- 16 Apr, 2015 14 commits
-
-
David Crawshaw authored
Change-Id: I402629b154ae36c879a8d5088cb0f71b075f31dc Reviewed-on: https://go-review.googlesource.com/8958Reviewed-by:
Josh Bleecher Snyder <josharian@gmail.com>
-
Shenghou Ma authored
Fixes #10116. Change-Id: I3b3f80791a1db4c2b7318f81a115972cd2237f05 Signed-off-by:
Shenghou Ma <minux@golang.org> Reviewed-on: https://go-review.googlesource.com/8784Reviewed-by:
David Crawshaw <crawshaw@golang.org>
-
Shenghou Ma authored
Change-Id: I3b3f80791a1db4c2b7318f81a115972cd2237f04 Signed-off-by:
-
David Crawshaw authored
We no longer need the EXC_BAD_ACCESS watcher as runtime/cgo contains a mach exception handler that catches it. And now lldb only intermittently reports process connection and exiting, so instead just look for the PASS from Go. Change-Id: I403266558f5a900e0b87ec1019d9baec88148d23 Reviewed-on: https://go-review.googlesource.com/8957Reviewed-by:
-
Matthew Dempsky authored
If the requested coding bit sizes don't result in a full binary tree, then reject the input as invalid. Exception: We still need to allow degenerate Huffman codings with a single 1-bit code to be compatible with zlib and files compressed with Go's compress/flate package. Update #10426. Change-Id: I171b98d12e65b4deb9f4031cd802407ebb5e266c Reviewed-on: https://go-review.googlesource.com/8922Reviewed-by:
Nigel Tao <nigeltao@golang.org>
-
Dave Day authored
Change-Id: I6089cebf7de73ce9c4e53ad3e8ef2673d970bda2 Reviewed-on: https://go-review.googlesource.com/8908Reviewed-by:
Rob Pike <r@golang.org>
-
Shenghou Ma authored
Change-Id: I3b3f80791a1db4c2b7318f81a115972cd2237f02 Signed-off-by:
-
Shenghou Ma authored
ld64 cannot handle BR26 reloc with non-zero addend. It incorrectly thinks that non-zero addend for BR26 means the code is not PIC, but those BR26 relocs should be fully resolved at link time. Change-Id: I3b3f80791a1db4c2b7318f81a115972cd2237f01 Signed-off-by:
-
Shenghou Ma authored
To support "SVC $0x80", which is needed for darwin/arm64. Change-Id: I3b3f80791a1db4c2b7318f81a115972cd2237f00 Signed-off-by:
-
Matthew Dempsky authored
This code's test coverage is ad hoc at best, and it's easy to make changes that accidentally regress invariants. This CL adds a "sanity" constant that can be changed to "true" during development to add extra runtime checking that the Huffman decoder tables are sane. Change-Id: I0d0ca53ad7c9566be18046d9b255e1a30059f28b Reviewed-on: https://go-review.googlesource.com/8974Reviewed-by:
-
Matthew Dempsky authored
When decoding Huffman codes, if an invalid bit sequence is discovered, reject the input instead of treating it as a 0-length code. Fixes #10426. Change-Id: Ie2f1a3a718afd7c6bee73a67480d4b84936c21c9 Reviewed-on: https://go-review.googlesource.com/8893 Run-TryBot: Matthew Dempsky <mdempsky@google.com> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by:
-
Dave Day authored
This also involves adding functions to typelinks along with a minor change to ensure they are sorted correctly. Change-Id: I054a79b6498a634cbccce17579f52c299733c2cf Reviewed-on: https://go-review.googlesource.com/1996Reviewed-by:
Ian Lance Taylor <iant@golang.org>
-
Nigel Tao authored
Change-Id: I211c0d33dc292c6a703d788f6d4d286107bcb6b0 Reviewed-on: https://go-review.googlesource.com/8906Reviewed-by:
-
Mikio Hara authored
