[dev.simd] all: merge master (9b2d39b) into dev.simd

Conflicts:

- src/internal/buildcfg/exp.go

Merge List:

+ 2025-09-22 9b2d39b75b cmd/compile/internal/ssa: match style and formatting
+ 2025-09-22 e23edf5e55 runtime: don't re-read metrics before check in TestReadMetricsSched
+ 2025-09-22 177cd8d763 log/slog: use a pooled json encoder
+ 2025-09-22 2353c15785 cmd/cgo/internal/test: skip TestMultipleAssign when using UCRT on Windows
+ 2025-09-22 32dfd69282 cmd/dist: disable FIPS 140-3 mode when testing maphash with purego
+ 2025-09-19 7f6ff5ec3e cmd/compile: fix doc word
+ 2025-09-19 9693b94be0 runtime: include stderr when objdump fails
+ 2025-09-19 8616981ce6 log/slog: optimize slog Level.String() to avoid fmt.Sprintf
+ 2025-09-19 b8af744360 testing: fix example for unexported identifier
+ 2025-09-19 51dc5bfe6c Revert "cmd/go: disable cgo by default if CC unset and  DefaultCC doesn't exist"
+ 2025-09-19 ee7bf06cb3 time: improve ParseDuration performance for invalid input
+ 2025-09-19 f9e61a9a32 cmd/compile: duplicate nil check to two branches of write barrier
+ 2025-09-18 3cf1aaf8b9 runtime: use futexes with 64-bit time on Linux
+ 2025-09-18 0ab038af62 cmd/compile/internal/abi: use clear built-in
+ 2025-09-18 00bf24fdca bytes: use clear in test
+ 2025-09-18 f9701d21d2 crypto: use clear built-in
+ 2025-09-18 a58afe44fa net: fix testHookCanceledDial race
+ 2025-09-18 3203a5da29 net/http: avoid connCount underflow race
+ 2025-09-18 8ca209ec39 context: don't return a non-nil from Err before Done is closed
+ 2025-09-18 3032894e04 runtime: make explicit nil check in heapSetTypeSmallHeader
+ 2025-09-17 ef05b66d61 cmd/internal/obj/riscv: add support for Zicond instructions
+ 2025-09-17 78ef487a6f cmd/compile: fix the issue of shift amount exceeding the valid range
+ 2025-09-17 77aac7bb75 runtime: don't enable heap randomization if MSAN or ASAN is enabled
+ 2025-09-17 465b85eb76 runtime: fix CheckScavengedBitsCleared with randomized heap base
+ 2025-09-17 909704b85e encoding/json/v2: fix typo in comment
+ 2025-09-17 3db5979e8c testing: use reflect.TypeAssert and reflect.TypeFor
+ 2025-09-17 6a8dbbecbf path/filepath: fix EvalSymlinks to return ENOTDIR on plan9
+ 2025-09-17 bffe7ad9f1 go/parser: Add TestBothLineAndLeadComment
+ 2025-09-17 02a888e820 go/ast: document that (*ast.File).Comments is sorted by position
+ 2025-09-16 594deca981 cmd/link: simplify PE relocations mapping
+ 2025-09-16 9df1a289ac go/parser: simplify expectSemi
+ 2025-09-16 72ba117bda internal/buildcfg: enable randomizedHeapBase64 by default
+ 2025-09-16 796ea3bc2e os/user: align test file name and build tags
+ 2025-09-16 a69395eab2 runtime/_mkmalloc: add a copy of cloneNode
+ 2025-09-16 cbdad4fc3c cmd/go: check pattern for utf8 validity before call regexp.MustCompile
+ 2025-09-16 c2d85eb999 cmd/go: disable cgo by default if CC unset and  DefaultCC doesn't exist
+ 2025-09-16 ac82fe68aa bytes,strings: remove reference to non-existent SplitFunc
+ 2025-09-16 0b26678db2 cmd/compile: fix mips zerorange implementation
+ 2025-09-16 e2cfc1eb3a cmd/internal/obj/riscv: improve handling of float point moves
+ 2025-09-16 281c632e6e crypto/x509/internal/macos: standardize package name
+ 2025-09-16 61dc7fe30d iter: document that calling yield after terminated range loop causes runtime panic

