Martin Möhrmann [Thu, 27 Oct 2016 21:31:38 +0000 (23:31 +0200)]
cmd/compile: move stringtoslicebytetmp to the backend
- removes the runtime function stringtoslicebytetmp
- removes the generation of calls to stringtoslicebytetmp from the frontend
- adds handling of OSTRARRAYBYTETMP in the backend
This reduces binary sizes and avoids function call overhead.
Matthew Dempsky [Thu, 27 Oct 2016 09:02:30 +0000 (02:02 -0700)]
cmd/compile: stop adding implicit OKEY nodes
Keys are uncommon in array and slice literals, and normalizing
OARRAYLIT and OSLICELIT nodes to always use OKEY ends up not reducing
complexity much. Instead, only create OKEY nodes to represent explicit
keys, and recalculate implicit keys when/where necessary.
Keith Randall [Tue, 25 Oct 2016 22:49:52 +0000 (15:49 -0700)]
cmd/compile: use masks instead of branches for slicing
When we do
var x []byte = ...
y := x[i:]
We can't just use y.ptr = x.ptr + i, as the new pointer may point to the
next object in memory after the backing array.
We used to fix this by doing:
y.cap = x.cap - i
delta := i
if y.cap == 0 {
delta = 0
}
y.ptr = x.ptr + delta
That generates a branch in what is otherwise straight-line code.
Better to do:
y.cap = x.cap - i
mask := (y.cap - 1) >> 63 // -1 if y.cap==0, 0 otherwise
y.ptr = x.ptr + i &^ mask
It's about the same number of instructions (~4, depending on what
parts are constant, and the target architecture), but it is all
inline. It plays nicely with CSE, and the mask can be computed
in parallel with the index (in cases where a multiply is required).
Robert Griesemer [Thu, 27 Oct 2016 18:20:20 +0000 (11:20 -0700)]
cmd/compile, go/parser: disallow "type T = p.T" - must use "=>"
I had added this originally so we can play with different notations
but it doesn't make sense to keep it around since gofmt will convert
a type alias declaration using "=" into one using "=>" anyhow. More
importantly, the spec doesn't permit it.
Change-Id: Icb010b5a9976aebf877e48b3ce9d7245559ca494
Reviewed-on: https://go-review.googlesource.com/32105
Run-TryBot: Robert Griesemer <gri@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Michael Hudson-Doyle [Wed, 26 Oct 2016 02:57:58 +0000 (15:57 +1300)]
cmd/compile, runtime: make the go.itab.* symbols module-local
Otherwise, the way the ELF dynamic linker works means that you can end up with
the same itab being passed to additab twice, leading to the itab linked list
having a cycle in it. Add a test to additab in runtime to catch this when it
happens, not some arbitrary and surprsing time later.
Fixes #17594
Change-Id: I6c82edcc9ac88ac188d1185370242dc92f46b1ad
Reviewed-on: https://go-review.googlesource.com/32131
Run-TryBot: Michael Hudson-Doyle <michael.hudson@canonical.com> Reviewed-by: David Crawshaw <crawshaw@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Russ Cox [Wed, 26 Oct 2016 18:32:43 +0000 (14:32 -0400)]
mime: preserve unnecessary backslash escapes as literals
When MSIE sends a full file path (in "intranet mode"), it does not
escape backslashes: "C:\dev\go\foo.txt", not "C:\\dev\\go\\foo.txt".
No known MIME generators emit unnecessary backslash escapes
for simple token characters like numbers and letters.
If we see an unnecessary backslash escape, assume it is from MSIE
and intended as a literal backslash. This makes Go servers deal better
with MSIE without affecting the way they handle conforming MIME
generators.
Russ Cox [Wed, 26 Oct 2016 20:48:53 +0000 (16:48 -0400)]
net/mail: allow empty quoted string name in address again
CL 12905 disallowed "Bob" <""@example.com> but inadvertently
also disallowed "" <bob@example.com>. Move the empty string
check to apply only in the addr-spec.
Adam Langley [Wed, 26 Oct 2016 17:05:03 +0000 (10:05 -0700)]
crypto/tls: add GetClientCertificate callback
Currently, the selection of a client certificate done internally based
on the limitations given by the server's request and the certifcates in
the Config. This means that it's not possible for an application to
control that selection based on details of the request.
This change adds a callback, GetClientCertificate, that is called by a
Client during the handshake and which allows applications to select the
best certificate at that time.
(Based on https://golang.org/cl/25570/ by Bernd Fix.)