Under some dial tests that require external network connectivity, we must prevent application traffic but must not interfere with control plane traffic such as DNS message exchange. But test helper function disableSocketConnect prevents both application and control plane traffic unconditionally and makes some dial tests with -ipv6 fail when CGO_ENABLED=0. This change makes disableSocketConnect take a look at not only address family but socket type for fixing some dial tests with -ipv6 when CGO_ENBALED=0. Change-Id: I32241d9592d31483424bb5e69cb4d56f3fc20312 Reviewed-on: https://go-review.googlesource.com/8743Reviewed-by:
-
- 15 Apr, 2015 22 commits
-
-
Michael Hudson-Doyle authored
Change-Id: I09e84161d106960a69972f5fc845a1e40c28e58f Reviewed-on: https://go-review.googlesource.com/8331 Run-TryBot: Ian Lance Taylor <iant@golang.org> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by:
-
Michael Hudson-Doyle authored
You can now do 'go install -buildmode=shared std' and get yourself a nice (33 meg) libstd.so (which is not useful until there is -linkshared support as well, of course). Change-Id: Ie9b7e7f72abc7d369a6e3ecc98903a9d197bd6e6 Reviewed-on: https://go-review.googlesource.com/8300 Run-TryBot: Ian Lance Taylor <iant@golang.org> TryBot-Result: Gobot Gobot <gobot@golang.org> Reviewed-by:
-
Ian Lance Taylor authored
Move the single file from internal/syscall to internal/syscall/unix, to match the golang.org/x/sys layout. Change-Id: I2fb2832b4cb22efc7666bd276f5401ac3e73dc40 Reviewed-on: https://go-review.googlesource.com/8972Reviewed-by:
Brad Fitzpatrick <bradfitz@golang.org> Reviewed-by:
-
David Crawshaw authored
Change-Id: I9ceceb29291ea9f5d7b675dfabd665c5e3618471 Reviewed-on: https://go-review.googlesource.com/8955Reviewed-by:
-
Michael Hudson-Doyle authored
Fixes #10449 Change-Id: I1dc2d0213e6a46f3609222d5460c1a54081e2471 Reviewed-on: https://go-review.googlesource.com/8931Reviewed-by:
-
Michael Hudson-Doyle authored
This is $GOPATH/pkg/linux_amd64 or similar. cmd/go already had a grotty calculation of this and I need to add another one for -buildmode=shared. Change-Id: Ied28c9b7cce671da8d45920e124a3e0c2501258a Reviewed-on: https://go-review.googlesource.com/8930Reviewed-by:
-
Michael Hudson-Doyle authored
Including having -r "" preventing rpath from being set at all. Change-Id: Ib40d7bf93a6e9ef21985c4a05b5703e4fbd1cd1b Reviewed-on: https://go-review.googlesource.com/8806Reviewed-by:
-
Josh Bleecher Snyder authored
It is faster to execute MOVQ AX,(DI) MOVQ AX,8(DI) MOVQ AX,16(DI) MOVQ AX,24(DI) ADDQ $32,DI than STOSQ STOSQ STOSQ STOSQ However, in order to be able to jump into the middle of a block of MOVQs, the call site needs to pre-adjust DI. If we're clearing a small area, the cost of that DI pre-adjustment isn't repaid. This CL switches the DUFFZERO implementation to use a hybrid strategy, in which small clears use STOSQ as before, but large clears use mostly MOVQ/ADDQ blocks. benchmark old ns/op new ns/op delta BenchmarkClearFat8 0.55 0.55 +0.00% BenchmarkClearFat12 0.82 0.83 +1.22% BenchmarkClearFat16 0.55 0.55 +0.00% BenchmarkClearFat24 0.82 0.82 +0.00% BenchmarkClearFat32 2.20 1.94 -11.82% BenchmarkClearFat40 1.92 1.66 -13.54% BenchmarkClearFat48 2.21 1.93 -12.67% BenchmarkClearFat56 3.03 2.20 -27.39% BenchmarkClearFat64 3.26 2.48 -23.93% BenchmarkClearFat72 3.57 2.76 -22.69% BenchmarkClearFat80 3.83 3.05 -20.37% BenchmarkClearFat88 4.14 3.30 -20.29% BenchmarkClearFat128 5.54 4.69 -15.34% BenchmarkClearFat256 9.95 9.09 -8.64% BenchmarkClearFat512 18.7 17.9 -4.28% BenchmarkClearFat1024 36.2 35.4 -2.21% Change-Id: Ic786406d9b3cab68d5a231688f9e66fcd1bd7103 Reviewed-on: https://go-review.googlesource.com/2585Reviewed-by:
Keith Randall <khr@golang.org>
-
Josh Bleecher Snyder authored
Val is used to hold constant values. Reg was the odd duck out. Generated using eg. No functional changes. Passes toolstash -cmp. Change-Id: Ic1de769a1f92bb02e09a4428d998b716f307e2f6 Reviewed-on: https://go-review.googlesource.com/8912Reviewed-by:
Russ Cox <rsc@golang.org>
-
Michael Hudson-Doyle authored
By removing type slice, renaming type sliceStruct to type slice and whacking until it compiles. Has a pleasing net reduction of conversions. Fixes #10188 Change-Id: I77202b8df637185b632fd7875a1fdd8d52c7a83c Reviewed-on: https://go-review.googlesource.com/8770Reviewed-by:
-
Michael Hudson-Doyle authored
We forgot to add the !gccgo tag to cpuid_386.s. Change-Id: I2de2ed92ac9686c9365cb37cd29121fa98c2bf37 Reviewed-on: https://go-review.googlesource.com/8960Reviewed-by:
Dave Cheney <dave@cheney.net> Reviewed-by:
-
Rob Pike authored
Change-Id: Ia0d7c2f08f8f124bcd62c7f9744227812d68903d Reviewed-on: https://go-review.googlesource.com/8941Reviewed-by:
-
Rob Pike authored
When a reflect.Value is passed to Printf (etc.), fmt called the String method, which does not disclose its contents. To get the contents, one could call Value.Interface(), but that is illegal if the Value is not exported or otherwise forbidden. This CL improves the situation with a trivial change to the fmt package: when we see a reflect.Value as an argument, we treat it exactly as we treat a reflect.Value we make inside the package. This means that we always print the contents of the Value as if _that_ was the argument to Printf. This is arguably a breaking change but I think it is a genuine improvement and no greater a break than many other tweaks we have made to formatted output from this package. Fixes #8965. Change-Id: Ifc2a4ce3c1134ad5160e101d2196c22f1542faab Reviewed-on: https://go-review.googlesource.com/8731Reviewed-by:
roger peppe <rogpeppe@gmail.com> Reviewed-by:
-
Rob Pike authored
This trivial addition to the io package makes it easy to control the buffer size and allocation properties of io.Copy. Change-Id: Ica1a6bd015e429d4e655bc0c6f66cea21c454acf Reviewed-on: https://go-review.googlesource.com/8730Reviewed-by:
-
Dave Cheney authored
Fixes #9855 Use an architectural zero register as the source for zeroing, if available. Change-Id: Ie5b4ba4e3d356c6f892bfd1cebd14d5152bdeeb0 Reviewed-on: https://go-review.googlesource.com/8722Reviewed-by:
-
Mikio Hara authored
Change-Id: I63053c45081f47ba736720a78d85c31c13767891 Reviewed-on: https://go-review.googlesource.com/8642Reviewed-by:
-
Mikio Hara authored
Change-Id: Icaac9a48a3b9a3c5542235162e21ab8303592965 Reviewed-on: https://go-review.googlesource.com/8641Reviewed-by:
-
Dave Cheney authored
Fixes #10450 runtime.cputicks is called from runtime.exitsyscall and must not split the stack. cputicks is implemented in several ways and the NOSPLIT annotation was missing from a few of these. Change-Id: I5cbbb4e5888c5d298fe2fef240782d0e49f59af8 Reviewed-on: https://go-review.googlesource.com/8939Reviewed-by:
Aram Hăvărneanu <aram@mgk.ro>
-
Alex Brainman authored
When Windows calls externalthreadhandler it expects to receive return value in AX. We don't set AX anywhere. Change that. Store ctrlhandler1 and profileloop1 return values into AX before returning from externalthreadhandler. Fixes #10215. Change-Id: Ied04542cc3ebe7d4a26660e970f9f78098143591 Reviewed-on: https://go-review.googlesource.com/8901Reviewed-by:
-
Nigel Tao authored
http://www.w3.org/TR/PNG/#5ChunkOrdering disallows them. Fixes #10423 Change-Id: I3399ce53dc8b41b1b5f0b906a5912e6efd80418f Reviewed-on: https://go-review.googlesource.com/8905Reviewed-by:
-
Robert Griesemer authored
- The go/importer package provides access to compiler-specific importers. - Adjusted go/internal/gcimporter and go/types as needed. - types.Check was removed - not much simpler than calling types.Config.Check. - Package "unsafe" is now handled by the type checker; importers are not called for it anymore. - In std lib tests, re-use importer for faster testing (no need to re-import previously imported packages). - Minor cleanups. The code still needs cleanups before submitting. Change-Id: Idd456da2e9641688fe056504367348926feb0755 Reviewed-on: https://go-review.googlesource.com/8767Reviewed-by:
Alan Donovan <adonovan@google.com> Run-TryBot: Robert Griesemer <gri@golang.org>
-
Josh Bleecher Snyder authored
Use environment variables to allow set-and-forget. Add a script to attempt to autodetect codesign info. Change-Id: Ic56b9c5f097b1a4117ebb89c408bc333d91f581d Reviewed-on: https://go-review.googlesource.com/8910Reviewed-by:
-
- 14 Apr, 2015 4 commits
-
-
Josh Bleecher Snyder authored
No functional changes. Passes toolstash -cmp. Change-Id: I4fba0c248645c3910ee3f7fc99dacafb676c5dc2 Reviewed-on: https://go-review.googlesource.com/8911Reviewed-by:
-
Austin Clements authored
A G will be preempted if it runs for 10ms without blocking. Currently this constant is hard-coded in retake. Move it to a global const. We'll use the time slice length in scheduling background GC. Change-Id: I79a979948af2fad3afe5df9d4af4062f166554b7 Reviewed-on: https://go-review.googlesource.com/8838Reviewed-by:
Rick Hudson <rlh@golang.org>
-
Austin Clements authored
mHeap_ReclaimList is asked to reclaim at least npages pages, but it counts the number of spans reclaimed, not the number of pages reclaimed. The number of spans reclaimed is strictly larger than the number of pages, so this is not strictly wrong, but it is forcing more reclamation than was intended by the caller, which delays large allocations. Fix this by increasing the count by the number of pages in the swept span, rather than just increasing it by 1. Fixes #9048. Change-Id: I5ae364a9837a6012e68fcd431bba000340cfd50c Reviewed-on: https://go-review.googlesource.com/8920Reviewed-by:
Dmitry Vyukov <dvyukov@google.com> Reviewed-by:
-
Austin Clements authored
Commit d7e0ad4b removed the next_gc manipulation from mSpan_Sweep, but left in the traceNextGC() for recording the updated next_gc value. Remove this now unnecessary call. Change-Id: I28e0de071661199be9810d7bdcc81ce50b5a58ae Reviewed-on: https://go-review.googlesource.com/8894Reviewed-by:
-
