[JitDiff X64] [Daniel-Svensson] Some System.Decimal performance improvements

[MihuBot]

Job completed in 14 minutes.
dotnet/runtime#99212

Diffs

Found 260 files with textual diffs.

Summary of Code Size diffs:
(Lower is better)

Total bytes of base: 39178912
Total bytes of diff: 39178748
Total bytes of delta: -164 (-0.00 % of base)
Total relative delta: -1.62
 diff is an improvement.
 relative diff is an improvement.


Total byte diff includes 179 bytes from reconciling methods
	Base had 2 unique methods, 139 unique bytes
	Diff had 7 unique methods, 318 unique bytes

Top file improvements (bytes):
 -164 : System.Private.CoreLib.dasm (-0.00 % of base)

1 total files with Code Size differences (1 improved, 0 regressed), 258 unchanged.

Top method regressions (bytes):
 95 (5.21 % of base) : System.Private.CoreLib.dasm - System.Decimal+DecCalc:VarDecDiv(byref,byref) (FullOpts)
 94 (Infinity of base) : System.Private.CoreLib.dasm - System.Math:<BigMul>g__SoftwareFallback|49_0(ulong,ulong,byref):ulong (FullOpts) (0 base, 1 diff methods)
 45 (Infinity of base) : System.Private.CoreLib.dasm - System.Math:<CopySign>g__SoftwareFallback|55_0(double,double):double (FullOpts) (0 base, 1 diff methods)
 45 (Infinity of base) : System.Private.CoreLib.dasm - System.Math:BigMul(uint,ulong,byref):ulong (FullOpts) (0 base, 1 diff methods)
 45 (Infinity of base) : System.Private.CoreLib.dasm - System.Math:BigMul(ulong,uint,byref):ulong (FullOpts) (0 base, 1 diff methods)
 44 (Infinity of base) : System.Private.CoreLib.dasm - System.Decimal+DecCalc:IncreaseScale(byref,uint) (FullOpts) (0 base, 1 diff methods)
 28 (Infinity of base) : System.Private.CoreLib.dasm - System.Decimal+DecCalc:Div64By32(ulong,uint):System.ValueTuple`2[uint,uint] (FullOpts) (0 base, 1 diff methods)
 17 (Infinity of base) : System.Private.CoreLib.dasm - System.Decimal+DecCalc:Div128By64(ulong,ulong):ulong (FullOpts) (0 base, 1 diff methods)
 4 (13.79 % of base) : System.Private.CoreLib.dasm - System.Decimal+DecCalc:IncreaseScale64(byref,uint) (FullOpts)

Top method improvements (bytes):
 -158 (-19.73 % of base) : System.Private.CoreLib.dasm - System.Decimal+DecCalc:VarDecMul(byref,byref) (FullOpts)
 -142 (-88.75 % of base) : System.Private.CoreLib.dasm - System.Decimal+DecCalc:Div96By64(byref,ulong):uint (FullOpts)
 -94 (-100.00 % of base) : System.Private.CoreLib.dasm - System.Math:<BigMul>g__SoftwareFallback|47_0(ulong,ulong,byref):ulong (FullOpts) (1 base, 0 diff methods)
 -45 (-100.00 % of base) : System.Private.CoreLib.dasm - System.Math:<CopySign>g__SoftwareFallback|53_0(double,double):double (FullOpts) (1 base, 0 diff methods)
 -40 (-39.22 % of base) : System.Private.CoreLib.dasm - System.Decimal+DecCalc:Div96By32(byref,uint):uint (FullOpts)
 -26 (-11.21 % of base) : System.Private.CoreLib.dasm - System.Decimal+DecCalc:VarDecCmpSub(byref,byref):int (FullOpts)
 -23 (-3.58 % of base) : System.Private.CoreLib.dasm - System.Decimal+DecCalc:VarDecFromR4(float,byref) (FullOpts)
 -21 (-1.63 % of base) : System.Private.CoreLib.dasm - System.Decimal+DecCalc:DecAddSub(byref,byref,ubyte) (FullOpts)
 -12 (-7.23 % of base) : System.Private.CoreLib.dasm - System.Decimal+DecCalc:Div128By96(byref,byref):uint (FullOpts)
 -9 (-4.07 % of base) : System.Private.CoreLib.dasm - System.Decimal+DecCalc:VarCyFromDec(byref):long (FullOpts)
 -9 (-1.46 % of base) : System.Private.CoreLib.dasm - System.Decimal+DecCalc:VarDecMod(byref,byref) (FullOpts)
 -2 (-4.08 % of base) : System.Private.CoreLib.dasm - System.Decimal+DecCalc:IncreaseScale(byref,uint):uint (FullOpts)

Top method regressions (percentages):
 17 (Infinity of base) : System.Private.CoreLib.dasm - System.Decimal+DecCalc:Div128By64(ulong,ulong):ulong (FullOpts) (0 base, 1 diff methods)
 28 (Infinity of base) : System.Private.CoreLib.dasm - System.Decimal+DecCalc:Div64By32(ulong,uint):System.ValueTuple`2[uint,uint] (FullOpts) (0 base, 1 diff methods)
 44 (Infinity of base) : System.Private.CoreLib.dasm - System.Decimal+DecCalc:IncreaseScale(byref,uint) (FullOpts) (0 base, 1 diff methods)
 94 (Infinity of base) : System.Private.CoreLib.dasm - System.Math:<BigMul>g__SoftwareFallback|49_0(ulong,ulong,byref):ulong (FullOpts) (0 base, 1 diff methods)
 45 (Infinity of base) : System.Private.CoreLib.dasm - System.Math:<CopySign>g__SoftwareFallback|55_0(double,double):double (FullOpts) (0 base, 1 diff methods)
 45 (Infinity of base) : System.Private.CoreLib.dasm - System.Math:BigMul(uint,ulong,byref):ulong (FullOpts) (0 base, 1 diff methods)
 45 (Infinity of base) : System.Private.CoreLib.dasm - System.Math:BigMul(ulong,uint,byref):ulong (FullOpts) (0 base, 1 diff methods)
 4 (13.79 % of base) : System.Private.CoreLib.dasm - System.Decimal+DecCalc:IncreaseScale64(byref,uint) (FullOpts)
 95 (5.21 % of base) : System.Private.CoreLib.dasm - System.Decimal+DecCalc:VarDecDiv(byref,byref) (FullOpts)

