[builtins] Optimize udivmodti4 for many platforms.

[danlark]

Summary

While benchmarking uint128 division we found out that it has huge latency for small divisors

https://reviews.llvm.org/D83027

Benchmark                                                   Time(ns)        CPU(ns)     Iterations
--------------------------------------------------------------------------------------------------
BM_DivideIntrinsic128UniformDivisor<unsigned __int128>            13.0           13.0     55000000  
BM_DivideIntrinsic128UniformDivisor<__int128>                     14.3           14.3     50000000  
BM_RemainderIntrinsic128UniformDivisor<unsigned __int128>         13.5           13.5     52000000  
BM_RemainderIntrinsic128UniformDivisor<__int128>                  14.1           14.1     50000000  
BM_DivideIntrinsic128SmallDivisor<unsigned __int128>             153            153        5000000  
BM_DivideIntrinsic128SmallDivisor<__int128>                      170            170        3000000  
BM_RemainderIntrinsic128SmallDivisor<unsigned __int128>          153            153        5000000  
BM_RemainderIntrinsic128SmallDivisor<__int128>                   155            155        5000000

This patch suggests a more optimized version of the division:

If the divisor is 64 bit, we can proceed with the divq instruction on x86 or constant multiplication mechanisms for other platforms. Once both divisor and dividend are not less than 2**64, we use branch free subtract algorithm, it has at most 64 cycles. After that our benchmarks improved significantly

Benchmark                                                   Time(ns)        CPU(ns)     Iterations
--------------------------------------------------------------------------------------------------
BM_DivideIntrinsic128UniformDivisor<unsigned __int128>            11.0           11.0     64000000  
BM_DivideIntrinsic128UniformDivisor<__int128>                     13.8           13.8     51000000  
BM_RemainderIntrinsic128UniformDivisor<unsigned __int128>         11.6           11.6     61000000  
BM_RemainderIntrinsic128UniformDivisor<__int128>                  13.7           13.7     52000000  
BM_DivideIntrinsic128SmallDivisor<unsigned __int128>              27.1           27.1     26000000  
BM_DivideIntrinsic128SmallDivisor<__int128>                       29.4           29.4     24000000  
BM_RemainderIntrinsic128SmallDivisor<unsigned __int128>           27.9           27.8     26000000  
BM_RemainderIntrinsic128SmallDivisor<__int128>                    29.1           29.1     25000000

If not using divq instrinsics, it is still much better

Benchmark                                                   Time(ns)        CPU(ns)     Iterations
--------------------------------------------------------------------------------------------------
BM_DivideIntrinsic128UniformDivisor<unsigned __int128>            12.2           12.2     58000000  
BM_DivideIntrinsic128UniformDivisor<__int128>                     13.5           13.5     52000000  
BM_RemainderIntrinsic128UniformDivisor<unsigned __int128>         12.7           12.7     56000000  
BM_RemainderIntrinsic128UniformDivisor<__int128>                  13.7           13.7     51000000  
BM_DivideIntrinsic128SmallDivisor<unsigned __int128>              30.2           30.2     24000000  
BM_DivideIntrinsic128SmallDivisor<__int128>                       33.2           33.2     22000000  
BM_RemainderIntrinsic128SmallDivisor<unsigned __int128>           31.4           31.4     23000000  
BM_RemainderIntrinsic128SmallDivisor<__int128>                    33.8           33.8     21000000

PowerPC benchmarks:

Was

BM_DivideIntrinsic128UniformDivisor<unsigned __int128>            22.3           22.3     32000000  
BM_DivideIntrinsic128UniformDivisor<__int128>                     23.8           23.8     30000000  
BM_RemainderIntrinsic128UniformDivisor<unsigned __int128>         22.5           22.5     32000000  
BM_RemainderIntrinsic128UniformDivisor<__int128>                  24.9           24.9     29000000  
BM_DivideIntrinsic128SmallDivisor<unsigned __int128>             394            394        2000000  
BM_DivideIntrinsic128SmallDivisor<__int128>                      397            397        2000000  
BM_RemainderIntrinsic128SmallDivisor<unsigned __int128>          399            399        2000000  
BM_RemainderIntrinsic128SmallDivisor<__int128>                   397            397        2000000

