Update to latest libprimesieve

lib/primesieve/include/primesieve/PreSieve_Tables.hpp

@@ -1,5 +1,5 @@
 ///
-/// @file   PreSieve_Table.hpp
+/// @file   PreSieve_Tables.hpp
 /// @brief  Pre-sieve multiples of small primes < 100 to speed up the
 ///         sieve of Eratosthenes. The idea is to allocate several
 ///         arrays (buffers_) and remove the multiples of small primes
@@ -104,7 +104,7 @@ int main()
       if (j + 1 < buffers[i].size())
         std::cout << (((j+1) % 20 == 0) ? "," : ", ");
     }
-    std::cout << " }," << std::endl;
+    std::cout << (i + 1 < buffers.size() ? " }," : " }") << std::endl;
   }
 
   std::cout << "}};" << std::endl;
@@ -10860,7 +10860,7 @@ const primesieve::Array<std::initializer_list<uint8_t>, 8> buffers =
     0xff, 0xff, 0xff, 0xfd, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xef, 0xff, 0xff, 0xff, 0xfb, 0xff, 0xf7, 0xff, 0xff, 0xbf,
     0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xfb, 0xff, 0x7f, 0xff, 0xff, 0xff, 0xfd, 0xff, 0xf7, 0xff, 0xff, 0xff,
     0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xfe, 0xef, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
-    0xff, 0xff, 0xff, 0x7f, 0xfe, 0xff, 0xff },
+    0xff, 0xff, 0xff, 0x7f, 0xfe, 0xff, 0xff }
 }};
 
 } // namespace

lib/primesieve/src/README.md

@@ -52,12 +52,14 @@
   buckets and only when there are no more buckets in the stock the
   MemoryPool will allocate new buckets.
 
-* **PreSieve** is used to pre-sieve multiples of small primes < 100
-  to speed up the sieve of Eratosthenes. Upon creation the
-  multiples of small primes are removed from multiple buffers. Then
-  whilst sieving, we perform a bitwise AND on the buffer arrays
-  and store the result in the sieve array. Pre-sieving provides a
-  speedup of up to 30% when sieving the primes < 10^10.
+* **PreSieve** is used to pre-sieve multiples of small primes < 100 to speed
+  up the sieve of Eratosthenes. The ```PreSieve_Tables.hpp``` header contains
+  8 static lookup tables which have been sieved using the primes < 100. Each
+  of these lookup tables has a size of about 32 kilobytes and the total size
+  of all pre-sieve lookup tables is about 200 kilobytes. Whilst sieving, we
+  perform a bitwise AND of the pre-sieve lookup tables and store the result in
+  the sieve array. Pre-sieving provides a speedup of up to 30% when sieving
+  the primes < 10^10.
 
 * **SievingPrimes** is used to generate the sieving primes ≤ sqrt(stop).
   SievingPrimes is used by the CountPrintPrimes and PrimeGenerator classes.