Adam Langley [Wed, 26 Oct 2016 19:30:30 +0000 (12:30 -0700)]
crypto/tls: add a SignatureScheme type.
The SignatureAndHashAlgorithm from TLS 1.2[1] is being changed to
SignatureScheme in TLS 1.3[2]. (The actual values are compatible
however.)
Since we expect to support TLS 1.3 in the future, we're already using
the name and style of SignatureScheme in the recently augmented
ClientHelloInfo. As this is public API, it seems that SignatureScheme
should have its own type and exported values, which is implemented in
this change.
Adam Langley [Wed, 26 Oct 2016 20:57:08 +0000 (13:57 -0700)]
crypto/x509: don't accept a root that already appears in a chain.
Since a root certificate is self-signed, it's a valid child of itself.
If a root certificate appeared both in the pool of intermediates and
roots the verification code could find a chain which included it twice:
first as an intermediate and then as a root. (Existing checks prevented
the code from looping any more.)
This change stops the exact same certificate from appearing twice in a
chain. This simplifies the results in the face of the common
configuration error of a TLS server returning a root certificate.
(This should also stop two different versions of the “same” root
appearing in a chain because the self-signature on one will not validate
for the other.)
Nigel Tao [Thu, 27 Oct 2016 02:36:48 +0000 (13:36 +1100)]
image/png: allow tRNS chunk without a PLTE chunk.
While https://www.w3.org/TR/PNG/#5ChunkOrdering says that tRNS's
ordering constraint is "After PLTE; before IDAT", it is legal for a tRNS
chunk to occur without a PLTE chunk at all, for greyscale and truecolor
transparency as opposed to palette-based transparency. See
https://www.w3.org/TR/PNG/#11transinfo
Fixes #17511.
Change-Id: I047b0b01d78a1cda65e00eeac229bb972cda431d
Reviewed-on: https://go-review.googlesource.com/32139 Reviewed-by: Rob Pike <r@golang.org>
Andrew Gerrand [Wed, 26 Oct 2016 04:37:15 +0000 (15:37 +1100)]
doc: s/race/ethnicity/ in the Code of Conduct
As agreed upon by the Code of Conduct working group, "race" may refer to
an attempt to classify people based on "defining characteristics",
regardless of how this people view themselves, while "ethnicity" refers
to how people identify themselves.
The Code of Conduct working group believes that the term "ethnicity"
will be more comprehensive and inclusive, and will better serve the Go
community.
Change-Id: I724b72cadb8cf29b4bac8f83017b0303feae3c94
Reviewed-on: https://go-review.googlesource.com/32133 Reviewed-by: Ian Lance Taylor <iant@golang.org> Reviewed-by: Russ Cox <rsc@golang.org>
Klaus Post [Sat, 15 Oct 2016 12:37:19 +0000 (14:37 +0200)]
compress/flate: make compression level 0 consistent
Tests for determinism was not working as intended since io.Copybuffer
uses the io.WriterTo if available.
This exposed that level 0 (no compression) changed output
based on the number of writes and buffers given to the
writer.
Previously, Write would emit a new raw block (BTYPE=00) for
every non-empty call to Write.
This CL fixes it such that a raw block is only emitted upon
the following conditions:
* A full window is obtained (every 65535 bytes)
* Flush is called
* Close is called
Change-Id: I807f866d97e2db7820f11febab30a96266a6cbf1
Reviewed-on: https://go-review.googlesource.com/31174
Run-TryBot: Joe Tsai <thebrokentoaster@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Joe Tsai <thebrokentoaster@gmail.com>
Klaus Post [Sat, 22 Oct 2016 16:50:02 +0000 (18:50 +0200)]
compress/flate: level 1 (best speed) match across blocks
This change makes deflate level 1 (best speed) match across
block boundaries. This comes at a small speed penalty,
but improves compression on almost all output.
Bill O'Farrell [Wed, 26 Oct 2016 22:35:12 +0000 (18:35 -0400)]
math/big: uses SIMD for some math big functions on s390x
The following benchmarks are improved by the amounts shown
(Others unaffected beyond the level of noise.)
Also adds a test to confirm non-SIMD implementation still correct,
even when run on SIMD-capable machine
Ben Burkert [Mon, 17 Oct 2016 21:47:48 +0000 (14:47 -0700)]
crypto/tls: add CloseWrite method to Conn
The CloseWrite method sends a close_notify alert record to the other
side of the connection. This record indicates that the sender has
finished sending on the connection. Unlike the Close method, the sender
may still read from the connection until it recieves a close_notify
record (or the underlying connection is closed). This is analogous to a
TCP half-close.
