There are a few cases where this can be useful. Apart from the obvious
(and silly)
100*n + 200*n
where we generate one IMUL instead of two, consider:
15*n + 31*n
Currently, the compiler strength-reduces both imuls, generating:
0x0000 00000 MOVQ "".n+8(SP), AX
0x0005 00005 MOVQ AX, CX
0x0008 00008 SHLQ $4, AX
0x000c 00012 SUBQ CX, AX
0x000f 00015 MOVQ CX, DX
0x0012 00018 SHLQ $5, CX
0x0016 00022 SUBQ DX, CX
0x0019 00025 ADDQ CX, AX
0x001c 00028 MOVQ AX, "".~r1+16(SP)
0x0021 00033 RET
But combining the imuls is both faster and shorter:
0x0000 00000 MOVQ "".n+8(SP), AX
0x0005 00005 IMULQ $46, AX
0x0009 00009 MOVQ AX, "".~r1+16(SP)
0x000e 00014 RET
even without strength-reduction.
Moreover, consider:
5*n + 7*(n+1) + 11*(n+2)
We already have a rule that rewrites 7(n+1) into 7n+7, so the
generated code (without imuls merging) looks like this:
0x0000 00000 MOVQ "".n+8(SP), AX
0x0005 00005 LEAQ (AX)(AX*4), CX
0x0009 00009 MOVQ AX, DX
0x000c 00012 NEGQ AX
0x000f 00015 LEAQ (AX)(DX*8), AX
0x0013 00019 ADDQ CX, AX
0x0016 00022 LEAQ (DX)(CX*2), CX
0x001a 00026 LEAQ 29(AX)(CX*1), AX
0x001f 00031 MOVQ AX, "".~r1+16(SP)
But with imuls merging, the 5n, 7n and 11n factors get merged, and the
generated code looks like this:
0x0000 00000 MOVQ "".n+8(SP), AX
0x0005 00005 IMULQ $23, AX
0x0009 00009 ADDQ $29, AX
0x000d 00013 MOVQ AX, "".~r1+16(SP)
0x0012 00018 RET
Which is both faster and shorter; that's also the exact same code that
clang and the intel c compiler generate for the above expression.
Change-Id: Ib4d5503f05d2f2efe31a1be14e2fe6cac33730a9
Reviewed-on: https://go-review.googlesource.com/55143
Reviewed-by: Keith Randall <khr@golang.org>
The DWARF entries for type-specific sudog entries used the
channel value type instead of a pointer-to-value type for the elem field.
Fixes#21094
R=go1.10
Change-Id: I3f63a5664f42b571f729931309f2c9f6f38ab031
Reviewed-on: https://go-review.googlesource.com/50170
Reviewed-by: Ian Lance Taylor <iant@golang.org>
If the source importer only encounters "soft" type checking errors
it can safely return the type-checked package because it will be
completely set up. This makes the source importer slightly more
robust in the presence of errors.
Fixes#20855.
Change-Id: I5af9ccdb30eee6bca7a0fab872f6057bde521bf3
Reviewed-on: https://go-review.googlesource.com/55730
Reviewed-by: Alan Donovan <adonovan@google.com>
pkgPath always received the empty string. Worse yet, it panicked if it
received anything else. This has been the case ever since newName was
introduced in early 2016.
Change-Id: I5f164305bd30c34455ef35e776c7616f303b37e4
Reviewed-on: https://go-review.googlesource.com/54331
Run-TryBot: Daniel Martí <mvdan@mvdan.cc>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: David Crawshaw <crawshaw@golang.org>
updates #13745
Multiprecision squaring can be done in a straightforward manner
with about half the multiplications of a basic multiplication
due to the symmetry of the operands. This change implements
basic squaring for nat types and uses it for Int multiplication
when the same variable is supplied to both arguments of
z.Mul(x, x). This has some overhead to allocate a temporary
variable to hold the cross products, shift them to double and
add them to the diagonal terms. There is a speed benefit in
the intermediate range when the overhead is neglible and the
asymptotic performance of karatsuba multiplication has not been
reached.