Change-Id: Ic06019efc855913632003f41eb10c746b3410b0a

8 files changed, 84 insertions, 22 deletions

diff --git a/src/cmd/compile/internal/abi/abiutils.go b/src/cmd/compile/internal/abi/abiutils.go
index cef7885815..7acab36e8d 100644
--- a/src/cmd/compile/internal/abi/abiutils.go
+++ b/src/cmd/compile/internal/abi/abiutils.go
@@ -664,9 +664,7 @@ func (state *assignState) tryAllocRegs(typ *types.Type) []RegIndex {
 func (pa *ABIParamAssignment) ComputePadding(storage []uint64) []uint64 {
 	nr := len(pa.Registers)
 	padding := storage[:nr]
-	for i := 0; i < nr; i++ {
-		padding[i] = 0
-	}
+	clear(padding)
 	if pa.Type.Kind() != types.TSTRUCT || nr == 0 {
 		return padding
 	}
diff --git a/src/cmd/compile/internal/mips/ggen.go b/src/cmd/compile/internal/mips/ggen.go
index 023f4d958e..394f015589 100644
--- a/src/cmd/compile/internal/mips/ggen.go
+++ b/src/cmd/compile/internal/mips/ggen.go
@@ -5,6 +5,7 @@
 package mips
 
 import (
+	"cmd/compile/internal/base"
 	"cmd/compile/internal/objw"
 	"cmd/compile/internal/types"
 	"cmd/internal/obj"
@@ -17,7 +18,7 @@ func zerorange(pp *objw.Progs, p *obj.Prog, off, cnt int64, _ *uint32) *obj.Prog
 	}
 
 	for cnt != 0 {
-		p = pp.Append(p, mips.AMOVW, obj.TYPE_REG, mips.REGZERO, 0, obj.TYPE_MEM, mips.REGSP, off)
+		p = pp.Append(p, mips.AMOVW, obj.TYPE_REG, mips.REGZERO, 0, obj.TYPE_MEM, mips.REGSP, base.Ctxt.Arch.FixedFrameSize+off)
 		cnt -= int64(types.PtrSize)
 		off += int64(types.PtrSize)
 	}
diff --git a/src/cmd/compile/internal/mips64/ggen.go b/src/cmd/compile/internal/mips64/ggen.go
index 0740d9abe8..740d68e335 100644
--- a/src/cmd/compile/internal/mips64/ggen.go
+++ b/src/cmd/compile/internal/mips64/ggen.go
@@ -17,7 +17,7 @@ func zerorange(pp *objw.Progs, p *obj.Prog, off, cnt int64, _ *uint32) *obj.Prog
 	}
 
 	for cnt != 0 {
-		p = pp.Append(p, mips.AMOVV, obj.TYPE_REG, mips.REGZERO, 0, obj.TYPE_MEM, mips.REGSP, off)
+		p = pp.Append(p, mips.AMOVV, obj.TYPE_REG, mips.REGZERO, 0, obj.TYPE_MEM, mips.REGSP, off+8)
 		cnt -= int64(types.PtrSize)
 		off += int64(types.PtrSize)
 	}
diff --git a/src/cmd/compile/internal/ssa/README.md b/src/cmd/compile/internal/ssa/README.md
index 8184f9c002..3626f5bb7b 100644
--- a/src/cmd/compile/internal/ssa/README.md
+++ b/src/cmd/compile/internal/ssa/README.md
@@ -48,8 +48,7 @@ However, certain types don't come from Go and are special; below we will cover
 
 Some operators contain an auxiliary field. The aux fields are usually printed as
 enclosed in `[]` or `{}`, and could be the constant op argument, argument type,
-etc.
-for example:
+etc. For example:
 
 	v13 (?) = Const64 <int> [1]
 
diff --git a/src/cmd/compile/internal/ssa/_gen/LOONG64.rules b/src/cmd/compile/internal/ssa/_gen/LOONG64.rules
index d0a64364d6..0d2384143c 100644
--- a/src/cmd/compile/internal/ssa/_gen/LOONG64.rules
+++ b/src/cmd/compile/internal/ssa/_gen/LOONG64.rules
@@ -717,7 +717,8 @@
 (SRLVconst [rc] (MOVBUreg x)) && rc >= 8 => (MOVVconst [0])
 