This is a rework of CL 25159 with fixes for the unstable test.
Joe Tsai [Wed, 26 Oct 2016 21:31:48 +0000 (14:31 -0700)]
unicode/utf8: optimize ValidRune
Re-writing the switch statement as a single boolean expression
reduces the number of branches that the compiler generates.
It is also arguably easier to read as a pair of numeric ranges
that valid runes can exist in.
No test changes since the existing test does a good job of
testing all of the boundaries.
This change was to gain back some performance after a correctness
fix done in http://golang.org/cl/32123.
The correctness fix (CL/32123) slowed down the benchmarks slightly:
benchmark old ns/op new ns/op delta
BenchmarkIndexRune/10-4 19.3 21.6 +11.92%
BenchmarkIndexRune/32-4 33.6 35.2 +4.76%
Since the fix relies on utf8.ValidRune, this CL improves benchmarks:
benchmark old ns/op new ns/op delta
BenchmarkIndexRune/10-4 21.6 20.0 -7.41%
BenchmarkIndexRune/32-4 35.2 33.5 -4.83%
Joe Tsai [Wed, 26 Oct 2016 21:18:37 +0000 (14:18 -0700)]
bytes, strings: fix regression in IndexRune
In all previous versions of Go, the behavior of IndexRune(s, r)
where r was utf.RuneError was that it would effectively return the
index of any invalid UTF-8 byte sequence (include RuneError).
Optimizations made in http://golang.org/cl/28537 and
http://golang.org/cl/28546 altered this undocumented behavior such
that RuneError would only match on the RuneError rune itself.
Although, the new behavior is arguably reasonable, it did break code
that depended on the previous behavior. Thus, we add special checks
to ensure that we preserve the old behavior.
There is a slight performance hit for correctness:
benchmark old ns/op new ns/op delta
BenchmarkIndexRune/10-4 19.3 21.6 +11.92%
BenchmarkIndexRune/32-4 33.6 35.2 +4.76%
This only occurs on small strings. The performance hit for larger strings
is neglible and not shown.
Carl Johnson [Wed, 26 Oct 2016 02:35:01 +0000 (22:35 -0400)]
net/http: Improve docs for Response.ParseForm
- Removes a subject-verb disagreement.
- Documents that PATCH requests also populate PostForm.
- Explains that r.PostForm is always set (but blank for GET etc.).
Than McIntosh [Mon, 24 Oct 2016 17:05:10 +0000 (13:05 -0400)]
cmd/compile: avoid truncating fieldname var locations
Don't include package path when creating LSyms for auto and param
variables during Prog generation, and update the DWARF emit routine
accordingly (remove the code that chops off package path from names in
DWARF var location expressions). Implementation suggested by mdempsky@.
The intent of this change is to have saner location expressions in cases
where the variable corresponds to a structure field. For example, the
SSA compiler's "decompose" phase can take a slice value and break it
apart into three scalar variables corresponding to the fields (slice "X"
gets split into "X.len", "X.cap", "X.ptr"). In such cases we want the
name in the location expression to omit the package path but preserve
the original variable name (e.g. "X").
Fixes #16338
Change-Id: Ibc444e7f3454b70fc500a33f0397e669d127daa1
Reviewed-on: https://go-review.googlesource.com/31819
Run-TryBot: Than McIntosh <thanm@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Matthew Dempsky <mdempsky@google.com>
David Chase [Fri, 21 Oct 2016 16:01:52 +0000 (12:01 -0400)]
cmd/compile: add explicit 'where' to EscStep data for explanations
Sometimes neither the src nor the dst of an escape edge
contains the line number appropriate to the edge, so add
a field so that can be set correctly.
Also updated some of the explanations to be less jargon-y
and perhaps more informative, and folded bug example into
test.
Cleaned up some of the function/method names in esc.go
and did a quick sanity check that each "bundling" function
was actually called often enough to justify its existence.
Austin Clements [Mon, 24 Oct 2016 18:20:07 +0000 (14:20 -0400)]
runtime: scan mark worker stacks like normal
Currently, markroot delays scanning mark worker stacks until mark
termination by putting the mark worker G directly on the rescan list
when it encounters one during the mark phase. Without this, since mark
workers are non-preemptible, two mark workers that attempt to scan
each other's stacks can deadlock.
However, this is annoyingly asymmetric and causes some real problems.