basicSqrThreshold = 20
karatsubaSqrThreshold = 400
Were set by running calibrate_test.go to measure timing differences
between the algorithms. Benchmarks for squaring:
name old time/op new time/op delta
IntSqr/1-4 51.5ns ±25% 25.1ns ± 7% -51.38% (p=0.008 n=5+5)
IntSqr/2-4 79.1ns ± 4% 72.4ns ± 2% -8.47% (p=0.008 n=5+5)
IntSqr/3-4 102ns ± 4% 97ns ± 5% ~ (p=0.056 n=5+5)
IntSqr/5-4 161ns ± 4% 163ns ± 7% ~ (p=0.952 n=5+5)
IntSqr/8-4 277ns ± 5% 267ns ± 6% ~ (p=0.087 n=5+5)
IntSqr/10-4 358ns ± 3% 360ns ± 4% ~ (p=0.730 n=5+5)
IntSqr/20-4 1.07µs ± 3% 1.01µs ± 6% ~ (p=0.056 n=5+5)
IntSqr/30-4 2.36µs ± 4% 1.72µs ± 2% -27.03% (p=0.008 n=5+5)
IntSqr/50-4 5.19µs ± 3% 3.88µs ± 4% -25.37% (p=0.008 n=5+5)
IntSqr/80-4 11.3µs ± 4% 8.6µs ± 3% -23.78% (p=0.008 n=5+5)
IntSqr/100-4 16.2µs ± 4% 12.8µs ± 3% -21.49% (p=0.008 n=5+5)
IntSqr/200-4 50.1µs ± 5% 44.7µs ± 3% -10.65% (p=0.008 n=5+5)
IntSqr/300-4 105µs ±11% 95µs ± 3% -9.50% (p=0.008 n=5+5)
IntSqr/500-4 231µs ± 5% 227µs ± 2% ~ (p=0.310 n=5+5)
IntSqr/800-4 496µs ± 9% 459µs ± 3% -7.40% (p=0.016 n=5+5)
IntSqr/1000-4 700µs ± 3% 710µs ± 5% ~ (p=0.841 n=5+5)
Show a speed up of 10-25% in the range where basicSqr is optimal,
improved single word squaring and no significant difference when
the fallback to standard multiplication is used.
Change-Id: Iae2c82ca91cf890823f91e5c83bbe9a2c534b72b
Reviewed-on: https://go-review.googlesource.com/53638
Reviewed-by: Robert Griesemer <gri@golang.org>
Run-TryBot: Robert Griesemer <gri@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
$ gotip tool -h says:
For more about each tool command, see 'go tool command -h'.
But it's better to suggest
go doc cmd/<command>
Fixes#18313
Change-Id: I0a36d585906a5e1879e5b7927d1b6173e97cb500
Reviewed-on: https://go-review.googlesource.com/55990
Reviewed-by: Ian Lance Taylor <iant@golang.org>
The current implementation of the extended Euclidean GCD algorithm
calculates both cosequences x and y inside the division loop. This
is unneccessary since the second Bezout coefficient can be obtained
at the end of calculation via a multiplication, subtraction and a
division. In case only one coefficient is needed, e.g. ModInverse
this calculation can be skipped entirely. This is a standard
optimization, see e.g.
"Handbook of Elliptic and Hyperelliptic Curve Cryptography"
Cohen et al pp 191
Available at:
http://cs.ucsb.edu/~koc/ccs130h/2013/EllipticHyperelliptic-CohenFrey.pdf
Updates #15833
Change-Id: I1e0d2e63567cfed97fd955048fe6373d36f22757
Reviewed-on: https://go-review.googlesource.com/50530
Reviewed-by: Robert Griesemer <gri@golang.org>
The current implementation uses a shift and add
loop to compute the product of x's exponent xe and
the integer part of y (yi) for yi up to 1<<63.
Since xe is an 11-bit exponent, this product can be
up to 74-bits and overflow both 32 and 64-bit int.
This change checks whether the accumulated exponent
will fit in the 11-bit float exponent of the output
and breaks out of the loop early if overflow is detected.