With this patch

BM_DivideIntrinsic128UniformDivisor<unsigned __int128>            21.7           21.7     33000000  
BM_DivideIntrinsic128UniformDivisor<__int128>                     23.0           23.0     31000000  
BM_RemainderIntrinsic128UniformDivisor<unsigned __int128>         21.9           21.9     33000000  
BM_RemainderIntrinsic128UniformDivisor<__int128>                  23.9           23.9     30000000  
BM_DivideIntrinsic128SmallDivisor<unsigned __int128>              32.7           32.6     23000000  
BM_DivideIntrinsic128SmallDivisor<__int128>                       33.4           33.4     21000000  
BM_RemainderIntrinsic128SmallDivisor<unsigned __int128>           31.1           31.1     22000000  
BM_RemainderIntrinsic128SmallDivisor<__int128>                    33.2           33.2     22000000

My email: [e-mail], I don't have commit rights

[danlark]

Use native division when high is zero.

[courbet]

Nice. I have not looked at the code yet, but I think it would make sense to add your benchmark to the compiler-rt unit tests.

[danlark]

Clang format

[danlark]

Style

[danlark]

Ping and maybe I need some help in introducing benchmarks (if we really need to) for this because I don't quite understand how and where it should be best located.

Also, there are some in test/builtins/timing but they seem to work only for MacOS which is pretty non-obvious why :\

[danlark]

Style

[danlark]

CHAR_BIT instead of 8

[danlark]

In D81809#2092173, @courbet wrote:

Nice. I have not looked at the code yet, but I think it would make sense to add your benchmark to the compiler-rt unit tests.

Added https://reviews.llvm.org/D83027

[danlark]

Ping

[danlark]

Fix clang-tidy warnings

[danlark]

Fix non x86-mode

[danlark]

clang-format

[danlark]

Fix clang-tidy readability warnings more

[MaskRay]

compiler-rt/lib/builtins/udivmodti4.c

22

Please use lower case, as the file originally uses. A large portion of LLVM does use uppercase and people find that annoying, but they don't want to fix which will cause lots of churns. However, for code which correctly uses lower case, there is no point converting to uppercase.

33

No need to comment __builtin_clzll

36

For comments which are one their own lines, the LLVM style usually appends a period.

42

The first sentence is sufficient. No need to repeat: "The behavior is undefined if the right operand is negative, or greater than or equal to the length in bits of the promoted left operand."

[MaskRay]

compiler-rt/lib/builtins/udivmodti4.c

151

Please keep lower-case and don't do unneeded variable renaming.

[danlark]

Fix comments

[danlark]

Update the real reference

[danlark]

Please, take a look

[courbet]

Sorry for taking that long, this is a dense review. Overall this looks good, and the renaming and splitting actually also improve the readability of the non-x86 part.
I only have style comments.

compiler-rt/lib/builtins/udivmodti4.c

23

How is this simple ? :) maybe default ?

32

what's norm ? Normalization ? In general, please avoid the abbreviations here, this is already quite dense.

115–122

// When the divisor fits in 64 bits, we can use an optimized path.

124–129

typo: remainder

[danlark]

Fix comments

[danlark]

Fix clang format

compiler-rt/lib/builtins/udivmodti4.c

 //===-- udivmodti4.c - Implement __udivmodti4 -----------------------------===//
 //
 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
 // See https://llvm.org/LICENSE.txt for license information.
 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
 //
 //===----------------------------------------------------------------------===//
 //
 // This file implements __udivmodti4 for the compiler_rt library.
 //
 //===----------------------------------------------------------------------===//
 
 #include "int_lib.h"
 
