[PATCH v7 03/12] RISC-V: Add CRC expander to generate faster CRC.

If the target is ZBC or ZBKC, it uses clmul instruction for the CRC
calculation.  Otherwise, if the target is ZBKB, generates table-based
CRC, but for reversing inputs and the output uses bswap and brev8
instructions.  Add new tests to check CRC generation for ZBC, ZBKC and
ZBKB targets.

gcc/

	* expr.cc (gf2n_poly_long_div_quotient): New function.
	* expr.h (gf2n_poly_long_div_quotient):  New function declaration.
	* hwint.cc (reflect_hwi): New function.
	* hwint.h (reflect_hwi): New function declaration.
	* config/riscv/bitmanip.md (crc_rev<ANYI1:mode><ANYI:mode>4): New
	expander for reversed CRC.
	(crc<SUBX1:mode><SUBX:mode>4): New expander for bit-forward CRC.
	* config/riscv/iterators.md (SUBX1, ANYI1): New iterators.
	* config/riscv/riscv-protos.h (generate_reflecting_code_using_brev):
	New function declaration.
	(expand_crc_using_clmul): Likewise.
	(expand_reversed_crc_using_clmul): Likewise.
	* config/riscv/riscv.cc (generate_reflecting_code_using_brev): New
	function.
	(expand_crc_using_clmul): Likewise.
	(expand_reversed_crc_using_clmul): Likewise.
	* config/riscv/riscv.md (UNSPEC_CRC, UNSPEC_CRC_REV):  New unspecs.

	* doc/sourcebuild.texi: Document new target selectors.

gcc/testsuite
	* lib/target-supports.exp (check_effective_target_riscv_zbc): New
	target supports predicate.
	(check_effective_target_riscv_zbkb): Likewise.
	(check_effective_target_riscv_zbkc): Likewise.
	(check_effective_target_zbc_ok): Likewise.
	(check_effective_target_zbkb_ok): Likewise.
	(check_effective_target_zbkc_ok): Likewise.
	(riscv_get_arch): Add zbkb and zbkc support.

	* gcc.target/riscv/crc-builtin-zbc32.c: New file.
	* gcc.target/riscv/crc-builtin-zbc64.c: Likewise.

	    Co-author: Jeff Law  <jlaw@ventanamicro.com>

13 files changed, 494 insertions, 1 deletions

diff --git a/gcc/config/riscv/bitmanip.md b/gcc/config/riscv/bitmanip.md
index 06ff698..23dc47e 100644
--- a/gcc/config/riscv/bitmanip.md
+++ b/gcc/config/riscv/bitmanip.md
@@ -1192,3 +1192,66 @@
   "TARGET_ZBC"
   "clmulr\t%0,%1,%2"
   [(set_attr "type" "clmul")])
+
+;; Reversed CRC 8, 16, 32 for TARGET_64
+(define_expand "crc_rev<ANYI1:mode><ANYI:mode>4"
+	;; return value (calculated CRC)
+  [(set (match_operand:ANYI 0 "register_operand" "=r")
+		      ;; initial CRC
+	(unspec:ANYI [(match_operand:ANYI 1 "register_operand" "r")
+		      ;; data
+		      (match_operand:ANYI1 2 "register_operand" "r")
+		      ;; polynomial without leading 1
+		      (match_operand:ANYI 3)]
+		      UNSPEC_CRC_REV))]
+  /* We don't support the case when data's size is bigger than CRC's size.  */
+  "<ANYI:MODE>mode >= <ANYI1:MODE>mode"
+{
+  /* If we have the ZBC or ZBKC extension (ie, clmul) and
+     it is possible to store the quotient within a single variable
+     (E.g.  CRC64's quotient may need 65 bits,
+     we can't keep it in 64 bit variable.)
+     then use clmul instruction to implement the CRC,
+     otherwise (TARGET_ZBKB) generate table based using brev.  */
+  if ((TARGET_ZBKC || TARGET_ZBC) && <ANYI:MODE>mode < word_mode)
+    expand_reversed_crc_using_clmul (<ANYI:MODE>mode, <ANYI1:MODE>mode,
+				     operands);
+  else if (TARGET_ZBKB)
+    /* Generate table-based CRC.
+       To reflect values use brev and bswap instructions.  */
+    expand_reversed_crc_table_based (operands[0], operands[1],
+				     operands[2], operands[3],
+				     GET_MODE (operands[2]),
+				     generate_reflecting_code_using_brev);
+  else
+    /* Generate table-based CRC.
+       To reflect values use standard reflecting algorithm.  */
+    expand_reversed_crc_table_based (operands[0], operands[1],
+				     operands[2], operands[3],
+				     GET_MODE (operands[2]),
+				     generate_reflecting_code_standard);
+  DONE;
+})
+
+;; CRC 8, 16, (32 for TARGET_64)
+(define_expand "crc<SUBX1:mode><SUBX:mode>4"
+	;; return value (calculated CRC)
+  [(set (match_operand:SUBX 0 "register_operand" "=r")
+		      ;; initial CRC
+	(unspec:SUBX [(match_operand:SUBX 1 "register_operand" "r")
+		      ;; data
+		      (match_operand:SUBX1 2 "register_operand" "r")
+		      ;; polynomial without leading 1
+		      (match_operand:SUBX 3)]
+		      UNSPEC_CRC))]
+  /* We don't support the case when data's size is bigger than CRC's size.  */
+  "(TARGET_ZBKC || TARGET_ZBC) && <SUBX:MODE>mode >= <SUBX1:MODE>mode"
+{
+  /* If we have the ZBC or ZBKC extension (ie, clmul) and
+     it is possible to store the quotient within a single variable
+     (E.g.  CRC64's quotient may need 65 bits,
+     we can't keep it in 64 bit variable.)
+     then use clmul instruction to implement the CRC.  */
+  expand_crc_using_clmul (<SUBX:MODE>mode, <SUBX1:MODE>mode, operands);
+  DONE;
+})
diff --git a/gcc/config/riscv/iterators.md b/gcc/config/riscv/iterators.md
index 0816594..fae8989 100644
--- a/gcc/config/riscv/iterators.md
+++ b/gcc/config/riscv/iterators.md
@@ -62,9 +62,15 @@
 ;; Iterator for hardware integer modes narrower than XLEN.
 (define_mode_iterator SUBX [QI HI (SI "TARGET_64BIT")])
 