The current handling of yi >= 1<<63 uses Exp(y * Log(x))
which incorrectly returns Nan for x<0. In addition,
for y this large, Exp(y * Log(x)) can be enumerated
to only overflow except when x == -1 since the
boundary cases computed exactly:
Pow(NextAfter(1.0, Inf(1)), 1<<63) == 2.72332... * 10^889
Pow(NextAfter(1.0, Inf(-1)), 1<<63) == 1.91624... * 10^-445
exceed the range of float64. So, the call can be
replaced with a simple case statement analgous to
y == Inf that correctly handles x < 0 as well.
Fixes#7394
Change-Id: I6f50dc951f3693697f9669697599860604323102
Reviewed-on: https://go-review.googlesource.com/48290
Reviewed-by: Robert Griesemer <gri@golang.org>
The Writer logic was not consistent about when an IO error would
persist across multiple calls on Writer's methods.
Thus, to make the error handling more consistent we always check
the persistent state of the error prior to every exported method
call, and return an error if set. Otherwise, it is the responsibility
of every exported method to persist any fatal errors that may occur.
As a simplification, we can remove the close field since that
information can be represented by simply storing ErrWriteAfterClose
in the err field.
Change-Id: I8746ca36b3739803e0373253450db69b3bd12f38
Reviewed-on: https://go-review.googlesource.com/55590
Run-TryBot: Joe Tsai <joetsai@digital-static.net>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
The GNU tar format defines the following type flags:
TypeGNULongName = 'L' // Next file has a long name
TypeGNULongLink = 'K' // Next file symlinks to a file w/ a long name
Anytime a string exceeds the field dedicated to store it, the GNU format
permits a fake "file" to be prepended where that file entry has a Typeflag
of 'L' or 'K' and the contents of the file is a NUL-terminated string.
Contrary to previous TODO comments,
the GNU format supports arbitrary strings (without NUL) rather UTF-8 strings.
The manual says the following:
<<<
The name, linkname, magic, uname, and gname are
null-terminated character strings
>>>
<<<
All characters in header blocks are represented
by using 8-bit characters in the local variant of ASCII.
>>>
From this description, we gather the following:
* We must forbid NULs in any GNU strings
* Any 8-bit value (other than NUL) is permitted
Since the modern world has moved to UTF-8, it is really difficult to
determine what a "local variant of ASCII" means. For this reason,
we treat strings as just an arbitrary binary string (without NUL)
and leave it to the user to determine the encoding of this string.
(Practically, it seems that UTF-8 is the typical encoding used
in GNU archives seen in the wild).
The implementation of GNU tar seems to confirm this interpretation
of the manual where it permits any arbitrary binary string to exist
within these fields so long as they do not contain the NUL character.
$ touch `echo -e "not\x80\x81\x82\x83utf8"`
$ gnutar -H gnu --tar -cvf gnu-not-utf8.tar $(echo -e "not\x80\x81\x82\x83utf8")
The fact that we permit arbitrary binary in GNU strings goes
hand-in-hand with the fact that GNU also permits a "base-256" encoding
of numeric fields, which is effectively two-complement binary.
Change-Id: Ic037ec6bed306d07d1312f0058594bd9b64d9880
Reviewed-on: https://go-review.googlesource.com/55573
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Joe Tsai <thebrokentoaster@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
In the existing implementation, if pattern is an empty string,
program calls a panic with the message which is a concatenation of
"http: invalid pattern " and pattern.
In this case, pattern is an empty, so the commit removes
this concatenation and the trailing space.
Fixes: #21102
Change-Id: I49f58b52d835311a6ac642de871eb15646e48a54
Reviewed-on: https://go-review.googlesource.com/50350
Reviewed-by: Dmitri Shuralyov <shurcool@gmail.com>
Reviewed-by: Tom Bergan <tombergan@google.com>
Run-TryBot: Tom Bergan <tombergan@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
The code was adding race.Errors to t.raceErrors before checking
Failed, but Failed was using t.raceErrors+race.Errors. We don't want
to change Failed, since that would affect tests themselves, so modify
the harness to not unnecessarily change t.raceErrors.
Updates #19851Fixes#21338
Change-Id: I7bfdf281f90e045146c92444f1370d55c45221d4
Reviewed-on: https://go-review.googlesource.com/54050
Reviewed-by: Joe Tsai <thebrokentoaster@gmail.com>
Run-TryBot: Joe Tsai <thebrokentoaster@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
This reduces the code footprint of code like:
println("foo=", foo, "bar=", bar)
which is fairly common in the runtime.
