# M5 CodeGen/Generic expected-fail manifest.
#
# Format: test<TAB>reason
#
# Every broad-sweep `unsupported` row is recorded here with the reason the
# M5 target cannot compile it.  generic.sh enforces both directions:
#   - each entry must be `unsupported`, not `pass` (stale) and never `crash`;
#   - every `unsupported` row in the sweep must have an entry here.
#
# A crash is a regression, never an expected failure.  Reasons are the actual
# diagnostic categories from the sweep; "M6" marks a feature deferred from
# the M5 subset (floats, 64-bit integers, general varargs, aggregates, GC).
2003-05-27-phifcmpd	target rejection: TMS34010 supports only integer (<=32 bit) and pointer arguments
2003-05-27-useboolinotherbb	target rejection: TMS34010 supports only integer (<=32 bit) and pointer arguments
2003-05-27-usefsubasbool	target rejection: TMS34010 supports only integer (<=32 bit) and pointer arguments
2003-05-28-ManyArgs	target rejection: TMS34010 supports only integer (<=32 bit) and pointer call arguments
2003-07-07-BadLongConst	target rejection: M2 supports only void/integer(<=32)/pointer returns
2003-07-29-BadConstSbyte	target rejection: TMS34010 supports only integer (<=32 bit) and pointer call arguments
2004-05-09-LiveVarPartialRegister	target rejection: M2 .data supports only integer, pointer and byte-array initializers
2005-10-21-longlonggtu	target rejection: M2 supports only void/integer(<=32)/pointer returns
2005-12-01-Crash	target rejection: TMS34010 supports only integer (<=32 bit) and pointer arguments
2005-12-12-ExpandSextInreg	target rejection: M2 supports only void/integer(<=32)/pointer returns
2006-02-12-InsertLibcall	target rejection: TMS34010 supports only integer (<=32 bit) and pointer arguments
2006-04-28-Sign-extend-bool	sign_extend_inreg on i1 is not selectable
2006-06-12-LowerSwitchCrash	target rejection: M2 supports only void/integer(<=32)/pointer returns
2006-07-03-schedulers	target rejection: TMS34010 supports only integer (<=32 bit) and pointer arguments
2006-09-02-LocalAllocCrash	64-bit shift (srl_parts) reached selection; i64 is deferred to M6
2007-04-08-MultipleFrameIndices	inline asm register constraint cannot be allocated
2007-04-27-InlineAsm-X-Dest	inline asm register constraint cannot be allocated
2007-04-27-LargeMemObject	inline asm memory constraint is not supported
2007-05-15-InfiniteRecursion	target rejection: M2 supports only void/integer(<=32)/pointer returns
2007-12-17-InvokeAsm	inline asm memory constraint is not supported
2008-01-25-dag-combine-mul	target rejection: TMS34010: 64-bit multiply reached code generation; the i64 path is reserved for M6. M5 supports 32-bit multiplication only.
2008-01-30-LoadCrash	target rejection: TMS34010 supports only integer (<=32 bit) and pointer arguments
2008-02-04-Ctlz	target rejection: TMS34010 supports only integer (<=32 bit) and pointer arguments
2008-02-20-MatchingMem	inline asm memory constraint is not supported
2008-02-25-NegateZero	target rejection: TMS34010 supports only integer (<=32 bit) and pointer call arguments
2008-02-26-NegatableCrash	target rejection: M2 supports only void/integer(<=32)/pointer returns
2008-08-07-PtrToInt-SmallerInt	target rejection: M2 .data supports only integer, pointer and byte-array initializers
2009-03-17-LSR-APInt	target rejection: TMS34010 supports only integer (<=32 bit) and pointer arguments
2009-03-29-SoftFloatVectorExtract	target rejection: M2 supports only void/integer(<=32)/pointer returns
2009-04-10-SinkCrash	target rejection: TMS34010 supports only integer (<=32 bit) and pointer arguments
2010-07-27-DAGCombineCrash	target rejection: M2 supports only void/integer(<=32)/pointer returns
2010-ZeroSizedArg	target rejection: TMS34010 supports only integer (<=32 bit) and pointer arguments
2013-03-20-APFloatCrash	target rejection: TMS34010 supports only integer (<=32 bit) and pointer arguments
APIntParam	target rejection: TMS34010 supports only integer (<=32 bit) and pointer arguments
APIntSextParam	target rejection: TMS34010 supports only integer (<=32 bit) and pointer arguments
APIntZextParam	target rejection: TMS34010 supports only integer (<=32 bit) and pointer arguments
ForceStackAlign	target rejection: M2 supports only void/integer(<=32)/pointer returns
GlobalISel/irtranslator-byte-type	target rejection: M2 supports only void/integer(<=32)/pointer returns
add-with-overflow-128	target rejection: TMS34010 supports only integer (<=32 bit) and pointer arguments
addr-label	blockaddress values are not supported
annotate	target rejection: M2 .data supports only integer, pointer and byte-array initializers
badarg6	target rejection: M2 supports only void/integer(<=32)/pointer returns
bfloat	target rejection: M2 supports only void/integer(<=32)/pointer returns
bfloat-op	missing libcall lowering (floats/vectors are deferred to M6)
bool-to-double	target rejection: M2 supports only void/integer(<=32)/pointer returns
bswap	target rejection: TMS34010 supports only integer (<=32 bit) and pointer call arguments
cast-fp	target rejection: M2 .data supports only integer, pointer and byte-array initializers
constindices	target rejection: TMS34010 supports only integer (<=32 bit) and pointer call arguments
dag-combine-ossfuzz-crash	target rejection: TMS34010: 64-bit multiply reached code generation; the i64 path is reserved for M6. M5 supports 32-bit multiplication only.
