[PATCH v2 0/4] LoongArch: Add some floating-point operations

[PATCH v2 0/4] LoongArch: Add some floating-point operations

[Xi Ruoyao]

These patches allow to expand the following builtins to floating point
instructions for LoongArch:

- __builtin_rint{,f}
- __builtin_{l,ll}rint{,f}
- __builtin_{l,ll}floor{,f}
- __builtin_{l,ll}ceil{,f}
- __builtin_scalb{n,ln}{,f}
- __builtin_logb{,f}

Bootstrapped and regtested on loongarch64-linux-gnu.  And a modified
Glibc using the builtins for rint{,f}, {l,ll}rint{,f}, and logb{,f}
also survived Glibc test suite.

Please review ASAP because GCC 13 stage 1 will end on Nov. 13th.

v1 -> v2: Only use ftint{rm,rp} instructions if floor and ceil are
allowed to raise inexact exception.

Xi Ruoyao (4):
  LoongArch: Rename frint_<fmt> to rint<mode>2
  LoongArch: Add ftint{,rm,rp}.{w,l}.{s,d} instructions
  LoongArch: Add fscaleb.{s,d} instructions as ldexp{sf,df}3
  LoongArch: Add flogb.{s,d} instructions and expand logb{sf,df}2

 gcc/config/loongarch/loongarch.md             | 95 ++++++++++++++++++-
 gcc/testsuite/gcc.target/loongarch/flogb.c    | 18 ++++
 gcc/testsuite/gcc.target/loongarch/frint.c    | 16 ++++
 gcc/testsuite/gcc.target/loongarch/fscaleb.c  | 48 ++++++++++
 .../gcc.target/loongarch/ftint-no-inexact.c   | 44 +++++++++
 gcc/testsuite/gcc.target/loongarch/ftint.c    | 44 +++++++++
 6 files changed, 261 insertions(+), 4 deletions(-)
 create mode 100644 gcc/testsuite/gcc.target/loongarch/flogb.c
 create mode 100644 gcc/testsuite/gcc.target/loongarch/frint.c
 create mode 100644 gcc/testsuite/gcc.target/loongarch/fscaleb.c
 create mode 100644 gcc/testsuite/gcc.target/loongarch/ftint-no-inexact.c
 create mode 100644 gcc/testsuite/gcc.target/loongarch/ftint.c

-- 
2.38.1

[PATCH v2 1/4] LoongArch: Rename frint_<fmt> to rint<mode>2

[Xi Ruoyao]

Use standard name so __builtin_rint{,f} can be expanded to one
instruction.

gcc/ChangeLog:

	* config/loongarch/loongarch.md (frint_<fmt>): Rename to ..
	(rint<mode>2): .. this.

gcc/testsuite/ChangeLog:

	* gcc.target/loongarch/frint.c: New test.
---
 gcc/config/loongarch/loongarch.md          |  4 ++--
 gcc/testsuite/gcc.target/loongarch/frint.c | 16 ++++++++++++++++
 2 files changed, 18 insertions(+), 2 deletions(-)
 create mode 100644 gcc/testsuite/gcc.target/loongarch/frint.c

diff --git a/gcc/config/loongarch/loongarch.md b/gcc/config/loongarch/loongarch.md
index bda34d0f3db..a14ab14ac24 100644
--- a/gcc/config/loongarch/loongarch.md
+++ b/gcc/config/loongarch/loongarch.md
@@ -2012,8 +2012,8 @@ (define_insn "lui_h_hi12"
   [(set_attr "type" "move")]
 )
 
-;; Convert floating-point numbers to integers
-(define_insn "frint_<fmt>"
+;; Round floating-point numbers to integers
+(define_insn "rint<mode>2"
   [(set (match_operand:ANYF 0 "register_operand" "=f")
 	(unspec:ANYF [(match_operand:ANYF 1 "register_operand" "f")]
 		      UNSPEC_FRINT))]
diff --git a/gcc/testsuite/gcc.target/loongarch/frint.c b/gcc/testsuite/gcc.target/loongarch/frint.c
new file mode 100644
index 00000000000..3ee6a8f973a
--- /dev/null
+++ b/gcc/testsuite/gcc.target/loongarch/frint.c
@@ -0,0 +1,16 @@
+/* { dg-do compile } */
+/* { dg-options "-mdouble-float" } */
+/* { dg-final { scan-assembler "frint\\.s" } } */
+/* { dg-final { scan-assembler "frint\\.d" } } */
+
+double
+my_rint (double a)
+{
+  return __builtin_rint (a);
+}
+
+float
+my_rintf (float a)
+{
+  return __builtin_rintf (a);
+}
-- 
2.38.1

[PATCH v2 2/4] LoongArch: Add ftint{,rm,rp}.{w,l}.{s,d} instructions

[Xi Ruoyao]

This allows to optimize the following builtins if -fno-math-errno:

- __builtin_lrint{,f}
- __builtin_lfloor{,f}
- __builtin_lceil{,f}

Inspired by
https://gcc.gnu.org/pipermail/gcc-patches/2022-November/605287.html.

ANYFI is added so the compiler won't try ftint.l.s if -mfpu=32.  If we
simply used GPR here an ICE would be triggered with __builtin_lrintf
and -mfpu=32.

ftint{rm,rp} instructions may raise inexact exception, so they can't be
used if -fno-trapping-math -fno-fp-int-builtin-inexact.

Note that the .w.{s,d} variants are not tested because we don't support
ILP32 for now.

gcc/ChangeLog:

	* config/loongarch/loongarch.md (UNSPEC_FTINT): New unspec.
	(UNSPEC_FTINTRM): Likewise.
	(UNSPEC_FTINTRP): Likewise.
	(LRINT): New define_int_iterator.
	(lrint_pattern): New define_int_attr.
	(lrint_submenmonic): Likewise.
	(lrint_allow_inexact): Likewise.
	(ANYFI): New define_mode_iterator.
	(lrint<ANYF><ANYFI>): New instruction template.

gcc/testsuite/ChangeLog:

	* gcc.target/loongarch/ftint.c: New test.
	* gcc.target/loongarch/ftint-no-inexact.c: New test.
---
 gcc/config/loongarch/loongarch.md             | 34 ++++++++++++++
 .../gcc.target/loongarch/ftint-no-inexact.c   | 44 +++++++++++++++++++
 gcc/testsuite/gcc.target/loongarch/ftint.c    | 44 +++++++++++++++++++
 3 files changed, 122 insertions(+)
 create mode 100644 gcc/testsuite/gcc.target/loongarch/ftint-no-inexact.c
 create mode 100644 gcc/testsuite/gcc.target/loongarch/ftint.c

diff --git a/gcc/config/loongarch/loongarch.md b/gcc/config/loongarch/loongarch.md
index a14ab14ac24..eb127c346a3 100644
--- a/gcc/config/loongarch/loongarch.md
+++ b/gcc/config/loongarch/loongarch.md
@@ -38,6 +38,9 @@ (define_c_enum "unspec" [
   UNSPEC_FMAX
   UNSPEC_FMIN
   UNSPEC_FCOPYSIGN
+  UNSPEC_FTINT
+  UNSPEC_FTINTRM
+  UNSPEC_FTINTRP
 
   ;; Override return address for exception handling.
   UNSPEC_EH_RETURN
@@ -374,6 +377,11 @@ (define_mode_iterator QHWD [QI HI SI (DI "TARGET_64BIT")])
 (define_mode_iterator ANYF [(SF "TARGET_HARD_FLOAT")
 			    (DF "TARGET_DOUBLE_FLOAT")])
 
+;; Iterator for fixed-point modes which can be hold by a hardware
+;; floating-point register.
+(define_mode_iterator ANYFI [(SI "TARGET_HARD_FLOAT")
+			     (DI "TARGET_DOUBLE_FLOAT")])
+
 ;; A mode for which moves involving FPRs may need to be split.
 (define_mode_iterator SPLITF
   [(DF "!TARGET_64BIT && TARGET_DOUBLE_FLOAT")
@@ -515,6 +523,19 @@ (define_code_attr fcond [(unordered "cun")
 (define_code_attr sel [(eq "masknez") (ne "maskeqz")])
 (define_code_attr selinv [(eq "maskeqz") (ne "masknez")])
 
+;; Iterator and attributes for floating-point to fixed-point conversion
+;; instructions.
+(define_int_iterator LRINT [UNSPEC_FTINT UNSPEC_FTINTRM UNSPEC_FTINTRP])
+(define_int_attr lrint_pattern [(UNSPEC_FTINT "lrint")
+				(UNSPEC_FTINTRM "lfloor")
+				(UNSPEC_FTINTRP "lceil")])
+(define_int_attr lrint_submenmonic [(UNSPEC_FTINT "")
+				    (UNSPEC_FTINTRM "rm")
+				    (UNSPEC_FTINTRP "rp")])
+(define_int_attr lrint_allow_inexact [(UNSPEC_FTINT "1")
+				      (UNSPEC_FTINTRM "0")
+				      (UNSPEC_FTINTRP "0")])
+
 ;;
 ;;  ....................
 ;;
@@ -2022,6 +2043,19 @@ (define_insn "rint<mode>2"
   [(set_attr "type" "fcvt")
    (set_attr "mode" "<MODE>")])
 