 // (x + x) << c -> x << c+1
-((SLLV|SLL)const [c] (ADDV x x)) => ((SLLV|SLL)const [c+1] x)
+((SLLV|SLL)const <t> [c] (ADDV x x)) && c < t.Size() * 8 - 1  => ((SLLV|SLL)const [c+1] x)
+((SLLV|SLL)const <t> [c] (ADDV x x)) && c >= t.Size() * 8 - 1 => (MOVVconst [0])
 
 // mul by constant
 (MULV _ (MOVVconst [0])) => (MOVVconst [0])
diff --git a/src/cmd/compile/internal/ssa/_gen/LOONG64Ops.go b/src/cmd/compile/internal/ssa/_gen/LOONG64Ops.go
index 0d5e0eb76f..a3db4def56 100644
--- a/src/cmd/compile/internal/ssa/_gen/LOONG64Ops.go
+++ b/src/cmd/compile/internal/ssa/_gen/LOONG64Ops.go
@@ -247,7 +247,7 @@ func init() {
 		{name: "SLL", argLength: 2, reg: gp21, asm: "SLL"},                        // arg0 << arg1, shift amount is mod 32
 		{name: "SLLV", argLength: 2, reg: gp21, asm: "SLLV"},                      // arg0 << arg1, shift amount is mod 64
 		{name: "SLLconst", argLength: 1, reg: gp11, asm: "SLL", aux: "Int64"},     // arg0 << auxInt, auxInt should be in the range 0 to 31.
-		{name: "SLLVconst", argLength: 1, reg: gp11, asm: "SLLV", aux: "Int64"},   // arg0 << auxInt
+		{name: "SLLVconst", argLength: 1, reg: gp11, asm: "SLLV", aux: "Int64"},   // arg0 << auxInt, auxInt should be in the range 0 to 63.
 		{name: "SRL", argLength: 2, reg: gp21, asm: "SRL"},                        // arg0 >> arg1, shift amount is mod 32
 		{name: "SRLV", argLength: 2, reg: gp21, asm: "SRLV"},                      // arg0 >> arg1, unsigned, shift amount is mod 64
 		{name: "SRLconst", argLength: 1, reg: gp11, asm: "SRL", aux: "Int64"},     // arg0 >> auxInt, auxInt should be in the range 0 to 31.
diff --git a/src/cmd/compile/internal/ssa/rewriteLOONG64.go b/src/cmd/compile/internal/ssa/rewriteLOONG64.go
index 6a9b723c8c..3990b2833b 100644
--- a/src/cmd/compile/internal/ssa/rewriteLOONG64.go
+++ b/src/cmd/compile/internal/ssa/rewriteLOONG64.go
@@ -6561,15 +6561,17 @@ func rewriteValueLOONG64_OpLOONG64SLLV(v *Value) bool {
 }
 func rewriteValueLOONG64_OpLOONG64SLLVconst(v *Value) bool {
 	v_0 := v.Args[0]
-	// match: (SLLVconst [c] (ADDV x x))
+	// match: (SLLVconst <t> [c] (ADDV x x))
+	// cond: c < t.Size() * 8 - 1
 	// result: (SLLVconst [c+1] x)
 	for {
+		t := v.Type
 		c := auxIntToInt64(v.AuxInt)