Top method improvements (percentages):
 -94 (-100.00 % of base) : System.Private.CoreLib.dasm - System.Math:<BigMul>g__SoftwareFallback|47_0(ulong,ulong,byref):ulong (FullOpts) (1 base, 0 diff methods)
 -45 (-100.00 % of base) : System.Private.CoreLib.dasm - System.Math:<CopySign>g__SoftwareFallback|53_0(double,double):double (FullOpts) (1 base, 0 diff methods)
 -142 (-88.75 % of base) : System.Private.CoreLib.dasm - System.Decimal+DecCalc:Div96By64(byref,ulong):uint (FullOpts)
 -40 (-39.22 % of base) : System.Private.CoreLib.dasm - System.Decimal+DecCalc:Div96By32(byref,uint):uint (FullOpts)
 -158 (-19.73 % of base) : System.Private.CoreLib.dasm - System.Decimal+DecCalc:VarDecMul(byref,byref) (FullOpts)
 -26 (-11.21 % of base) : System.Private.CoreLib.dasm - System.Decimal+DecCalc:VarDecCmpSub(byref,byref):int (FullOpts)
 -12 (-7.23 % of base) : System.Private.CoreLib.dasm - System.Decimal+DecCalc:Div128By96(byref,byref):uint (FullOpts)
 -2 (-4.08 % of base) : System.Private.CoreLib.dasm - System.Decimal+DecCalc:IncreaseScale(byref,uint):uint (FullOpts)
 -9 (-4.07 % of base) : System.Private.CoreLib.dasm - System.Decimal+DecCalc:VarCyFromDec(byref):long (FullOpts)
 -23 (-3.58 % of base) : System.Private.CoreLib.dasm - System.Decimal+DecCalc:VarDecFromR4(float,byref) (FullOpts)
 -21 (-1.63 % of base) : System.Private.CoreLib.dasm - System.Decimal+DecCalc:DecAddSub(byref,byref,ubyte) (FullOpts)
 -9 (-1.46 % of base) : System.Private.CoreLib.dasm - System.Decimal+DecCalc:VarDecMod(byref,byref) (FullOpts)

21 total methods with Code Size differences (12 improved, 9 regressed), 230455 unchanged.

--------------------------------------------------------------------------------

Artifacts:

• build-artifacts-main.zip (652 MB)
• build-clr-checked-main.zip (87 MB)
• build-artifacts-pr.zip (653 MB)
• build-clr-checked-pr.zip (87 MB)
• diff-frameworks.txt (6 KB)
• jit-diffs-frameworks.zip (259 MB)

[MihuBot]

Top method regressions

95 (5.21 % of base) - System.Decimal+DecCalc:VarDecDiv(byref,byref)