+;; Convert floating-point numbers to integers
+(define_insn "<lrint_pattern><ANYF:mode><ANYFI:mode>2"
+  [(set (match_operand:ANYFI 0 "register_operand" "=f")
+	(unspec:ANYFI [(match_operand:ANYF 1 "register_operand" "f")]
+		      LRINT))]
+  "TARGET_HARD_FLOAT &&
+   (<lrint_allow_inexact>
+    || flag_fp_int_builtin_inexact
+    || !flag_trapping_math)"
+  "ftint<lrint_submenmonic>.<ANYFI:ifmt>.<ANYF:fmt> %0,%1"
+  [(set_attr "type" "fcvt")
+   (set_attr "mode" "<ANYF:MODE>")])
+
 ;; Load the low word of operand 0 with operand 1.
 (define_insn "load_low<mode>"
   [(set (match_operand:SPLITF 0 "register_operand" "=f,f")
diff --git a/gcc/testsuite/gcc.target/loongarch/ftint-no-inexact.c b/gcc/testsuite/gcc.target/loongarch/ftint-no-inexact.c
new file mode 100644
index 00000000000..88b83a9c056
--- /dev/null
+++ b/gcc/testsuite/gcc.target/loongarch/ftint-no-inexact.c
@@ -0,0 +1,44 @@
+/* { dg-do compile } */
+/* { dg-options "-mabi=lp64d -mdouble-float -fno-math-errno -fno-fp-int-builtin-inexact" } */
+/* { dg-final { scan-assembler "ftint\\.l\\.s" } } */
+/* { dg-final { scan-assembler "ftint\\.l\\.d" } } */
+/* { dg-final { scan-assembler-not "ftintrm\\.l\\.s" } } */
+/* { dg-final { scan-assembler-not "ftintrm\\.l\\.d" } } */
+/* { dg-final { scan-assembler-not "ftintrp\\.l\\.s" } } */
+/* { dg-final { scan-assembler-not "ftintrp\\.l\\.d" } } */
+
+long
+my_lrint (double a)
+{
+  return __builtin_lrint (a);
+}
+
+long
+my_lrintf (float a)
+{
+  return __builtin_lrintf (a);
+}
+
+long
+my_lfloor (double a)
+{
+  return __builtin_lfloor (a);
+}
+
+long
+my_lfloorf (float a)
+{
+  return __builtin_lfloorf (a);
+}
+
+long
+my_lceil (double a)
+{
+  return __builtin_lceil (a);
+}
+
+long
+my_lceilf (float a)
+{
+  return __builtin_lceilf (a);
+}
diff --git a/gcc/testsuite/gcc.target/loongarch/ftint.c b/gcc/testsuite/gcc.target/loongarch/ftint.c
new file mode 100644
index 00000000000..7a326a454d8
--- /dev/null
+++ b/gcc/testsuite/gcc.target/loongarch/ftint.c
@@ -0,0 +1,44 @@
+/* { dg-do compile } */
+/* { dg-options "-mabi=lp64d -mdouble-float -fno-math-errno -ffp-int-builtin-inexact" } */
+/* { dg-final { scan-assembler "ftint\\.l\\.s" } } */
+/* { dg-final { scan-assembler "ftint\\.l\\.d" } } */
+/* { dg-final { scan-assembler "ftintrm\\.l\\.s" } } */
+/* { dg-final { scan-assembler "ftintrm\\.l\\.d" } } */
+/* { dg-final { scan-assembler "ftintrp\\.l\\.s" } } */
+/* { dg-final { scan-assembler "ftintrp\\.l\\.d" } } */
+
+long
+my_lrint (double a)
+{
+  return __builtin_lrint (a);
+}
+
+long
+my_lrintf (float a)
+{
+  return __builtin_lrintf (a);
+}
+
+long
+my_lfloor (double a)
+{
+  return __builtin_lfloor (a);
+}
+
+long
+my_lfloorf (float a)
+{
+  return __builtin_lfloorf (a);
+}
+
+long
+my_lceil (double a)
+{
+  return __builtin_lceil (a);
+}
+
+long
+my_lceilf (float a)
+{
+  return __builtin_lceilf (a);
+}
-- 
2.38.1

[PATCH v2 3/4] LoongArch: Add fscaleb.{s,d} instructions as ldexp{sf,df}3

[Xi Ruoyao]

This allows optimizing __builtin_ldexp{,f} and __builtin_scalbn{,f} with
-fno-math-errno.

IMODE is added because we can't hard code SI for operand 2: fscaleb.d
instruction always take the high half of both source registers into
account.  See my_ldexp_long in the test case.

gcc/ChangeLog:

	* config/loongarch/loongarch.md (UNSPEC_FSCALEB): New unspec.
	(type): Add fscaleb.
	(IMODE): New mode attr.
	(ldexp<mode>3): New instruction template.

gcc/testsuite/ChangeLog:

	* gcc.target/loongarch/fscaleb.c: New test.
---
 gcc/config/loongarch/loongarch.md            | 26 ++++++++++-
 gcc/testsuite/gcc.target/loongarch/fscaleb.c | 48 ++++++++++++++++++++
 2 files changed, 72 insertions(+), 2 deletions(-)
 create mode 100644 gcc/testsuite/gcc.target/loongarch/fscaleb.c

diff --git a/gcc/config/loongarch/loongarch.md b/gcc/config/loongarch/loongarch.md
index eb127c346a3..c141c9adde2 100644
--- a/gcc/config/loongarch/loongarch.md
+++ b/gcc/config/loongarch/loongarch.md
@@ -41,6 +41,7 @@ (define_c_enum "unspec" [
   UNSPEC_FTINT
   UNSPEC_FTINTRM
   UNSPEC_FTINTRP
+  UNSPEC_FSCALEB
 
   ;; Override return address for exception handling.
   UNSPEC_EH_RETURN
@@ -220,6 +221,7 @@ (define_attr "qword_mode" "no,yes"
 ;; fcmp		floating point compare
 ;; fcopysign	floating point copysign
 ;; fcvt		floating point convert
+;; fscaleb	floating point scale
 ;; fsqrt	floating point square root
 ;; frsqrt       floating point reciprocal square root
 ;; multi	multiword sequence (or user asm statements)
@@ -231,8 +233,8 @@ (define_attr "type"
   "unknown,branch,jump,call,load,fpload,fpidxload,store,fpstore,fpidxstore,
    prefetch,prefetchx,condmove,mgtf,mftg,const,arith,logical,
    shift,slt,signext,clz,trap,imul,idiv,move,
-   fmove,fadd,fmul,fmadd,fdiv,frdiv,fabs,fneg,fcmp,fcopysign,fcvt,fsqrt,
-   frsqrt,accext,accmod,multi,atomic,syncloop,nop,ghost"
+   fmove,fadd,fmul,fmadd,fdiv,frdiv,fabs,fneg,fcmp,fcopysign,fcvt,fscaleb,
+   fsqrt,frsqrt,accext,accmod,multi,atomic,syncloop,nop,ghost"
   (cond [(eq_attr "jirl" "!unset") (const_string "call")
 	 (eq_attr "got" "load") (const_string "load")
 
@@ -418,6 +420,10 @@ (define_mode_attr UNITMODE [(SF "SF") (DF "DF")])
 ;; the controlling mode.
 (define_mode_attr HALFMODE [(DF "SI") (DI "SI") (TF "DI")])
 
+;; This attribute gives the integer mode that has the same size of a
+;; floating-point mode.
+(define_mode_attr IMODE [(SF "SI") (DF "DI")])
+
 ;; This code iterator allows signed and unsigned widening multiplications
 ;; to use the same template.
 (define_code_iterator any_extend [sign_extend zero_extend])
@@ -1014,7 +1020,23 @@ (define_insn "copysign<mode>3"
   "fcopysign.<fmt>\t%0,%1,%2"
   [(set_attr "type" "fcopysign")
    (set_attr "mode" "<UNITMODE>")])
+\f
+;;
+;;  ....................
+;;
+;;	FLOATING POINT SCALE
+;;
+;;  ....................
 