 		if v_0.Op != OpLOONG64ADDV {
 			break
 		}
 		x := v_0.Args[1]
-		if x != v_0.Args[0] {
+		if x != v_0.Args[0] || !(c < t.Size()*8-1) {
 			break
 		}
 		v.reset(OpLOONG64SLLVconst)
@@ -6577,6 +6579,23 @@ func rewriteValueLOONG64_OpLOONG64SLLVconst(v *Value) bool {
 		v.AddArg(x)
 		return true
 	}
+	// match: (SLLVconst <t> [c] (ADDV x x))
+	// cond: c >= t.Size() * 8 - 1
+	// result: (MOVVconst [0])
+	for {
+		t := v.Type
+		c := auxIntToInt64(v.AuxInt)
+		if v_0.Op != OpLOONG64ADDV {
+			break
+		}
+		x := v_0.Args[1]
+		if x != v_0.Args[0] || !(c >= t.Size()*8-1) {
+			break
+		}
+		v.reset(OpLOONG64MOVVconst)
+		v.AuxInt = int64ToAuxInt(0)
+		return true
+	}
 	// match: (SLLVconst [c] (MOVVconst [d]))
 	// result: (MOVVconst [d<<uint64(c)])
 	for {
@@ -6593,15 +6612,17 @@ func rewriteValueLOONG64_OpLOONG64SLLVconst(v *Value) bool {
 }
 func rewriteValueLOONG64_OpLOONG64SLLconst(v *Value) bool {
 	v_0 := v.Args[0]
-	// match: (SLLconst [c] (ADDV x x))
+	// match: (SLLconst <t> [c] (ADDV x x))
+	// cond: c < t.Size() * 8 - 1
 	// result: (SLLconst [c+1] x)
 	for {
+		t := v.Type
 		c := auxIntToInt64(v.AuxInt)
 		if v_0.Op != OpLOONG64ADDV {
 			break
 		}
 		x := v_0.Args[1]
-		if x != v_0.Args[0] {
+		if x != v_0.Args[0] || !(c < t.Size()*8-1) {
 			break
 		}
 		v.reset(OpLOONG64SLLconst)
@@ -6609,6 +6630,23 @@ func rewriteValueLOONG64_OpLOONG64SLLconst(v *Value) bool {
 		v.AddArg(x)
 		return true
 	}
+	// match: (SLLconst <t> [c] (ADDV x x))
+	// cond: c >= t.Size() * 8 - 1
+	// result: (MOVVconst [0])
+	for {
+		t := v.Type
+		c := auxIntToInt64(v.AuxInt)
+		if v_0.Op != OpLOONG64ADDV {
+			break
+		}
+		x := v_0.Args[1]
+		if x != v_0.Args[0] || !(c >= t.Size()*8-1) {
+			break
+		}
+		v.reset(OpLOONG64MOVVconst)
+		v.AuxInt = int64ToAuxInt(0)
+		return true
+	}
 	return false
 }
 func rewriteValueLOONG64_OpLOONG64SRA(v *Value) bool {
diff --git a/src/cmd/compile/internal/ssa/writebarrier.go b/src/cmd/compile/internal/ssa/writebarrier.go
index 9ef3667d51..ec6901f13e 100644
--- a/src/cmd/compile/internal/ssa/writebarrier.go
+++ b/src/cmd/compile/internal/ssa/writebarrier.go
@@ -303,6 +303,15 @@ func writebarrier(f *Func) {
 		mem := stores[0].MemoryArg()
 		pos := stores[0].Pos
 