First, markroot does not own the G at that point, so it's not
technically safe to add it to the rescan list. I haven't been able to
find a specific problem this could cause, but I suspect it's the root
cause of issue #17099. Second, this will interfere with the hybrid
barrier, since there is no stack rescanning during mark termination
with the hybrid barrier.
This commit switches to a different approach. We move the mark
worker's call to gcDrain to the system stack and set the mark worker's
status to _Gwaiting for the duration of the drain to indicate that
it's preemptible. This lets another mark worker scan its G stack while
the drain is running on the system stack. We don't return to the G
stack until we can switch back to _Grunning, which ensures we don't
race with a stack scan. This lets us eliminate the special case for
mark worker stack scans and scan them just like any other goroutine.
The only subtlety to this approach is that we have to disable stack
shrinking for mark workers; they could be referring to captured
variables from the G stack, so it's not safe to move their stacks.
Rob Pike [Wed, 26 Oct 2016 02:14:20 +0000 (19:14 -0700)]
doc/effectivego: reword confusing sentence
For some reason git won't let me write
doc/effective_go.html: reword confusing sentence
or even
doc/effective_go: reword confusing sentence
as the subject line for this CL, but that's not important. The
actual CL just rewrites one sentence and adds an option to grep in
the associated example.
Fixes #15875
Change-Id: Iee159ea751caf4b73eacf3dfc86e29032646373f
Reviewed-on: https://go-review.googlesource.com/32110 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Austin Clements [Mon, 24 Oct 2016 01:53:44 +0000 (21:53 -0400)]
runtime: fix bad pointer with 0 stack barriers
Currently, if the number of stack barriers for a stack is 0, we'll
create a zero-length slice that points just past the end of the stack
allocation. This bad pointer causes GC panics.
Fix this by creating a nil slice if the stack barrier count is 0.
In practice, the only way this can happen is if
GODEBUG=gcstackbarrieroff=1 is set because even the minimum size stack
reserves space for two stack barriers.
Change-Id: I3527c9a504c445b64b81170ee285a28594e7983d
Reviewed-on: https://go-review.googlesource.com/31762 Reviewed-by: Rick Hudson <rlh@golang.org>
Austin Clements [Mon, 3 Oct 2016 20:18:17 +0000 (16:18 -0400)]
runtime: debug code to panic when marking a free object
This adds debug code enabled in gccheckmark mode that panics if we
attempt to mark an unallocated object. This is a common issue with the
hybrid barrier when we're manipulating uninitialized memory that
contains stale pointers. This also tends to catch bugs that will lead
to "sweep increased allocation count" crashes closer to the source of
the bug.
Change-Id: I443ead3eac6f316a46f50b106078b524cac317f4
Reviewed-on: https://go-review.googlesource.com/31761 Reviewed-by: Rick Hudson <rlh@golang.org>
Austin Clements [Fri, 21 Oct 2016 02:45:18 +0000 (22:45 -0400)]
runtime: simplify reflectcall write barriers
Currently reflectcall has a subtle dance with write barriers where the
assembly code copies the result values from the stack to the in-heap
argument frame without write barriers and then calls into the runtime
after the fact to invoke the necessary write barriers.
For the hybrid barrier (and for ROC), we need to switch to a
*pre*-write write barrier, which is very difficult to do with the
current setup. We could tie ourselves in knots of subtle reasoning
about why it's okay in this particular case to have a post-write write
barrier, but this commit instead takes a different approach. Rather
than making things more complex, this simplifies reflection calls so
that the argument copy is done in Go using normal bulk write barriers.
The one difficulty with this approach is that calling into Go requires
putting arguments on the stack, but the call* functions "donate" their
entire stack frame to the called function. We can get away with this
now because the copy avoids using the stack and has copied the results
out before we clobber the stack frame to call into the write barrier.
The solution in this CL is to call another function, passing arguments
in registers instead of on the stack, and let that other function
reserve more stack space and setup the arguments for the runtime.
This approach seemed to work out the best. I also tried making the
call* functions reserve 32 extra bytes of frame for the write barrier
arguments and adjust SP up by 32 bytes around the call. However, even
with the necessary changes to the assembler to correct the spdelta
table, the runtime was still having trouble with the frame layout (and
the changes to the assembler caused many other things that do strange
things with the SP to fail to assemble). The approach I took doesn't
require any funny business with the SP.
Russ Cox [Fri, 21 Oct 2016 16:41:15 +0000 (12:41 -0400)]
cmd/go: report missing vendor visibility error
The logic for saving the list of packages was not always
preferring to keep error messages around correctly.
