Keith Randall [Mon, 13 Sep 2021 16:17:22 +0000 (09:17 -0700)]
reflect: fix _faststr optimization
CL 345486 introduced an optimization to reflect's map accesses
which is not quite correct. We can't use the optimized code if the
value type is >128 bytes.
See cmd/compile/internal/walk/walk.go:mapfast
Fixes #48357
Change-Id: I8e3c7858693083dd4393a8de48ca5fa47bab66f2
Reviewed-on: https://go-review.googlesource.com/c/go/+/349593
Trust: Keith Randall <khr@golang.org>
Trust: Joe Tsai <joetsai@digital-static.net>
Trust: Josh Bleecher Snyder <josharian@gmail.com>
Trust: Martin Möhrmann <martin@golang.org>
Run-TryBot: Keith Randall <khr@golang.org>
Run-TryBot: Joe Tsai <joetsai@digital-static.net> Reviewed-by: Joe Tsai <joetsai@digital-static.net> Reviewed-by: Josh Bleecher Snyder <josharian@gmail.com> Reviewed-by: Martin Möhrmann <martin@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Martin Möhrmann [Sat, 28 Aug 2021 15:54:10 +0000 (17:54 +0200)]
strings: add Clone function
The new strings.Clone function copies the input string
without the returned cloned string referencing the
input strings memory.
goarch: amd64
cpu: Intel(R) Core(TM) i5-1038NG7 CPU @ 2.00GHz
name time/op
Clone-8 24.2ns ± 2%
name alloc/op
Clone-8 48.0B ± 0%
name allocs/op
Clone-8 1.00 ± 0%
Update #45038
Fixes #40200
Change-Id: Id9116c21c14328ec3931ef9a67a2e4f30ff301f9
Reviewed-on: https://go-review.googlesource.com/c/go/+/345849
Trust: Martin Möhrmann <martin@golang.org>
Run-TryBot: Martin Möhrmann <martin@golang.org>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Joe Tsai <joetsai@digital-static.net> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Than McIntosh [Thu, 1 Jul 2021 13:27:46 +0000 (09:27 -0400)]
cmd/link: minor code cleanup in dwarf gen
Minor code cleanup to get rid of a few unused parameters and return
values in the linker's dwarf generation code. No functional changes.
Change-Id: I1a68ebe0f08d8d32ca7adfdd2fb9db573a4fd5f5
Reviewed-on: https://go-review.googlesource.com/c/go/+/332070
Trust: Than McIntosh <thanm@google.com> Reviewed-by: Cherry Mui <cherryyz@google.com>
Run-TryBot: Than McIntosh <thanm@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Robert Findley [Thu, 9 Sep 2021 22:41:30 +0000 (18:41 -0400)]
go/types: remove some unnecessary loading/expansion of Named types
For Identical an u.nify, only type arguments and pointer identity is
needed.
Change-Id: Id4018d2a53044fa20fd26d28890f28b37b6d6d70
Reviewed-on: https://go-review.googlesource.com/c/go/+/349409
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Robert Griesemer <gri@golang.org>
Joe Tsai [Wed, 4 Aug 2021 08:22:45 +0000 (01:22 -0700)]
bufio: add Writer.AvailableBuffer
This adds a new Writer.AvailableBuffer method that returns
an empty buffer with a possibly non-empty capacity for use
with append-like APIs.
The typical usage pattern is something like:
b := bw.AvailableBuffer()
b = appendValue(b, v)
bw.Write(b)
It allows logic combining append-like APIs with bufio.Writer to avoid
needing to allocate and manage buffers themselves and allows the
append-like APIs to directly write into the buffer for a bufio.Writer.
Fixes #47527
Change-Id: I9cd169f3f8e8c7cd40818caf3daf1944c826fc66
Reviewed-on: https://go-review.googlesource.com/c/go/+/345569
Trust: Joe Tsai <joetsai@digital-static.net>
Run-TryBot: Joe Tsai <joetsai@digital-static.net>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Joe Tsai [Fri, 27 Aug 2021 02:13:22 +0000 (19:13 -0700)]
reflect: optimize for maps with string keys
Over 80% of all Go map types use a string as the key.
The Go runtime already has a specialized implementation for such maps
in runtime/map_faststr.go. However, the Go reflection implementation
has not historically made use of that implementation.
This CL plumbs the appropriate logic to be accessible from Go reflection
so that it can benefit as well.
Change-Id: I5495d48948d8caf2d004a03ae1820ab5f8729670
Reviewed-on: https://go-review.googlesource.com/c/go/+/345486
Trust: Joe Tsai <joetsai@digital-static.net>
Run-TryBot: Joe Tsai <joetsai@digital-static.net>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
Joe Tsai [Wed, 4 Aug 2021 07:57:07 +0000 (00:57 -0700)]
bufio: make Reader.Reset and Writer.Reset work on the zero value
For batch allocation reasons, it would be useful to nest a
bufio.Reader or bufio.Writer in a struct as a value,
rather than a pointer. When the Reset method is called,
have it use the default buffer size if the buffer is nil.
Fixes #45374
Change-Id: I80df18a13575431428a42ed150a1579de1282637
Reviewed-on: https://go-review.googlesource.com/c/go/+/345570
Trust: Joe Tsai <joetsai@digital-static.net>
Run-TryBot: Joe Tsai <joetsai@digital-static.net>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
WANG Xuerui [Sun, 1 Aug 2021 16:56:25 +0000 (00:56 +0800)]
doc/asm: fix HTML markup
Change-Id: I33bde4835d3b83fafd55beea483f6236c4c62840
Reviewed-on: https://go-review.googlesource.com/c/go/+/338990 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Trust: Than McIntosh <thanm@google.com>
Florin Papa [Fri, 7 May 2021 00:17:59 +0000 (17:17 -0700)]
debug/elf: retain original error message when getSymbols fails.
The original error is currently discarded, and that makes it difficult
to know what failed, in case we want to retry only certain errors.