+;; Iterator for hardware integer modes narrower than XLEN, same as SUBX.
+(define_mode_iterator SUBX1 [QI HI (SI "TARGET_64BIT")])
+
 ;; Iterator for hardware-supported integer modes.
 (define_mode_iterator ANYI [QI HI SI (DI "TARGET_64BIT")])
 
+;; Iterator for hardware integer modes narrower than XLEN, same as ANYI.
+(define_mode_iterator ANYI1 [QI HI SI (DI "TARGET_64BIT")])
+
 (define_mode_iterator ANYI_DOUBLE_TRUNC [HI SI (DI "TARGET_64BIT")])
 
 (define_mode_iterator ANYI_QUAD_TRUNC [SI (DI "TARGET_64BIT")])
diff --git a/gcc/config/riscv/riscv-protos.h b/gcc/config/riscv/riscv-protos.h
index 500b357..99ee6ef 100644
--- a/gcc/config/riscv/riscv-protos.h
+++ b/gcc/config/riscv/riscv-protos.h
@@ -175,6 +175,9 @@ extern bool riscv_reg_frame_related (rtx);
 extern void riscv_split_sum_of_two_s12 (HOST_WIDE_INT, HOST_WIDE_INT *,
 					HOST_WIDE_INT *);
 extern bool riscv_vector_float_type_p (const_tree type);
+extern void generate_reflecting_code_using_brev (rtx *);
+extern void expand_crc_using_clmul (scalar_mode, scalar_mode, rtx *);
+extern void expand_reversed_crc_using_clmul (scalar_mode, scalar_mode, rtx *);
 
 /* Routines implemented in riscv-c.cc.  */
 void riscv_cpu_cpp_builtins (cpp_reader *);
diff --git a/gcc/config/riscv/riscv.cc b/gcc/config/riscv/riscv.cc
index efdb1d3..7a1724d 100644
--- a/gcc/config/riscv/riscv.cc
+++ b/gcc/config/riscv/riscv.cc
@@ -13502,6 +13502,163 @@ riscv_use_by_pieces_infrastructure_p (unsigned HOST_WIDE_INT size,
   return default_use_by_pieces_infrastructure_p (size, alignment, op, speed_p);
 }
 