 #ifdef CRT_HAS_128BIT
 
+// Returns the 128 bit division result by 64 bit. Result must fit in 64 bits.
+// Remainder stored in r.
+// Taken and adjusted from libdivide libdivide_128_div_64_to_64 division
+// fallback.
+UNUSED
+static inline du_int udiv128by64to64simple(du_int u1, du_int u0, du_int v,
+                                           du_int *r) {
+  const unsigned n_udword_bits = sizeof(du_int) * CHAR_BIT;
+  const du_int b = (1ULL << (n_udword_bits / 2)); // Number base (32 bits)
+  du_int un1, un0;                                // Norm. dividend LSD's
+  du_int vn1, vn0;                                // Norm. divisor digits
+  du_int q1, q0;                                  // Quotient digits
+  du_int un64, un21, un10;                        // Dividend digit pairs
+  du_int rhat;                                    // A remainder
+  si_int s;                                       // Shift amount for norm
+
+  // count leading zeros
+  s = __builtin_clzll(v);
+  if (s > 0) {
+    // Normalize divisor
+    v = v << s;
+    un64 = (u1 << s) | (u0 >> (n_udword_bits - s));
+    un10 = u0 << s; // Shift dividend left
+  } else {
+    // Avoid undefined behavior of (u0 >> 64).
+    // The behavior is undefined if the right operand is
+    // negative, or greater than or equal to the length
+    // in bits of the promoted left operand.
+    un64 = u1;
+    un10 = u0;
+  }
+
+  // Break divisor up into two 32-bit digits
+  vn1 = v >> (n_udword_bits / 2);
+  vn0 = v & 0xFFFFFFFF;
+
+  // Break right half of dividend into two digits
+  un1 = un10 >> (n_udword_bits / 2);
+  un0 = un10 & 0xFFFFFFFF;
+
+  // Compute the first quotient digit, q1
+  q1 = un64 / vn1;
+  rhat = un64 - q1 * vn1;
+
+  while (q1 >= b || q1 * vn0 > b * rhat + un1) {
+    q1 = q1 - 1;
+    rhat = rhat + vn1;
+    if (rhat >= b)
+      break;
+  }
+
+  // Multiply and subtract
+  un21 = un64 * b + un1 - q1 * v;
+
+  // Compute the second quotient digit
+  q0 = un21 / vn1;
+  rhat = un21 - q0 * vn1;
+
+  while (q0 >= b || q0 * vn0 > b * rhat + un0) {
+    q0 = q0 - 1;
+    rhat = rhat + vn1;
+    if (rhat >= b)
+      break;
+  }
+
+  *r = (un21 * b + un0 - q0 * v) >> s;
+  return q1 * b + q0;
+}
+
+static inline du_int udiv128by64to64(du_int u1, du_int u0, du_int v,
+                                     du_int *r) {
+#if defined(__x86_64__)
+  du_int result;
+  __asm__("divq %[v]"
+          : "=a"(result), "=d"(*r)
+          : [ v ] "r"(v), "a"(u0), "d"(u1));
+  return result;
+#else
+  return udiv128by64to64simple(u1, u0, v, r);
+#endif
+}
+
 // Effects: if rem != 0, *rem = a % b
 // Returns: a / b
 
