed-on: https://go-review.googlesource.com/c/go/+/217303 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Cherry Zhang [Tue, 28 Jan 2020 23:18:58 +0000 (18:18 -0500)]
[dev.link] cmd/link: make symbol's global index unique
Currently, when mapping symbol's local index to global index, for
duplicated and overwriting/overwritten symbols, each appearance
of the symbol gets a global index, with one being the "primary",
and others "redirect" to it through the overwrite map. Basically,
the local-global index mapping is one to one, with overwrite/
dedup happening in global index level.
This has a few drawbacks:
- All symbol accesses effectively need to query the overwrite
map. This may hurt performance.
- For multi-level overwrites, (Y overwrites X, Z overwrites Y),
this can get quite complicated, and we have to follow the
redirection recursively.
- Failed to follow or to update the overwrite map leads to bugs.
In this CL, we change the index mapping mechanism so that each
symbol get a unique global index. Multiple appearances of the
same symbol get the same index. Now the local-global index
mapping is N to one. Overwrite/dedup happens directly in the
local-global mapping.
We keep both mapping directions in arrays. Each object carries
an array for its local-global mapping. The loader carries an
array mapping global index to the "primary" local index, which is
the one we should load from. This way, we can get rid of the
overwrite map, and index conversions are simply array accesses.
TODO: we still make reservation of the index space upfront, and
leave holes for dup symbols. Maybe get rid of the reservation and
holes.
Filippo Valsorda [Tue, 28 Jan 2020 22:37:58 +0000 (14:37 -0800)]
doc/go1.14: add crypto/tls release notes
A few minor changes that didn't feel worth mentioning:
* CL 205059: support leaving Certificates/GetCertificate nil if
GetConfigForClient is set
* CL 205059: send the unrecognized_name alert when there are no
available certificates
* CL 205068: add correct names for CHACHA20_POLY1305 ciphersuite constants
* CL 204046: fix CreateCRL for Ed25519 CAs
* CL 205058: add CertificateRequestInfo.Version
Change-Id: Ie820fb0c6842e669edde031132c7cda5b09e7682
Reviewed-on: https://go-review.googlesource.com/c/go/+/216759
Run-TryBot: Filippo Valsorda <filippo@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Cherry Zhang [Fri, 31 Jan 2020 19:45:52 +0000 (14:45 -0500)]
[dev.link] all: merge branch 'master' into dev.link
It has been a while we have not done this.
Merge conflict resolution:
- deleted/rewritten code modified on master
- CL 214286, ported in CL 217317
(cmd/internal/obj/objfile.go)
- CL 210678, it already includes a fix to new code
(cmd/link/internal/ld/deadcode.go)
- CL 209317, applied in this CL
(cmd/link/internal/loadelf/ldelf.go)
Dmitri Shuralyov [Fri, 31 Jan 2020 18:21:57 +0000 (13:21 -0500)]
doc/go1.14: move "Minor changes to the library" heading up
This heading was below the minor changes to the standard library.
It should be on top.
Many of the minor changes have been documented, so remove the broad
TODO comment. It is still a TODO to highlight more prominent changes
to the library, if there are any, under the "Core library" heading.
Updates #36878
Change-Id: If7fd9af9a933af917523e33fd2922c5f3c02c98b
Reviewed-on: https://go-review.googlesource.com/c/go/+/217277 Reviewed-by: Alexander Rakoczy <alex@golang.org> Reviewed-by: Carlos Amedee <carlos@golang.org>
Cherry Zhang [Thu, 30 Jan 2020 03:10:51 +0000 (22:10 -0500)]
[dev.link] cmd/link: fix payload pointer liveness
Currently, the symbol updater uses a pointer pointing to the
loader's payloads array. If the payloads slice grows (and moves),
the pointer may become stale and no longer point to the symbol's
actual payload. Specifically, consider
sb, sym := l.MakeSymbolUpdater(...)
// add a bunch of external symbols, which grows payload slice
sb.SetType(t)
l.SymType(sym) // may not return t
sb.SetType on line 3 may not have the desired effect, as
sb.extSymPayload may no longer point to the right payload. As a
result, the type we get on line 4 may be not the one we set.
Fix this by making the payload's address permanent. Once it is
allocated it will never move.
Robert Griesemer [Fri, 31 Jan 2020 04:13:02 +0000 (20:13 -0800)]
go/types: unexport Checker.LookupFieldOrMethod
Implementation changes in go/types for #6977 required that internal
LookupFieldOrMethod calls had access to the current *Checker. In
order to make quick progress, I added a *Checker receiver to the
function LookupFieldOrMethod (thus making it a method), and added
a new function LookupFieldOrMethod. The plan was always to rename
that function (Checker.LookupFieldOrMethod) such that it wouldn't
be exported; with the obvious name being Checker.lookupFieldOrMethod.
But that name was already in use which is why I postponed the rename.