+/* Generate instruction sequence
+   which reflects the value of the OP using bswap and brev8 instructions.
+   OP's mode may be less than word_mode, to get the correct number,
+   after reflecting we shift right the value by SHIFT_VAL.
+   E.g. we have 1111 0001, after reflection (target 32-bit) we will get
+   1000 1111 0000 0000, if we shift-out 16 bits,
+   we will get the desired one: 1000 1111.  */
+
+void
+generate_reflecting_code_using_brev (rtx *op)
+{
+  machine_mode op_mode = GET_MODE (*op);
+  HOST_WIDE_INT shift_val = (BITS_PER_WORD
+			     - GET_MODE_BITSIZE (op_mode).to_constant ());
+  riscv_expand_op (BSWAP, word_mode, *op, *op, *op);
+  riscv_expand_op (LSHIFTRT, word_mode, *op, *op,
+		   gen_int_mode (shift_val, word_mode));
+  if (TARGET_64BIT)
+    emit_insn (gen_riscv_brev8_di (*op, *op));
+  else
+    emit_insn (gen_riscv_brev8_si (*op, *op));
+}
+
+
+/* Generate assembly to calculate CRC using clmul instruction.
+   The following code will be generated when the CRC and data sizes are equal:
+     li      a4,quotient
+     li      a5,polynomial
+     xor     a0,a1,a0
+     clmul   a0,a0,a4
+     srli    a0,a0,crc_size
+     clmul   a0,a0,a5
+     slli    a0,a0,word_mode_size - crc_size
+     srli    a0,a0,word_mode_size - crc_size
+     ret
+   crc_size may be 8, 16, 32.
+   Some instructions will be added for the cases when CRC's size is larger than
+   data's size.
+   OPERANDS[1] is input CRC,
+   OPERANDS[2] is data (message),
+   OPERANDS[3] is the polynomial without the leading 1.  */
+
+void
+expand_crc_using_clmul (scalar_mode crc_mode, scalar_mode data_mode,
+			rtx *operands)
+{
+  /* Check and keep arguments.  */
+  gcc_assert (!CONST_INT_P (operands[0]));
+  gcc_assert (CONST_INT_P (operands[3]));
+  unsigned short crc_size = GET_MODE_BITSIZE (crc_mode);
+  gcc_assert (crc_size <= 32);
+  unsigned short data_size = GET_MODE_BITSIZE (data_mode);
+
+  /* Calculate the quotient.  */
+  unsigned HOST_WIDE_INT
+      q = gf2n_poly_long_div_quotient (UINTVAL (operands[3]), crc_size);
+
+  rtx crc_extended = gen_rtx_ZERO_EXTEND (word_mode, operands[1]);
+  rtx crc = gen_reg_rtx (word_mode);
+  if (crc_size > data_size)
+    riscv_expand_op (LSHIFTRT, word_mode, crc, crc_extended,
+		     gen_int_mode (crc_size - data_size, word_mode));
+  else
+    crc = gen_rtx_ZERO_EXTEND (word_mode, operands[1]);
+  rtx t0 = gen_reg_rtx (word_mode);
+  riscv_emit_move (t0, gen_int_mode (q, word_mode));
+  rtx t1 = gen_reg_rtx (word_mode);
+  riscv_emit_move (t1, operands[3]);
+
+  rtx a0 = gen_reg_rtx (word_mode);
+  rtx data = gen_rtx_ZERO_EXTEND (word_mode, operands[2]);
+  riscv_expand_op (XOR, word_mode, a0, crc, data);
+
+  if (TARGET_64BIT)
+    emit_insn (gen_riscv_clmul_di (a0, a0, t0));
+  else
+    emit_insn (gen_riscv_clmul_si (a0, a0, t0));
+
+  riscv_expand_op (LSHIFTRT, word_mode, a0, a0,
+		   gen_int_mode (crc_size, word_mode));
+  if (TARGET_64BIT)
+    emit_insn (gen_riscv_clmul_di (a0, a0, t1));