-// Translated from Figure 3-40 of The PowerPC Compiler Writer's Guide
-
 COMPILER_RT_ABI tu_int __udivmodti4(tu_int a, tu_int b, tu_int *rem) {
-  const unsigned n_udword_bits = sizeof(du_int) * CHAR_BIT;
   const unsigned n_utword_bits = sizeof(tu_int) * CHAR_BIT;
-  utwords n;
-  n.all = a;
-  utwords d;
-  d.all = b;
-  utwords q;
-  utwords r;
-  unsigned sr;
-  // special cases, X is unknown, K != 0
-  if (n.s.high == 0) {
-    if (d.s.high == 0) {
-      // 0 X
-      // ---
-      // 0 X
-      if (rem)
-        *rem = n.s.low % d.s.low;
-      return n.s.low / d.s.low;
-    }
-    // 0 X
-    // ---
-    // K X
-    if (rem)
-      *rem = n.s.low;
-    return 0;
-  }
-  // n.s.high != 0
-  if (d.s.low == 0) {
-    if (d.s.high == 0) {
-      // K X
-      // ---
-      // 0 0
-      if (rem)
-        *rem = n.s.high % d.s.low;
-      return n.s.high / d.s.low;
-    }
-    // d.s.high != 0
-    if (n.s.low == 0) {
-      // K 0
-      // ---
-      // K 0
-      if (rem) {
-        r.s.high = n.s.high % d.s.high;
-        r.s.low = 0;
-        *rem = r.all;
-      }
-      return n.s.high / d.s.high;
-    }
-    // K K
-    // ---
-    // K 0
-    if ((d.s.high & (d.s.high - 1)) == 0) /* if d is a power of 2 */ {
-      if (rem) {
-        r.s.low = n.s.low;
-        r.s.high = n.s.high & (d.s.high - 1);
-        *rem = r.all;
-      }
-      return n.s.high >> __builtin_ctzll(d.s.high);
-    }
-    // K K
-    // ---
-    // K 0
-    sr = __builtin_clzll(d.s.high) - __builtin_clzll(n.s.high);
-    // 0 <= sr <= n_udword_bits - 2 or sr large
-    if (sr > n_udword_bits - 2) {
-      if (rem)
-        *rem = n.all;
-      return 0;
-    }
-    ++sr;
-    // 1 <= sr <= n_udword_bits - 1
-    // q.all = n.all << (n_utword_bits - sr);
-    q.s.low = 0;
-    q.s.high = n.s.low << (n_udword_bits - sr);
-    // r.all = n.all >> sr;
-    r.s.high = n.s.high >> sr;
-    r.s.low = (n.s.high << (n_udword_bits - sr)) | (n.s.low >> sr);
-  } else /* d.s.low != 0 */ {
-    if (d.s.high == 0) {
-      // K X
-      // ---
-      // 0 K
-      if ((d.s.low & (d.s.low - 1)) == 0) /* if d is a power of 2 */ {
-        if (rem)
-          *rem = n.s.low & (d.s.low - 1);
-        if (d.s.low == 1)
-          return n.all;
-        sr = __builtin_ctzll(d.s.low);
-        q.s.high = n.s.high >> sr;
-        q.s.low = (n.s.high << (n_udword_bits - sr)) | (n.s.low >> sr);
-        return q.all;
-      }
-      // K X
-      // ---
-      // 0 K
-      sr = 1 + n_udword_bits + __builtin_clzll(d.s.low) -
-           __builtin_clzll(n.s.high);
-      // 2 <= sr <= n_utword_bits - 1
-      // q.all = n.all << (n_utword_bits - sr);
-      // r.all = n.all >> sr;
-      if (sr == n_udword_bits) {
-        q.s.low = 0;
-        q.s.high = n.s.low;
-        r.s.high = 0;
-        r.s.low = n.s.high;
-      } else if (sr < n_udword_bits) /* 2 <= sr <= n_udword_bits - 1 */ {
-        q.s.low = 0;
-        q.s.high = n.s.low << (n_udword_bits - sr);
-        r.s.high = n.s.high >> sr;
-        r.s.low = (n.s.high << (n_udword_bits - sr)) | (n.s.low >> sr);
-      } else /* n_udword_bits + 1 <= sr <= n_utword_bits - 1 */ {
-        q.s.low = n.s.low << (n_utword_bits - sr);
-        q.s.high = (n.s.high << (n_utword_bits - sr)) |
-                   (n.s.low >> (sr - n_udword_bits));