+(define_insn "ldexp<mode>3"
+  [(set (match_operand:ANYF 0 "register_operand" "=f")
+	(unspec:ANYF [(match_operand:ANYF    1 "register_operand" "f")
+		      (match_operand:<IMODE> 2 "register_operand" "f")]
+		     UNSPEC_FSCALEB))]
+  "TARGET_HARD_FLOAT"
+  "fscaleb.<fmt>\t%0,%1,%2"
+  [(set_attr "type" "fscaleb")
+   (set_attr "mode" "<UNITMODE>")])
 \f
 ;;
 ;;  ...................
diff --git a/gcc/testsuite/gcc.target/loongarch/fscaleb.c b/gcc/testsuite/gcc.target/loongarch/fscaleb.c
new file mode 100644
index 00000000000..f18470fbb8f
--- /dev/null
+++ b/gcc/testsuite/gcc.target/loongarch/fscaleb.c
@@ -0,0 +1,48 @@
+/* { dg-do compile } */
+/* { dg-options "-O2 -mabi=lp64d -mdouble-float -fno-math-errno" } */
+/* { dg-final { scan-assembler-times "fscaleb\\.s" 3 } } */
+/* { dg-final { scan-assembler-times "fscaleb\\.d" 4 } } */
+/* { dg-final { scan-assembler-times "slli\\.w" 1 } } */
+
+double
+my_scalbln (double a, long b)
+{
+  return __builtin_scalbln (a, b);
+}
+
+double
+my_scalbn (double a, int b)
+{
+  return __builtin_scalbn (a, b);
+}
+
+double
+my_ldexp (double a, int b)
+{
+  return __builtin_ldexp (a, b);
+}
+
+float
+my_scalblnf (float a, long b)
+{
+  return __builtin_scalblnf (a, b);
+}
+
+float
+my_scalbnf (float a, int b)
+{
+  return __builtin_scalbnf (a, b);
+}
+
+float
+my_ldexpf (float a, int b)
+{
+  return __builtin_ldexpf (a, b);
+}
+
+/* b must be sign-extended */
+double
+my_ldexp_long (double a, long b)
+{
+  return __builtin_ldexp (a, b);
+}
-- 
2.38.1

[PATCH v2 4/4] LoongArch: Add flogb.{s,d} instructions and expand logb{sf,df}2

[Xi Ruoyao]

On LoongArch, flogb instructions extract the exponent of a non-negative
floating point value, but produces NaN for negative values.  So we need
to add a fabs instruction when we expand logb.

gcc/ChangeLog:

	* config/loongarch/loongarch.md (UNSPEC_FLOGB): New unspec.
	(type): Add flogb.
	(logb_non_negative<mode>2): New instruction template.
	(logb<mode>2): New define_expand.

gcc/testsuite/ChangeLog:

	* gcc.target/loongarch/flogb.c: New test.
---
 gcc/config/loongarch/loongarch.md          | 35 ++++++++++++++++++++--
 gcc/testsuite/gcc.target/loongarch/flogb.c | 18 +++++++++++
 2 files changed, 51 insertions(+), 2 deletions(-)
 create mode 100644 gcc/testsuite/gcc.target/loongarch/flogb.c

diff --git a/gcc/config/loongarch/loongarch.md b/gcc/config/loongarch/loongarch.md
index c141c9adde2..682ab961741 100644
--- a/gcc/config/loongarch/loongarch.md
+++ b/gcc/config/loongarch/loongarch.md
@@ -42,6 +42,7 @@ (define_c_enum "unspec" [
   UNSPEC_FTINTRM
   UNSPEC_FTINTRP
   UNSPEC_FSCALEB
+  UNSPEC_FLOGB
 
   ;; Override return address for exception handling.
   UNSPEC_EH_RETURN
@@ -217,6 +218,7 @@ (define_attr "qword_mode" "no,yes"
 ;; fdiv		floating point divide
 ;; frdiv	floating point reciprocal divide
 ;; fabs		floating point absolute value
+;; flogb	floating point exponent extract
 ;; fneg		floating point negation
 ;; fcmp		floating point compare
 ;; fcopysign	floating point copysign
@@ -233,8 +235,8 @@ (define_attr "type"
   "unknown,branch,jump,call,load,fpload,fpidxload,store,fpstore,fpidxstore,
    prefetch,prefetchx,condmove,mgtf,mftg,const,arith,logical,
    shift,slt,signext,clz,trap,imul,idiv,move,
-   fmove,fadd,fmul,fmadd,fdiv,frdiv,fabs,fneg,fcmp,fcopysign,fcvt,fscaleb,
-   fsqrt,frsqrt,accext,accmod,multi,atomic,syncloop,nop,ghost"
+   fmove,fadd,fmul,fmadd,fdiv,frdiv,fabs,flogb,fneg,fcmp,fcopysign,fcvt,
+   fscaleb,fsqrt,frsqrt,accext,accmod,multi,atomic,syncloop,nop,ghost"
   (cond [(eq_attr "jirl" "!unset") (const_string "call")
 	 (eq_attr "got" "load") (const_string "load")
 
@@ -1039,6 +1041,35 @@ (define_insn "ldexp<mode>3"
    (set_attr "mode" "<UNITMODE>")])
 \f
 ;;
+;;  ....................
+;;
+;;	FLOATING POINT EXPONENT EXTRACT
+;;
+;;  ....................
+
+(define_insn "logb_non_negative<mode>2"
+  [(set (match_operand:ANYF 0 "register_operand" "=f")
+	(unspec:ANYF [(match_operand:ANYF 1 "register_operand" "f")]
+		     UNSPEC_FLOGB))]
+  "TARGET_HARD_FLOAT"
+  "flogb.<fmt>\t%0,%1"
+  [(set_attr "type" "flogb")
+   (set_attr "mode" "<UNITMODE>")])
+
+(define_expand "logb<mode>2"
+  [(set (match_operand:ANYF 0 "register_operand")
+	(unspec:ANYF [(abs:ANYF (match_operand:ANYF 1 "register_operand"))]
+		     UNSPEC_FLOGB))]
+  "TARGET_HARD_FLOAT"
+{
+  rtx tmp = gen_reg_rtx (<MODE>mode);
+
+  emit_insn (gen_abs<mode>2 (tmp, operands[1]));
+  emit_insn (gen_logb_non_negative<mode>2 (operands[0], tmp));
+  DONE;
+})
+\f
+;;
 ;;  ...................
 ;;
 ;;  Count leading zeroes.
diff --git a/gcc/testsuite/gcc.target/loongarch/flogb.c b/gcc/testsuite/gcc.target/loongarch/flogb.c
new file mode 100644
index 00000000000..1daefe54e13
--- /dev/null
+++ b/gcc/testsuite/gcc.target/loongarch/flogb.c
@@ -0,0 +1,18 @@
+/* { dg-do compile } */
+/* { dg-options "-mdouble-float -fno-math-errno" } */
+/* { dg-final { scan-assembler "fabs\\.s" } } */
+/* { dg-final { scan-assembler "fabs\\.d" } } */
+/* { dg-final { scan-assembler "flogb\\.s" } } */
+/* { dg-final { scan-assembler "flogb\\.d" } } */
+
+double
+my_logb (double a)
+{
+  return __builtin_logb (a);
+}
+
+float
+my_logbf (float a)
+{
+  return __builtin_logbf (a);
+}
-- 
2.38.1

Re: [PATCH v2 2/4] LoongArch: Add ftint{,rm,rp}.{w,l}.{s,d} instructions

[Lulu Cheng]

在 2022/11/9 下午9:53, Xi Ruoyao 写道:
> +;; Convert floating-point numbers to integers
> +(define_insn "<lrint_pattern><ANYF:mode><ANYFI:mode>2"
> +  [(set (match_operand:ANYFI 0 "register_operand" "=f")
> +	(unspec:ANYFI [(match_operand:ANYF 1 "register_operand" "f")]
> +		      LRINT))]
> +  "TARGET_HARD_FLOAT &&
> +   (<lrint_allow_inexact>
> +    || flag_fp_int_builtin_inexact
> +    || !flag_trapping_math)"
>
+    || !flag_trapping_math

I think this condition is backwards.

Re: [PATCH v2 1/4] LoongArch: Rename frint_<fmt> to rint<mode>2

[Lulu Cheng]

LGTM!

Thanks.