+  else
+    emit_insn (gen_riscv_clmul_si (a0, a0, t1));
+
+  if (crc_size > data_size)
+    {
+      rtx crc_part = gen_reg_rtx (word_mode);
+      riscv_expand_op (ASHIFT, word_mode, crc_part, operands[1],
+		       gen_int_mode (data_size, word_mode));
+      riscv_expand_op (XOR, word_mode, a0, a0, crc_part);
+    }
+  riscv_emit_move (operands[0], gen_lowpart (crc_mode, a0));
+}
+
+/* Generate assembly to calculate reversed CRC using clmul instruction.
+   OPERANDS[1] is input CRC,
+   OPERANDS[2] is data (message),
+   OPERANDS[3] is the polynomial without the leading 1.  */
+
+void
+expand_reversed_crc_using_clmul (scalar_mode crc_mode, scalar_mode data_mode,
+				 rtx *operands)
+{
+  /* Check and keep arguments.  */
+  gcc_assert (!CONST_INT_P (operands[0]));
+  gcc_assert (CONST_INT_P (operands[3]));
+  unsigned short crc_size = GET_MODE_BITSIZE (crc_mode);
+  gcc_assert (crc_size <= 32);
+  unsigned short data_size = GET_MODE_BITSIZE (data_mode);
+  rtx polynomial = operands[3];
+
+  /* Calculate the quotient.  */
+  unsigned HOST_WIDE_INT
+  q = gf2n_poly_long_div_quotient (UINTVAL (polynomial), crc_size);
+  /* Reflect the calculated quotient.  */
+  q = reflect_hwi (q, crc_size + 1);
+  rtx t0 = gen_reg_rtx (word_mode);
+  riscv_emit_move (t0, gen_int_mode (q, word_mode));
+
+  /* Reflect the polynomial.  */
+  unsigned HOST_WIDE_INT
+  ref_polynomial = reflect_hwi (UINTVAL (polynomial),
+				crc_size);
+  rtx t1 = gen_reg_rtx (word_mode);
+  riscv_emit_move (t1, gen_int_mode (ref_polynomial << 1, word_mode));
+
+  rtx crc = gen_rtx_ZERO_EXTEND (word_mode, operands[1]);
+  rtx data = gen_rtx_ZERO_EXTEND (word_mode, operands[2]);
+  rtx a0 = gen_reg_rtx (word_mode);
+  riscv_expand_op (XOR, word_mode, a0, crc, data);
+
+  if (TARGET_64BIT)
+    emit_insn (gen_riscv_clmul_di (a0, a0, t0));
+  else
+    emit_insn (gen_riscv_clmul_si (a0, a0, t0));
+
+  rtx num_shift = gen_int_mode (GET_MODE_BITSIZE (word_mode) - data_size,
+				word_mode);
+  riscv_expand_op (ASHIFT, word_mode, a0, a0, num_shift);
+
+  if (TARGET_64BIT)
+    emit_insn (gen_riscv_clmulh_di (a0, a0, t1));
+  else
+    emit_insn (gen_riscv_clmulh_si (a0, a0, t1));
+
+  if (crc_size > data_size)
+    {
+      rtx data_size_shift = gen_int_mode (data_size, word_mode);
+      rtx crc_part = gen_reg_rtx (word_mode);
+      riscv_expand_op (LSHIFTRT, word_mode, crc_part, crc, data_size_shift);
+      riscv_expand_op (XOR, word_mode, a0, a0, crc_part);
+    }
+
+  riscv_emit_move (operands[0], gen_lowpart (crc_mode, a0));
+}
+
 /* Initialize the GCC target structure.  */
 #undef TARGET_ASM_ALIGNED_HI_OP
 #define TARGET_ASM_ALIGNED_HI_OP "\t.half\t"
diff --git a/gcc/config/riscv/riscv.md b/gcc/config/riscv/riscv.md
index afde529..5f0749a 100644
--- a/gcc/config/riscv/riscv.md
+++ b/gcc/config/riscv/riscv.md
@@ -95,6 +95,10 @@
   ;; XTheadFmv moves
   UNSPEC_XTHEADFMV
   UNSPEC_XTHEADFMV_HW
+
+  ;; CRC unspecs
+  UNSPEC_CRC
+  UNSPEC_CRC_REV
 ])
 
