Re: [swift-dev] Swift incremental compile profiling

Hey Sam, One thought: if you have an app with mixed objc and swift code, then the app-bridge.h and app-swift.h files might be creating massive choke points in your dependency graph. I have no idea how optimized the bridging functionality is but it has always seemed like a potentially weak part

Re: [swift-dev] [swift] master, stable-swift-3.0-merge-branch: [SourceKit] Use DARWIN_TOOLCHAIN_VERSION if provided, as the version to use for the XPC service name. (f242326)

> On Apr 7, 2016, at 9:34 PM, Argyrios Kyrtzidis via swift-commits > wrote: > > Repository : github.com/apple/swift > On branches: master,stable-swift-3.0-merge-branch Hey, to whomever it may concern: commit notification e-mails started going out like this this morning, with "stable-swift-3.0-

Re: [swift-dev] A type-checking performance case study

On Fri, Apr 8, 2016 at 11:47 AM, John McCall via swift-dev wrote: > I strongly recommend that we pursue declaration hiding as a language change. I completely agree with John. Not having this rule is not just a performance problem, it also leaks into the user model. For example: - If you have a

Re: [swift-dev] Failing Ubuntu 14.04 Build

Accidentally (I believe) posted to swift-build-dev. I do indeed see the error on 15.10 as well now with "missing required module 'SwiftGblic'", for whatever reason the swift binary being built cannot do a simple 'import Glibc' without this error coming out: Welcome to Swift version 3.0-dev (LLVM

[swift-dev] A type-checking performance case study

The standard library team asked the compiler team to take a look at the performance of the type-checker. Their observation was that several in-progress additions to the standard library were causing a disproportionate increase in the time it required to compile a simple-seeming expression, and

Re: [swift-dev] Why are we re-linking?

on Fri Apr 08 2016, Chris Bieneman wrote: > In Clang this is handled by having a CMake option control whether or > not the SCM revision is embedded in the binary > (CLANG_APPEND_VC_REV). Clang defaults the option to 'Off' so that > engineering workflows don't need to override it, but it is frequ

Re: [swift-dev] [swift-lldb-dev] [Swift CI] Build Still Failing: OSS - Swift Package - Ubuntu 14.04 (master) #943

Investigating now. Filed this to track: Build failure: Swift CI packaging: OSS - Swift Package - Ubuntu 14.04 (master) #943 > On Apr 8, 2016, at 9:18 AM, Mishal Shah via swift-lldb-dev > wrote: > > Hi Todd, > > We are seeing failure in LLDB, could this be related to LLDB merge from > yes

Re: [swift-dev] [Swift CI] Build Still Failing: OSS - Swift Package - Ubuntu 14.04 (master) #943

Hi Todd, We are seeing failure in LLDB, could this be related to LLDB merge from yesterday? FAILED: source/API/CMakeFiles/liblldb.dir/__/__/scripts/LLDBWrapPython.cpp.o /usr/bin/clang++ -DGTEST_HAS_RTTI=0 -DHAVE_PROCESS_VM_READV -DHAVE_ROUND -DLIBXML2_DEFINED -D__STDC_CONSTANT_MACROS -D__S

Re: [swift-dev] Why are we re-linking?

In Clang this is handled by having a CMake option control whether or not the SCM revision is embedded in the binary (CLANG_APPEND_VC_REV). Clang defaults the option to 'Off' so that engineering workflows don't need to override it, but it is frequently enabled on bots and in distributions. The b

Re: [swift-dev] Building Toolchain Fails

On 08 Apr 2016, at 01:43, Daniel Dunbar wrote: > > I believe this was a transient failure on trunk, please update and retry. It failed multiple times over many hours of update-checkout + build-toolchain. But it works now! :) /Daniel ___ swift-dev ma