Eventually I forgot to clean it up. This CL fixes that with the
following renames:
Change-Id: Icfafd0de9a19841ba5bd87142730fe7323204491
Reviewed-on: https://go-review.googlesource.com/c/go/+/217134
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Daniel Martí <mvdan@mvdan.cc> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Keith Randall [Thu, 30 Jan 2020 18:49:12 +0000 (10:49 -0800)]
doc: document new hash/maphash package
Update #36878
Update #28322
Change-Id: I793c7c4dbdd23fdecd715500e90b7cc0cbe4cea5
Reviewed-on: https://go-review.googlesource.com/c/go/+/217099 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Filippo Valsorda [Thu, 30 Jan 2020 22:18:42 +0000 (14:18 -0800)]
cmd/go.mod: sync x/crypto with std
go get golang.org/x/crypto@v0.0.0-20200128174031-69ecbb4d6d5d
go mod vendor
git checkout -- vendor/golang.org/x/sys/unix/asm_linux_riscv64.s \
vendor/golang.org/x/tools/go/analysis/passes/asmdecl/asmdecl.go
Robert Griesemer [Thu, 30 Jan 2020 20:26:20 +0000 (12:26 -0800)]
math/big: update comment on Int.GCD
Per the suggestion https://golang.org/cl/216200/2/doc/go1.14.html#423.
Updates #28878.
Change-Id: I654d2d114409624219a0041916f0a4030efc7573
Reviewed-on: https://go-review.googlesource.com/c/go/+/217104 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Robert Griesemer [Thu, 30 Jan 2020 19:34:24 +0000 (11:34 -0800)]
doc/go1.14: fix minor typo (update release notes)
Follow-up on https://golang.org/cl/216200/2/doc/go1.14.html#423 .
Updates #36878.
Change-Id: I693a9eb05c6f1f42721a92fda46a4f3449defa24
Reviewed-on: https://go-review.googlesource.com/c/go/+/217100 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Change-Id: I40594be85ee0a0d4b35bacc90104568d2b8a4761
Reviewed-on: https://go-review.googlesource.com/c/go/+/216997 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Hana Kim [Wed, 29 Jan 2020 22:09:28 +0000 (17:09 -0500)]
doc/go1.14.html: describe the runtime/pprof change
golang.org/cl/204636
golang.org/cl/205097
Updates #36874
Change-Id: I773868fd027e9cc2187f0a738900f0fcb7711635
Reviewed-on: https://go-review.googlesource.com/c/go/+/216877 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Bryan C. Mills [Wed, 29 Jan 2020 14:14:50 +0000 (09:14 -0500)]
go/build: update TestImportDirNotExist to accept more detailed error strings
In CL 203820, we switched go/build to use the caller's working
directory for the main module (rather than srcDir), so that go/build
resolution now respects the requirements and replacements of the main
module. When the passed-in srcDir is empty, as of that CL we use "go
list" instead of falling back to in-process (GOPATH-mode) path lookup.
Unfortunately, that broke go/build.TestImportDirNotExist when
GO111MODULE=on: the test was looking for the specific error message
produced by the in-process lookup.
This change relaxes the test to accept the error message produced by
"go list" when srcDir is empty.
Joel Sing [Wed, 29 Jan 2020 09:14:18 +0000 (20:14 +1100)]
cmd/internal/obj/riscv,cmd/link: shorten the riscv64 call sequence
Now that the other dependent offset has been identified, we can remove the
unnecessary ADDI instruction from the riscv64 call sequence (reducing it
to AUIPC+JALR, rather than the previous AUIPC+ADDI+JALR).
Joel Sing [Wed, 29 Jan 2020 09:06:54 +0000 (20:06 +1100)]
cmd/compile,cmd/link: fix and re-enable open-coded defers on riscv64
The R_CALLRISCV relocation marker is on the JALR instruction, however the actual
relocation is currently two instructions previous for the AUIPC+ADDI sequence.
Adjust the platform dependent offset accordingly and re-enable open-coded defers.
Fixes #36786.
Change-Id: I71597c193c447930fbe94ce44b7355e89ae877bb
Reviewed-on: https://go-review.googlesource.com/c/go/+/216797
Run-TryBot: Joel Sing <joel@sing.id.au>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Cherry Zhang <cherryyz@google.com>
Daniel Martí [Wed, 22 Jan 2020 12:20:02 +0000 (12:20 +0000)]
doc: add the change to json.Compact in the 1.14 changelog
Fixes #36690.
Change-Id: Id4234ab9467270d51f0411375b71ece7f41269b2
Reviewed-on: https://go-review.googlesource.com/c/go/+/215817 Reviewed-by: Ian Lance Taylor <iant@golang.org> Reviewed-by: Jay Conrod <jayconrod@google.com>
Brad Fitzpatrick [Tue, 28 Jan 2020 21:40:08 +0000 (21:40 +0000)]
doc/go1.14: document RISC-V support
Fixes #36708
Updates #27532
Change-Id: I9e3bb92d15825e2c4a505e1aea41b2897f18e0ff
Reviewed-on: https://go-review.googlesource.com/c/go/+/216757 Reviewed-by: Joel Sing <joel@sing.id.au> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Ian Lance Taylor [Tue, 28 Jan 2020 00:35:28 +0000 (16:35 -0800)]
net/http: don't treat an alternate protocol as a known round tripper
As of CL 175857, the client code checks for known round tripper
implementations, and uses simpler cancellation code when it finds one.