 (define_c_enum "unspecv" [
diff --git a/gcc/doc/sourcebuild.texi b/gcc/doc/sourcebuild.texi
index b7167b3..54e485a 100644
--- a/gcc/doc/sourcebuild.texi
+++ b/gcc/doc/sourcebuild.texi
@@ -2548,6 +2548,24 @@ Test target architecture has support for the zacas extension.
 @item riscv_zalrsc
 Test target architecture has support for the zalrsc extension.
 
+@item riscv_zbc
+Test target architecture has support for the zbc extension.
+
+@item riscv_zbc_ok
+Test target architecture can execute code with zbc extension enabled.
+
+@item riscv_zbkb
+Test target architecture has support for the zbkb extension.
+
+@item riscv_zbkb_ok
+Test target architecture can execute code with zbkb extension enabled.
+
+@item riscv_zbkc
+Test target architecture has support for the zbkc extension.
+
+@item riscv_zbkc_ok
+Test target architecture can execute code with zbkc extension enabled.
+
 @item riscv_ztso
 Test target architecture has support for the ztso extension.
 
diff --git a/gcc/expr.cc b/gcc/expr.cc
index de25437..70f2ece 100644
--- a/gcc/expr.cc
+++ b/gcc/expr.cc
@@ -14178,6 +14178,33 @@ int_expr_size (const_tree exp)
   return tree_to_shwi (size);
 }
 
+/* Return the quotient of polynomial long division of x^2N by POLYNOMIAL
+   in GF (2^N).
+   Author: Richard Sandiford <richard.sandiford@arm.com>  */
+
+unsigned HOST_WIDE_INT
+gf2n_poly_long_div_quotient (unsigned HOST_WIDE_INT polynomial,
+			     unsigned short n)
+{
+  /* The result has degree N, so needs N + 1 bits.  */
+  gcc_assert (n < 64);
+
+  /* Perform a division step for the x^2N coefficient.  At this point the
+     quotient and remainder have N implicit trailing zeros.  */
+  unsigned HOST_WIDE_INT quotient = 1;
+  unsigned HOST_WIDE_INT remainder = polynomial;
+
+  /* Process the coefficients for x^(2N-1) down to x^N, with each step
+     reducing the number of implicit trailing zeros by one.  */
+  for (unsigned int i = 0; i < n; ++i)
+    {
+      bool coeff = remainder & (HOST_WIDE_INT_1U << (n - 1));
+      quotient = (quotient << 1) | coeff;
+      remainder = (remainder << 1) ^ (coeff ? polynomial : 0);
+    }
+  return quotient;
+}
+
 /* Calculate CRC for the initial CRC and given POLYNOMIAL.
    CRC_BITS is CRC size.  */
 
diff --git a/gcc/expr.h b/gcc/expr.h
index 5dd059d..fe93ee7 100644
--- a/gcc/expr.h
+++ b/gcc/expr.h
@@ -377,6 +377,11 @@ extern rtx expr_size (tree);
 extern bool mem_ref_refers_to_non_mem_p (tree);
 extern bool non_mem_decl_p (tree);
 
+/* Return the quotient of the polynomial long division of x^2N by POLYNOMIAL
+   in GF (2^N).  */
+extern unsigned HOST_WIDE_INT
+gf2n_poly_long_div_quotient (unsigned HOST_WIDE_INT, unsigned short);
+
 /* Generate table-based CRC.  */
 extern void generate_reflecting_code_standard (rtx *);
 extern void expand_crc_table_based (rtx, rtx, rtx, rtx, machine_mode);
diff --git a/gcc/hwint.cc b/gcc/hwint.cc
index e5c3619..adb30e2 100644
--- a/gcc/hwint.cc
+++ b/gcc/hwint.cc
@@ -188,3 +188,21 @@ least_common_multiple (HOST_WIDE_INT a, HOST_WIDE_INT b)
 {
   return mul_hwi (abs_hwi (a) / gcd (a, b), abs_hwi (b));
 }
+
+/* Reflect (reverse) the bits of a given VALUE within a specified BITWIDTH.  */
+
+unsigned HOST_WIDE_INT
+reflect_hwi (unsigned HOST_WIDE_INT value, unsigned bitwidth)
+{
+  unsigned HOST_WIDE_INT reflected_value = 0;
+  /* Loop through each bit in the specified BITWIDTH.  */
+  for (size_t i = 0; i < bitwidth; i++)
+    {
+      reflected_value <<= 1;
+      /* Add the least significant bit of the current value to the
+	 reflected value.  */
+      reflected_value |= (value & 1);
+      value >>= 1;
+    }
+  return reflected_value;
+}
\ No newline at end of file
diff --git a/gcc/hwint.h b/gcc/hwint.h
index 82416bf..f69b61d 100644
--- a/gcc/hwint.h
+++ b/gcc/hwint.h
@@ -284,6 +284,7 @@ extern HOST_WIDE_INT gcd (HOST_WIDE_INT, HOST_WIDE_INT);
 extern HOST_WIDE_INT pos_mul_hwi (HOST_WIDE_INT, HOST_WIDE_INT);
 extern HOST_WIDE_INT mul_hwi (HOST_WIDE_INT, HOST_WIDE_INT);
 extern HOST_WIDE_INT least_common_multiple (HOST_WIDE_INT, HOST_WIDE_INT);
+extern unsigned HOST_WIDE_INT reflect_hwi (unsigned HOST_WIDE_INT, unsigned);
 