 ; Assembly listing for method System.Decimal+DecCalc:VarDecDiv(byref,byref) (FullOpts)
 ; Emitting BLENDED_CODE for X64 with AVX512 - Unix
 ; FullOpts code
 ; optimized code
 ; rbp based frame
 ; fully interruptible
 ; No PGO data
-; 0 inlinees with PGO data; 54 single block inlinees; 11 inlinees without PGO data
+; 0 inlinees with PGO data; 74 single block inlinees; 18 inlinees without PGO data
 ; Final local variable assignments
 ;
-; V00 arg0 [V00,T12] ( 15, 10 ) byref -> rbx single-def
-; V01 arg1 [V01,T19] ( 13, 9.50) byref -> r15 single-def
-; V02 loc0 [V02 ] ( 41, 65.50) struct (16) [rbp-0x38] do-not-enreg[XSF] addr-exposed ld-addr-op <System.Decimal+DecCalc+Buf12>
-; V03 loc1 [V03,T02] ( 10, 40 ) int -> [rbp-0x3C] 
-; V04 loc2 [V04,T01] ( 22, 56 ) int -> registers 
+; V00 arg0 [V00,T17] ( 15, 10 ) byref -> rbx single-def
+; V01 arg1 [V01,T26] ( 13, 9.50) byref -> r15 single-def
+; V02 loc0 [V02 ] ( 55, 93.50) struct (16) [rbp-0x38] do-not-enreg[XSF] addr-exposed ld-addr-op <System.Decimal+DecCalc+Buf12>
+; V03 loc1 [V03,T02] ( 8, 32 ) int -> [rbp-0x3C] 
+; V04 loc2 [V04,T01] ( 22, 56 ) int -> r8 
 ; V05 loc3 [V05,T00] ( 40, 70.50) int -> r14 ld-addr-op
-; V06 loc4 [V06,T21] ( 5, 8 ) ubyte -> r13 
+; V06 loc4 [V06,T29] ( 5, 8 ) ubyte -> r13 
 ; V07 loc5 [V07,T03] ( 16, 29 ) int -> rcx 
-; V08 loc6 [V08,T20] ( 7, 10.50) int -> r12 
-; V09 loc7 [V09,T11] ( 7, 14 ) int -> [rbp-0x40] 
-; V10 loc8 [V10,T10] ( 4, 16 ) int -> rax 
+; V08 loc6 [V08,T28] ( 9, 8 ) int -> r12 
+; V09 loc7 [V09,T16] ( 6, 13.50) int -> [rbp-0x40] 
+;* V10 loc8 [V10 ] ( 0, 0 ) int -> zero-ref 
 ;* V11 loc9 [V11 ] ( 0, 0 ) struct (16) zero-ref ld-addr-op <System.ReadOnlySpan`1[uint]>
-; V12 loc10 [V12 ] ( 30, 53.50) struct (16) [rbp-0x50] do-not-enreg[XSF] addr-exposed ld-addr-op <System.Decimal+DecCalc+Buf16>
-; V13 loc11 [V13,T25] ( 7, 7 ) long -> r12 
-; V14 loc12 [V14,T36] ( 3, 1.50) long -> rax 
+; V12 loc10 [V12 ] ( 32, 58 ) struct (16) [rbp-0x50] do-not-enreg[XSF] addr-exposed ld-addr-op <System.Decimal+DecCalc+Buf16>
+; V13 loc11 [V13,T37] ( 6, 6.50) long -> r12 
+; V14 loc12 [V14,T49] ( 3, 1.50) long -> r8 
 ; V15 loc13 [V15 ] ( 8, 7.50) struct (16) [rbp-0x60] do-not-enreg[XSF] addr-exposed ld-addr-op <System.Decimal+DecCalc+Buf12>
 ; V16 loc14 [V16,T08] ( 18, 23 ) int -> rcx ld-addr-op
 ; V17 loc15 [V17,T09] ( 15, 18 ) long -> rdi ld-addr-op
-; V18 loc16 [V18,T37] ( 3, 1.50) long -> rdi 
-; V19 loc17 [V19,T38] ( 3, 1.50) int -> rdi 
+; V18 loc16 [V18,T50] ( 3, 1.50) long -> rdi 
+; V19 loc17 [V19,T51] ( 3, 1.50) int -> rdi 
 ;# V20 OutArgs [V20 ] ( 1, 1 ) struct ( 0) [rsp+0x00] do-not-enreg[XS] addr-exposed "OutgoingArgSpace"
 ;* V21 tmp1 [V21 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
 ;* V22 tmp2 [V22 ] ( 0, 0 ) long -> zero-ref "impAppendStmt"
 ;* V23 tmp3 [V23 ] ( 0, 0 ) long -> zero-ref "impAppendStmt"
 ;* V24 tmp4 [V24 ] ( 0, 0 ) long -> zero-ref "impAppendStmt"
 ;* V25 tmp5 [V25 ] ( 0, 0 ) long -> zero-ref "impAppendStmt"
-; V26 tmp6 [V26,T22] ( 2, 8 ) int -> rax "dup spill"
+; V26 tmp6 [V26,T30] ( 2, 8 ) int -> r8 "dup spill"
 ;* V27 tmp7 [V27 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
 ;* V28 tmp8 [V28 ] ( 0, 0 ) long -> zero-ref "impAppendStmt"
-; V29 tmp9 [V29,T32] ( 3, 3 ) long -> rax "dup spill"
-; V30 tmp10 [V30,T23] ( 2, 8 ) int -> rax "dup spill"
+; V29 tmp9 [V29,T47] ( 3, 3 ) long -> r8 "dup spill"
+; V30 tmp10 [V30,T31] ( 2, 8 ) int -> r8 "dup spill"
 ;* V31 tmp11 [V31 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
-; V32 tmp12 [V32,T24] ( 2, 8 ) int -> rax "dup spill"
-; V33 tmp13 [V33,T04] ( 3, 24 ) long -> rdi "dup spill"
-; V34 tmp14 [V34,T46] ( 3, 0 ) ref -> rbx class-hnd exact single-def "NewObj constructor temp" <System.DivideByZeroException>
+; V32 tmp12 [V32,T32] ( 2, 8 ) int -> r8 "dup spill"
+;* V33 tmp13 [V33 ] ( 0, 0 ) struct ( 8) zero-ref multireg-ret "dup spill" <System.ValueTuple`2[uint,uint]>
+; V34 tmp14 [V34,T61] ( 3, 0 ) ref -> rbx class-hnd exact single-def "NewObj constructor temp" <System.DivideByZeroException>
 ;* V35 tmp15 [V35 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
-; V36 tmp16 [V36,T26] ( 3, 6 ) int -> rcx "Inline return value spill temp"
-; V37 tmp17 [V37,T13] ( 3, 12 ) int -> rcx "Inlining Arg"
-;* V38 tmp18 [V38 ] ( 0, 0 ) struct (16) zero-ref "ReadOnlySpan<T> for CreateSpan<T>" <System.ReadOnlySpan`1[uint]>
-;* V39 tmp19 [V39 ] ( 0, 0 ) ubyte -> zero-ref "Inline return value spill temp"
-; V40 tmp20 [V40,T05] ( 3, 24 ) long -> rdi "dup spill"
-;* V41 tmp21 [V41 ] ( 0, 0 ) long -> zero-ref "Inline stloc first use temp"
-; V42 tmp22 [V42,T27] ( 3, 6 ) int -> rdx "Inline stloc first use temp"
-;* V43 tmp23 [V43 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
-; V44 tmp24 [V44,T33] ( 2, 2 ) long -> rdi "Inlining Arg"
-; V45 tmp25 [V45,T28] ( 3, 6 ) int -> rax "Inline return value spill temp"
-; V46 tmp26 [V46,T14] ( 3, 12 ) int -> rax "Inlining Arg"
-;* V47 tmp27 [V47 ] ( 0, 0 ) struct (16) zero-ref "ReadOnlySpan<T> for CreateSpan<T>" <System.ReadOnlySpan`1[uint]>
-; V48 tmp28 [V48,T15] ( 3, 12 ) long -> rdi "Inline stloc first use temp"
-;* V49 tmp29 [V49 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
-;* V50 tmp30 [V50 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
-;* V51 tmp31 [V51 ] ( 0, 0 ) ubyte -> zero-ref "Inline return value spill temp"
-; V52 tmp32 [V52,T06] ( 3, 24 ) long -> rdx "dup spill"
-;* V53 tmp33 [V53 ] ( 0, 0 ) long -> zero-ref "Inline stloc first use temp"
-; V54 tmp34 [V54,T29] ( 3, 6 ) int -> rdx "Inline stloc first use temp"
-;* V55 tmp35 [V55 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
-; V56 tmp36 [V56,T30] ( 3, 6 ) int -> rax "Inline return value spill temp"
-; V57 tmp37 [V57,T16] ( 3, 12 ) int -> rax "Inlining Arg"
-;* V58 tmp38 [V58 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
-;* V59 tmp39 [V59 ] ( 0, 0 ) struct (16) zero-ref "ReadOnlySpan<T> for CreateSpan<T>" <System.ReadOnlySpan`1[uint]>