在 2022/11/9 下午9:53, Xi Ruoyao 写道:
> Use standard name so __builtin_rint{,f} can be expanded to one
> instruction.
>
> gcc/ChangeLog:
>
> 	* config/loongarch/loongarch.md (frint_<fmt>): Rename to ..
> 	(rint<mode>2): .. this.
>
> gcc/testsuite/ChangeLog:
>
> 	* gcc.target/loongarch/frint.c: New test.
> ---
>   gcc/config/loongarch/loongarch.md          |  4 ++--
>   gcc/testsuite/gcc.target/loongarch/frint.c | 16 ++++++++++++++++
>   2 files changed, 18 insertions(+), 2 deletions(-)
>   create mode 100644 gcc/testsuite/gcc.target/loongarch/frint.c
>
> diff --git a/gcc/config/loongarch/loongarch.md b/gcc/config/loongarch/loongarch.md
> index bda34d0f3db..a14ab14ac24 100644
> --- a/gcc/config/loongarch/loongarch.md
> +++ b/gcc/config/loongarch/loongarch.md
> @@ -2012,8 +2012,8 @@ (define_insn "lui_h_hi12"
>     [(set_attr "type" "move")]
>   )
>   
> -;; Convert floating-point numbers to integers
> -(define_insn "frint_<fmt>"
> +;; Round floating-point numbers to integers
> +(define_insn "rint<mode>2"
>     [(set (match_operand:ANYF 0 "register_operand" "=f")
>   	(unspec:ANYF [(match_operand:ANYF 1 "register_operand" "f")]
>   		      UNSPEC_FRINT))]
> diff --git a/gcc/testsuite/gcc.target/loongarch/frint.c b/gcc/testsuite/gcc.target/loongarch/frint.c
> new file mode 100644
> index 00000000000..3ee6a8f973a
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/loongarch/frint.c
> @@ -0,0 +1,16 @@
> +/* { dg-do compile } */
> +/* { dg-options "-mdouble-float" } */
> +/* { dg-final { scan-assembler "frint\\.s" } } */
> +/* { dg-final { scan-assembler "frint\\.d" } } */
> +
> +double
> +my_rint (double a)
> +{
> +  return __builtin_rint (a);
> +}
> +
> +float
> +my_rintf (float a)
> +{
> +  return __builtin_rintf (a);
> +}

Re: [PATCH v2 2/4] LoongArch: Add ftint{,rm,rp}.{w,l}.{s,d} instructions

[Xi Ruoyao]

On Thu, 2022-11-10 at 14:41 +0800, Lulu Cheng wrote:
> 
> 在 2022/11/9 下午9:53, Xi Ruoyao 写道:
> > +;; Convert floating-point numbers to integers
> > +(define_insn "<lrint_pattern><ANYF:mode><ANYFI:mode>2"
> > +  [(set (match_operand:ANYFI 0 "register_operand" "=f")
> > +       (unspec:ANYFI [(match_operand:ANYF 1 "register_operand"
> > "f")]
> > +                     LRINT))]
> > +  "TARGET_HARD_FLOAT &&
> > +   (<lrint_allow_inexact>
> > +    || flag_fp_int_builtin_inexact
> > +    || !flag_trapping_math)"
> > 
> +    || !flag_trapping_math
> 
> I think this condition is backwards.

I copied the logic from aarch64.md:6702.

Joseph: can you confirm that -ftrapping-math allows floor and ceil to
raise inexact exception?  The man page currently says:

The default is -ffp-int-builtin-inexact, allowing the exception to be
raised, unless C2X or a later C standard is selected.  This option does 
                                                       ^^^^^^^^^^^
nothing unless -ftrapping-math is in effect.

To me it's not very clear that "this option" stands for -fno-fp-int-
builtin-inexact or -ffp-int-builtin-inexact.

-- 
Xi Ruoyao <xry111@xry111.site>
School of Aerospace Science and Technology, Xidian University

Re: [PATCH v2 2/4] LoongArch: Add ftint{,rm,rp}.{w,l}.{s,d} instructions

[Joseph Myers]

[-- Attachment #1: Type: text/plain, Size: 1450 bytes --]

On Thu, 10 Nov 2022, Xi Ruoyao via Gcc-patches wrote:

> Joseph: can you confirm that -ftrapping-math allows floor and ceil to
> raise inexact exception?  The man page currently says:
> 
> The default is -ffp-int-builtin-inexact, allowing the exception to be
> raised, unless C2X or a later C standard is selected.  This option does 
>                                                        ^^^^^^^^^^^
> nothing unless -ftrapping-math is in effect.
> 
> To me it's not very clear that "this option" stands for -fno-fp-int-
> builtin-inexact or -ffp-int-builtin-inexact.

The -ftrapping-math option (which is on by default) means that we care 
about whether operations raise exceptions: they should raise exceptions if 
and only if the relevant standard permit them to do so.

The combination of -ftrapping-math with -fno-fp-int-builtin-inexact means 
the listed built-in functions must not raise "inexact".

If -fno-trapping-math is used, then we don't care about whether exceptions 
are raised or not (for any floating-point operations, not just those 
functions).  So given -fno-trapping-math, there is no difference between 
-fno-fp-int-builtin-inexact and -ffp-int-builtin-inexact.

If -ffp-int-builtin-inexact (default before C2X), we don't care about 
whether those functions raise "inexact" (but still care about other 
exceptions and exceptions for other operations, unless 
-fno-trapping-math).

-- 
Joseph S. Myers
joseph@codesourcery.com

Re: [PATCH v2 2/4] LoongArch: Add ftint{,rm,rp}.{w,l}.{s,d} instructions

[Xi Ruoyao]

Lulu:

So I think the code is correct:

+   (<lrint_allow_inexact>
+    || flag_fp_int_builtin_inexact
+    || !flag_trapping_math)"

<lrint_allow_inexact> is 1 for lrint, 0 for lceil and lfloor.  As N3054
says:

   The lrint and llrint functions provide floating-to-integer conversion as prescribed by IEC 60559.
   They round according to the current rounding direction. If the rounded value is outside the range of
   the return type, the numeric result is unspecified and the "invalid" floating-point exception is raised.
   When they raise no other floating-point exception and the result differs from the argument, they
   raise the "inexact" floating-point exception.
   
If flag_fp_int_builtin_inexact is set, we allow lceil and lfloor to
raise "inexact".

If flag_trapping_math is not set (it's set by default and can only be
unset explicitly with -fno-trapping-math: it's not even implied by -
ffast-math), we don't care about whether exceptions are raised or not.

So lceil and lfloor can be used if -ffp-int-builtin-inexact, or -fno-
trapping-math.

On Fri, 2022-11-11 at 00:07 +0000, Joseph Myers wrote:
> On Thu, 10 Nov 2022, Xi Ruoyao via Gcc-patches wrote:
> 
> > Joseph: can you confirm that -ftrapping-math allows floor and ceil to
> > raise inexact exception?  The man page currently says:
> > 
> > The default is -ffp-int-builtin-inexact, allowing the exception to be
> > raised, unless C2X or a later C standard is selected.  This option does 
> >                                                        ^^^^^^^^^^^
> > nothing unless -ftrapping-math is in effect.
> > 
> > To me it's not very clear that "this option" stands for -fno-fp-int-
> > builtin-inexact or -ffp-int-builtin-inexact.
> 
> The -ftrapping-math option (which is on by default) means that we care
> about whether operations raise exceptions: they should raise exceptions if 
> and only if the relevant standard permit them to do so.
> 
> The combination of -ftrapping-math with -fno-fp-int-builtin-inexact means 
> the listed built-in functions must not raise "inexact".
> 
> If -fno-trapping-math is used, then we don't care about whether exceptions 
> are raised or not (for any floating-point operations, not just those 
> functions).  So given -fno-trapping-math, there is no difference between 
> -fno-fp-int-builtin-inexact and -ffp-int-builtin-inexact.
> 
> If -ffp-int-builtin-inexact (default before C2X), we don't care about 
> whether those functions raise "inexact" (but still care about other 
> exceptions and exceptions for other operations, unless 
> -fno-trapping-math).

-- 
Xi Ruoyao <xry111@xry111.site>
School of Aerospace Science and Technology, Xidian University

Re: [PATCH v2 2/4] LoongArch: Add ftint{,rm,rp}.{w,l}.{s,d} instructions

[Lulu Cheng]

I have no more questions.

Thanks.

在 2022/11/11 下午1:29, Xi Ruoyao 写道:
> Lulu:
>
> So I think the code is correct:
>
> +   (<lrint_allow_inexact>
> +    || flag_fp_int_builtin_inexact
> +    || !flag_trapping_math)"
>
> <lrint_allow_inexact> is 1 for lrint, 0 for lceil and lfloor.  As N3054
> says:
>
>     The lrint and llrint functions provide floating-to-integer conversion as prescribed by IEC 60559.
>     They round according to the current rounding direction. If the rounded value is outside the range of
>     the return type, the numeric result is unspecified and the "invalid" floating-point exception is raised.
>     When they raise no other floating-point exception and the result differs from the argument, they
>     raise the "inexact" floating-point exception.
>     
> If flag_fp_int_builtin_inexact is set, we allow lceil and lfloor to
> raise "inexact".
>
> If flag_trapping_math is not set (it's set by default and can only be
> unset explicitly with -fno-trapping-math: it's not even implied by -
> ffast-math), we don't care about whether exceptions are raised or not.
>
> So lceil and lfloor can be used if -ffp-int-builtin-inexact, or -fno-
> trapping-math.
>
> On Fri, 2022-11-11 at 00:07 +0000, Joseph Myers wrote:
>> On Thu, 10 Nov 2022, Xi Ruoyao via Gcc-patches wrote:
>>
>>> Joseph: can you confirm that -ftrapping-math allows floor and ceil to
>>> raise inexact exception?  The man page currently says:
>>>
>>> The default is -ffp-int-builtin-inexact, allowing the exception to be
>>> raised, unless C2X or a later C standard is selected.  This option does
>>>                                                         ^^^^^^^^^^^
>>> nothing unless -ftrapping-math is in effect.
>>>
>>> To me it's not very clear that "this option" stands for -fno-fp-int-
>>> builtin-inexact or -ffp-int-builtin-inexact.
>> The -ftrapping-math option (which is on by default) means that we care
>> about whether operations raise exceptions: they should raise exceptions if
>> and only if the relevant standard permit them to do so.
>>
>> The combination of -ftrapping-math with -fno-fp-int-builtin-inexact means
>> the listed built-in functions must not raise "inexact".
>>
>> If -fno-trapping-math is used, then we don't care about whether exceptions
>> are raised or not (for any floating-point operations, not just those
>> functions).  So given -fno-trapping-math, there is no difference between
>> -fno-fp-int-builtin-inexact and -ffp-int-builtin-inexact.
>>
>> If -ffp-int-builtin-inexact (default before C2X), we don't care about
>> whether those functions raise "inexact" (but still care about other
>> exceptions and exceptions for other operations, unless
>> -fno-trapping-math).