Change-Id: Id7e927ec242464249c4dfa5cda0f264adef3c898
Reviewed-on: https://go-review.googlesource.com/c/go/+/317851
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Trust: Than McIntosh <thanm@google.com>
Ethan Reesor [Fri, 20 Aug 2021 23:57:45 +0000 (18:57 -0500)]
testing: increase alternation precedence
Updates handling of go test flags -run and -bench to give alternation
precendence over the / delimiter. Currently, `A/B|C/D` is effectively
`A/(B|C)/D` - with this change, it changes to effectively `(A/B)|(C/D)`.
Fixes #39904
Change-Id: Iebe5efd8d91c72eed6351bd63b4689b0fcb0ed0f
Reviewed-on: https://go-review.googlesource.com/c/go/+/343883
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Trust: Than McIntosh <thanm@google.com>
Robert Griesemer [Thu, 9 Sep 2021 22:43:19 +0000 (15:43 -0700)]
cmd/compile/internal/syntax: correct follow token for type parameter lists
When parsing a type parameter declaration, parts of the code still
expected a ) as closing token. Use the correct follow token ) or ]
depending on parameter list kind.
Also, consistently use tokstring (not tok.String()) for user-facing
(error) messages.
Follow-up on comment in CL 348730.
For #43527.
Change-Id: Ib1d4feb526771a1668a54c3bb7a671f6c8a65940
Reviewed-on: https://go-review.googlesource.com/c/go/+/348742
Trust: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Robert Findley <rfindley@google.com>
Michael Munday [Thu, 9 Sep 2021 22:47:14 +0000 (23:47 +0100)]
cmd/compile: add support for Abs and Copysign intrinsics on riscv64
Also, add the FABSS and FABSD pseudo instructions to the assembler.
The compiler could use FSGNJX[SD] directly but there doesn't seem
to be much advantage to doing so and the pseudo instructions are
easier to understand.
Change-Id: Ie8825b8aa8773c69cc4f07a32ef04abf4061d80d
Reviewed-on: https://go-review.googlesource.com/c/go/+/348989
Trust: Michael Munday <mike.munday@lowrisc.org>
Run-TryBot: Michael Munday <mike.munday@lowrisc.org>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Joel Sing <joel@sing.id.au>
In some rewrite rules for arm64 bitfield optimizations, the
bitfield lsb value and the bitfield width value are related
to datasize, some of them use datasize directly to check the
bitfield lsb value is valid, to get the bitfiled width value,
but some of them call isARM64BFMask() and arm64BFWidth()
functions. In order to be consistent, this patch changes them
all to use datasize.
Besides, this patch sorts the codegen test cases.
Run the "toolstash-check -all" command and find one inconsistent code
is as the following.
The above inconsistence is due to this patch changing the range of the
field lsb value in "UBFIZ" optimization rules from "lc+(32|16|8)<64" to
"lc<64", so that the following code is generated as "UBFIZ". The logical
of changed code is still correct.
The code of src/math/fma.go:160:
const uvinf = 0x7FF0000000000000
func FMA(a, b uint32) float64 {
ps := a+b
return Float64frombits(uint64(ps)<<63 | uvinf)
}
The new assembly code:
TEXT "".FMA(SB), LEAF|NOFRAME|ABIInternal, $0-16
MOVWU "".a(FP), R0
MOVWU "".b+4(FP), R1
ADD R1, R0, R0
UBFIZ $63, R0, $1, R0
ORR $9218868437227405312, R0, R0
MOVD R0, "".~r2+8(FP)
RET (R30)
Robert Griesemer [Thu, 9 Sep 2021 22:55:05 +0000 (15:55 -0700)]
cmd/compile: fix test case for unified IR (fix build)
For #48301.
Change-Id: Ie5f57dcce86773c06c5140abf13a6cfff79eb323
Reviewed-on: https://go-review.googlesource.com/c/go/+/348743
Trust: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org> Reviewed-by: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Add explicit conversions to avoid warnings from -Wsign-conversion and
-Wshorten-64-to-32. Also avoid runtime errors from -fsanitize=undefined.
Fixes #48121
Change-Id: I29dc8d976884fc42826392c10f1e1759bb1a3989
Reviewed-on: https://go-review.googlesource.com/c/go/+/348739
Trust: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Cherry Mui <cherryyz@google.com>
Robert Griesemer [Thu, 9 Sep 2021 04:30:01 +0000 (21:30 -0700)]
cmd/compile/internal/syntax: better error message for missing type constraint
For #43527.
Change-Id: I8c706e68572286d5675383eb2dfd75b5618b646b
Reviewed-on: https://go-review.googlesource.com/c/go/+/348730
Trust: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Robert Findley <rfindley@google.com>
korzhao [Sat, 28 Aug 2021 21:21:17 +0000 (05:21 +0800)]
time: propagate "," separator for fractional seconds into Format
In CL 300996 that fixed issue #6189, we made Parse recognize
"," as a separator for fractional seconds.
However, we didn't modify Format to propagate the separator
verbatim from Parse. Without this change, we break prior
functionality that relied on a comma being used in Format.
Fixes #48037
Change-Id: I6565a25e8657ca3747a58b25acba58f27cdcddc0
Reviewed-on: https://go-review.googlesource.com/c/go/+/345438
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org> Reviewed-by: Emmanuel Odeke <emmanuel@orijtech.com>
Trust: Cherry Mui <cherryyz@google.com>
cmd/compile: fix implement for closure in a global assignment
If closure in a global assignment and has a method receiver.
We should assign receiver as a global variable, not a local variable.
Fixes #48225
Change-Id: I8f65dd6e8baf66a5eff24028d28ad0a594091add
Reviewed-on: https://go-review.googlesource.com/c/go/+/348512
Run-TryBot: Dan Scales <danscales@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Dan Scales <danscales@google.com> Reviewed-by: Keith Randall <khr@golang.org>
Trust: Dan Scales <danscales@google.com>
Paul E. Murphy [Tue, 23 Mar 2021 21:17:00 +0000 (16:17 -0500)]
cmd/link: enable internal linker in more cases for ppc64le
The internal linker is capable of linking the ppc64le linux
race detector and approved cgo packages.
Likewise, ppc64/linux and ppc64/aix do not support the race
detector. Thus, extra code to enforce external linking when
using the race detector on ppc64/ppc64le can be removed
entirely.
