zstd: bump avo pin, regenerate arm64 asm (#1167)

* zstd: bump avo pin, regenerate arm64 with BMI2 lowering

Bumps the pinned avo commit (honeycombio/avo) past today's fixes: a
flag-liveness bug and a 32-bit CMPL/TESTL fold bug (neither actually reachable
from this codebase - seqdec's generated output is byte-identical without
them, and CMPL/TESTL aren't used here at all), plus a new EQ/NE-only
sub-32-bit CMPW/TESTB lowering that fse_decoder's buildDtable needs.

Checked whether the sub-word compare/test fold (the general form of which the
new avo now refuses to lower) actually produced wrong output in the
already-shipped fse_decoder_arm64.s: it didn't. All three call sites happened
to operate on registers already clean above the compared width (via an
upstream sign/zero-extending load, or a preceding bounds check), verified by
tracing register provenance by hand. Real risk, but mitigated by accident
rather than by design - which is exactly why the stricter check is worth
having going forward.

The new avo also prefers the BMI2 twin over the generic one when lowering to
arm64 (arm64 has native flag-free equivalents for the BMI2 idioms), so
seqdec_arm64.s now lowers sequenceDecs_decode_bmi2/decodeSync_bmi2 instead of
the generic variants. Symbol names are unaffected (arm64Name strips the
_bmi2/_amd64 suffix either way).

Validated: go vet and go test ./zstd/... pass natively on arm64; FuzzDecodeAll
(26.4M execs), FuzzDecoder (9.6M execs), and FuzzDecAllNoBMI2 pass with zero
crashers.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

* zstd: bump avo pin, prefer generic (non-BMI2) twin on arm64

Benchmarking the BMI2-preferred seqdec on real hardware (Apple M4, Cortex-A72)
showed it's 13-50% slower than the generic twin, not faster as assumed -- see
the follow-up avo commit for why. Bump past that fix; avo's arm64 lowering now
defaults to the generic twin (ARM64PreferBMI2 defaults false), which this repo
doesn't override. seqdec_arm64.s is now byte-identical to what #1160 shipped,
aside from the skip-reason comment text; fse_decoder_arm64.s is unchanged
(the CMPW/TESTB fix has no BMI2 twin to choose between).

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

Author: lizthegrey

zstd/_generate/go.mod

@@ -6,7 +6,7 @@ toolchain go1.24.2
 
 require (
 	github.com/klauspost/compress v1.15.15
-	github.com/mmcloughlin/avo v0.6.0
+	github.com/mmcloughlin/avo v0.6.1-0.20260709170202-520357590ac6
 )
 
 require (
@@ -17,4 +17,4 @@ require (
 
 replace github.com/klauspost/compress => ../..
 
-replace github.com/mmcloughlin/avo => github.com/honeycombio/avo v0.6.1-0.20260629192935-ff52dcb5f88d
+replace github.com/mmcloughlin/avo => github.com/honeycombio/avo v0.6.1-0.20260709170202-520357590ac6

zstd/_generate/go.sum

@@ -1,7 +1,7 @@
 github.com/google/go-cmp v0.6.0 h1:ofyhxvXcZhMsU5ulbFiLKl/XBFqE1GSq7atu8tAmTRI=
 github.com/google/go-cmp v0.6.0/go.mod h1:17dUlkBOakJ0+DkrSSNjCkIjxS6bF9zb3elmeNGIjoY=
-github.com/honeycombio/avo v0.6.1-0.20260629192935-ff52dcb5f88d h1:946aSI2nbyDYa8iTjFmwoM4S4TKmE8heVLGlchLj7vU=
-github.com/honeycombio/avo v0.6.1-0.20260629192935-ff52dcb5f88d/go.mod h1:+Pk+j3KceMwAIGb32HZSCvkMCzJvxy6xCRk+urz7iPw=
+github.com/honeycombio/avo v0.6.1-0.20260709170202-520357590ac6 h1:NG37EIVIXn6fWkKN8xIUqLZ7huCzt+17QTbW2bEc9vA=
+github.com/honeycombio/avo v0.6.1-0.20260709170202-520357590ac6/go.mod h1:+Pk+j3KceMwAIGb32HZSCvkMCzJvxy6xCRk+urz7iPw=
 golang.org/x/mod v0.27.0 h1:kb+q2PyFnEADO2IEF935ehFUXlWiNjJWtRNgBLSfbxQ=
 golang.org/x/mod v0.27.0/go.mod h1:rWI627Fq0DEoudcK+MBkNkCe0EetEaDSwJJkCcjpazc=
 golang.org/x/sync v0.16.0 h1:ycBJEhp9p4vXvUZNszeOq0kGTPghopOL8q0fq3vstxw=

zstd/fse_decoder_arm64.s

@@ -28,7 +28,10 @@ TEXT ·buildDtable_asm(SB), $0-24
 init_main_loop:
 	ADD  R8<<1, R1, R15
 	MOVH (R15), R9
-	CMN  $1, R9
+	AND  $0xffff, R9, R15
+	MOVD $-1, R16
+	AND  $0xffff, R16, R16
+	CMP  R16, R15
 	BNE  do_not_update_high_threshold
 	ADD  R7<<3, R5, R15
 	MOVB R8, 1(R15)
@@ -126,9 +129,13 @@ build_table_main_table:
 	RET
 
 build_table_check1_ok:
-	TST  R1, R1
+	AND  $0xff, R1, R15
+	AND  $0xff, R1, R16
+	TST  R16, R15
 	BNE  build_table_check2_ok
-	CMP  R6, R7
+	AND  $0xffff, R7, R15
+	AND  $0xffff, R6, R16
+	CMP  R16, R15
 	BNE  build_table_check2_ok
 	MOVD ctx+8(FP), R0
 	MOVD R7, 24(R0)

zstd/seqdec_arm64.s

@@ -670,9 +670,9 @@ error_overread:
 	MOVD R16, ret+24(FP)
 	RET
 
-// skipped sequenceDecs_decode_bmi2 (BMI2 not available on arm64)
+// skipped sequenceDecs_decode_bmi2 (generic twin preferred on arm64)
 
-// skipped sequenceDecs_decode_56_bmi2 (BMI2 not available on arm64)
+// skipped sequenceDecs_decode_56_bmi2 (generic twin preferred on arm64)
 
 // func sequenceDecs_executeSimple_amd64(ctx *executeAsmContext) bool
 // Requires: SSE
@@ -1966,7 +1966,7 @@ error_not_enough_space:
 	MOVD R16, ret+24(FP)
 	RET
 
-// skipped sequenceDecs_decodeSync_bmi2 (BMI2 not available on arm64)
+// skipped sequenceDecs_decodeSync_bmi2 (generic twin preferred on arm64)
 
 // func sequenceDecs_decodeSync_safe_amd64(s *sequenceDecs, br *bitReader, ctx *decodeSyncAsmContext) int
 // Requires: CMOV, SSE
@@ -2702,4 +2702,4 @@ error_not_enough_space:
 	MOVD R16, ret+24(FP)
 	RET
 
-// skipped sequenceDecs_decodeSync_safe_bmi2 (BMI2 not available on arm64)
+// skipped sequenceDecs_decodeSync_safe_bmi2 (generic twin preferred on arm64)