VIXL Release 1.2

Refer to the README.md and LICENCE files for details.

.gitignore

@@ -1,11 +1,9 @@
 # ignore python compiled object
 *.pyc
-# ignore vi temporary files
-*.swo
-*.swp
 .sconsign.dblite
 obj/
 cctest*
 bench_*
 libvixl*
 example-*
+vixl_stats.csv

README.md

@@ -1,4 +1,4 @@
-VIXL: AArch64 Runtime Code Generation Library Version 1.1
+VIXL: AArch64 Runtime Code Generation Library Version 1.2
 =========================================================
 
 Contents:
@@ -54,7 +54,7 @@ were deemed unnecessary:
 
  * No Advanced SIMD support.
  * Limited rounding mode support for floating point.
- * No support for synchronisation instructions.
+ * Limited support for synchronisation instructions.
  * Limited support for system instructions.
  * A few miscellaneous integer and floating point instructions are missing.
 

doc/changelog.md

@@ -1,6 +1,25 @@
 VIXL Change Log
 ===============
 
+* 1.2
+    + Added support for `fmadd`, `fnmadd`, `fnmsub`, `fminnm`, `fmaxnm`,
+      `frinta`, `fcvtau` and `fcvtas`.
+    + Added support for assembling and disassembling `isb`, `dsb` and `dmb`.
+    + Added support for automatic inversion of compare instructions when using
+      negative immediates.
+    + Added support for using `movn` when generating immediates.
+    + Added explicit flag-setting 'S' instructions, and removed
+      `SetFlags` and `LeaveFlags` arguments.
+    + Added support for `Movk` in macro assembler.
+    + Added support for W register parameters to `Tbz` and `Tbnz`.
+    + Added support for using immediate operands with `Csel`.
+    + Added new debugger syntax for memory inspection.
+    + Fixed `smull`, `fmsub` and `sdiv` simulation.
+    + Fixed sign extension for W->X conversions using `sxtb`, `sxth` and `sxtw`.
+    + Prevented code generation for certain side-effect free operations,
+      such as `add r, r, #0`, in the macro assembler.
+    + Other small bug fixes.
+
 * 1.1
     + Improved robustness of instruction decoder and disassembler.
     + Added support for double-to-float conversions using `fcvt`.