Fixes #21961
Change-Id: I10db14f65ee616ee3291e17409e8333e3af7d4df
Reviewed-on: https://go-review.googlesource.com/c/go/+/304459
Run-TryBot: Paul Murphy <murp@ibm.com>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: Lynn Boger <laboger@linux.vnet.ibm.com> Reviewed-by: Cherry Mui <cherryyz@google.com>
Matthew Dempsky [Thu, 9 Sep 2021 18:17:39 +0000 (11:17 -0700)]
test: add compiler regress tests for #46461
gri@ reports that types2 now correctly handles when type parameters
recursively refer back to the parameterized type, so we might as well
add tests to exercise that. Unified IR also correctly handles
importing and exporting these types, but -G=3 currently does not.
Updates #46461.
Change-Id: I272102aa08c40c980b9aeeca9f834291dfbbcc3e
Reviewed-on: https://go-review.googlesource.com/c/go/+/348738
Trust: Matthew Dempsky <mdempsky@google.com>
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Robert Griesemer <gri@golang.org>
Dan Scales [Wed, 8 Sep 2021 15:41:54 +0000 (08:41 -0700)]
cmd/compile: fix case where init info of OAS node is dropped
When an OAS node is converted to an OSELRECV2 node in tcSelect(), the
possible DCL node in the Init field was being dropped, since a
completely new node was being created and the Init field was not set. I
don't expect n.Init() to be set for the ORECV case, but the code now
deals with that too.
Fixed bug in both tcSelect() and transformSelect().
Fixes #48289
Change-Id: I09918a70f7cbaa4aa9a17546169f908a8787df15
Reviewed-on: https://go-review.googlesource.com/c/go/+/348569
Trust: Dan Scales <danscales@google.com>
Run-TryBot: Dan Scales <danscales@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com> Reviewed-by: Matthew Dempsky <mdempsky@google.com> Reviewed-by: Robert Griesemer <gri@golang.org>
To be consistent with CL 348376, spell out 'RecvTypeParams' in go/types
and types2 API.
Updates #47916
Change-Id: If8b3fd4274ccb944bd0ff04d7007e94e5fba61c1
Reviewed-on: https://go-review.googlesource.com/c/go/+/348810
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Robert Griesemer <gri@golang.org>
Michael Anthony Knyszek [Fri, 3 Sep 2021 21:35:59 +0000 (21:35 +0000)]
reflect: add hooks for dealing with narrow width floats
Currently on amd64 and arm64, float32 values just live in the bottom 32
bits of the register, so reflect simply places them there in a RegArgs
for reflectcall to load them. This works fine because both of these
platforms don't care what the upper 32 bits are, and have instructions
to operate on float32 values specifically that we use. In sum, the
representation of the float32 in memory is identical to that of the
representation in a register.
On other platforms, however, the representation of FP values differ
depending on whether they're in memory or in a register. For instance,
on ppc64, all floating point values get promoted to a float64 when
loaded to a register (i.e. there's only one set of FP instructions). As
another example, on riscv64, narrow-width floats in registers need to be
NaN-boxed.
What all this means is that for supporting the register ABI on these
platforms, reflect needs to do a little extra work to ensure that the
representation of FP values in a RegArgs matches the representation it
takes on in a register. For this purpose, this change abstracts away the
action of storing values into a RegArgs a little bit and adds a
platform-specific hook which currently does nothing but copy the value.
For #40724.
Change-Id: I65dcc7d86d5602a584f86026ac204564617f4c5a
Reviewed-on: https://go-review.googlesource.com/c/go/+/347566
Trust: Michael Knyszek <mknyszek@google.com>
Run-TryBot: Michael Knyszek <mknyszek@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Cherry Mui <cherryyz@google.com>
Damien Neil [Fri, 6 Aug 2021 17:33:08 +0000 (10:33 -0700)]
net: deprecate (net.Error).Temporary
Fixes #45729.
Change-Id: I819dd0cc79fc23baac46cafd78bb80f5133c992b
Reviewed-on: https://go-review.googlesource.com/c/go/+/340261
Trust: Damien Neil <dneil@google.com>
Trust: Bryan C. Mills <bcmills@google.com>
Run-TryBot: Damien Neil <dneil@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Keith Randall [Thu, 9 Sep 2021 15:09:57 +0000 (08:09 -0700)]
cmd/compile: stenciled conversions might be NOPs
A generic conversion might be required for when converting T->interface{}.
When stenciled with T=interface{}, then that conversion doesn't need
to do anything.
Fixes #48276
Change-Id: Ife65d01c99fbd0895cb7eec79df9e93e752b1fa5
Reviewed-on: https://go-review.googlesource.com/c/go/+/348736
Trust: Keith Randall <khr@golang.org>
Run-TryBot: Keith Randall <khr@golang.org> Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
TryBot-Result: Go Bot <gobot@golang.org>
Matthew Dempsky [Wed, 8 Sep 2021 21:03:53 +0000 (14:03 -0700)]
test: re-enable AsmCheck tests for types2-based frontends
run.go has logic for being able to run tests with various -G flags,
but not all test types (in particular, "asmcheck" tests) support
configuring non-default -G levels. The intention was that these tests
would continue running in the default mode (at the time -G=0), but at
some point it seems like we accidentally disabled them all
together (if it ever worked correctly in the first place).
Fixes #48247.
Change-Id: I13917cb0012cbe522d29b23b888de6136872ead4
Reviewed-on: https://go-review.googlesource.com/c/go/+/348671
Trust: Matthew Dempsky <mdempsky@google.com>
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Michael Munday <mike.munday@lowrisc.org>
Robert Findley [Wed, 8 Sep 2021 23:03:37 +0000 (19:03 -0400)]
go/types: reduce number of delayed functions
This is a port of CL 348018 to go/types. It differs from that CL due to
the way that field lists are represented in go/ast.
Change-Id: Ib5a0243b44d0bf9e95d039f624c668f8c329f8fa
Reviewed-on: https://go-review.googlesource.com/c/go/+/348691
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Robert Griesemer <gri@golang.org>
Robert Findley [Wed, 8 Sep 2021 22:27:31 +0000 (18:27 -0400)]
go/types: handle recursive type parameter constraints
This is a port of CL 348090 to go/types. Notably, unlike in types2,
declareTypeParams was previously setting the default constraint to the
empty interface, not nil, because this was missed in CL 335034 (no
changes were made to declareTypeParams). This CL fixes this discrepancy.