The missed error led to an internal consistency failure later.
Russ Cox [Tue, 25 Oct 2016 19:44:14 +0000 (15:44 -0400)]
cmd/vet: diagnose non-space-separated struct tag like `json:"x",xml:"y"`
This is not strictly illegal but it probably should be (too late)
and doesn't mean what it looks like it means:
the second key-value pair has the key ",xml".
Fixes #14466.
Change-Id: I174bccc23fd28affeb87f57f77c6591634ade641
Reviewed-on: https://go-review.googlesource.com/32031
Run-TryBot: Russ Cox <rsc@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Rob Pike <r@golang.org> Reviewed-by: Quentin Smith <quentin@golang.org>
Michael Munday [Fri, 21 Oct 2016 12:39:39 +0000 (08:39 -0400)]
cmd/compile: improve s390x SSA rules for logical ops
This CL introduces some minor changes to match rules more closely
to the instructions they are targeting. s390x logical operation
with immediate instructions typically leave some bits in the
target register unchanged. This means for example that an XOR
with -1 requires 2 instructions. It is better in cases such as
this to create a constant and leave it visible to the compiler
so that it can be reused rather than hiding it in the assembler.
This CL also tweaks the rules a bit to ensure that constants are
folded when possible.
Change-Id: I1c6dee31ece00fc3c5fdf6a24f1abbc91dd2db2a
Reviewed-on: https://go-review.googlesource.com/31754 Reviewed-by: Keith Randall <khr@golang.org>
Alexander Morozov [Tue, 25 Oct 2016 21:37:54 +0000 (14:37 -0700)]
runtime: fix comments in time.go
Change-Id: I5c501f598f41241e6d7b21d98a126827a3c3ad9a
Reviewed-on: https://go-review.googlesource.com/32018 Reviewed-by: Ian Lance Taylor <iant@golang.org>
Reason for revert: Reverting the revert, which will re-enable the convI2E optimization. We originally reverted the convI2E optimization because it was making the builder fail, but the underlying cause was later determined to be unrelated.
Original CL: https://go-review.googlesource.com/31260
Revert CL: https://go-review.googlesource.com/31310
Real bug: https://go-review.googlesource.com/c/25159
Real fix: https://go-review.googlesource.com/c/31316
Change-Id: I17237bb577a23a7675a5caab970ccda71a4124f2
Reviewed-on: https://go-review.googlesource.com/32023
Run-TryBot: Keith Randall <khr@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org> Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Joe Tsai [Tue, 18 Oct 2016 21:56:19 +0000 (14:56 -0700)]
net/http: fix redirect logic to handle mutations of cookies
In the situation where the Client.Jar is set and the Request.Header
has cookies manually inserted, the redirect logic needs to be
able to apply changes to cookies from "Set-Cookie" headers to both
the Jar and the manually inserted Header cookies.
Since Header cookies lack information about the original domain
and path, the logic in this CL simply removes cookies from the
initial Header if any subsequent "Set-Cookie" matches. Thus,
in the event of cookie conflicts, the logic preserves the behavior
prior to change made in golang.org/cl/28930.
Austin Clements [Thu, 6 Oct 2016 01:22:33 +0000 (21:22 -0400)]
runtime: eliminate allspans snapshot
Now that sweeping and span marking use the sweep list, there's no need
for the work.spans snapshot of the allspans list. This change
eliminates the few remaining uses of it, which are either dead code or
can use allspans directly, and removes work.spans and its support
functions.
Austin Clements [Wed, 5 Oct 2016 22:32:21 +0000 (18:32 -0400)]
runtime: make markrootSpans time proportional to in-use spans
Currently markrootSpans iterates over all spans ever allocated to find
the in-use spans. Since we now have a list of in-use spans, change it
to iterate over that instead.
This, combined with the previous change, fixes #9265. Before these two
changes, blowing up the heap to 8GB and then shrinking it to a 0MB
live set caused the small-heap portion of the test to run 60x slower
than without the initial blowup. With these two changes, the time is
indistinguishable.
Austin Clements [Wed, 5 Oct 2016 21:50:39 +0000 (17:50 -0400)]
runtime: make sweep time proportional to in-use spans
Currently sweeping walks the list of all spans, which means the work
in sweeping is proportional to the maximum number of spans ever used.
If the heap was once large but is now small, this causes an
amortization failure: on a small heap, GCs happen frequently, but a
full sweep still has to happen in each GC cycle, which means we spent
a lot of time in sweeping.