denormal-fp-math-cl-opt	target rejection: M2 supports only void/integer(<=32)/pointer returns
expand-experimental-reductions	target rejection: M2 supports only void/integer(<=32)/pointer returns
extractelement-shuffle	target rejection: TMS34010 supports only integer (<=32 bit) and pointer arguments
fast-math-flags	IR verifier rejects the input (fast-math flags on an integer call)
fneg-fabs	target rejection: M2 supports only void/integer(<=32)/pointer returns
fp-to-int-invalid	missing libcall lowering (floats/vectors are deferred to M6)
fp128-exp10-libcall	target rejection: M2 supports only void/integer(<=32)/pointer returns
fp128-math-libcalls	target rejection: M2 supports only void/integer(<=32)/pointer returns
fp_to_int	target rejection: TMS34010 supports only integer (<=32 bit) and pointer arguments
fpoperations	target rejection: M2 supports only void/integer(<=32)/pointer returns
fpowi-promote	target rejection: M2 supports only void/integer(<=32)/pointer returns
half	target rejection: M2 supports only void/integer(<=32)/pointer returns
half-op	missing libcall lowering (floats/vectors are deferred to M6)
i128-addsub	target rejection: TMS34010 supports only integer (<=32 bit) and pointer arguments
i128-arith	target rejection: M2 supports only void/integer(<=32)/pointer returns
icmp-illegal	target rejection: TMS34010 supports only integer (<=32 bit) and pointer arguments
inline-asm-mem-clobber	inline asm register constraint cannot be allocated
inline-asm-special-strings	inline asm syntax is not supported by the TMS34010 parser
intrinsics	target rejection: M2 supports only void/integer(<=32)/pointer returns
isunord	target rejection: TMS34010 supports only integer (<=32 bit) and pointer arguments
llvm-ct-intrinsics	target rejection: TMS34010 supports only integer (<=32 bit) and pointer arguments
multiple-return-values-cross-block-with-invoke	target rejection: M2 supports only void/integer(<=32)/pointer returns
nested-select	target rejection: TMS34010 supports only integer (<=32 bit) and pointer arguments
overloaded-intrinsic-name	gc.statepoint calls are not supported
pr12507	target rejection: TMS34010 supports only integer (<=32 bit) and pointer arguments
pr24662	target rejection: M2 supports only void/integer(<=32)/pointer returns
pr33094	target rejection: M2 supports only void/integer(<=32)/pointer returns
pr49872	target rejection: TMS34010 supports only integer (<=32 bit) and pointer arguments
print-arith-fp	target rejection: M2 .data supports only integer, pointer and byte-array initializers
ptr-annotate	target rejection: M2 .data supports only integer, pointer and byte-array initializers
replace-intrinsics-with-veclib	target rejection: M2 supports only void/integer(<=32)/pointer returns
replace-intrinsics-with-veclib-darwin-libsystem-m	target rejection: M2 supports only void/integer(<=32)/pointer returns
select	target rejection: TMS34010 supports only integer (<=32 bit) and pointer arguments
select-cc	target rejection: M2 supports only void/integer(<=32)/pointer returns
shift-int64	target rejection: M2 supports only void/integer(<=32)/pointer returns
storetrunc-fp	target rejection: TMS34010 supports only integer (<=32 bit) and pointer arguments
trap	target rejection: M2 supports only direct calls to global functions and indirect calls through a function pointer
v-split	missing libcall lowering (floats/vectors are deferred to M6)
vector	missing libcall lowering (floats/vectors are deferred to M6)
vector-casts	missing libcall lowering (floats/vectors are deferred to M6)
zero-sized-array	target rejection: TMS34010 supports only integer (<=32 bit) and pointer arguments