Re: [PATCH v2 3/4] LoongArch: Add fscaleb.{s,d} instructions as ldexp{sf,df}3

[Lulu Cheng]

在 2022/11/9 下午9:53, Xi Ruoyao 写道:
> This allows optimizing __builtin_ldexp{,f} and __builtin_scalbn{,f} with
> -fno-math-errno.
>
> IMODE is added because we can't hard code SI for operand 2: fscaleb.d
> instruction always take the high half of both source registers into
> account.  See my_ldexp_long in the test case.
>
> gcc/ChangeLog:
>
> 	* config/loongarch/loongarch.md (UNSPEC_FSCALEB): New unspec.
> 	(type): Add fscaleb.
> 	(IMODE): New mode attr.
> 	(ldexp<mode>3): New instruction template.
>
> gcc/testsuite/ChangeLog:
>
> 	* gcc.target/loongarch/fscaleb.c: New test.
> ---
>   gcc/config/loongarch/loongarch.md            | 26 ++++++++++-
>   gcc/testsuite/gcc.target/loongarch/fscaleb.c | 48 ++++++++++++++++++++
>   2 files changed, 72 insertions(+), 2 deletions(-)
>   create mode 100644 gcc/testsuite/gcc.target/loongarch/fscaleb.c
>
> diff --git a/gcc/config/loongarch/loongarch.md b/gcc/config/loongarch/loongarch.md
> index eb127c346a3..c141c9adde2 100644
> --- a/gcc/config/loongarch/loongarch.md
> +++ b/gcc/config/loongarch/loongarch.md
> @@ -41,6 +41,7 @@ (define_c_enum "unspec" [
>     UNSPEC_FTINT
>     UNSPEC_FTINTRM
>     UNSPEC_FTINTRP
> +  UNSPEC_FSCALEB
>   
>     ;; Override return address for exception handling.
>     UNSPEC_EH_RETURN
> @@ -220,6 +221,7 @@ (define_attr "qword_mode" "no,yes"
>   ;; fcmp		floating point compare
>   ;; fcopysign	floating point copysign
>   ;; fcvt		floating point convert
> +;; fscaleb	floating point scale
>   ;; fsqrt	floating point square root
>   ;; frsqrt       floating point reciprocal square root
>   ;; multi	multiword sequence (or user asm statements)
> @@ -231,8 +233,8 @@ (define_attr "type"
>     "unknown,branch,jump,call,load,fpload,fpidxload,store,fpstore,fpidxstore,
>      prefetch,prefetchx,condmove,mgtf,mftg,const,arith,logical,
>      shift,slt,signext,clz,trap,imul,idiv,move,
> -   fmove,fadd,fmul,fmadd,fdiv,frdiv,fabs,fneg,fcmp,fcopysign,fcvt,fsqrt,
> -   frsqrt,accext,accmod,multi,atomic,syncloop,nop,ghost"
> +   fmove,fadd,fmul,fmadd,fdiv,frdiv,fabs,fneg,fcmp,fcopysign,fcvt,fscaleb,
> +   fsqrt,frsqrt,accext,accmod,multi,atomic,syncloop,nop,ghost"
>     (cond [(eq_attr "jirl" "!unset") (const_string "call")
>   	 (eq_attr "got" "load") (const_string "load")
>   
> @@ -418,6 +420,10 @@ (define_mode_attr UNITMODE [(SF "SF") (DF "DF")])
>   ;; the controlling mode.
>   (define_mode_attr HALFMODE [(DF "SI") (DI "SI") (TF "DI")])
>   
> +;; This attribute gives the integer mode that has the same size of a
> +;; floating-point mode.
> +(define_mode_attr IMODE [(SF "SI") (DF "DI")])
> +
>   ;; This code iterator allows signed and unsigned widening multiplications
>   ;; to use the same template.
>   (define_code_iterator any_extend [sign_extend zero_extend])
> @@ -1014,7 +1020,23 @@ (define_insn "copysign<mode>3"
>     "fcopysign.<fmt>\t%0,%1,%2"
>     [(set_attr "type" "fcopysign")
>      (set_attr "mode" "<UNITMODE>")])
> +\f
> +;;
> +;;  ....................
> +;;
> +;;	FLOATING POINT SCALE
> +;;
> +;;  ....................
>   
> +(define_insn "ldexp<mode>3"
> +  [(set (match_operand:ANYF 0 "register_operand" "=f")
> +	(unspec:ANYF [(match_operand:ANYF    1 "register_operand" "f")
> +		      (match_operand:<IMODE> 2 "register_operand" "f")]
> +		     UNSPEC_FSCALEB))]
> +  "TARGET_HARD_FLOAT"
> +  "fscaleb.<fmt>\t%0,%1,%2"
> +  [(set_attr "type" "fscaleb")
> +   (set_attr "mode" "<UNITMODE>")])
>   \f
>   ;;
>   ;;  ...................
> diff --git a/gcc/testsuite/gcc.target/loongarch/fscaleb.c b/gcc/testsuite/gcc.target/loongarch/fscaleb.c
> new file mode 100644
> index 00000000000..f18470fbb8f
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/loongarch/fscaleb.c
> @@ -0,0 +1,48 @@
> +/* { dg-do compile } */
> +/* { dg-options "-O2 -mabi=lp64d -mdouble-float -fno-math-errno" } */
> +/* { dg-final { scan-assembler-times "fscaleb\\.s" 3 } } */
> +/* { dg-final { scan-assembler-times "fscaleb\\.d" 4 } } */
> +/* { dg-final { scan-assembler-times "slli\\.w" 1 } } */
> +
> +double
> +my_scalbln (double a, long b)
> +{
> +  return __builtin_scalbln (a, b);
> +}
> +
> +double
> +my_scalbn (double a, int b)
> +{
> +  return __builtin_scalbn (a, b);
> +}
> +
> +
> +float
> +my_scalblnf (float a, long b)
> +{
> +  return __builtin_scalblnf (a, b);
> +}
> +
> +float
> +my_scalbnf (float a, int b)
> +{
> +  return __builtin_scalbnf (a, b);
> +}
> +
>
I think scalbln/scalblnf/scalbn/scalbnf these four builtin test function 
with the macro __FLT_RADIX__ control.

These functions are tested only if the macro __FLT_RADIX__ has a value of 2.

Re: [PATCH v2 3/4] LoongArch: Add fscaleb.{s,d} instructions as ldexp{sf,df}3

[Xi Ruoyao]