-;* V60 tmp40 [V60 ] ( 0, 0 ) ubyte -> zero-ref "Inline return value spill temp"
-; V61 tmp41 [V61,T07] ( 3, 24 ) long -> rdx "dup spill"
-;* V62 tmp42 [V62 ] ( 0, 0 ) long -> zero-ref "Inline stloc first use temp"
-; V63 tmp43 [V63,T31] ( 3, 6 ) int -> rdx "Inline stloc first use temp"
-;* V64 tmp44 [V64 ] ( 0, 0 ) ubyte -> zero-ref "Inline return value spill temp"
-; V65 tmp45 [V65,T17] ( 3, 12 ) long -> rsi "Inline stloc first use temp"
-; V66 tmp46 [V66,T18] ( 3, 12 ) int -> rdx "Inline stloc first use temp"
-;* V67 tmp47 [V67 ] ( 0, 0 ) ubyte -> zero-ref "Inline return value spill temp"
-; V68 tmp48 [V68,T39] ( 3, 1.50) long -> rsi "Inline stloc first use temp"
-; V69 tmp49 [V69,T40] ( 3, 1.50) int -> rdx "Inline stloc first use temp"
-;* V70 tmp50 [V70 ] ( 0, 0 ) ubyte -> zero-ref "Inline return value spill temp"
-; V71 tmp51 [V71,T41] ( 3, 1.50) long -> rsi "Inline stloc first use temp"
-; V72 tmp52 [V72,T42] ( 3, 1.50) int -> rdx "Inline stloc first use temp"
-;* V73 tmp53 [V73 ] ( 0, 0 ) ubyte -> zero-ref "Inline return value spill temp"
-; V74 tmp54 [V74,T43] ( 3, 1.50) long -> rsi "Inline stloc first use temp"
-; V75 tmp55 [V75,T44] ( 3, 1.50) int -> rdx "Inline stloc first use temp"
-;* V76 tmp56 [V76 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
-;* V77 tmp57 [V77 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
-;* V78 tmp58 [V78 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
+;* V36 tmp16 [V36 ] ( 0, 0 ) int -> zero-ref "Inline return value spill temp"
+; V37 tmp17 [V37,T44] ( 6, 3 ) int -> rdx "Inline stloc first use temp"
+;* V38 tmp18 [V38 ] ( 0, 0 ) struct ( 8) zero-ref do-not-enreg[SR] multireg-ret "Return value temp for multireg return" <System.ValueTuple`2[uint,uint]>
+;* V39 tmp19 [V39 ] ( 0, 0 ) struct ( 8) zero-ref multireg-ret "Inline stloc first use temp" <System.ValueTuple`2[uint,uint]>
+;* V40 tmp20 [V40 ] ( 0, 0 ) struct ( 8) zero-ref do-not-enreg[SR] multireg-ret "Return value temp for multireg return" <System.ValueTuple`2[uint,uint]>
+;* V41 tmp21 [V41 ] ( 0, 0 ) struct ( 8) zero-ref do-not-enreg[SR] multireg-ret "Return value temp for multireg return" <System.ValueTuple`2[uint,uint]>
+; V42 tmp22 [V42,T38] ( 3, 6 ) int -> r8 "Inline return value spill temp"
+; V43 tmp23 [V43,T18] ( 3, 12 ) int -> r8 "Inlining Arg"
+;* V44 tmp24 [V44 ] ( 0, 0 ) struct (16) zero-ref "ReadOnlySpan<T> for CreateSpan<T>" <System.ReadOnlySpan`1[uint]>
+; V45 tmp25 [V45,T33] ( 2, 8 ) long -> rsi ld-addr-op "Inline ldloca(s) first use temp"
+; V46 tmp26 [V46,T19] ( 3, 12 ) long -> rdx "Inline stloc first use temp"
+;* V47 tmp27 [V47 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+;* V48 tmp28 [V48 ] ( 0, 0 ) long -> zero-ref "Inline return value spill temp"
+;* V49 tmp29 [V49 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+; V50 tmp30 [V50 ] ( 2, 8 ) long -> [rbp-0x68] do-not-enreg[X] addr-exposed ld-addr-op "Inline ldloca(s) first use temp"
+; V51 tmp31 [V51,T12] ( 2, 16 ) long -> rdx "impAppendStmt"
+;* V52 tmp32 [V52 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+;* V53 tmp33 [V53 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
+;* V54 tmp34 [V54 ] ( 0, 0 ) struct ( 8) zero-ref do-not-enreg[SBR] multireg-ret "Inline return value spill temp" <System.ValueTuple`2[uint,uint]>
+; V55 tmp35 [V55,T04] ( 3, 24 ) long -> rax "Inlining Arg"
+;* V56 tmp36 [V56 ] ( 0, 0 ) struct ( 8) zero-ref do-not-enreg[SR] multireg-ret "Return value temp for multireg return" <System.ValueTuple`2[uint,uint]>
+;* V57 tmp37 [V57 ] ( 0, 0 ) ubyte -> zero-ref "Inline return value spill temp"
+; V58 tmp38 [V58,T05] ( 3, 24 ) long -> rsi "dup spill"
+;* V59 tmp39 [V59 ] ( 0, 0 ) long -> zero-ref "Inline stloc first use temp"
+; V60 tmp40 [V60,T39] ( 3, 6 ) int -> rdi "Inline stloc first use temp"
+;* V61 tmp41 [V61 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+; V62 tmp42 [V62,T48] ( 2, 2 ) long -> rax "Inlining Arg"
+;* V63 tmp43 [V63 ] ( 0, 0 ) long -> zero-ref "Inline return value spill temp"
+;* V64 tmp44 [V64 ] ( 0, 0 ) byref -> zero-ref "Inline stloc first use temp"
+;* V65 tmp45 [V65 ] ( 0, 0 ) long -> zero-ref "impAppendStmt"
+;* V66 tmp46 [V66 ] ( 0, 0 ) struct (16) zero-ref do-not-enreg[SBR] multireg-ret "Return value temp for multireg return" <System.ValueTuple`2[ulong,ulong]>
+;* V67 tmp47 [V67 ] ( 0, 0 ) struct (16) zero-ref multireg-ret "Inline stloc first use temp" <System.ValueTuple`2[ulong,ulong]>
+;* V68 tmp48 [V68 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+; V69 tmp49 [V69,T40] ( 3, 6 ) int -> r8 "Inline return value spill temp"
+; V70 tmp50 [V70,T20] ( 3, 12 ) int -> r8 "Inlining Arg"
+;* V71 tmp51 [V71 ] ( 0, 0 ) struct (16) zero-ref "ReadOnlySpan<T> for CreateSpan<T>" <System.ReadOnlySpan`1[uint]>
+; V72 tmp52 [V72,T34] ( 2, 8 ) long -> rsi ld-addr-op "Inline ldloca(s) first use temp"
+; V73 tmp53 [V73,T21] ( 3, 12 ) long -> rdx "Inline stloc first use temp"
+;* V74 tmp54 [V74 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+;* V75 tmp55 [V75 ] ( 0, 0 ) long -> zero-ref "Inline return value spill temp"
+;* V76 tmp56 [V76 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+; V77 tmp57 [V77 ] ( 2, 8 ) long -> [rbp-0x70] do-not-enreg[X] addr-exposed ld-addr-op "Inline ldloca(s) first use temp"
+; V78 tmp58 [V78,T13] ( 2, 16 ) long -> rdx "impAppendStmt"
 ;* V79 tmp59 [V79 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
 ;* V80 tmp60 [V80 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
-;* V81 tmp61 [V81 ] ( 0, 0 ) byref -> zero-ref "field V11._reference (fldOffset=0x0)" P-INDEP
-;* V82 tmp62 [V82 ] ( 0, 0 ) int -> zero-ref "field V11._length (fldOffset=0x8)" P-INDEP
-;* V83 tmp63 [V83 ] ( 0, 0 ) byref -> zero-ref "field V38._reference (fldOffset=0x0)" P-INDEP
-;* V84 tmp64 [V84 ] ( 0, 0 ) int -> zero-ref "field V38._length (fldOffset=0x8)" P-INDEP
-;* V85 tmp65 [V85 ] ( 0, 0 ) byref -> zero-ref "field V47._reference (fldOffset=0x0)" P-INDEP
-;* V86 tmp66 [V86 ] ( 0, 0 ) int -> zero-ref "field V47._length (fldOffset=0x8)" P-INDEP