Change-Id: I0fa54a660ba14c6cbefa81a27ab7eb193df3be20
Reviewed-on: https://go-review.googlesource.com/c/go/+/348690
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Robert Griesemer <gri@golang.org>
Robert Findley [Wed, 8 Sep 2021 22:16:36 +0000 (18:16 -0400)]
go/types: detect constraint type inference cycles
This is a port of CL 347300 to go/types. The test was adjusted to match
the differing error positioning in go/types: errors are placed on the
ast.CallExpr.Fun, rather than the Lparen.
Change-Id: I3dee5d7d75bae18817cc1f04ab0d357d3a6a8198
Reviewed-on: https://go-review.googlesource.com/c/go/+/348689
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Robert Griesemer <gri@golang.org>
syscall: drop fallback to pipe in Pipe on linux/arm
Follow-up for CL 346789
The minimum required Linux kernel version for Go 1.18 will be changed to
2.6.32, see #45964. The pipe2 syscall was added in 2.6.27, so the
fallback to use pipe in Pipe on linux/arm can be removed.
For #45964
Change-Id: I8b18244ca1f849f10e90565b4fef80ce777fef69
Reviewed-on: https://go-review.googlesource.com/c/go/+/347349
Trust: Tobias Klauser <tobias.klauser@gmail.com>
Run-TryBot: Tobias Klauser <tobias.klauser@gmail.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
cmd/compile: fix unified IR panic when expanding nested inline function
When reading body of inlining function, which has another inlined
function in the body, the reader still add this inlined function to
todoBodies, which it shouldn't because the inlined function was read
already.
To fix this, introduce new flag to signal that we are done construting
all functions in todoBodies, thus the addBody shouldn't add anything
to todoBodies then.
Updates #48094
Change-Id: I45105dd518f0a7b69c6dcbaf23b957623f271203
Reviewed-on: https://go-review.googlesource.com/c/go/+/347529
Trust: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Run-TryBot: Cuong Manh Le <cuong.manhle.vn@gmail.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Ian Lance Taylor [Wed, 8 Sep 2021 23:21:33 +0000 (16:21 -0700)]
internal/poll: report open fds when TestSplicePipePool fails
For #48066
Change-Id: I1152a1c15756df35b71b27d3e7025d97da9e70b0
Reviewed-on: https://go-review.googlesource.com/c/go/+/348579
Trust: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Emmanuel Odeke <emmanuel@orijtech.com>
cmd/compile: fix wrong instantiated type for embedded receiver
In case of embedded field, if the receiver was fully instantiated, we
must use its instantiated type, instead of passing the type params of
the base receiver.
Fixes #47797
Fixes #48253
Change-Id: I97613e7e669a72605137e82406f7bf5fbb629378
Reviewed-on: https://go-review.googlesource.com/c/go/+/348549
Trust: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Trust: Dan Scales <danscales@google.com>
Run-TryBot: Cuong Manh Le <cuong.manhle.vn@gmail.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Dan Scales <danscales@google.com>
fanzha02 [Tue, 24 Aug 2021 04:46:54 +0000 (12:46 +0800)]
cmd/internal/obj/arm64: add checks for incorrect use of REGTMP register
The current assembler uses REGTMP as a temporary destination register,
when optimizing one instruction into a multi-instruction sequence. But
in some cases, when the source register is REGTMP, this behavior is
incorrect.
For example:
ADD $0x1234567, R27, R3
The current assembler encodes it into
MOVD $17767, R27
MOVK $(291<<16), R27
ADD R27, R27, R3
It is illegal to overwrite R27. This CL adds the related checks.
Add test cases.
Change-Id: I0af373d9fd23d8f067c093778dd4cc76748faf38
Reviewed-on: https://go-review.googlesource.com/c/go/+/344689 Reviewed-by: Cherry Mui <cherryyz@google.com>
Run-TryBot: Cherry Mui <cherryyz@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Trust: fannie zhang <Fannie.Zhang@arm.com>
Matthew Dempsky [Wed, 8 Sep 2021 00:24:50 +0000 (17:24 -0700)]
cmd/compile: remove 'ext' fields from unified IR reader/writer types
This is a vestigial artifact of how I initially split apart the public
and private data for objects. But now objects are split into more
parts, and it's proven easier to just keep them as separate variables.
So it's time to cleanup the initial public/private code to follow the
same approach.
Change-Id: I3976b19fb433cbe21d299d3799ec616f9e59561e
Reviewed-on: https://go-review.googlesource.com/c/go/+/348412
Trust: Matthew Dempsky <mdempsky@google.com>
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Matthew Dempsky [Tue, 7 Sep 2021 20:23:08 +0000 (13:23 -0700)]
cmd/compile: simplify value coding for unified IR
In indexed export, values are always exported along with their type
and are encoded in a type-sensitive manner, because this matches how
cmd/compile handled constants internally.
However, go/types intentionally differs from this, decoupling type
from value representation. As unified IR strives to be more
go/types-centric, it makes sense to embrace this and make values a
more first-class encoding.
Change-Id: If21d849c4f610358bd776d5665469d180bcd5f6e
Reviewed-on: https://go-review.googlesource.com/c/go/+/348014
Trust: Matthew Dempsky <mdempsky@google.com>
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Matthew Dempsky [Wed, 8 Sep 2021 22:59:11 +0000 (15:59 -0700)]
cmd/compile: extrapolate $GOROOT in unified IR
This ensures that diagnostics for files within $GOROOT continue to be
reported using their full filepath, rather than the abbreviated
filepath. Notably, this is necessary for test/run.go, which has tests
that expect to see the full filepath.
Updates #48247.
Change-Id: I440e2c6dd6109ca059d81cee49e476bba805d703
Reviewed-on: https://go-review.googlesource.com/c/go/+/348670
Trust: Matthew Dempsky <mdempsky@google.com>
Run-TryBot: Matthew Dempsky <mdempsky@google.com> Reviewed-by: Robert Griesemer <gri@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Robert Griesemer [Wed, 8 Sep 2021 23:03:57 +0000 (16:03 -0700)]
go/types, types2: add Environment to Config
Port to types2 and adjust compiler accordingly.
