Make numerous improvements

- Python static hello world now 1.8mb
- Python static fully loaded now 10mb
- Python HTTPS client now uses MbedTLS
- Python REPL now completes import stmts
- Increase stack size for Python for now
- Begin synthesizing posixpath and ntpath
- Restore Python \N{UNICODE NAME} support
- Restore Python NFKD symbol normalization
- Add optimized code path for Intel SHA-NI
- Get more Python unit tests passing faster
- Get Python help() pagination working on NT
- Python hashlib now supports MbedTLS PBKDF2
- Make memcpy/memmove/memcmp/bcmp/etc. faster
- Add Mersenne Twister and Vigna to LIBC_RAND
- Provide privileged __printf() for error code
- Fix zipos opendir() so that it reports ENOTDIR
- Add basic chmod() implementation for Windows NT
- Add Cosmo's best functions to Python cosmo module
- Pin function trace indent depth to that of caller
- Show memory diagram on invalid access in MODE=dbg
- Differentiate stack overflow on crash in MODE=dbg
- Add stb_truetype and tools for analyzing font files
- Upgrade to UNICODE 13 and reduce its binary footprint
- COMPILE.COM now logs resource usage of build commands
- Start implementing basic poll() support on bare metal
- Set getauxval(AT_EXECFN) to GetModuleFileName() on NT
- Add descriptions to strerror() in non-TINY build modes
- Add COUNTBRANCH() macro to help with micro-optimizations
- Make error / backtrace / asan / memory code more unbreakable
- Add fast perfect C implementation of μ-Law and a-Law audio codecs
- Make strtol() functions consistent with other libc implementations
- Improve Linenoise implementation (see also github.com/jart/bestline)
- COMPILE.COM now suppresses stdout/stderr of successful build commands

libc/rand/rand64.c

@@ -17,8 +17,10 @@
 │ PERFORMANCE OF THIS SOFTWARE.                                                │
 ╚─────────────────────────────────────────────────────────────────────────────*/
 #include "libc/bits/bits.h"
+#include "libc/bits/xadd.h"
 #include "libc/calls/calls.h"
 #include "libc/nexgen32e/rdtsc.h"
+#include "libc/nexgen32e/x86feature.h"
 #include "libc/rand/rand.h"
 #include "libc/runtime/runtime.h"
 #include "libc/sysv/consts/auxv.h"
@@ -28,46 +30,45 @@ static uint64_t thepool;
 /**
  * Returns nondeterministic random data.
  *
- * This random number seed generator blends information from:
+ * This function automatically seeds itself on startup and reseeds
+ * itself after fork() and vfork(). It takes about nanosecond to run.
+ * That makes it much slower than vigna() and rand() but much faster
+ * than rdrand() and rdseed().
  *
- * - rdtsc() hardware clock
- * - getpid() process identifier
- * - getauxval(AT_RANDOM) on Linux
- *
- * It's 100% guaranteed to not hard block the system.
- *
- * @see rngset(), getrandom()
+ * @see rdseed(), rdrand(), rand(), random(), rngset()
+ * @note based on vigna's algorithm
  * @asyncsignalsafe
  * @vforksafe
  */
 uint64_t rand64(void) {
+  bool cf;
   register uint64_t t;
-  t = thepool;
-  t ^= getpid() * 11400714819643198487ull + 123456789123456789;
-  t ^= t << 13;
-  t ^= t >> 7;
-  t ^= t << 17;
-  t ^= rdtsc() * 11400714819643198487ull + 123456789123456789;
-  t ^= t << 13;
-  t ^= t >> 7;
-  t ^= t << 17;
-  thepool ^= t;
-  return t;
+  if (X86_HAVE(RDSEED)) {
+    asm volatile(CFLAG_ASM("rdseed\t%1")
+                 : CFLAG_CONSTRAINT(cf), "=r"(t)
+                 : /* no inputs */
+                 : "cc");
+    if (cf) {
+      thepool ^= t;
+      return t;
+    }
+  }
+  t = _xadd(&thepool, 0x9e3779b97f4a7c15);
+  t ^= (getpid() * 0x1001111111110001ull + 0xdeaadead) >> 31;
+  t = (t ^ (t >> 30)) * 0xbf58476d1ce4e5b9;
+  t = (t ^ (t >> 27)) * 0x94d049bb133111eb;
+  return t ^ (t >> 31);
 }
 
 static textstartup void rand64_init(int argc, char **argv, char **envp,
                                     intptr_t *auxv) {
-  uint64_t t;
-  t = kStartTsc * 88172645463325252 + 123456789123456789;
-  if (AT_RANDOM) {
-    for (; auxv[0]; auxv += 2) {
-      if (auxv[0] == AT_RANDOM) {
-        t ^= READ64LE((const char *)auxv[1]);
-        break;
-      }
+  for (; auxv[0]; auxv += 2) {
+    if (auxv[0] == AT_RANDOM) {
+      thepool = READ64LE((const char *)auxv[1]);
+      return;
     }
   }
-  thepool = t;
+  thepool = kStartTsc;
 }
 
 const void *const g_rand64_init[] initarray = {rand64_init};