From e47c672a80a2e159e1b01779f33062894e62d85b Mon Sep 17 00:00:00 2001 From: JF Bastien Date: Fri, 12 Jun 2015 14:39:36 +0200 Subject: Short document name bikeshed. --- AstSemantics.md | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'AstSemantics.md') diff --git a/AstSemantics.md b/AstSemantics.md index 05d2092..f295559 100644 --- a/AstSemantics.md +++ b/AstSemantics.md @@ -130,7 +130,7 @@ which leads to the following advantages: splitting, which can reduce throughput overhead, at the cost of increasing code size (potentially very significantly in pathological cases). * The - [signature-restricted proper tail-call](EssentialPostMVPFeatures.md#signature-restricted-proper-tail-calls) + [signature-restricted proper tail-call](PostMVP.md#signature-restricted-proper-tail-calls) feature would allow efficient compilation of arbitrary irreducible control flow. @@ -158,12 +158,12 @@ to easily be folded into the hardware load instruction *and* for groups of loads with the same base and different offsets to easily share a single bounds check. In the MVP, the indices are 32-bit unsigned integers. With -[64-bit integers](EssentialPostMVPFeatures.md#64-bit-integers) and +[64-bit integers](PostMVP.md#64-bit-integers) and [>4GiB heaps](FutureFeatures.md#heaps-bigger-than-4gib), these nodes would also accept 64-bit unsigned integers. In the MVP, heaps are not shared between threads. When -[threads](EssentialPostMVPFeatures.md#threads) are added as a feature, the basic +[threads](PostMVP.md#threads) are added as a feature, the basic `LoadHeap`/`StoreHeap` nodes will have the most relaxed semantics specified in the memory model and new heap-access nodes will be added with atomic and ordering guarantees. -- cgit v1.2.3