Pseudo-intrinsics for ARM: Guard vroti_epi32/64

MacOS native (clang) gcc wouldn't allow rotations larger than width
so we simply mask the argument correctly.

Closes #4585

doc/INSTALL

@@ -153,16 +153,21 @@ that probably work fine but these instructions are for "Homebrew":
 
   1. Install Homebrew:
 	http://mxcl.github.io/homebrew/
-  2. Install Homebrew's gcc and openssl:
+  2. Install Homebrew's gcc and OpenSSL:
 	brew install gcc openssl
-  3. Make sure /usr/local/bin precedes /usr/bin in your $PATH
-  4. Configure, possibly adding a CC option for pointing to a specific gcc:
-        ./configure CC="gcc-6"
+  3. Consider adding homebrew directories such as /usr/local/bin and/or
+     /opt/homebrew/bin to your $PATH.
+  4. Configure, possibly adding a CC option for pointing to a specific gcc and
+     using whatever LDFLAGS and CPPFLAGS was recommended when you installed
+     Homebrew's OpenSSL:
+        ./configure CC="/opt/homebrew/bin/gcc-10" \
+          LDFLAGS="-L/opt/homebrew/opt/openssl@1.1/lib" \
+          CPPFLAGS="-I/opt/homebrew/opt/openssl@1.1/include"
   5. Clean old files and make:
         make -s clean && make -sj4
 
-After the above, you should be able to get an optimal build with AVX and/or
-whatever extra features your CPU has got.
+After the above, you should be able to get an optimal build with AVX/ASIMD
+or whatever extra features your CPU has got.
 
 If you get weird problems including things like "error: unknown type name
 'dispatch_block_t'" on 10.10 Yosemite, you might need to apply a patch for
@@ -171,7 +176,7 @@ From John's "src" directory:
 
   sudo patch -b -N -p0 < unused/Yosemite.patch
 
-The patch is not needed for 10.11 "El Capitan".
+The patch is not needed for 10.11 "El Capitan" or later!
 
 NOTE: The above command will create backup files. If you ever want to restore
 everything as it were originally:

doc/NEWS

@@ -163,11 +163,13 @@ Major changes from 1.9.0-jumbo-1 (May 2019) in this bleeding-edge version:
 - Final fixes for zip2john selection of 1-byte or 2-byte early-reject checks,
   as well as exactly what source to use for that check, after seeing several
   issues and digging into the specs.  Note that a re-run of zip2john is needed
-  for correct cracking. [magnum; 2021]
+  for correct cracking.  [magnum; 2021]
 
 - Allow any node range divisible by fork (or MPI) count.  [Solar, magnum; 2021]
 
-- Fix autoconf selection of arch.h for Apple M1.  [magnum; 2021]
+- Fix autoconf selection of arch.h for Apple M1.  Improve ARM pseudo-intrinsics
+  portability (would fail on clang).  Allow __ARM_FEATURE_UNALIGNED to set our
+  ARCH_ALLOWS_UNALIGNED macro.  [magnum; 2021]
 
 
 Major changes from 1.8.0-jumbo-1 (December 2014) to 1.9.0-jumbo-1 (May 2019):

src/pseudo_intrinsics.h

@@ -66,10 +66,10 @@ typedef union {
 #define VLOADU_EMULATED         1
 #define vor(x, y)               (vtype)vorrq_u32((x).v32, (y).v32)
 #define vorn(x, y)              (vtype)vornq_u32((x).v32, (y).v32)
-#define vroti_epi32(x, i)       (i > 0 ? (vtype)vsliq_n_u32(vshrq_n_u32((x).v32, 32 - (i)), (x).v32, i) : \
-                                         (vtype)vsriq_n_u32(vshlq_n_u32((x).v32, 32 + (i)), (x).v32, -(i)))
-#define vroti_epi64(x, i)       (i > 0 ? (vtype)vsliq_n_u64(vshrq_n_u64((x).v64, 64 - (i)), (x).v64, i) : \
-                                         (vtype)vsriq_n_u64(vshlq_n_u64((x).v64, 64 + (i)), (x).v64, -(i)))
+#define vroti_epi32(x, i)       (i > 0 ? (vtype)vsliq_n_u32(vshrq_n_u32((x).v32, 32 - ((i) & 31)), (x).v32, (i) & 31) : \
+                                 (vtype)vsriq_n_u32(vshlq_n_u32((x).v32, (32 + (i)) & 31), (x).v32, (-(i)) & 31))
+#define vroti_epi64(x, i)       (i > 0 ? (vtype)vsliq_n_u64(vshrq_n_u64((x).v64, 64 - ((i) & 63)), (x).v64, (i) & 63) : \
+                                 (vtype)vsriq_n_u64(vshlq_n_u64((x).v64, (64 + (i)) & 63), (x).v64, (-(i)) & 63))
 #define vroti16_epi32           vroti_epi32
 #define vset1_epi32(i)          (vtype)vdupq_n_u32(i)
 #define vset1_epi64(i)          (vtype)vdupq_n_u64(i)