On Sat, 2022-11-12 at 11:54 +0800, Lulu Cheng wrote:
> 
> 在 2022/11/9 下午9:53, Xi Ruoyao 写道:
> > This allows optimizing __builtin_ldexp{,f} and __builtin_scalbn{,f}
> > with
> > -fno-math-errno.
> > 
> > IMODE is added because we can't hard code SI for operand 2:
> > fscaleb.d
> > instruction always take the high half of both source registers into
> > account.  See my_ldexp_long in the test case.
> > 
> > gcc/ChangeLog:
> > 
> >         * config/loongarch/loongarch.md (UNSPEC_FSCALEB): New
> > unspec.
> >         (type): Add fscaleb.
> >         (IMODE): New mode attr.
> >         (ldexp<mode>3): New instruction template.
> > 
> > gcc/testsuite/ChangeLog:
> > 
> >         * gcc.target/loongarch/fscaleb.c: New test.
> > ---
> >   gcc/config/loongarch/loongarch.md            | 26 ++++++++++-
> >   gcc/testsuite/gcc.target/loongarch/fscaleb.c | 48
> > ++++++++++++++++++++
> >   2 files changed, 72 insertions(+), 2 deletions(-)
> >   create mode 100644 gcc/testsuite/gcc.target/loongarch/fscaleb.c
> > 
> > diff --git a/gcc/config/loongarch/loongarch.md
> > b/gcc/config/loongarch/loongarch.md
> > index eb127c346a3..c141c9adde2 100644
> > --- a/gcc/config/loongarch/loongarch.md
> > +++ b/gcc/config/loongarch/loongarch.md
> > @@ -41,6 +41,7 @@ (define_c_enum "unspec" [
> >     UNSPEC_FTINT
> >     UNSPEC_FTINTRM
> >     UNSPEC_FTINTRP
> > +  UNSPEC_FSCALEB
> >   
> >     ;; Override return address for exception handling.
> >     UNSPEC_EH_RETURN
> > @@ -220,6 +221,7 @@ (define_attr "qword_mode" "no,yes"
> >   ;; fcmp               floating point compare
> >   ;; fcopysign  floating point copysign
> >   ;; fcvt               floating point convert
> > +;; fscaleb     floating point scale
> >   ;; fsqrt      floating point square root
> >   ;; frsqrt       floating point reciprocal square root
> >   ;; multi      multiword sequence (or user asm statements)
> > @@ -231,8 +233,8 @@ (define_attr "type"
> >    
> > "unknown,branch,jump,call,load,fpload,fpidxload,store,fpstore,fpidxs
> > tore,
> >      prefetch,prefetchx,condmove,mgtf,mftg,const,arith,logical,
> >      shift,slt,signext,clz,trap,imul,idiv,move,
> > -  
> > fmove,fadd,fmul,fmadd,fdiv,frdiv,fabs,fneg,fcmp,fcopysign,fcvt,fsqrt
> > ,
> > -   frsqrt,accext,accmod,multi,atomic,syncloop,nop,ghost"
> > +  
> > fmove,fadd,fmul,fmadd,fdiv,frdiv,fabs,fneg,fcmp,fcopysign,fcvt,fscal
> > eb,
> > +   fsqrt,frsqrt,accext,accmod,multi,atomic,syncloop,nop,ghost"
> >     (cond [(eq_attr "jirl" "!unset") (const_string "call")
> >          (eq_attr "got" "load") (const_string "load")
> >   
> > @@ -418,6 +420,10 @@ (define_mode_attr UNITMODE [(SF "SF") (DF
> > "DF")])
> >   ;; the controlling mode.
> >   (define_mode_attr HALFMODE [(DF "SI") (DI "SI") (TF "DI")])
> >   
> > +;; This attribute gives the integer mode that has the same size of
> > a
> > +;; floating-point mode.
> > +(define_mode_attr IMODE [(SF "SI") (DF "DI")])
> > +
> >   ;; This code iterator allows signed and unsigned widening
> > multiplications
> >   ;; to use the same template.
> >   (define_code_iterator any_extend [sign_extend zero_extend])
> > @@ -1014,7 +1020,23 @@ (define_insn "copysign<mode>3"
> >     "fcopysign.<fmt>\t%0,%1,%2"
> >     [(set_attr "type" "fcopysign")
> >      (set_attr "mode" "<UNITMODE>")])
> > +\f
> > +;;
> > +;;  ....................
> > +;;
> > +;;     FLOATING POINT SCALE
> > +;;
> > +;;  ....................
> >   
> > +(define_insn "ldexp<mode>3"
> > +  [(set (match_operand:ANYF 0 "register_operand" "=f")
> > +       (unspec:ANYF [(match_operand:ANYF    1 "register_operand"
> > "f")
> > +                     (match_operand:<IMODE> 2 "register_operand"
> > "f")]
> > +                    UNSPEC_FSCALEB))]
> > +  "TARGET_HARD_FLOAT"
> > +  "fscaleb.<fmt>\t%0,%1,%2"
> > +  [(set_attr "type" "fscaleb")
> > +   (set_attr "mode" "<UNITMODE>")])
> >   \f
> >   ;;
> >   ;;  ...................
> > diff --git a/gcc/testsuite/gcc.target/loongarch/fscaleb.c
> > b/gcc/testsuite/gcc.target/loongarch/fscaleb.c
> > new file mode 100644
> > index 00000000000..f18470fbb8f
> > --- /dev/null
> > +++ b/gcc/testsuite/gcc.target/loongarch/fscaleb.c
> > @@ -0,0 +1,48 @@
> > +/* { dg-do compile } */
> > +/* { dg-options "-O2 -mabi=lp64d -mdouble-float -fno-math-errno" }
> > */
> > +/* { dg-final { scan-assembler-times "fscaleb\\.s" 3 } } */
> > +/* { dg-final { scan-assembler-times "fscaleb\\.d" 4 } } */
> > +/* { dg-final { scan-assembler-times "slli\\.w" 1 } } */
> > +
> > +double
> > +my_scalbln (double a, long b)
> > +{
> > +  return __builtin_scalbln (a, b);
> > +}
> > +
> > +double
> > +my_scalbn (double a, int b)
> > +{
> > +  return __builtin_scalbn (a, b);
> > +}
> > +
> > +
> > +float
> > +my_scalblnf (float a, long b)
> > +{
> > +  return __builtin_scalblnf (a, b);
> > +}
> > +
> > +float
> > +my_scalbnf (float a, int b)
> > +{
> > +  return __builtin_scalbnf (a, b);
> > +}
> > +
> > 
> I think scalbln/scalblnf/scalbn/scalbnf these four builtin test
> function 
> with the macro __FLT_RADIX__ control.
> 
> These functions are tested only if the macro __FLT_RADIX__ has a value
> of 2.

LoongArch does not use RESET_FLOAT_FORMAT on SFmode, so __FLT_RADIX__ is
always 2.

-- 
Xi Ruoyao <xry111@xry111.site>
School of Aerospace Science and Technology, Xidian University

Re: [PATCH v2 4/4] LoongArch: Add flogb.{s,d} instructions and expand logb{sf,df}2

[Lulu Cheng]

LGTM!

Thanks.

在 2022/11/9 下午9:53, Xi Ruoyao 写道:
> On LoongArch, flogb instructions extract the exponent of a non-negative
> floating point value, but produces NaN for negative values.  So we need
> to add a fabs instruction when we expand logb.
>
> gcc/ChangeLog:
>
> 	* config/loongarch/loongarch.md (UNSPEC_FLOGB): New unspec.
> 	(type): Add flogb.
> 	(logb_non_negative<mode>2): New instruction template.
> 	(logb<mode>2): New define_expand.
>
> gcc/testsuite/ChangeLog:
>
> 	* gcc.target/loongarch/flogb.c: New test.
> ---
>   gcc/config/loongarch/loongarch.md          | 35 ++++++++++++++++++++--
>   gcc/testsuite/gcc.target/loongarch/flogb.c | 18 +++++++++++
>   2 files changed, 51 insertions(+), 2 deletions(-)
>   create mode 100644 gcc/testsuite/gcc.target/loongarch/flogb.c
>
> diff --git a/gcc/config/loongarch/loongarch.md b/gcc/config/loongarch/loongarch.md
> index c141c9adde2..682ab961741 100644
> --- a/gcc/config/loongarch/loongarch.md
> +++ b/gcc/config/loongarch/loongarch.md
> @@ -42,6 +42,7 @@ (define_c_enum "unspec" [
>     UNSPEC_FTINTRM
>     UNSPEC_FTINTRP
>     UNSPEC_FSCALEB
> +  UNSPEC_FLOGB
>   
>     ;; Override return address for exception handling.
>     UNSPEC_EH_RETURN
> @@ -217,6 +218,7 @@ (define_attr "qword_mode" "no,yes"
>   ;; fdiv		floating point divide
>   ;; frdiv	floating point reciprocal divide
>   ;; fabs		floating point absolute value
> +;; flogb	floating point exponent extract
>   ;; fneg		floating point negation
>   ;; fcmp		floating point compare
>   ;; fcopysign	floating point copysign
> @@ -233,8 +235,8 @@ (define_attr "type"
>     "unknown,branch,jump,call,load,fpload,fpidxload,store,fpstore,fpidxstore,
>      prefetch,prefetchx,condmove,mgtf,mftg,const,arith,logical,
>      shift,slt,signext,clz,trap,imul,idiv,move,
> -   fmove,fadd,fmul,fmadd,fdiv,frdiv,fabs,fneg,fcmp,fcopysign,fcvt,fscaleb,
> -   fsqrt,frsqrt,accext,accmod,multi,atomic,syncloop,nop,ghost"
> +   fmove,fadd,fmul,fmadd,fdiv,frdiv,fabs,flogb,fneg,fcmp,fcopysign,fcvt,
> +   fscaleb,fsqrt,frsqrt,accext,accmod,multi,atomic,syncloop,nop,ghost"
>     (cond [(eq_attr "jirl" "!unset") (const_string "call")
>   	 (eq_attr "got" "load") (const_string "load")
>   
> @@ -1039,6 +1041,35 @@ (define_insn "ldexp<mode>3"
>      (set_attr "mode" "<UNITMODE>")])
>   \f
>   ;;
> +;;  ....................
> +;;
> +;;	FLOATING POINT EXPONENT EXTRACT
> +;;
> +;;  ....................
> +
> +(define_insn "logb_non_negative<mode>2"
> +  [(set (match_operand:ANYF 0 "register_operand" "=f")
> +	(unspec:ANYF [(match_operand:ANYF 1 "register_operand" "f")]
> +		     UNSPEC_FLOGB))]
> +  "TARGET_HARD_FLOAT"
> +  "flogb.<fmt>\t%0,%1"
> +  [(set_attr "type" "flogb")
> +   (set_attr "mode" "<UNITMODE>")])
> +
> +(define_expand "logb<mode>2"
> +  [(set (match_operand:ANYF 0 "register_operand")
> +	(unspec:ANYF [(abs:ANYF (match_operand:ANYF 1 "register_operand"))]
> +		     UNSPEC_FLOGB))]
> +  "TARGET_HARD_FLOAT"
> +{
> +  rtx tmp = gen_reg_rtx (<MODE>mode);
> +
> +  emit_insn (gen_abs<mode>2 (tmp, operands[1]));
> +  emit_insn (gen_logb_non_negative<mode>2 (operands[0], tmp));
> +  DONE;
> +})
> +\f
> +;;
>   ;;  ...................
>   ;;
>   ;;  Count leading zeroes.
> diff --git a/gcc/testsuite/gcc.target/loongarch/flogb.c b/gcc/testsuite/gcc.target/loongarch/flogb.c
> new file mode 100644
> index 00000000000..1daefe54e13
> --- /dev/null
> +++ b/gcc/testsuite/gcc.target/loongarch/flogb.c
> @@ -0,0 +1,18 @@
> +/* { dg-do compile } */
> +/* { dg-options "-mdouble-float -fno-math-errno" } */
> +/* { dg-final { scan-assembler "fabs\\.s" } } */
> +/* { dg-final { scan-assembler "fabs\\.d" } } */
> +/* { dg-final { scan-assembler "flogb\\.s" } } */
> +/* { dg-final { scan-assembler "flogb\\.d" } } */
> +
> +double
> +my_logb (double a)
> +{
> +  return __builtin_logb (a);
> +}
> +
> +float
> +my_logbf (float a)
> +{
> +  return __builtin_logbf (a);
> +}

