@@ -24,22 +24,18 @@ them.
24
24
Non-comprehensive list of changes in this release
25
25
=================================================
26
26
27
- * The optimizer will now convert calls to ``memcmp `` into a calls to ``bcmp `` in
28
- some circumstances. Users who are building freestanding code (not depending on
29
- the platform's libc) without specifying ``-ffreestanding `` may need to either
30
- pass ``-fno-builtin-bcmp ``, or provide a ``bcmp `` function.
31
-
32
27
* Two new extension points, namely ``EP_FullLinkTimeOptimizationEarly `` and
33
28
``EP_FullLinkTimeOptimizationLast `` are available for plugins to specialize
34
29
the legacy pass manager full LTO pipeline.
35
30
36
- * ** llvm-objcopy/llvm-strip ** got support for COFF object files/executables,
31
+ * `` llvm-objcopy/llvm-strip `` got support for COFF object files/executables,
37
32
supporting the most common copying/stripping options.
38
33
39
34
* The CMake parameter ``CLANG_ANALYZER_ENABLE_Z3_SOLVER `` has been replaced by
40
35
``LLVM_ENABLE_Z3_SOLVER ``.
41
36
42
- * The RISCV target is no longer "experimental" (see below for more details).
37
+ * The RISCV target is no longer "experimental" (see
38
+ `Changes to the RISCV Target `_ below for more details).
43
39
44
40
* The ORCv1 JIT API has been deprecated. Please see
45
41
`Transitioning from ORCv1 to ORCv2 <ORCv2.html#transitioning-from-orcv1-to-orcv2 >`_.
@@ -57,8 +53,13 @@ Noteworthy optimizations
57
53
able to cast away 'const'. This is (and has always been) undefined
58
54
behavior, but up until now had not been actively utilized for optimization
59
55
purposes in this exact way. For more information, please see:
60
- `bug 42763 <https://bugs.llvm.org/show_bug.cgi?id=42763>_ ` and
61
- `post commit discussion <http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20190422/646945.html>_ `.
56
+ `bug 42763 <https://bugs.llvm.org/show_bug.cgi?id=42763 >`_ and
57
+ `post commit discussion <http://lists.llvm.org/pipermail/llvm-commits/Week-of-Mon-20190422/646945.html >`_.
58
+
59
+ * The optimizer will now convert calls to ``memcmp `` into a calls to ``bcmp `` in
60
+ some circumstances. Users who are building freestanding code (not depending on
61
+ the platform's libc) without specifying ``-ffreestanding `` may need to either
62
+ pass ``-fno-builtin-bcmp ``, or provide a ``bcmp `` function.
62
63
63
64
* LLVM will now pattern match wide scalar values stored by a succession of
64
65
narrow stores. For example, Clang will compile the following function that
@@ -206,7 +207,7 @@ Changes to the PowerPC Target
206
207
207
208
* Improve precision of square root reciprocal estimate
208
209
209
- * Enabled MachinePipeliner support for P9 with -ppc-enable-pipeliner.
210
+ * Enabled MachinePipeliner support for P9 with `` -ppc-enable-pipeliner `` .
210
211
211
212
* MMX/SSE/SSE2 intrinsics headers have been ported to PowerPC using Altivec.
212
213
@@ -220,11 +221,11 @@ Changes to the PowerPC Target
220
221
221
222
* Enhancements of hardware loops interaction with LSR.
222
223
223
- * New builtins added, eg: __builtin_setrnd.
224
+ * New builtins added, eg: `` __builtin_setrnd `` .
224
225
225
226
* Various codegen improvements for both scalar and vector code
226
227
227
- * Various new exploitations and bug fixes, eg : exploited P9 maddld.
228
+ * Various new exploitations and bug fixes, e.g : exploited P9 `` maddld `` .
228
229
229
230
230
231
Changes to the SystemZ Target
@@ -295,7 +296,7 @@ External Open Source Projects Using LLVM 9
295
296
Mull - Mutation Testing tool for C and C++
296
297
------------------------------------------
297
298
298
- `Mull <https://github.com/mull-project/mull >`_ is a LLVM-based tool for
299
+ `Mull <https://github.com/mull-project/mull >`_ is an LLVM-based tool for
299
300
mutation testing with a strong focus on C and C++ languages.
300
301
301
302
Zig Programming Language
0 commit comments