Change-Id: I2e72b151ef834977dca64cb2e62cedcac4e46062
Reviewed-on: https://go-review.googlesource.com/c/go/+/348578
Trust: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org> Reviewed-by: Matthew Dempsky <mdempsky@google.com> Reviewed-by: Robert Findley <rfindley@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
A common bug during development is to introduce rewrite rule cycles.
This is annoying because it takes a while to notice that
make.bash is a bit too slow this time, and to remember why.
And then you have to manually arrange to debug.
Make this all easier by automating it.
Detect cycles, and when we detect one, print the sequence
of rewrite rules that occur within a single cycle before crashing.
Robert Findley [Wed, 8 Sep 2021 19:51:10 +0000 (15:51 -0400)]
go/ast: rename MultiIndexExpr to IndexListExpr
As discussed in #47781, IndexListExpr is one character shorter and has
the advantage of being next to IndexExpr in documentation.
Updates #47781
Change-Id: I709d5c1a79b4f9aebcd6445e4ab0cd6dae45bab7
Reviewed-on: https://go-review.googlesource.com/c/go/+/348609
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com> Reviewed-by: Robert Griesemer <gri@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Michael Matloob [Fri, 3 Sep 2021 22:11:00 +0000 (18:11 -0400)]
cmd/go: print offending -mod value in workspace mode
Workspace mode only allows -mod to be set to 'readonly'. When returning
the error for other values of -mod, specify the the value in the error
so it's easier to see what's going on when the go command is invoked
through another tool.
For #45713, #48186
Change-Id: I3ba206a71c7a20c18aeccfa48f3c9dc935a180a4
Reviewed-on: https://go-review.googlesource.com/c/go/+/347593
Trust: Michael Matloob <matloob@golang.org>
Trust: Daniel Martí <mvdan@mvdan.cc>
Run-TryBot: Michael Matloob <matloob@golang.org>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Jay Conrod <jayconrod@google.com> Reviewed-by: Daniel Martí <mvdan@mvdan.cc> Reviewed-by: Bryan C. Mills <bcmills@google.com>
Michael Matloob [Fri, 3 Sep 2021 17:33:37 +0000 (13:33 -0400)]
cmd/go: improve the creation and editing of go.work files
This change changes go.work files so that directory paths are clearly
distinguished from module paths by either being rooted absolute paths or
starting with '.' or '..' path elements if they are relative paths.
go mod initwork now checks that the go.work file doesn't already exist
before creating it, and gomod initwork and gomod editwork look up the
module path corresponding to a directory and write it to the directory
directive's comment.
For #45713
Change-Id: I6983779059b7de6fc83d359280ceffb263f6b641
Reviewed-on: https://go-review.googlesource.com/c/go/+/347591
Trust: Michael Matloob <matloob@golang.org>
Run-TryBot: Michael Matloob <matloob@golang.org>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Jay Conrod <jayconrod@google.com>
Robert Findley [Wed, 8 Sep 2021 13:58:44 +0000 (09:58 -0400)]
go/types: temporarily pin the Checker to Interface during checking
While type checking expressions involving interface types, it is
possible that their type set is used before delayed actions are
processed. As a result, computeInterfaceTypeSet is called with a nil
checker, and errors in the interface type definition result in panics
(see #48234).
To avoid the panics, store a *Checker on Interface for use in between
checking of the interface type expression and processing of delayed
actions.
Fixes #48234
Change-Id: I5509bc1c01b55edac52446b9e075fbe8fcc01874
Reviewed-on: https://go-review.googlesource.com/c/go/+/348371
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Robert Griesemer <gri@golang.org>
Robert Findley [Wed, 8 Sep 2021 15:57:04 +0000 (11:57 -0400)]
go/types: spell out 'Type' in type parameter APIs
As discussed on the go/types proposal (#47916), we should spell out the
word 'Type', rather than using 'T'.
Change-Id: I5f51255eedc07fea61f909b7ecb3093a7fab765e
Reviewed-on: https://go-review.googlesource.com/c/go/+/348376
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com> Reviewed-by: Robert Griesemer <gri@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Robert Findley [Wed, 8 Sep 2021 15:46:58 +0000 (11:46 -0400)]
go/ast: rename TParams fields to TypeParams
As discussed in the ast proposal (#47781), there's not really a strong
reason to avoid spelling out 'Type'.
Change-Id: I0ba1bf03b112ea60509a78a89a050a302779d9d0
Reviewed-on: https://go-review.googlesource.com/c/go/+/348375
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com> Reviewed-by: Robert Griesemer <gri@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Robert Findley [Fri, 3 Sep 2021 15:22:18 +0000 (11:22 -0400)]
go/types: move NewTypeParam off of Checker
This aligns with the API proposal.
Change-Id: I9967a317196392ffa5ddbe5391d7aba5f6e7bad2
Reviewed-on: https://go-review.googlesource.com/c/go/+/347561
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Robert Griesemer <gri@golang.org>
Robert Findley [Fri, 3 Sep 2021 15:17:37 +0000 (11:17 -0400)]
go/types: move typeHash to environment.go
This is a pure code move, with no other changes.
Change-Id: Id31f1f960d3208dc614556de89bf39b7ca77df3a
Reviewed-on: https://go-review.googlesource.com/c/go/+/347560
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Robert Griesemer <gri@golang.org>
Robert Findley [Thu, 2 Sep 2021 22:39:58 +0000 (18:39 -0400)]
go/types: implement deduplication of instances using the Environment
Implement deduplication of type instances via the Environment by
recording a map of unique IDs for *Named types. This avoids an issue
with the existing type hash, where qualified names alone were not
sufficient to differentiate two *Named types that have the same fully
qualified name but which are distinct pointers. It also allows us to
drop the scope accounting for local types.
A subtle bug is also fixed in subst.go, where the instance t was passed
to typeHash rather than t.orig.
Change-Id: I85639ccc1c9bfee470babd2fc85375484c8ed0b9
Reviewed-on: https://go-review.googlesource.com/c/go/+/344390
Trust: Robert Findley <rfindley@google.com>
Run-TryBot: Robert Findley <rfindley@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Robert Griesemer <gri@golang.org>
Robert Griesemer [Tue, 7 Sep 2021 20:58:49 +0000 (13:58 -0700)]
cmd/compile/internal/types2: handle recursive type parameter constraints
Check type constraints after the respective type parameter list
has been associated with a parameterized type so that recursive
type parameter constraints "see" a parameterized type.