Re: [PATCH v2 3/4] LoongArch: Add fscaleb.{s,d} instructions as ldexp{sf,df}3

[Lulu Cheng]

在 2022/11/12 下午12:40, Xi Ruoyao 写道:
> On Sat, 2022-11-12 at 11:54 +0800, Lulu Cheng wrote:
>> 在 2022/11/9 下午9:53, Xi Ruoyao 写道:
>>> This allows optimizing __builtin_ldexp{,f} and __builtin_scalbn{,f}
>>> with
>>> -fno-math-errno.
>>>
>>> IMODE is added because we can't hard code SI for operand 2:
>>> fscaleb.d
>>> instruction always take the high half of both source registers into
>>> account.  See my_ldexp_long in the test case.
>>>
>>> gcc/ChangeLog:
>>>
>>>          * config/loongarch/loongarch.md (UNSPEC_FSCALEB): New
>>> unspec.
>>>          (type): Add fscaleb.
>>>          (IMODE): New mode attr.
>>>          (ldexp<mode>3): New instruction template.
>>>
>>> gcc/testsuite/ChangeLog:
>>>
>>>          * gcc.target/loongarch/fscaleb.c: New test.
>>> ---
>>>    gcc/config/loongarch/loongarch.md            | 26 ++++++++++-
>>>    gcc/testsuite/gcc.target/loongarch/fscaleb.c | 48
>>> ++++++++++++++++++++
>>>    2 files changed, 72 insertions(+), 2 deletions(-)
>>>    create mode 100644 gcc/testsuite/gcc.target/loongarch/fscaleb.c
>>>
>>> diff --git a/gcc/config/loongarch/loongarch.md
>>> b/gcc/config/loongarch/loongarch.md
>>> index eb127c346a3..c141c9adde2 100644
>>> --- a/gcc/config/loongarch/loongarch.md
>>> +++ b/gcc/config/loongarch/loongarch.md
>>> @@ -41,6 +41,7 @@ (define_c_enum "unspec" [
>>>      UNSPEC_FTINT
>>>      UNSPEC_FTINTRM
>>>      UNSPEC_FTINTRP
>>> +  UNSPEC_FSCALEB
>>>    
>>>      ;; Override return address for exception handling.
>>>      UNSPEC_EH_RETURN
>>> @@ -220,6 +221,7 @@ (define_attr "qword_mode" "no,yes"
>>>    ;; fcmp               floating point compare
>>>    ;; fcopysign  floating point copysign
>>>    ;; fcvt               floating point convert
>>> +;; fscaleb     floating point scale
>>>    ;; fsqrt      floating point square root
>>>    ;; frsqrt       floating point reciprocal square root
>>>    ;; multi      multiword sequence (or user asm statements)
>>> @@ -231,8 +233,8 @@ (define_attr "type"
>>>     
>>> "unknown,branch,jump,call,load,fpload,fpidxload,store,fpstore,fpidxs
>>> tore,
>>>       prefetch,prefetchx,condmove,mgtf,mftg,const,arith,logical,
>>>       shift,slt,signext,clz,trap,imul,idiv,move,
>>> -
>>> fmove,fadd,fmul,fmadd,fdiv,frdiv,fabs,fneg,fcmp,fcopysign,fcvt,fsqrt
>>> ,
>>> -   frsqrt,accext,accmod,multi,atomic,syncloop,nop,ghost"
>>> +
>>> fmove,fadd,fmul,fmadd,fdiv,frdiv,fabs,fneg,fcmp,fcopysign,fcvt,fscal
>>> eb,
>>> +   fsqrt,frsqrt,accext,accmod,multi,atomic,syncloop,nop,ghost"
>>>      (cond [(eq_attr "jirl" "!unset") (const_string "call")
>>>           (eq_attr "got" "load") (const_string "load")
>>>    
>>> @@ -418,6 +420,10 @@ (define_mode_attr UNITMODE [(SF "SF") (DF
>>> "DF")])
>>>    ;; the controlling mode.
>>>    (define_mode_attr HALFMODE [(DF "SI") (DI "SI") (TF "DI")])
>>>    
>>> +;; This attribute gives the integer mode that has the same size of
>>> a
>>> +;; floating-point mode.
>>> +(define_mode_attr IMODE [(SF "SI") (DF "DI")])
>>> +
>>>    ;; This code iterator allows signed and unsigned widening
>>> multiplications
>>>    ;; to use the same template.
>>>    (define_code_iterator any_extend [sign_extend zero_extend])
>>> @@ -1014,7 +1020,23 @@ (define_insn "copysign<mode>3"
>>>      "fcopysign.<fmt>\t%0,%1,%2"
>>>      [(set_attr "type" "fcopysign")
>>>       (set_attr "mode" "<UNITMODE>")])
>>> +\f
>>> +;;
>>> +;;  ....................
>>> +;;
>>> +;;     FLOATING POINT SCALE
>>> +;;
>>> +;;  ....................
>>>    
>>> +(define_insn "ldexp<mode>3"
>>> +  [(set (match_operand:ANYF 0 "register_operand" "=f")
>>> +       (unspec:ANYF [(match_operand:ANYF    1 "register_operand"
>>> "f")
>>> +                     (match_operand:<IMODE> 2 "register_operand"
>>> "f")]
>>> +                    UNSPEC_FSCALEB))]
>>> +  "TARGET_HARD_FLOAT"
>>> +  "fscaleb.<fmt>\t%0,%1,%2"
>>> +  [(set_attr "type" "fscaleb")
>>> +   (set_attr "mode" "<UNITMODE>")])
>>>    \f
>>>    ;;
>>>    ;;  ...................
>>> diff --git a/gcc/testsuite/gcc.target/loongarch/fscaleb.c
>>> b/gcc/testsuite/gcc.target/loongarch/fscaleb.c
>>> new file mode 100644
>>> index 00000000000..f18470fbb8f
>>> --- /dev/null
>>> +++ b/gcc/testsuite/gcc.target/loongarch/fscaleb.c
>>> @@ -0,0 +1,48 @@
>>> +/* { dg-do compile } */
>>> +/* { dg-options "-O2 -mabi=lp64d -mdouble-float -fno-math-errno" }
>>> */
>>> +/* { dg-final { scan-assembler-times "fscaleb\\.s" 3 } } */
>>> +/* { dg-final { scan-assembler-times "fscaleb\\.d" 4 } } */
>>> +/* { dg-final { scan-assembler-times "slli\\.w" 1 } } */
>>> +
>>> +double
>>> +my_scalbln (double a, long b)
>>> +{
>>> +  return __builtin_scalbln (a, b);
>>> +}
>>> +
>>> +double
>>> +my_scalbn (double a, int b)
>>> +{
>>> +  return __builtin_scalbn (a, b);
>>> +}
>>> +
>>> +
>>> +float
>>> +my_scalblnf (float a, long b)
>>> +{
>>> +  return __builtin_scalblnf (a, b);
>>> +}
>>> +
>>> +float
>>> +my_scalbnf (float a, int b)
>>> +{
>>> +  return __builtin_scalbnf (a, b);
>>> +}
>>> +
>>>
>> I think scalbln/scalblnf/scalbn/scalbnf these four builtin test
>> function
>> with the macro __FLT_RADIX__ control.
>>
>> These functions are tested only if the macro __FLT_RADIX__ has a value
>> of 2.
> LoongArch does not use RESET_FLOAT_FORMAT on SFmode, so __FLT_RADIX__ is
> always 2.
Ok, I have no more questions