Prior to this change, this makes function calls to print each of:
"foo=", " ", foo, " ", "bar=", " ", bar, "\n"
After this change, this prints:
"foo= ", foo, " bar= ", bar, "\n"
This shrinks the hello world binary by 0.4%.
More importantly, this improves the instruction
density of important runtime routines.
Change-Id: I8971bdf5382fbaaf4a82bad4442f9da07c28d395
Reviewed-on: https://go-review.googlesource.com/55098
Run-TryBot: Josh Bleecher Snyder <josharian@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Rather than emitting spaces and newlines for println
as we walk the expression, construct it all up front.
This enables further optimizations.
This requires using printstring instead of print in
the implementation of printsp and printnl,
on pain of infinite recursion.
That's ok; it's more efficient anyway, and just as simple.
While we're here, do it for other print routines as well.
Change-Id: I61d7df143810e00710c4d4d948d904007a7fd190
Reviewed-on: https://go-review.googlesource.com/55097
Run-TryBot: Josh Bleecher Snyder <josharian@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Matthew Dempsky <mdempsky@google.com>
Writing to selectdone on the stack of another goroutine meant a
pretty subtle dance between the select code and the stack copying
code. Instead move the selectdone variable into the g struct.
Change-Id: Id246aaf18077c625adef7ca2d62794afef1bdd1b
Reviewed-on: https://go-review.googlesource.com/53390
Reviewed-by: Austin Clements <austin@google.com>
Run-TryBot: Austin Clements <austin@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
I noticed that we don't set an itab's function pointers at compile
time. Instead, we currently do it at executable startup.
Set the function pointers at compile time instead. This shortens
startup time. It has no effect on normal binary size. Object files
will have more relocations, but that isn't a big deal.
For PIE there are additional pointers that will need to be adjusted at
load time. There are already other pointers in an itab that need to be
adjusted, so the cache line will already be paged in. There might be
some binary size overhead to mark these pointers. The "go test -c
-buildmode=pie net/http" binary is 0.18% bigger.
Update #20505
Change-Id: I267c82489915b509ff66e512fc7319b2dd79b8f7
Reviewed-on: https://go-review.googlesource.com/44341
Run-TryBot: Keith Randall <khr@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Martin Möhrmann <moehrmann@google.com>
Currently, GC captures the start-the-world time stamp after
startTheWorldWithSema returns. This is problematic for two reasons:
1. It's possible to get preempted between startTheWorldWithSema
starting the world and calling nanotime.
2. startTheWorldWithSema does several clean-up tasks after the world
is up and running that on rare occasions can take upwards of 10ms.
Since the runtime uses the start-the-world time stamp to compute the
STW duration, both of these can significantly inflate the reported STW
duration.
Fix this by having startTheWorldWithSema itself call nanotime once the
world is started.
Change-Id: I114630234fb73c9dabae50a2ef1884661f2459db
Reviewed-on: https://go-review.googlesource.com/55410
Run-TryBot: Austin Clements <austin@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Rick Hudson <rlh@golang.org>
Detected by BoGo test FragmentAcrossChangeCipherSpec-Server-Packed.
Change-Id: I9a76697b9cdeb010642766041971de5c7e533481
Reviewed-on: https://go-review.googlesource.com/48811
Reviewed-by: Adam Langley <agl@golang.org>
Run-TryBot: Adam Langley <agl@golang.org>
The processClientKeyExchange and processServerKeyExchange functions unmarshal an
encoded EC point and explicitly check whether the point is on the curve. The explicit
check can be omitted because elliptic.Unmarshal fails if the point is not on the curve
and the returned error would always be the same.
Fixes#20496
Change-Id: I5231a655eace79acee2737dd036a0c255ed42dbb
Reviewed-on: https://go-review.googlesource.com/44311
Reviewed-by: Adam Langley <agl@golang.org>
Reviewed-by: Avelino <t@avelino.xxx>
Run-TryBot: Adam Langley <agl@golang.org>
asn1.NullRawValue was used according to RFC 3279 2.2.1. Without this tag,
the output didn't match openssl.