Fixes #45550.
Fixes #47796.
Change-Id: Iac74610ca017a78013820624230c857395506aff
Reviewed-on: https://go-review.googlesource.com/c/go/+/348090
Trust: Robert Griesemer <gri@golang.org> Reviewed-by: Robert Findley <rfindley@google.com>
Pull in the latest published version of github.com/google/pprof
that is available at this time in the Go 1.18 development cycle.
Done with:
go get -d github.com/google/pprof@latest
go mod tidy
go mod vendor
For #36905.
Change-Id: Ib25aa38365ec70a0bed2a8a6527e5823ab9f9ded
Reviewed-on: https://go-review.googlesource.com/c/go/+/348410
Trust: Dmitri Shuralyov <dmitshur@golang.org>
Run-TryBot: Dmitri Shuralyov <dmitshur@golang.org>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Michael Pratt <mpratt@google.com>
Michael Munday [Wed, 8 Sep 2021 14:14:53 +0000 (14:14 +0000)]
test/codegen: fix package name for test case
The codegen tests are currently skipped (see #48247). The test
added in CL 346050 did not compile because it was in the main
package but did not contain a main function. Changing the package
to 'codegen' fixes the issue.
Updates #48247.
Change-Id: I0a0eaca8e6a7d7b335606d2c76a204ac0c12e6d5
Reviewed-on: https://go-review.googlesource.com/c/go/+/348392
Trust: Michael Munday <mike.munday@lowrisc.org>
Run-TryBot: Michael Munday <mike.munday@lowrisc.org> Reviewed-by: Cherry Mui <cherryyz@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Michael Munday [Wed, 8 Sep 2021 14:07:12 +0000 (14:07 +0000)]
test/codegen: fix compilation of bitfield tests
The codegen tests are currently skipped (see #48247) and the
bitfield tests do not actually compile due to a duplicate function
name (sbfiz5) added in CL 267602. Renaming the function fixes the
issue.
Updates #48247.
Change-Id: I626fd5ef13732dc358e73ace9ddcc4cbb6ae5b21
Reviewed-on: https://go-review.googlesource.com/c/go/+/348391
Trust: Michael Munday <mike.munday@lowrisc.org>
Run-TryBot: Michael Munday <mike.munday@lowrisc.org>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Cherry Mui <cherryyz@google.com>
Michael Munday [Wed, 8 Sep 2021 13:57:07 +0000 (13:57 +0000)]
test/codegen: remove broken riscv64 test
This test is not executed by default (see #48247) and does not
actually pass. It was added in CL 346689. The code generation
changes made in that CL only change how instructions are assembled,
they do not actually affect the output of the compiler. This test
is unfortunately therefore invalid and will never pass.
Updates #48247.
Change-Id: I0c807e4a111336e5a097fe4e3af2805f9932a87f
Reviewed-on: https://go-review.googlesource.com/c/go/+/348390
Trust: Michael Munday <mike.munday@lowrisc.org>
Run-TryBot: Michael Munday <mike.munday@lowrisc.org> Reviewed-by: Cherry Mui <cherryyz@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Change-Id: I9dc736c2c5256f7d9e80fd9c52c6725ecf0b8001
Reviewed-on: https://go-review.googlesource.com/c/go/+/348409
Trust: Dmitri Shuralyov <dmitshur@golang.org>
Trust: Alexander Rakoczy <alex@golang.org>
Run-TryBot: Dmitri Shuralyov <dmitshur@golang.org>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Alexander Rakoczy <alex@golang.org>
Paul E. Murphy [Tue, 23 Mar 2021 20:52:49 +0000 (15:52 -0500)]
cmd/link: rework .TOC. handling for ppc64le
Setup .TOC. to point to the same place for all objects. Today, the linker
assumes all call relocations can use the local function entry point of
imported object files. This requires a consistent pointer across all
objects.
This intentionally computes the .TOC. pointer in all linking configurations.
In some cases the .TOC. is not used today (e.g linking position-dependent go
only code). It is harmless and simple to compute in all cases, so just
do it for easier maintenance.
Notably, .TOC. is used in some cases when static linking is requested on
ppc64le/linux:
* Position-independent C code using a PC-rel relocation against .TOC.. cgo
generated C object files are usually compiled PIC even if the go binary
itself is not.
* Anything which causes PLT stub generation. The stubs always generate
a .TOC. relative relocation.
* The race detector. Today, this links in an externally compiled archive which
contains position-independent object files.
Similarly, position-independent linking is always punted to the external
linker on ppc64 today.
Joel Sing [Wed, 25 Aug 2021 15:15:23 +0000 (01:15 +1000)]
cmd/link/internal/riscv64,cmd/internal/obj/riscv: make error messages consistent
Be consistent with both style/wording and formatting of error messages.
Change-Id: I23013d904333a67a8f17dd364bb5015f54f419c0
Reviewed-on: https://go-review.googlesource.com/c/go/+/345050
Trust: Joel Sing <joel@sing.id.au> Reviewed-by: Cherry Mui <cherryyz@google.com>
Use cursym.WriteInt rather than building up a slice of bytes and then writing them
out via PutUint32. This also allows for variable instruction sizes, which will be
needed when support for compressed (2 byte length) instructions is added.
Change-Id: I17c9ffa52d27c91a24e161317e3db28e224804ae
Reviewed-on: https://go-review.googlesource.com/c/go/+/344460
Trust: Joel Sing <joel@sing.id.au> Reviewed-by: Cherry Mui <cherryyz@google.com>
Joel Sing [Sat, 21 Aug 2021 09:06:25 +0000 (09:06 +0000)]
cmd/internal/obj/riscv: absorb rewriteMOV into instruction generation
Replace the now poorly named rewriteMOV function with a markRelocs function,
absorbing the MOV validation into the instruction generation code.