 /* Like ctz_hwi, except 0 when x == 0.  */
 
diff --git a/gcc/testsuite/gcc.target/riscv/crc-builtin-zbc32.c b/gcc/testsuite/gcc.target/riscv/crc-builtin-zbc32.c
new file mode 100644
index 0000000..20d7d25
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/crc-builtin-zbc32.c
@@ -0,0 +1,21 @@
+/* { dg-do compile { target { riscv32*-*-* } } } */
+/* { dg-options "-march=rv32gc_zbc" } */
+
+#include <stdint-gcc.h>
+
+int8_t crc8_data8 ()
+{
+  return __builtin_crc8_data8 (0x34, 'a', 0x12);
+}
+
+int16_t crc16_data8 ()
+{
+  return __builtin_crc16_data8 (0x1234, 'a', 0x1021);
+}
+
+int16_t crc16_data16 ()
+{
+  return __builtin_crc16_data16 (0x1234, 0x3214, 0x1021);
+}
+
+/* { dg-final { scan-assembler-times "clmul\t" 6 } } */
\ No newline at end of file
diff --git a/gcc/testsuite/gcc.target/riscv/crc-builtin-zbc64.c b/gcc/testsuite/gcc.target/riscv/crc-builtin-zbc64.c
new file mode 100644
index 0000000..d99a78d
--- /dev/null
+++ b/gcc/testsuite/gcc.target/riscv/crc-builtin-zbc64.c
@@ -0,0 +1,66 @@
+/* { dg-do compile { target { riscv64*-*-* } } } */
+/* { dg-options "-march=rv64gc_zbc" } */
+
+#include <stdint-gcc.h>
+
+int8_t crc8_data8 ()
+{
+  return __builtin_crc8_data8 (0x34, 'a', 0x12);
+}
+
+int16_t crc16_data8 ()
+{
+  return __builtin_crc16_data8 (0x1234, 'a', 0x1021);
+}
+
+int16_t crc16_data16 ()
+{
+  return __builtin_crc16_data16 (0x1234, 0x3214, 0x1021);
+}
+
+int32_t crc32_data8 ()
+{
+  return __builtin_crc32_data8 (0xffffffff, 0x32, 0x4002123);
+}
+
+int32_t crc32_data16 ()
+{
+  return __builtin_crc32_data16 (0xffffffff, 0x3232, 0x4002123);
+}
+
+int32_t crc32_data32 ()
+{
+  return __builtin_crc32_data32 (0xffffffff, 0x123546ff, 0x4002123);
+}
+
+int8_t rev_crc8_data8 ()
+{
+  return __builtin_rev_crc8_data8 (0x34, 'a', 0x12);
+}
+
+int16_t rev_crc16_data8 ()
+{
+  return __builtin_rev_crc16_data8 (0x1234, 'a', 0x1021);
+}
+
+int16_t rev_crc16_data16 ()
+{
+  return __builtin_rev_crc16_data16 (0x1234, 0x3214, 0x1021);
+}
+
+int32_t rev_crc32_data8 ()
+{
+  return __builtin_rev_crc32_data8 (0xffffffff, 0x32, 0x4002123);
+}
+
+int32_t rev_crc32_data16 ()
+{
+  return __builtin_rev_crc32_data16 (0xffffffff, 0x3232, 0x4002123);
+}
+
+int32_t rev_crc32_data32 ()
+{
+  return __builtin_rev_crc32_data32 (0xffffffff, 0x123546ff, 0x4002123);
+}
+/* { dg-final { scan-assembler-times "clmul\t" 18 } } */
+/* { dg-final { scan-assembler-times "clmulh" 6 } } */
diff --git a/gcc/testsuite/lib/target-supports.exp b/gcc/testsuite/lib/target-supports.exp
index 95acd09..f2edbef 100644
--- a/gcc/testsuite/lib/target-supports.exp
+++ b/gcc/testsuite/lib/target-supports.exp
@@ -2065,6 +2065,39 @@ proc check_effective_target_riscv_zvbb { } {
     }]
 }
 