Fixes#19972
Change-Id: Ia52ddb810888837f913dbd65c4e1328f6c8084bb
Reviewed-on: https://go-review.googlesource.com/40730
Reviewed-by: Adam Langley <agl@golang.org>
Run-TryBot: Adam Langley <agl@golang.org>
Run-TryBot: Emmanuel Odeke <emm.odeke@gmail.com>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
golang.org/cl/55130 added utimensat for Solaris but didn't use it in
UtimesNano (despite indicating otherwise in the commit message). Fix
this by also using utimensat for UtimesNano on Solaris.
Because all versions of Solaris suppported by Go support utimensat,
there is no need for the fallback logic and utimensat can be called
unconditionally.
This issue was pointed out by Shawn Walker-Salas.
Updates #16480
Change-Id: I114338113a6da3cfcb8bca950674bdc8f5a7a9e5
Reviewed-on: https://go-review.googlesource.com/55141
Run-TryBot: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Some IDEs (e.g. intellij IDE based ones) create the .idea folder
to store project specific settings. This is irrelevant to Go project
that does not assume any specific IDEs, but interferes with git.
Change-Id: I0c93d9a3f7edff095fbe0c7a53b06c92b391c970
Reviewed-on: https://go-review.googlesource.com/53770
Reviewed-by: Ian Lance Taylor <iant@golang.org>
The destination slice does not need to be created at all. The source
slice itself can be used as the destination because the decode loop
increments by one and then the 'seen' byte is not used anymore. Therefore
the decoded byte can be stored in that index of the source slice itself.
This trick cannot be applied to EncodeString() because in that case,
the destination slice is large than the source. And for a single byte
in the source slice, two bytes in the destination slice is written.
func BenchmarkDecodeString(b *testing.B) {
for i := 0; i < b.N; i++ {
DecodeString("0123456789abcdef")
}
}
name old time/op new time/op delta
DecodeString 71.0ns ± 6% 58.0ns ± 0% -18.28% (p=0.008 n=5+5)
name old alloc/op new alloc/op delta
DecodeString 16.0B ± 0% 8.0B ± 0% -50.00% (p=0.008 n=5+5)
name old allocs/op new allocs/op delta
DecodeString 1.00 ± 0% 1.00 ± 0% ~ (all equal)
Change-Id: Id98db4e712444557a804155457a4dd8d1b8b416d
Reviewed-on: https://go-review.googlesource.com/55611
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
CL golang.org/cl/55130 messed up the definition of _AT_FDCWD on
dragonfly.
This fixes the following test failure on dragonfly/amd64:
--- FAIL: TestPackageMainTestImportsArchiveNotBinary (0.00s)
go_test.go:192: chtimes ./testdata/src/main_test/m.go: bad file descriptor
Change-Id: I4c96983769e6b02d714859dc838875c3c0f1be50
Reviewed-on: https://go-review.googlesource.com/55690
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
The stxr/stxrw/stxrb/stxrh instructions belong to STLXR-like instructions
set and they require special handling. The current code has no special
handling for those instructions.
The fix adds the special handling for those instructions.
Uncomment stxr/stxrw/stxrb/stxrh test cases.
Fixes#21397
Change-Id: I31cee29dd6b30b1c25badd5c7574dda7a01bf016
Reviewed-on: https://go-review.googlesource.com/54951
Run-TryBot: Cherry Zhang <cherryyz@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
Improve static branch prediction in arm64 wrapper prologue
by making the unusual case branch forwards. (Most other
architectures implement this optimization.)
Additionally, replace a CMP+BNE pair with a CBNZ
to save one instruction.
Change-Id: Id970038b34b4aaec18c101d62e2ee00f3e32a761
Reviewed-on: https://go-review.googlesource.com/54070
Run-TryBot: Cherry Zhang <cherryyz@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
Otherwise, some test flags don't work.
Change-Id: Iacf3930d0eec28e4d690cd382adbb2ecf866a0e2
Reviewed-on: https://go-review.googlesource.com/55615
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Add support for generating TBZ/TBNZ instructions.
The bit-test-and-branch pattern shows up in a number of
important places, including the runtime (gc bitmaps).