-;* V87 tmp67 [V87 ] ( 0, 0 ) byref -> zero-ref "field V59._reference (fldOffset=0x0)" P-INDEP
-;* V88 tmp68 [V88 ] ( 0, 0 ) int -> zero-ref "field V59._length (fldOffset=0x8)" P-INDEP
-; V89 tmp69 [V89,T34] ( 2, 2 ) long -> rdi "Cast away GC"
-; V90 tmp70 [V90,T35] ( 2, 2 ) long -> rdi "argument with side effect"
-; V91 cse0 [V91,T45] ( 3, 1.50) int -> rcx "CSE #11: conservative"
+; V81 tmp61 [V81,T35] ( 2, 8 ) long -> rsi ld-addr-op "Inline ldloca(s) first use temp"
+;* V82 tmp62 [V82 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+;* V83 tmp63 [V83 ] ( 0, 0 ) long -> zero-ref "Inline return value spill temp"
+;* V84 tmp64 [V84 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+; V85 tmp65 [V85 ] ( 2, 8 ) long -> [rbp-0x78] do-not-enreg[X] addr-exposed ld-addr-op "Inline ldloca(s) first use temp"
+; V86 tmp66 [V86,T14] ( 2, 16 ) long -> rdi "impAppendStmt"
+;* V87 tmp67 [V87 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+;* V88 tmp68 [V88 ] ( 0, 0 ) ubyte -> zero-ref "Inline return value spill temp"
+; V89 tmp69 [V89,T06] ( 3, 24 ) long -> rdx "dup spill"
+;* V90 tmp70 [V90 ] ( 0, 0 ) long -> zero-ref "Inline stloc first use temp"
+; V91 tmp71 [V91,T41] ( 3, 6 ) int -> rdx "Inline stloc first use temp"
+;* V92 tmp72 [V92 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+; V93 tmp73 [V93,T42] ( 3, 6 ) int -> r8 "Inline return value spill temp"
+; V94 tmp74 [V94,T22] ( 3, 12 ) int -> r8 "Inlining Arg"
+;* V95 tmp75 [V95 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+;* V96 tmp76 [V96 ] ( 0, 0 ) struct (16) zero-ref "ReadOnlySpan<T> for CreateSpan<T>" <System.ReadOnlySpan`1[uint]>
+; V97 tmp77 [V97,T36] ( 2, 8 ) long -> rsi ld-addr-op "Inline ldloca(s) first use temp"
+;* V98 tmp78 [V98 ] ( 0, 0 ) long -> zero-ref "Inline stloc first use temp"
+;* V99 tmp79 [V99 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+;* V100 tmp80 [V100 ] ( 0, 0 ) long -> zero-ref "Inline return value spill temp"
+;* V101 tmp81 [V101 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+; V102 tmp82 [V102 ] ( 2, 8 ) long -> [rbp-0x80] do-not-enreg[X] addr-exposed ld-addr-op "Inline ldloca(s) first use temp"
+; V103 tmp83 [V103,T15] ( 2, 16 ) long -> rdi "impAppendStmt"
+;* V104 tmp84 [V104 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+;* V105 tmp85 [V105 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
+;* V106 tmp86 [V106 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+;* V107 tmp87 [V107 ] ( 0, 0 ) ubyte -> zero-ref "Inline return value spill temp"
+; V108 tmp88 [V108,T07] ( 3, 24 ) long -> rdx "dup spill"
+;* V109 tmp89 [V109 ] ( 0, 0 ) long -> zero-ref "Inline stloc first use temp"
+; V110 tmp90 [V110,T43] ( 3, 6 ) int -> rdx "Inline stloc first use temp"
+;* V111 tmp91 [V111 ] ( 0, 0 ) ubyte -> zero-ref "Inline return value spill temp"
+; V112 tmp92 [V112,T23] ( 3, 12 ) long -> rsi "Inline stloc first use temp"
+; V113 tmp93 [V113,T24] ( 3, 12 ) int -> rdx "Inline stloc first use temp"
+;* V114 tmp94 [V114 ] ( 0, 0 ) ubyte -> zero-ref "Inline return value spill temp"
+; V115 tmp95 [V115,T52] ( 3, 1.50) long -> rsi "Inline stloc first use temp"
+; V116 tmp96 [V116,T53] ( 3, 1.50) int -> rdx "Inline stloc first use temp"
+;* V117 tmp97 [V117 ] ( 0, 0 ) ubyte -> zero-ref "Inline return value spill temp"
+; V118 tmp98 [V118,T54] ( 3, 1.50) long -> rsi "Inline stloc first use temp"
+; V119 tmp99 [V119,T55] ( 3, 1.50) int -> rdx "Inline stloc first use temp"
+;* V120 tmp100 [V120 ] ( 0, 0 ) ubyte -> zero-ref "Inline return value spill temp"
+; V121 tmp101 [V121,T56] ( 3, 1.50) long -> rsi "Inline stloc first use temp"
+; V122 tmp102 [V122,T57] ( 3, 1.50) int -> rdx "Inline stloc first use temp"
+;* V123 tmp103 [V123 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
+;* V124 tmp104 [V124 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
+;* V125 tmp105 [V125 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
+;* V126 tmp106 [V126 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+;* V127 tmp107 [V127 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
+;* V128 tmp108 [V128 ] ( 0, 0 ) byref -> zero-ref "field V11._reference (fldOffset=0x0)" P-INDEP
+;* V129 tmp109 [V129 ] ( 0, 0 ) int -> zero-ref "field V11._length (fldOffset=0x8)" P-INDEP
+; V130 tmp110 [V130,T25] ( 3, 12 ) int -> rax "field V33.Item1 (fldOffset=0x0)" P-INDEP
+; V131 tmp111 [V131,T27] ( 3, 8.50) int -> rdx "field V33.Item2 (fldOffset=0x4)" P-INDEP
+; V132 tmp112 [V132,T45] ( 6, 3 ) int -> rax "field V39.Item1 (fldOffset=0x0)" P-INDEP
+; V133 tmp113 [V133,T46] ( 6, 3 ) int -> rdx "field V39.Item2 (fldOffset=0x4)" P-INDEP
+;* V134 tmp114 [V134 ] ( 0, 0 ) byref -> zero-ref "field V44._reference (fldOffset=0x0)" P-INDEP
+;* V135 tmp115 [V135 ] ( 0, 0 ) int -> zero-ref "field V44._length (fldOffset=0x8)" P-INDEP
+;* V136 tmp116 [V136 ] ( 0, 0 ) int -> zero-ref "field V54.Item1 (fldOffset=0x0)" P-DEP
+;* V137 tmp117 [V137 ] ( 0, 0 ) int -> zero-ref "field V54.Item2 (fldOffset=0x4)" P-DEP
+;* V138 tmp118 [V138 ] ( 0, 0 ) long -> zero-ref "field V66.Item1 (fldOffset=0x0)" P-DEP
+;* V139 tmp119 [V139 ] ( 0, 0 ) long -> zero-ref "field V66.Item2 (fldOffset=0x8)" P-DEP
+; V140 tmp120 [V140,T59] ( 2, 1 ) long -> rax "field V67.Item1 (fldOffset=0x0)" P-INDEP
+; V141 tmp121 [V141,T60] ( 2, 1 ) long -> rdx "field V67.Item2 (fldOffset=0x8)" P-INDEP
+;* V142 tmp122 [V142 ] ( 0, 0 ) byref -> zero-ref "field V71._reference (fldOffset=0x0)" P-INDEP
+;* V143 tmp123 [V143 ] ( 0, 0 ) int -> zero-ref "field V71._length (fldOffset=0x8)" P-INDEP
+;* V144 tmp124 [V144 ] ( 0, 0 ) byref -> zero-ref "field V96._reference (fldOffset=0x0)" P-INDEP
+;* V145 tmp125 [V145 ] ( 0, 0 ) int -> zero-ref "field V96._length (fldOffset=0x8)" P-INDEP
+; V146 cse0 [V146,T58] ( 3, 1.50) int -> rdi "CSE #14: conservative"
+; V147 cse1 [V147,T10] ( 4, 16 ) long -> rdi "CSE #03: moderate"
+; V148 cse2 [V148,T11] ( 4, 16 ) long -> rdi "CSE #10: moderate"
 ;
-; Lcl frame size = 56
+; Lcl frame size = 88
 