Change-Id: I6ae94553f7f300862c40310343ef0be8f44a4b0f
Reviewed-on: https://go-review.googlesource.com/c/go/+/344459
Trust: Joel Sing <joel@sing.id.au> Reviewed-by: Cherry Mui <cherryyz@google.com>
Joel Sing [Sat, 21 Aug 2021 04:30:29 +0000 (04:30 +0000)]
cmd/internal/obj/riscv: simplify rewriteMOV
Rewrite and simplify the rewriteMOV function in preparation for eliminating it
entirely. Improve some error messages in the process.
Change-Id: Id9a77be5174d46cc23651930c2e9068ee6555690
Reviewed-on: https://go-review.googlesource.com/c/go/+/344458
Trust: Joel Sing <joel@sing.id.au> Reviewed-by: Cherry Mui <cherryyz@google.com>
Run-TryBot: Cherry Mui <cherryyz@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
Keith Randall [Tue, 7 Sep 2021 23:24:49 +0000 (16:24 -0700)]
reflect: add back signaling NaN test
I was a little too agressive in CL 258957 (removing 387 support) in
removing a signaling NaN test that should probably still exist.
I should have just removed the 387 skip, not the entire test.
Change-Id: I97ca53f190eb0de00e43b96629b2cae677e7dacb
Reviewed-on: https://go-review.googlesource.com/c/go/+/348209
Trust: Keith Randall <khr@golang.org>
Run-TryBot: Keith Randall <khr@golang.org>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Ian Lance Taylor [Tue, 7 Sep 2021 23:10:31 +0000 (16:10 -0700)]
reflect: correct typoo in comment
Change-Id: Ic3c69906a8afde63278f173df0427b1c460c9f3e
Reviewed-on: https://go-review.googlesource.com/c/go/+/348189
Trust: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org> Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Dan Scales [Tue, 31 Aug 2021 15:01:44 +0000 (08:01 -0700)]
cmd/compile: add CONVIFACE nodes needed in generic code due to assignments
Added new function earlyTransformAssign() to add needed CONVIFACE nodes
due to assignments in generic functions.
Fixes #48049
Change-Id: I7cd9cee6ecf34ed2ef0743d1b17645b9f520fa00
Reviewed-on: https://go-review.googlesource.com/c/go/+/347914
Trust: Dan Scales <danscales@google.com>
Run-TryBot: Dan Scales <danscales@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
Keith Randall [Tue, 7 Sep 2021 22:13:51 +0000 (15:13 -0700)]
reflect: add test for passing float32 signaling NaNs
Update #40724
Change-Id: I110cdb7c4a2c5db6b85ca951143430555261abf3
Reviewed-on: https://go-review.googlesource.com/c/go/+/348017
Trust: Keith Randall <khr@golang.org>
Run-TryBot: Keith Randall <khr@golang.org>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Dan Scales [Tue, 7 Sep 2021 17:54:17 +0000 (10:54 -0700)]
cmd/compile: remove now-unneeded SetHasTParam() for cached ptr element
We had code in NewPtr() to set the HasTParam/HasShape flag as needed for
the cached ptr element if it wasn't set correctly based on its Elem.
This was causing the race mentioned in the issue.
But that setting code is no longer needed, as long as we call
SetRParams() soon after calling NewIncompleteNamedType(), before
creating/translating the underlying type (which we do). The
HasTParam/HasShape attribute can only come from setting of rparams or a
direct typeparam/shape somewhere in the underlying type, both of which
don't depend on recursion, etc. (as long as the rparams are set early).
Added a check that HasTParam/HasShape are set correctly for the cached
pointer/slice elems in NewPtr() and NewSlice().
Fixes #48191
Change-Id: Ide7d82efb77ae97901e75b2e6c65bd1bfc25e0ee
Reviewed-on: https://go-review.googlesource.com/c/go/+/348089
Trust: Dan Scales <danscales@google.com>
Run-TryBot: Dan Scales <danscales@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
In CL 345572, we used the reflect.Value.SetLen method to avoid
extra memory allocation for reflect.Value.Slice.
This also applies to function decodeSlice
name old time/op new time/op delta
DecodeStringsSlice-12 96.5µs ±12% 63.0µs ± 8% -34.68% (p=0.000 n=9+10)
name old alloc/op new alloc/op delta
DecodeStringsSlice-12 89.3kB ± 0% 65.3kB ± 0% -26.89% (p=0.000 n=10+10)
name old allocs/op new allocs/op delta
DecodeStringsSlice-12 3.18k ± 0% 2.18k ± 0% -31.47% (p=0.000 n=10+10)
Change-Id: Ifdb43716cc90a265962dec022704a5571f447fd8
Reviewed-on: https://go-review.googlesource.com/c/go/+/347533 Reviewed-by: Joe Tsai <joetsai@digital-static.net> Reviewed-by: Rob Pike <r@golang.org>
Trust: Joe Tsai <joetsai@digital-static.net>
Run-TryBot: Joe Tsai <joetsai@digital-static.net>
TryBot-Result: Go Bot <gobot@golang.org>
Dan Scales [Mon, 6 Sep 2021 23:25:43 +0000 (16:25 -0700)]
cmd/compile: make sure imported instantiated types have their methods created
We should be putting a newly instantiated imported type in
Instantiate/doInst onto the instTypeList, so its methods/dictionaries
are instantiated. To do this, we needed a more general way to add a
type to instTypeList, so add NeedInstType(), analogous to
NeedRuntimeType(). This has the extra advantage that now all types
created by the type substituter are added to instTypeList without any
extra code, which was easy to forget. doInst() now correctly calls
NeedInstType().
This is a bit aggressive, since a fully instantiated type in a generic
function/method may never be used, if the generic method is never
instantiated in the local package. But it should be fairly uncommon for
a generic method to mention a fully instantiated type (but it does
happen in this bug).
Fixes both cases mentioned in the bug.
Fixed #48185
Change-Id: I19b5012dfac17e306c8005f8595a648b0ab280d0
Reviewed-on: https://go-review.googlesource.com/c/go/+/347909
Trust: Dan Scales <danscales@google.com>
Trust: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Run-TryBot: Dan Scales <danscales@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
Ruslan Andreev [Tue, 15 Jun 2021 14:04:30 +0000 (14:04 +0000)]
cmd/compile: add prefetch intrinsic support
This CL provide new intrinsics to emit prefetch instructions for AMD64
and ARM64 platforms:
Prefetch - prefetches data from memory address to cache;
PrefetchStreamed - prefetches data from memory address, with a hint
that this data is being streamed.