Before this change, there were 3 TB[N]?Z instructions in the Go tool,
all of which were in hand-written assembly. After this change, there
are 285. Also, the go1 benchmark binary gets about 4.5kB smaller.
Fixes#21361
Change-Id: I170c138b852754b9b8df149966ca5e62e6dfa771
Reviewed-on: https://go-review.googlesource.com/54470
Run-TryBot: Philip Hofer <phofer@umich.edu>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Cherry Zhang <cherryyz@google.com>
TestSizes has been added in CL 55551. This test is
failing on Plan 9 because executables don't have
a DWARF symbol table.
Fixes#21453.
Change-Id: I560611b49aea5417e8c5ac0cec6c7882bd9f8335
Reviewed-on: https://go-review.googlesource.com/55692
Run-TryBot: David du Colombier <0intro@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Martin Möhrmann <moehrmann@google.com>
struct32 and struct40 structs are already declared, remove them to
make runtime tests build.
Change-Id: I3814f2b850dcb15c4002a3aa22e2a9326e5a5e53
Reviewed-on: https://go-review.googlesource.com/55614
Reviewed-by: Martin Möhrmann <moehrmann@google.com>
The logic for USTAR was disabled because a previous implementation of
Writer had a wrong understanding of the differences between USTAR and GNU,
causing the prefix field is incorrectly be populated in GNU files.
Now that this issue has been fixed, we can re-enable the logic for USTAR
path splitting, which allows Writer to use the USTAR for a wider range
of possible inputs.
Updates #9683
Updates #12594
Updates #17630
Change-Id: I9fe34e5df63f99c6dd56fee3a7e7e4d6ec3995c9
Reviewed-on: https://go-review.googlesource.com/55574
Run-TryBot: Joe Tsai <thebrokentoaster@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Move all sentinel errors to common.go since some of them are
returned by both the reader and writer and remove errInvalidHeader
since it not used.
Also, consistently use the "tar: " prefix for errors.
Change-Id: I0afb185bbf3db80dfd9595321603924454a4c2f9
Reviewed-on: https://go-review.googlesource.com/55650
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Move variable declarations closer to their first uses.
Use an additional string variable s0 for error reporting that references
the original input string. This allows the variable s to be modified.
Change-Id: I4725152490ca1dc10c1161ad8ad2f4ae8933493f
Reviewed-on: https://go-review.googlesource.com/55138
Reviewed-by: Joe Tsai <thebrokentoaster@gmail.com>
Reviewed-by: Marvin Stenger <marvin.stenger94@gmail.com>
Run-TryBot: Joe Tsai <thebrokentoaster@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Instead of comparing if the number of elements will
not fit into memory check if the memory size of the
slices backing memory is higher then the memory limit.
This avoids a division or maxElems lookup.
With et.size > 0:
uintptr(newcap) > maxSliceCap(et.size)
-> uintptr(int(capmem / et.size)) > _MaxMem / et.size
-> capmem / et.size > _MaxMem / et.size
-> capmem > _MaxMem
Note that due to integer division from capmem > _MaxMem
it does not follow that uintptr(newcap) > maxSliceCap(et.size).
Consolidated runtime GrowSlice benchmarks by using sub-benchmarks and
added more struct sizes to show performance improvement when division
is avoided for element sizes larger than 32 bytes.
AMD64:
GrowSlice/Byte 38.9ns ± 2% 38.9ns ± 1% ~ (p=0.974 n=20+20)
GrowSlice/Int 58.3ns ± 3% 58.0ns ± 2% ~ (p=0.154 n=20+19)
GrowSlice/Ptr 95.7ns ± 2% 95.1ns ± 2% -0.60% (p=0.034 n=20+20)
GrowSlice/Struct/24 95.4ns ± 1% 93.9ns ± 1% -1.54% (p=0.000 n=19+19)
GrowSlice/Struct/32 110ns ± 1% 108ns ± 1% -1.76% (p=0.000 n=19+20)
GrowSlice/Struct/40 138ns ± 1% 128ns ± 1% -7.09% (p=0.000 n=20+20)
Change-Id: I1c37857c74ea809da373e668791caffb6a5cbbd3
Reviewed-on: https://go-review.googlesource.com/53471
Run-TryBot: Martin Möhrmann <moehrmann@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>