 G_M21333_IG01:
 push rbp
 push r15
 push r14
 push r13
 push r12
 push rbx
- sub rsp, 56
- lea rbp, [rsp+0x60]
+ sub rsp, 88
+ lea rbp, [rsp+0x80]
 mov rbx, rdi
 mov r15, rsi
-						;; size=25 bbWeight=1 PerfScore 7.25
+						;; size=28 bbWeight=1 PerfScore 7.25
 G_M21333_IG02:
- mov edi, dword ptr [rbx]
- sub edi, dword ptr [r15]
- shr edi, 16
- movsx r14, dil
+ mov eax, dword ptr [rbx]
+ sub eax, dword ptr [r15]
+ shr eax, 16
+ movsx r14, al
 xor r13d, r13d
- mov edi, dword ptr [r15+0x04]
- or edi, dword ptr [r15+0x0C]
- jne SHORT G_M21333_IG06
+ mov eax, dword ptr [r15+0x04]
+ or eax, dword ptr [r15+0x0C]
+ jne SHORT G_M21333_IG04
 						;; size=25 bbWeight=1 PerfScore 12.00
 G_M21333_IG03:
 mov r12d, dword ptr [r15+0x08]
 test r12d, r12d
- je G_M21333_IG45
- mov rdi, qword ptr [rbx+0x08]
- mov qword ptr [rbp-0x38], rdi
- mov edi, dword ptr [rbx+0x04]
- mov dword ptr [rbp-0x30], edi
- lea rdi, [rbp-0x38]
- mov esi, r12d
- mov rax, 0xD1FFAB1E ; code for System.Decimal+DecCalc:Div96By32(byref,uint):uint
- call [rax]System.Decimal+DecCalc:Div96By32(byref,uint):uint
-						;; size=46 bbWeight=0.50 PerfScore 6.62
+ je G_M21333_IG47
+ mov rax, qword ptr [rbx+0x08]
+ mov qword ptr [rbp-0x38], rax
+ mov eax, dword ptr [rbx+0x04]
+ mov dword ptr [rbp-0x30], eax
+ xor edx, edx
+ cmp dword ptr [rbp-0x30], 0
+ jne G_M21333_IG23
+ cmp dword ptr [rbp-0x34], r12d
+ jae G_M21333_IG24
+ mov edx, dword ptr [rbp-0x34]
+ xor eax, eax
+ mov dword ptr [rbp-0x34], eax
+ jmp G_M21333_IG25
+ align [0 bytes for IG37]
+						;; size=62 bbWeight=0.50 PerfScore 9.88
 G_M21333_IG04:
- mov dword ptr [rbp-0x40], eax
- test eax, eax
- jne G_M21333_IG25
-						;; size=11 bbWeight=4 PerfScore 9.00
-G_M21333_IG05:
- test r14d, r14d
- jge G_M21333_IG33
- mov ecx, r14d
- neg ecx
- cmp ecx, 9
- jge G_M21333_IG28
- jmp G_M21333_IG29
-						;; size=28 bbWeight=2 PerfScore 10.00
-G_M21333_IG06:
 mov ecx, dword ptr [r15+0x04]
 test ecx, ecx
- jne SHORT G_M21333_IG07
+ jne SHORT G_M21333_IG05
 mov ecx, dword ptr [r15+0x0C]
 						;; size=12 bbWeight=0.50 PerfScore 2.62
-G_M21333_IG07:
- xor eax, eax
- lzcnt eax, ecx
- mov rdi, qword ptr [rbx+0x08]
- shlx rdi, rdi, rax
- mov qword ptr [rbp-0x50], rdi
- mov edi, dword ptr [rbx+0x0C]
- mov esi, dword ptr [rbx+0x04]
- shl rsi, 32
- add rdi, rsi
- mov ecx, eax
- neg ecx
- add ecx, 32
- and ecx, 63
- shrx rdi, rdi, rcx
- mov qword ptr [rbp-0x48], rdi
- mov rdi, qword ptr [r15+0x08]
- shlx r12, rdi, rax
+G_M21333_IG05:
+ xor r8d, r8d
+ lzcnt r8d, ecx
+ mov rax, qword ptr [rbx+0x08]
+ shlx rax, rax, r8
+ mov qword ptr [rbp-0x50], rax
+ mov eax, dword ptr [rbx+0x0C]
+ mov edx, dword ptr [rbx+0x04]
+ shl rdx, 32
+ add rax, rdx
+ mov edi, r8d
+ neg edi
+ add edi, 32
+ and edi, 63
+ shrx rax, rax, rdi
+ mov qword ptr [rbp-0x48], rax
+ mov rax, qword ptr [r15+0x08]
+ shlx r12, rax, r8
 cmp dword ptr [r15+0x04], 0
- jne SHORT G_M21333_IG10
- xor edi, edi
- mov dword ptr [rbp-0x30], edi
- lea rdi, bword ptr [rbp-0x4C]
- mov rsi, r12
- mov rax, 0xD1FFAB1E ; code for System.Decimal+DecCalc:Div96By64(byref,ulong):uint
- call [rax]System.Decimal+DecCalc:Div96By64(byref,ulong):uint
- mov dword ptr [rbp-0x34], eax
- lea rdi, [rbp-0x50]
- mov rsi, r12
- mov rax, 0xD1FFAB1E ; code for System.Decimal+DecCalc:Div96By64(byref,ulong):uint
- call [rax]System.Decimal+DecCalc:Div96By64(byref,ulong):uint
- mov dword ptr [rbp-0x38], eax
-						;; size=116 bbWeight=0.50 PerfScore 15.38
-G_M21333_IG08:
+ jne SHORT G_M21333_IG08
+ xor eax, eax
+ mov dword ptr [rbp-0x30], eax
+ mov rax, qword ptr [rbp-0x50]
+ mov rdx, qword ptr [rbp-0x48]
+ div rdx:rax, r12
+ mov qword ptr [rbp-0x50], rdx
+ mov qword ptr [rbp-0x38], rax
+						;; size=94 bbWeight=0.50 PerfScore 42.88
+G_M21333_IG06:
 cmp qword ptr [rbp-0x50], 0
- jne G_M21333_IG19
+ jne G_M21333_IG17
 						;; size=11 bbWeight=4 PerfScore 12.00
-G_M21333_IG09:
+G_M21333_IG07:
 test r14d, r14d
- jge G_M21333_IG33
- mov eax, r14d
- neg eax
- cmp eax, 9
- jge G_M21333_IG21
- jmp G_M21333_IG22
-						;; size=28 bbWeight=2 PerfScore 10.00