This patch also provides prefetch calls pointed by RSC inside scanobject
and greyobject of GC mark logic.
Performance results provided by Michael:
https://perf.golang.org/search?q=upload:20210901.9
Benchmark parameters:
tree2 -heapsize=1000000000 -cpus=8
tree -n=18
parser
peano
If the condition is a bool constant, there's no need to walk both
branches.
Passes toolstash -cmp.
Change-Id: I4ee5e3553ce07c2213efba0d33d869b4a1b57783
Reviewed-on: https://go-review.googlesource.com/c/go/+/347911 Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Trust: Keith Randall <khr@golang.org>
Run-TryBot: Keith Randall <khr@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Dan Scales [Tue, 7 Sep 2021 14:46:27 +0000 (07:46 -0700)]
cmd/compile: fix type substituter to copy Funarg value for structs
We were missing copying the Funarg value when substituting for a struct
type.
Change-Id: Id0c2d9e55fb15987acb9edba6f74cf57cfd3417e
Reviewed-on: https://go-review.googlesource.com/c/go/+/347913
Run-TryBot: Dan Scales <danscales@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Keith Randall <khr@golang.org>
Trust: Dan Scales <danscales@google.com>
Matthew Dempsky [Tue, 7 Sep 2021 17:51:14 +0000 (10:51 -0700)]
cmd/compile: fix lazy loading in reader2
Calling reader2.obj fully loads the referenced object, which is
necessary in general; but for reading the package index, we just need
to setup the name->index mapping. This CL adds this, so that lazy
loading works as intended.
Change-Id: Ie51d59e2247d99b46f9dc69fba7ce89e2584b7c4
Reviewed-on: https://go-review.googlesource.com/c/go/+/348011
Trust: Matthew Dempsky <mdempsky@google.com>
Trust: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Run-TryBot: Matthew Dempsky <mdempsky@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Cuong Manh Le <cuong.manhle.vn@gmail.com>
For lazy import resolution, there's reentrancy issue with (*Named).load
method, when "t.resolve(t)" can lead us to the same named type, thus
(*Named).load is called recursively, causing the deadlock.
The main problem is that when instantinate a type, we calculate the type
hashing, including TParams. Calling t.TParams().Len() triggers the
reentrancy call to "(*Named).load".
To fix this, just not checking TParams().Len() if we are hashing.
Updates #48185
Change-Id: Ie34842d7b10fad5d11fbcf75bb1c64a89deac6b8
Reviewed-on: https://go-review.googlesource.com/c/go/+/347534
Trust: Cuong Manh Le <cuong.manhle.vn@gmail.com>
Trust: Robert Griesemer <gri@golang.org>
Run-TryBot: Cuong Manh Le <cuong.manhle.vn@gmail.com> Reviewed-by: Matthew Dempsky <mdempsky@google.com> Reviewed-by: Robert Findley <rfindley@google.com> Reviewed-by: Robert Griesemer <gri@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
Vyacheslav Pachkov [Mon, 23 Aug 2021 08:40:32 +0000 (08:40 +0000)]
cmd/link: resolve magic value in gdbscript section generation
According to the .debug_gdb_scripts section specification
[https://sourceware.org/gdb/onlinedocs/gdb/dotdebug_005fgdb_005fscripts-section.html],
each entry begins with a non-null prefix byte that specifies the kind of entry.
This commit resolves a question about magic byte and replaces a
hardcoded value with a meaningful constant "GdbScriptPythonFileId"
inside writegdbscript function.
Change-Id: I456c742bcb539a5853b9e2a6811033f35c37e7d4
GitHub-Last-Rev: 2f1c4cb9eec582a6e381fa866738aa17278005c2
GitHub-Pull-Request: golang/go#47646
Reviewed-on: https://go-review.googlesource.com/c/go/+/341391
Run-TryBot: Than McIntosh <thanm@google.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Cherry Mui <cherryyz@google.com> Reviewed-by: Than McIntosh <thanm@google.com>
cmd/compile: make sure that the names created for instantiated type are the same
Now we have two functions that create names for instantiated types.
They are inconsistent when dealing with byte/rune type.
This CL makes instTypeName2 reuse the code of typecheck.InstTypeName
Fixes #48198
Change-Id: I4c216b532cba6618ef9b63fd0b76e8f1c0ed7a75
Reviewed-on: https://go-review.googlesource.com/c/go/+/347491 Reviewed-by: Dan Scales <danscales@google.com>
Trust: Dan Scales <danscales@google.com>
Trust: Keith Randall <khr@golang.org>
Run-TryBot: Dan Scales <danscales@google.com>
TryBot-Result: Go Bot <gobot@golang.org>
syscall: use dup3 in forkAndExecInChild1 on all Linux platforms
The minimum required Linux kernel version for Go 1.18 will be changed to
2.6.32, see #45964. The dup3 syscall was added in 2.6.27, so the
fallback to use the dup2 syscall in forkAndExecInChild1 on some
platforms can be removed.
For #45964
Change-Id: I8e04d7b5b2488990a061a080ed64ea303ad048b8
Reviewed-on: https://go-review.googlesource.com/c/go/+/347350
Trust: Tobias Klauser <tobias.klauser@gmail.com>
Run-TryBot: Tobias Klauser <tobias.klauser@gmail.com> Reviewed-by: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Go Bot <gobot@golang.org>
WANG Xuerui [Mon, 6 Sep 2021 07:17:11 +0000 (15:17 +0800)]
cmd/internal/sys: fix placement of loong64 definition
In rebasing the patch series up to CL 339015, the branches were messed
up by me, and changes from v3 to v4 of CL 339009 was lost. Fix the
ordering to restore alphabetical order per original review.
Change-Id: I8e57c96e996c4f962cab684a9d305a8dbdeea43b
Reviewed-on: https://go-review.googlesource.com/c/go/+/347731
Run-TryBot: Tobias Klauser <tobias.klauser@gmail.com>
TryBot-Result: Go Bot <gobot@golang.org> Reviewed-by: Ian Lance Taylor <iant@golang.org>
Trust: Meng Zhuo <mzh@golangcn.org>