+# Return 1 if the target arch supports the Zbc extension, 0 otherwise.
+# Cache the result.
+
+proc check_effective_target_riscv_zbc { } {
+    return [check_no_compiler_messages riscv_ext_zbc assembly {
+       #ifndef __riscv_zbc
+       #error "Not __riscv_zbc"
+       #endif
+    }]
+}
+
+# Return 1 if the target arch supports the Zbkb extension, 0 otherwise.
+# Cache the result.
+
+proc check_effective_target_riscv_zbkb { } {
+    return [check_no_compiler_messages riscv_ext_zbkb assembly {
+       #ifndef __riscv_zbkb
+       #error "Not __riscv_zbkb"
+       #endif
+    }]
+}
+
+# Return 1 if the target arch supports the Zbkc extension, 0 otherwise.
+# Cache the result.
+
+proc check_effective_target_riscv_zbkc { } {
+    return [check_no_compiler_messages riscv_ext_zbkc assembly {
+       #ifndef __riscv_zbkc
+       #error "Not __riscv_zbkc"
+       #endif
+    }]
+}
+
 # Return 1 if the target arch supports the XTheadVector extension, 0 otherwise.
 # Cache the result.
 
@@ -2153,6 +2186,77 @@ proc check_effective_target_riscv_zvfh_ok { } {
     return 0
 }
 
+# Return 1 if we can execute code when using dg-add-options riscv_zbc
+
+proc check_effective_target_riscv_zbc_ok { } {
+    # If the target already supports zbc without any added options,
+    # we may assume we can execute just fine.
+    if { [check_effective_target_riscv_zbc] } {
+	return 1
+    }
+
+    # check if we can execute zbc insns with the given hardware or
+    # simulator
+    set gcc_march [riscv_get_arch]
+    if { [check_runtime ${gcc_march}_zbc_exec {
+	int main()
+	{
+	    asm ("clmul a0,a0,a1");
+	    asm ("clmulh a0,a0,a1");
+	    return 0;
+	} } "-march=${gcc_march}"] } {
+	    return 1
+	}
+    return 0
+}
+
+# Return 1 if we can execute code when using dg-add-options riscv_zbkb
+
+proc check_effective_target_riscv_zbkb_ok { } {
+    # If the target already supports zbkb without any added options,
+    # we may assume we can execute just fine.
+    if { [check_effective_target_riscv_zbkb] } {
+	return 1
+    }
+
+    # check if we can execute zbkb insns with the given hardware or
+    # simulator
+    set gcc_march [riscv_get_arch]
+    if { [check_runtime ${gcc_march}_zbkb_exec {
+	int main()
+	{
+	    asm ("brev8 a0,a0");
+	    return 0;
+	} } "-march=${gcc_march}"] } {
+	    return 1
+	}
+    return 0
+}
+
+# Return 1 if we can execute code when using dg-add-options riscv_zbkc
+
+proc check_effective_target_riscv_zbkc_ok { } {
+    # If the target already supports zbkc without any added options,
+    # we may assume we can execute just fine.
+    if { [check_effective_target_riscv_zbkc] } {
+	return 1
+    }
+
+    # check if we can execute zbkc insns with the given hardware or
+    # simulator
+    set gcc_march [riscv_get_arch]
+    if { [check_runtime ${gcc_march}_zbkc_exec {
+	int main()
+	{
+	    asm ("clmul a0,a0,a1");
+	    asm ("clmulh a0,a0,a1");
+	    return 0;
+	} } "-march=${gcc_march}"] } {
+	    return 1
+	}
+    return 0
+}
+
 # Return 1 if we can execute code when using dg-add-options riscv_zvbb
 
 proc check_effective_target_riscv_zvbb_ok { } {
@@ -2206,7 +2310,7 @@ proc check_effective_target_riscv_v_misalign_ok { } {
 proc riscv_get_arch { } {
     set gcc_march ""
     # ??? do we neeed to add more extensions to the list below?
-    foreach ext { i e m a f d q c b v zicsr zifencei zfh zba zbb zbc zbs zvbb zvfh ztso zaamo zalrsc zabha zacas } {
+    foreach ext { i e m a f d q c b v zicsr zifencei zfh zba zbb zbc zbkb zbkc zbs zvbb zvfh ztso zaamo zalrsc zabha zacas } {
 	if { [check_no_compiler_messages  riscv_ext_$ext assembly [string map [list DEF __riscv_$ext] {
 		#ifndef DEF
 		#error "Not DEF"