Re: [PATCH v2 0/4] LoongArch: Add some floating-point operations

[Xi Ruoyao]

On Wed, 2022-11-09 at 21:53 +0800, Xi Ruoyao wrote:
> These patches allow to expand the following builtins to floating point
> instructions for LoongArch:
> 
> - __builtin_rint{,f}
> - __builtin_{l,ll}rint{,f}
> - __builtin_{l,ll}floor{,f}
> - __builtin_{l,ll}ceil{,f}
> - __builtin_scalb{n,ln}{,f}
> - __builtin_logb{,f}
> 
> Bootstrapped and regtested on loongarch64-linux-gnu.  And a modified
> Glibc using the builtins for rint{,f}, {l,ll}rint{,f}, and logb{,f}
> also survived Glibc test suite.
> 
> Please review ASAP because GCC 13 stage 1 will end on Nov. 13th.
> 
> v1 -> v2: Only use ftint{rm,rp} instructions if floor and ceil are
> allowed to raise inexact exception.
> 
> Xi Ruoyao (4):
>   LoongArch: Rename frint_<fmt> to rint<mode>2
>   LoongArch: Add ftint{,rm,rp}.{w,l}.{s,d} instructions
>   LoongArch: Add fscaleb.{s,d} instructions as ldexp{sf,df}3
>   LoongArch: Add flogb.{s,d} instructions and expand logb{sf,df}2
> 
>  gcc/config/loongarch/loongarch.md             | 95
> ++++++++++++++++++-
>  gcc/testsuite/gcc.target/loongarch/flogb.c    | 18 ++++
>  gcc/testsuite/gcc.target/loongarch/frint.c    | 16 ++++
>  gcc/testsuite/gcc.target/loongarch/fscaleb.c  | 48 ++++++++++
>  .../gcc.target/loongarch/ftint-no-inexact.c   | 44 +++++++++
>  gcc/testsuite/gcc.target/loongarch/ftint.c    | 44 +++++++++
>  6 files changed, 261 insertions(+), 4 deletions(-)
>  create mode 100644 gcc/testsuite/gcc.target/loongarch/flogb.c
>  create mode 100644 gcc/testsuite/gcc.target/loongarch/frint.c
>  create mode 100644 gcc/testsuite/gcc.target/loongarch/fscaleb.c
>  create mode 100644 gcc/testsuite/gcc.target/loongarch/ftint-no-
> inexact.c
>  create mode 100644 gcc/testsuite/gcc.target/loongarch/ftint.c
> 

Pushed r13-3922.

I'll be busy in the following week.  Will do the work on Glibc side
after Nov. 20.

-- 
Xi Ruoyao <xry111@xry111.site>
School of Aerospace Science and Technology, Xidian University

Re: [PATCH v2 0/4] LoongArch: Add some floating-point operations

[Lulu Cheng]

在 2022/11/12 下午3:08, Xi Ruoyao 写道:
> On Wed, 2022-11-09 at 21:53 +0800, Xi Ruoyao wrote:
>> These patches allow to expand the following builtins to floating point
>> instructions for LoongArch:
>>
>> - __builtin_rint{,f}
>> - __builtin_{l,ll}rint{,f}
>> - __builtin_{l,ll}floor{,f}
>> - __builtin_{l,ll}ceil{,f}
>> - __builtin_scalb{n,ln}{,f}
>> - __builtin_logb{,f}
>>
>> Bootstrapped and regtested on loongarch64-linux-gnu.  And a modified
>> Glibc using the builtins for rint{,f}, {l,ll}rint{,f}, and logb{,f}
>> also survived Glibc test suite.
>>
>> Please review ASAP because GCC 13 stage 1 will end on Nov. 13th.
>>
>> v1 -> v2: Only use ftint{rm,rp} instructions if floor and ceil are
>> allowed to raise inexact exception.
>>
>> Xi Ruoyao (4):
>>    LoongArch: Rename frint_<fmt> to rint<mode>2
>>    LoongArch: Add ftint{,rm,rp}.{w,l}.{s,d} instructions
>>    LoongArch: Add fscaleb.{s,d} instructions as ldexp{sf,df}3
>>    LoongArch: Add flogb.{s,d} instructions and expand logb{sf,df}2
>>
>>   gcc/config/loongarch/loongarch.md             | 95
>> ++++++++++++++++++-
>>   gcc/testsuite/gcc.target/loongarch/flogb.c    | 18 ++++
>>   gcc/testsuite/gcc.target/loongarch/frint.c    | 16 ++++
>>   gcc/testsuite/gcc.target/loongarch/fscaleb.c  | 48 ++++++++++
>>   .../gcc.target/loongarch/ftint-no-inexact.c   | 44 +++++++++
>>   gcc/testsuite/gcc.target/loongarch/ftint.c    | 44 +++++++++
>>   6 files changed, 261 insertions(+), 4 deletions(-)
>>   create mode 100644 gcc/testsuite/gcc.target/loongarch/flogb.c
>>   create mode 100644 gcc/testsuite/gcc.target/loongarch/frint.c
>>   create mode 100644 gcc/testsuite/gcc.target/loongarch/fscaleb.c
>>   create mode 100644 gcc/testsuite/gcc.target/loongarch/ftint-no-
>> inexact.c
>>   create mode 100644 gcc/testsuite/gcc.target/loongarch/ftint.c
>>
> Pushed r13-3922.
>
> I'll be busy in the following week.  Will do the work on Glibc side
> after Nov. 20.
>
I will send the patch of prefetch later, please help to check it:-)

Re: [PATCH v2 0/4] LoongArch: Add some floating-point operations

[tangxiaolin]

How about I do this work on Glibc?



在 2022/11/12 下午3:08, Xi Ruoyao 写道:
> On Wed, 2022-11-09 at 21:53 +0800, Xi Ruoyao wrote:
>> These patches allow to expand the following builtins to floating point
>> instructions for LoongArch:
>>
>> - __builtin_rint{,f}
>> - __builtin_{l,ll}rint{,f}
>> - __builtin_{l,ll}floor{,f}
>> - __builtin_{l,ll}ceil{,f}
>> - __builtin_scalb{n,ln}{,f}
>> - __builtin_logb{,f}
>>
>> Bootstrapped and regtested on loongarch64-linux-gnu.  And a modified
>> Glibc using the builtins for rint{,f}, {l,ll}rint{,f}, and logb{,f}
>> also survived Glibc test suite.
>>
>> Please review ASAP because GCC 13 stage 1 will end on Nov. 13th.
>>
>> v1 -> v2: Only use ftint{rm,rp} instructions if floor and ceil are
>> allowed to raise inexact exception.
>>
>> Xi Ruoyao (4):
>>    LoongArch: Rename frint_<fmt> to rint<mode>2
>>    LoongArch: Add ftint{,rm,rp}.{w,l}.{s,d} instructions
>>    LoongArch: Add fscaleb.{s,d} instructions as ldexp{sf,df}3
>>    LoongArch: Add flogb.{s,d} instructions and expand logb{sf,df}2
>>
>>   gcc/config/loongarch/loongarch.md             | 95
>> ++++++++++++++++++-
>>   gcc/testsuite/gcc.target/loongarch/flogb.c    | 18 ++++
>>   gcc/testsuite/gcc.target/loongarch/frint.c    | 16 ++++
>>   gcc/testsuite/gcc.target/loongarch/fscaleb.c  | 48 ++++++++++
>>   .../gcc.target/loongarch/ftint-no-inexact.c   | 44 +++++++++
>>   gcc/testsuite/gcc.target/loongarch/ftint.c    | 44 +++++++++
>>   6 files changed, 261 insertions(+), 4 deletions(-)
>>   create mode 100644 gcc/testsuite/gcc.target/loongarch/flogb.c
>>   create mode 100644 gcc/testsuite/gcc.target/loongarch/frint.c
>>   create mode 100644 gcc/testsuite/gcc.target/loongarch/fscaleb.c
>>   create mode 100644 gcc/testsuite/gcc.target/loongarch/ftint-no-
>> inexact.c
>>   create mode 100644 gcc/testsuite/gcc.target/loongarch/ftint.c
>>
> Pushed r13-3922.
>
> I'll be busy in the following week.  Will do the work on Glibc side
> after Nov. 20.
>