-G_M21333_IG10:
+ jge G_M21333_IG35
+ mov r8d, r14d
+ neg r8d
+ cmp r8d, 9
+ jge G_M21333_IG19
+ jmp G_M21333_IG20
+						;; size=30 bbWeight=2 PerfScore 10.00
+G_M21333_IG08:
 mov qword ptr [rbp-0x60], r12
- mov edi, dword ptr [r15+0x0C]
- mov esi, dword ptr [r15+0x04]
- shl rsi, 32
- add rdi, rsi
- shrx rdi, rdi, rcx
+ mov esi, dword ptr [r15+0x0C]
+ mov eax, dword ptr [r15+0x04]
+ shl rax, 32
+ add rsi, rax
+ shrx rdi, rsi, rdi
 mov dword ptr [rbp-0x58], edi
 lea rdi, [rbp-0x50]
 lea rsi, [rbp-0x60]
 mov rax, 0xD1FFAB1E ; code for System.Decimal+DecCalc:Div128By96(byref,byref):uint
 call [rax]System.Decimal+DecCalc:Div128By96(byref,byref):uint
 mov edi, eax
 mov qword ptr [rbp-0x38], rdi
 xor edi, edi
 mov dword ptr [rbp-0x30], edi
 						;; size=58 bbWeight=0.50 PerfScore 7.00
-G_M21333_IG11:
+G_M21333_IG09:
 mov edi, dword ptr [rbp-0x48]
 or rdi, qword ptr [rbp-0x50]
- jne SHORT G_M21333_IG13
+ jne SHORT G_M21333_IG11
 						;; size=9 bbWeight=4 PerfScore 16.00
-G_M21333_IG12:
+G_M21333_IG10:
 test r14d, r14d
- jge G_M21333_IG33
- mov eax, r14d
- neg eax
- cmp eax, 9
- jge G_M21333_IG15
- jmp G_M21333_IG16
-						;; size=28 bbWeight=2 PerfScore 10.00
-G_M21333_IG13:
+ jge G_M21333_IG35
+ mov r8d, r14d
+ neg r8d
+ cmp r8d, 9
+ jge G_M21333_IG13
+ jmp G_M21333_IG14
+						;; size=30 bbWeight=2 PerfScore 10.00
+G_M21333_IG11:
 mov r13d, 1
 cmp r14d, 28
- je SHORT G_M21333_IG14
+ je SHORT G_M21333_IG12
 lea rdi, [rbp-0x38]
 mov esi, r14d
 mov rax, 0xD1FFAB1E ; code for System.Decimal+DecCalc:SearchScale(byref,int):int
 call [rax]System.Decimal+DecCalc:SearchScale(byref,int):int
- test eax, eax
- jne SHORT G_M21333_IG16
-						;; size=35 bbWeight=2 PerfScore 13.50
-G_M21333_IG14:
+ mov r8d, eax
+ test r8d, r8d
+ jne SHORT G_M21333_IG14
+						;; size=39 bbWeight=2 PerfScore 14.00
+G_M21333_IG12:
 cmp dword ptr [rbp-0x48], 0
- jl G_M21333_IG27
+ jl G_M21333_IG30
 mov ecx, dword ptr [rbp-0x4C]
 shr ecx, 31
- mov rax, qword ptr [rbp-0x50]
- add rax, rax
- mov qword ptr [rbp-0x50], rax
+ mov r8, qword ptr [rbp-0x50]
+ add r8, r8
+ mov qword ptr [rbp-0x50], r8
 mov edi, dword ptr [rbp-0x48]
 lea edi, [rcx+2*rdi]
 mov dword ptr [rbp-0x48], edi
 mov edi, dword ptr [rbp-0x48]
 cmp edi, dword ptr [rbp-0x58]
- ja G_M21333_IG27
+ ja G_M21333_IG30
 mov edi, dword ptr [rbp-0x48]
 cmp edi, dword ptr [rbp-0x58]
- jne G_M21333_IG33
+ jne G_M21333_IG35
 mov rdi, qword ptr [rbp-0x50]
 cmp rdi, qword ptr [rbp-0x60]
- ja G_M21333_IG27
+ ja G_M21333_IG30
 mov rdi, qword ptr [rbp-0x50]
 cmp rdi, qword ptr [rbp-0x60]
- jne G_M21333_IG33
+ jne G_M21333_IG35
 test byte ptr [rbp-0x38], 1
- je G_M21333_IG33
- jmp G_M21333_IG27
+ je G_M21333_IG35
+ jmp G_M21333_IG30
 						;; size=103 bbWeight=0.50 PerfScore 15.12
-G_M21333_IG15:
- mov eax, 9
-						;; size=5 bbWeight=2 PerfScore 0.50
-G_M21333_IG16:
- cmp eax, 10
- jae G_M21333_IG46
- mov edi, eax
+G_M21333_IG13:
+ mov r8d, 9
+						;; size=6 bbWeight=2 PerfScore 0.50
+G_M21333_IG14:
+ cmp r8d, 10
+ jae G_M21333_IG48
+ mov edi, r8d
 mov rsi, 0xD1FFAB1E ; static handle
 mov r12d, dword ptr [rsi+4*rdi]
- add r14d, eax
+ add r14d, r8d
 lea rdi, [rbp-0x38]
 mov esi, r12d
 mov rax, 0xD1FFAB1E ; code for System.Decimal+DecCalc:IncreaseScale(byref,uint):uint
 call [rax]System.Decimal+DecCalc:IncreaseScale(byref,uint):uint
 test eax, eax
- jne G_M21333_IG44
- lea rdi, [rbp-0x50]
- mov esi, r12d
- mov rax, 0xD1FFAB1E ; code for System.Decimal+DecCalc:IncreaseScale(byref,uint):uint
- call [rax]System.Decimal+DecCalc:IncreaseScale(byref,uint):uint
- mov dword ptr [rbp-0x44], eax
+ jne G_M21333_IG46
+ mov rdx, qword ptr [rbp-0x50]
+ mov edi, r12d
+ lea rsi, [rbp-0x80]
+ mulx rdi, rax, rdi
+ mov qword ptr [rsi], rax
+ mov rsi, qword ptr [rbp-0x80]
+ mov qword ptr [rbp-0x50], rsi
+ mov esi, dword ptr [rbp-0x48]
+ mov eax, r12d
+ imul rsi, rax
+ add rdi, rsi
+ mov qword ptr [rbp-0x48], rdi
 lea rdi, [rbp-0x50]
 lea rsi, [rbp-0x60]
 mov rax, 0xD1FFAB1E ; code for System.Decimal+DecCalc:Div128By96(byref,byref):uint
 call [rax]System.Decimal+DecCalc:Div128By96(byref,byref):uint