However, this code was not considering the case of a request that uses
a user-defined protocol, where the user-defined protocol was
registered with the transport to use a different round tripper.
The effect was that round trippers that worked with earlier
releases would not see the expected cancellation semantics with tip.
Fixes #36820
Change-Id: I60e75b5d0badcfb9fde9d73a966ba1d3f7aa42b1
Reviewed-on: https://go-review.googlesource.com/c/go/+/216618
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Brad Fitzpatrick [Tue, 28 Jan 2020 22:57:18 +0000 (22:57 +0000)]
cmd/dist: remove riscv64 from set of incomplete ports
Fixes #27532
Fixes #36853
Updates #28944
Change-Id: I4d0f212deb361c941ce7e5999e237a951c89a296
Reviewed-on: https://go-review.googlesource.com/c/go/+/216758 Reviewed-by: Josh Bleecher Snyder <josharian@gmail.com> Reviewed-by: Ian Lance Taylor <iant@golang.org> Reviewed-by: Joel Sing <joel@sing.id.au>
Michael Anthony Knyszek [Tue, 28 Jan 2020 19:59:19 +0000 (19:59 +0000)]
runtime: ensure that searchAddr always refers to inUse memory
This change formalizes an assumption made by the page allocator, which
is that (*pageAlloc).searchAddr should never refer to memory that is not
represented by (*pageAlloc).inUse. The portion of address space covered
by (*pageAlloc).inUse reflects the parts of the summary arrays which are
guaranteed to mapped, and so looking at any summary which is not
reflected there may cause a segfault.
In fact, this can happen today. This change thus also removes a
micro-optimization which is the only case which may cause
(*pageAlloc).searchAddr to point outside of any region covered by
(*pageAlloc).inUse, and adds a test verifying that the current segfault
can no longer occur.
Dmitri Shuralyov [Tue, 28 Jan 2020 18:20:57 +0000 (13:20 -0500)]
src/go.mod: import x/crypto/cryptobyte security fix for 32-bit archs
cryptobyte: fix panic due to malformed ASN.1 inputs on 32-bit archs
When int is 32 bits wide (on 32-bit architectures like 386 and arm), an
overflow could occur, causing a panic, due to malformed ASN.1 being
passed to any of the ASN1 methods of String.
Tested on linux/386 and darwin/amd64.
This fixes CVE-2020-7919 and was found thanks to the Project Wycheproof
test vectors.
x/crypto/cryptobyte is used in crypto/x509 for parsing certificates.
Malformed certificates might cause a panic during parsing on 32-bit
architectures (like arm and 386).
Joel Sing [Wed, 22 Jan 2020 16:38:31 +0000 (03:38 +1100)]
cmd/internal/obj/riscv: use signed immediates for U-instructions
On RISCV64, the U-instructions (AUIPC and LUI) take 20 bits, append 12 bits
of zeros and sign extend to 64-bits. As such, the 20 bit immediate value is
signed not unsigned.
Duplicate the register definitions and names to avoid importing the
cmd/internal/obj/riscv64 package. This makes it possible to build compiler rules
with a stable Go tool chain.
Cherry Zhang [Fri, 24 Jan 2020 19:11:04 +0000 (14:11 -0500)]
cmd/compile: on PPC64, fold offset into some loads/stores only when offset is 4-aligned
On PPC64, MOVWload, MOVDload, and MOVDstore are assembled to a
"DS from" instruction which requiers the offset is a multiple of
4. Only fold offset to such instructions if it is a multiple of 4.
Fixes #36723.
"GOARCH=ppc64 GOOS=linux go build -gcflags=all=-d=ssa/check/on std cmd"
passes now.
Change-Id: I67f2a6ac02f0d33d470f68ff54936c289a4c765b
Reviewed-on: https://go-review.googlesource.com/c/go/+/216379 Reviewed-by: Carlos Eduardo Seo <cseo@linux.vnet.ibm.com>
Bryan C. Mills [Fri, 24 Jan 2020 15:26:24 +0000 (10:26 -0500)]
doc/go1.14: note that module support is ready for production use
The public proxy and checksum database launched with Go 1.13 have been
running smoothly, pkg.go.dev is serving module-aware documentation,
and in 1.14 we have improved the vendoring workflow and finished
Subversion support to reach parity with GOPATH mode for users of those
features, updated documentation (including the “How to Write Go Code”
intro document) and published blog posts describing common modes of
usage, and improved the migration path for existing v2+ modules by
making version resolution less aggressive about "+incompatible" major
versions.
We (always) have more fit-and-finish work to do, but at this point we
believe that module mode will provide a better user experience than
GOPATH mode for most users, including in production use.