-        r.s.high = 0;
-        r.s.low = n.s.high >> (sr - n_udword_bits);
-      }
+  utwords dividend;
+  dividend.all = a;
+  utwords divisor;
+  divisor.all = b;
+  utwords quotient;
+  utwords remainder;
+  if (divisor.all > dividend.all) {
+      *rem = dividend.all;
+  // Divide 128 bit by 64 bit.
+  if (divisor.s.high == 0) {
+    remainder.s.high = 0;
+    if (dividend.s.high < divisor.s.low) {
+      // The result fits in 64 bits.
+      quotient.s.low = udiv128by64to64(dividend.s.high, dividend.s.low,
+                                       divisor.s.low, &remainder.s.low);
+      quotient.s.high = 0;
     } else {
-      // K X
-      // ---
-      // K K
-      sr = __builtin_clzll(d.s.high) - __builtin_clzll(n.s.high);
-      // 0 <= sr <= n_udword_bits - 1 or sr large
-      if (sr > n_udword_bits - 1) {
-        if (rem)
-          *rem = n.all;
-        return 0;
-      }
-      ++sr;
-      // 1 <= sr <= n_udword_bits
-      // q.all = n.all << (n_utword_bits - sr);
-      // r.all = n.all >> sr;
-      q.s.low = 0;
-      if (sr == n_udword_bits) {
-        q.s.high = n.s.low;
-        r.s.high = 0;
-        r.s.low = n.s.high;
-      } else {
-        r.s.high = n.s.high >> sr;
-        r.s.low = (n.s.high << (n_udword_bits - sr)) | (n.s.low >> sr);
-        q.s.high = n.s.low << (n_udword_bits - sr);
-      }
-    }
-  }
-  // Not a special case
-  // q and r are initialized with:
-  // q.all = n.all << (n_utword_bits - sr);
-  // r.all = n.all >> sr;
-  // 1 <= sr <= n_utword_bits - 1
-  su_int carry = 0;
-  for (; sr > 0; --sr) {
-    // r:q = ((r:q)  << 1) | carry
-    r.s.high = (r.s.high << 1) | (r.s.low >> (n_udword_bits - 1));
-    r.s.low = (r.s.low << 1) | (q.s.high >> (n_udword_bits - 1));
-    q.s.high = (q.s.high << 1) | (q.s.low >> (n_udword_bits - 1));
-    q.s.low = (q.s.low << 1) | carry;
-    // carry = 0;
-    // if (r.all >= d.all)
+      // First, divide with the high part to get the remaider in dividend.s.high
+      // After that dividend.s.high < divisor.s.low.
+      quotient.s.high = dividend.s.high / divisor.s.low;
+      dividend.s.high = dividend.s.high % divisor.s.low;
+      quotient.s.low = udiv128by64to64(dividend.s.high, dividend.s.low,
+                                       divisor.s.low, &remainder.s.low);
+    }
+    if (rem)
+      *rem = remainder.all;
+    return quotient.all;
+  }
+  // 0 <= shift <= 63
+  si_int shift =
+      __builtin_clzll(divisor.s.high) - __builtin_clzll(dividend.s.high);
+  divisor.all <<= shift;
+  quotient.s.high = 0;
+  quotient.s.low = 0;
+  for (; shift >= 0; --shift) {
+    quotient.s.low <<= 1;
+    // Branch free version of.
+    // if (dividend.all >= divisor.all)
     // {
-    //     r.all -= d.all;
-    //      carry = 1;
+    //    dividend.all -= divisor.all;
+    //    carry = 1;
     // }
-    const ti_int s = (ti_int)(d.all - r.all - 1) >> (n_utword_bits - 1);
-    carry = s & 1;
-    r.all -= d.all & s;
+    const ti_int s =
+        (ti_int)(divisor.all - dividend.all - 1) >> (n_utword_bits - 1);
+    quotient.s.low |= s & 1;
+    dividend.all -= divisor.all & s;
+    divisor.all >>= 1;
   }
-  q.all = (q.all << 1) | carry;
   if (rem)
-    *rem = r.all;
-  return q.all;
+    *rem = dividend.all;
+  return quotient.all;
 }
 
 #endif // CRT_HAS_128BIT