llvm.org GIT mirror llvm / 1415ff0
[ReleaseNotes] Mention various windows related changes in 7.0 git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_70@339646 91177308-0d34-0410-b5e6-96231b3b80d8 Martin Storsjo 1 year, 2 months ago
1 changed file(s) with 34 addition(s) and 6 deletion(s). Raw diff Collapse all Expand all
120120 not to be compliant, and higher optimization levels will still emit some
121121 information in v4 format.
122122
123 * Added support for the ``.rva`` assembler directive for COFF targets.
124
125 * The :program:`llvm-rc` tool (Windows Resource Compiler) has been improved
126 a bit. There are still known missing features, but it is generally usable
127 in many cases. (The tool still doesn't preprocess input files automatically,
128 but it can now handle leftover C declarations in preprocessor output, if
129 given output from a preprocessor run externally.)
130
131 * CodeView debug info can now be emitted MinGW configurations, if requested.
132
123133 * Note..
124134
125135 .. NOTE
143153
144154 * invariant.group metadata can now refer only empty metadata nodes.
145155
146 Changes to the ARM Backend
147 --------------------------
148
149 During this release ...
150
156 Changes to the AArch64 Target
157 -----------------------------
158
159 * The ``.inst`` assembler directive is now usable on both COFF and Mach-O
160 targets, in addition to ELF.
161
162 * Support for most remaining COFF relocations have been added.
163
164 * Support for TLS on Windows has been added.
165
166 Changes to the ARM Target
167 -------------------------
168
169 * The ``.inst`` assembler directive is now usable on both COFF and Mach-O
170 targets, in addition to ELF. For Thumb, it can now also automatically
171 deduce the instruction size, without having to specify it with
172 e.g. ``.inst.w`` as before.
151173
152174 Changes to the MIPS Target
153175 --------------------------
183205 Changes to the X86 Target
184206 -------------------------
185207
186 During this release ...
208 * The calling convention for the ``f80`` data type on MinGW targets has been
209 fixed. Normally, the calling convention for this type is handled within clang,
210 but if an intrinsic is used, which LLVM expands into a libcall, the
211 proper calling convention needs to be supported in LLVM as well. (Note,
212 on Windows, this data type is only used for long doubles in MinGW
213 environments - in MSVC environments, long doubles are the same size as
214 normal doubles.)
187215
188216 Changes to the AMDGPU Target
189217 -----------------------------