Fix this by creating a separate list consisting of just the in-use
spans to be swept, so sweeping is proportional to the number of in-use
spans (which is proportional to the live heap). Specifically, we
create two lists: a list of unswept in-use spans and a list of swept
in-use spans. At the start of the sweep cycle, the swept list becomes
the unswept list and the new swept list is empty. Allocating a new
in-use span adds it to the swept list. Sweeping moves spans from the
unswept list to the swept list.
This fixes the amortization problem because a shrinking heap moves
spans off the unswept list without adding them to the swept list,
reducing the time required by the next sweep cycle.
Updates #9265. This fix eliminates almost all of the time spent in
sweepone; however, markrootSpans has essentially the same bug, so now
the test program from this issue spends all of its time in
markrootSpans.
Austin Clements [Tue, 4 Oct 2016 20:22:41 +0000 (16:22 -0400)]
runtime: use len(h.spans) to indicate mapped region
Currently we set the len and cap of h.spans to the full reserved
region of the address space and track the actual mapped region
separately in h.spans_mapped. Since we have both the len and cap at
our disposal, change things so len(h.spans) tracks how much of the
spans array is mapped and eliminate h.spans_mapped. This simplifies
mheap and means we'll get nice "index out of bounds" exceptions if we
do try to go off the end of the spans rather than a SIGSEGV.
Austin Clements [Tue, 4 Oct 2016 20:03:00 +0000 (16:03 -0400)]
runtime: consolidate h_spans and mheap_.spans
Like h_allspans and mheap_.allspans, these were two ways of referring
to the spans array from when the runtime was split between C and Go.
Clean this up by making mheap_.spans a slice and eliminating h_spans.
Austin Clements [Tue, 4 Oct 2016 19:56:19 +0000 (15:56 -0400)]
runtime: eliminate mheap.nspan and use range loops
This was necessary in the C days when allspans was an mspan**, but now
that allspans is a Go slice, this is redundant with len(allspans) and
we can use range loops over allspans.
Austin Clements [Tue, 4 Oct 2016 19:51:31 +0000 (15:51 -0400)]
runtime: consolidate h_allspans and mheap_.allspans
These are two ways to refer to the allspans array that hark back to
when the runtime was split between C and Go. Clean this up by making
mheap_.allspans a slice and eliminating h_allspans.
Cherry Zhang [Thu, 13 Oct 2016 10:57:00 +0000 (06:57 -0400)]
cmd/compile: add a writebarrier phase in SSA
When the compiler insert write barriers, the frontend makes
conservative decisions at an early stage. This may have false
positives which result in write barriers for stack writes.
A new phase, writebarrier, is added to the SSA backend, to delay
the decision and eliminate false positives. The frontend still
makes conservative decisions. When building SSA, instead of
emitting runtime calls directly, it emits WB ops (StoreWB,
MoveWB, etc.), which will be expanded to branches and runtime
calls in writebarrier phase. Writes to static locations on stack
are detected and write barriers are removed.
All write barriers of stack writes found by the script from
issue #17330 are eliminated (except two false positives).
Fixes #17330.
Change-Id: I9bd66333da9d0ceb64dcaa3c6f33502798d1a0f8
Reviewed-on: https://go-review.googlesource.com/31131 Reviewed-by: Austin Clements <austin@google.com> Reviewed-by: David Chase <drchase@google.com> Reviewed-by: Keith Randall <khr@golang.org>
Cherry Zhang [Mon, 24 Oct 2016 02:32:16 +0000 (22:32 -0400)]
cmd/internal/obj: save link register in leaf function with non-empty frame on PPC64, ARM64, S390X
The runtime traceback code assumes non-empty frame has link
link register saved on LR architectures. Make sure it is so in
the assember.
Also make sure that LR is stored before update SP, so the traceback
code will not see a half-updated stack frame if a signal comes
during the execution of function prologue.
Fixes #17381.
Change-Id: I668b04501999b7f9b080275a2d1f8a57029cbbb3
Reviewed-on: https://go-review.googlesource.com/31760 Reviewed-by: Michael Munday <munday@ca.ibm.com>
Lynn Boger [Tue, 25 Oct 2016 20:13:07 +0000 (15:13 -0500)]
cmd/objdump: updates from golang.org/x/arch/ppc64/ppc64asm
Update the ppc64x disassembly code for use by objdump
from golang.org/x/arch/ppc64/ppc64asm commit fcea5ea.
Enable the objdump testcase for external linking on ppc64le
make a minor fix to the expected output.