+		// If there is a nil check before the WB store, duplicate it to
+		// the two branches, where the store and the WB load occur. So
+		// they are more likely be removed by late nilcheck removal (which
+		// is block-local).
+		var nilcheck, nilcheckThen, nilcheckEnd *Value
+		if a := stores[0].Args[0]; a.Op == OpNilCheck && a.Args[1] == mem {
+			nilcheck = a
+		}
+
 		// If the source of a MoveWB is volatile (will be clobbered by a
 		// function call), we need to copy it to a temporary location, as
 		// marshaling the args of wbMove might clobber the value we're
@@ -377,6 +386,10 @@ func writebarrier(f *Func) {
 		// For each write barrier store, append write barrier code to bThen.
 		memThen := mem
 
+		if nilcheck != nil {
+			nilcheckThen = bThen.NewValue2(nilcheck.Pos, OpNilCheck, nilcheck.Type, nilcheck.Args[0], memThen)
+		}
+
 		// Note: we can issue the write barrier code in any order. In particular,
 		// it doesn't matter if they are in a different order *even if* they end
 		// up referring to overlapping memory regions. For instance if an OpStore
@@ -447,6 +460,9 @@ func writebarrier(f *Func) {
 				// take care of the vast majority of these. We could
 				// patch this up in the signal handler, or use XCHG to
 				// combine the read and the write.
+				if ptr == nilcheck {
+					ptr = nilcheckThen
+				}
 				oldVal := bThen.NewValue2(pos, OpLoad, types.Types[types.TUINTPTR], ptr, memThen)
 				// Save old value to write buffer.
 				addEntry(pos, oldVal)
@@ -459,9 +475,12 @@ func writebarrier(f *Func) {
 		// Now do the rare cases, Zeros and Moves.
 		for _, w := range stores {
 			pos := w.Pos
+			dst := w.Args[0]
+			if dst == nilcheck {
+				dst = nilcheckThen
+			}
 			switch w.Op {
 			case OpZeroWB:
-				dst := w.Args[0]
 				typ := reflectdata.TypeLinksym(w.Aux.(*types.Type))
 				// zeroWB(&typ, dst)
 				taddr := b.NewValue1A(pos, OpAddr, b.Func.Config.Types.Uintptr, typ, sb)
@@ -469,7 +488,6 @@ func writebarrier(f *Func) {
 				f.fe.Func().SetWBPos(pos)
 				nWBops--
 			case OpMoveWB:
-				dst := w.Args[0]
 				src := w.Args[1]
 				if isVolatile(src) {
 					for _, c := range volatiles {
@@ -491,24 +509,29 @@ func writebarrier(f *Func) {
 		// merge memory
 		mem = bEnd.NewValue2(pos, OpPhi, types.TypeMem, mem, memThen)
 
+		if nilcheck != nil {
+			nilcheckEnd = bEnd.NewValue2(nilcheck.Pos, OpNilCheck, nilcheck.Type, nilcheck.Args[0], mem)
+		}
+
 		// Do raw stores after merge point.
 		for _, w := range stores {
 			pos := w.Pos
+			dst := w.Args[0]
+			if dst == nilcheck {
+				dst = nilcheckEnd
+			}
 			switch w.Op {
 			case OpStoreWB:
-				ptr := w.Args[0]
 				val := w.Args[1]
 				if buildcfg.Experiment.CgoCheck2 {
 					// Issue cgo checking code.
-					mem = wbcall(pos, bEnd, cgoCheckPtrWrite, sp, mem, ptr, val)
+					mem = wbcall(pos, bEnd, cgoCheckPtrWrite, sp, mem, dst, val)
 				}
-				mem = bEnd.NewValue3A(pos, OpStore, types.TypeMem, w.Aux, ptr, val, mem)
+				mem = bEnd.NewValue3A(pos, OpStore, types.TypeMem, w.Aux, dst, val, mem)
 			case OpZeroWB:
-				dst := w.Args[0]
 				mem = bEnd.NewValue2I(pos, OpZero, types.TypeMem, w.AuxInt, dst, mem)
 				mem.Aux = w.Aux
 			case OpMoveWB:
-				dst := w.Args[0]
 				src := w.Args[1]
 				if isVolatile(src) {
 					for _, c := range volatiles {
@@ -529,9 +552,8 @@ func writebarrier(f *Func) {
 			case OpVarDef, OpVarLive:
 				mem = bEnd.NewValue1A(pos, w.Op, types.TypeMem, w.Aux, mem)
 			case OpStore:
-				ptr := w.Args[0]
 				val := w.Args[1]
-				mem = bEnd.NewValue3A(pos, OpStore, types.TypeMem, w.Aux, ptr, val, mem)
+				mem = bEnd.NewValue3A(pos, OpStore, types.TypeMem, w.Aux, dst, val, mem)
 			}
 		}
 
@@ -557,6 +579,9 @@ func writebarrier(f *Func) {
 				f.freeValue(w)
 			}
 		}
+		if nilcheck != nil && nilcheck.Uses == 0 {
+			nilcheck.reset(OpInvalid)
+		}
 
 		// put values after the store sequence into the end block
 		bEnd.Values = append(bEnd.Values, after...)