 mov ecx, eax
 mov edx, ecx
 add rdx, qword ptr [rbp-0x38]
 mov qword ptr [rbp-0x38], rdx
 mov edi, ecx
 cmp rdx, rdi
- jae G_M21333_IG11
-						;; size=120 bbWeight=4 PerfScore 94.00
-G_M21333_IG17:
+ jae G_M21333_IG09
+						;; size=144 bbWeight=4 PerfScore 123.00
+G_M21333_IG15:
 mov edx, dword ptr [rbp-0x30]
 inc edx
 mov dword ptr [rbp-0x30], edx
 test edx, edx
- jne G_M21333_IG11
+ jne G_M21333_IG09
 						;; size=16 bbWeight=2 PerfScore 7.00
-G_M21333_IG18:
+G_M21333_IG16:
 mov rdx, qword ptr [rbp-0x50]
 or rdx, qword ptr [rbp-0x48]
 setne dl
 movzx rdx, dl
 lea rdi, [rbp-0x38]
 mov esi, r14d
 mov rax, 0xD1FFAB1E ; code for System.Decimal+DecCalc:OverflowUnscale(byref,int,ubyte):int
 call [rax]System.Decimal+DecCalc:OverflowUnscale(byref,int,ubyte):int
 mov r14d, eax
- jmp G_M21333_IG33
- align [0 bytes for IG35]
+ jmp G_M21333_IG35
 						;; size=41 bbWeight=0.50 PerfScore 5.25
-G_M21333_IG19:
+G_M21333_IG17:
 mov r13d, 1
 cmp r14d, 28
- je SHORT G_M21333_IG20
+ je SHORT G_M21333_IG18
 lea rdi, [rbp-0x38]
 mov esi, r14d
 mov rax, 0xD1FFAB1E ; code for System.Decimal+DecCalc:SearchScale(byref,int):int
 call [rax]System.Decimal+DecCalc:SearchScale(byref,int):int
- test eax, eax
- jne SHORT G_M21333_IG22
-						;; size=35 bbWeight=2 PerfScore 13.50
-G_M21333_IG20:
- mov rax, qword ptr [rbp-0x50]
- test rax, rax
- jl G_M21333_IG27
- add rax, rax
- cmp rax, r12
- ja G_M21333_IG27
- jne G_M21333_IG33
+ mov r8d, eax
+ test r8d, r8d
+ jne SHORT G_M21333_IG20
+						;; size=39 bbWeight=2 PerfScore 14.00
+G_M21333_IG18:
+ mov r8, qword ptr [rbp-0x50]
+ test r8, r8
+ jl G_M21333_IG30
+ add r8, r8
+ cmp r8, r12
+ ja G_M21333_IG30
+ jne G_M21333_IG35
 test byte ptr [rbp-0x38], 1
- je G_M21333_IG33
- jmp G_M21333_IG27
+ je G_M21333_IG35
+ jmp G_M21333_IG30
 						;; size=46 bbWeight=0.50 PerfScore 4.88
-G_M21333_IG21:
- mov eax, 9
-						;; size=5 bbWeight=2 PerfScore 0.50
-G_M21333_IG22:
- cmp eax, 10
- jae G_M21333_IG46
+G_M21333_IG19:
+ mov r8d, 9
+						;; size=6 bbWeight=2 PerfScore 0.50
+G_M21333_IG20:
+ cmp r8d, 10
+ jae G_M21333_IG48
+ mov edx, r8d
+ mov rdi, 0xD1FFAB1E ; static handle
+ mov edx, dword ptr [rdi+4*rdx]
+ mov eax, edx
+ add r14d, r8d
+ mov rdx, qword ptr [rbp-0x38]
 mov edi, eax
- mov rsi, 0xD1FFAB1E ; static handle
- mov edi, dword ptr [rsi+4*rdi]
- mov ecx, edi
- add r14d, eax
- lea rdi, [rbp-0x38]
- mov dword ptr [rbp-0x3C], ecx
- mov esi, ecx
- mov rax, 0xD1FFAB1E ; code for System.Decimal+DecCalc:IncreaseScale(byref,uint):uint
- call [rax]System.Decimal+DecCalc:IncreaseScale(byref,uint):uint
- test eax, eax
- jne G_M21333_IG44
- mov edi, dword ptr [rbp-0x50]
- mov eax, dword ptr [rbp-0x3C]
- mov esi, eax
- imul rdi, rsi
- mov dword ptr [rbp-0x50], edi
- mov esi, dword ptr [rbp-0x4C]
- imul rsi, rax
- shr rdi, 32
- add rdi, rsi
- mov qword ptr [rbp-0x4C], rdi
+ lea rsi, [rbp-0x70]
+ mulx rdx, rax, rdi
+ mov qword ptr [rsi], rax
+ mov rsi, qword ptr [rbp-0x70]
+ mov qword ptr [rbp-0x38], rsi
+ mov esi, dword ptr [rbp-0x30]
+ imul rsi, rdi
+ add rdx, rsi
+ mov dword ptr [rbp-0x30], edx
+ shr rdx, 32
+ test edx, edx
+ jne G_M21333_IG46
+ mov rdx, qword ptr [rbp-0x50]
+ lea rsi, [rbp-0x78]
+ mulx rdi, rax, rdi
+ mov qword ptr [rsi], rax
+ mov rsi, qword ptr [rbp-0x78]
+ mov dword ptr [rbp-0x48], edi
+ mov qword ptr [rbp-0x50], rsi
 lea rdi, [rbp-0x50]
 mov rsi, r12
 mov rax, 0xD1FFAB1E ; code for System.Decimal+DecCalc:Div96By64(byref,ulong):uint
 call [rax]System.Decimal+DecCalc:Div96By64(byref,ulong):uint
 mov ecx, eax
 mov edx, ecx
 add rdx, qword ptr [rbp-0x38]
 mov qword ptr [rbp-0x38], rdx
 mov edi, ecx
 cmp rdx, rdi
- jae G_M21333_IG08
-						;; size=133 bbWeight=4 PerfScore 118.00
-G_M21333_IG23:
+ jae G_M21333_IG06
+						;; size=151 bbWeight=4 PerfScore 142.00
+G_M21333_IG21:
 mov edx, dword ptr [rbp-0x30]
 inc edx
 mov dword ptr [rbp-0x30], edx
 test edx, edx
- jne G_M21333_IG08
+ jne G_M21333_IG06
 						;; size=16 bbWeight=2 PerfScore 7.00
-G_M21333_IG24:
+G_M21333_IG22:
 cmp qword ptr [rbp-0x50], 0
 setne dl
 movzx rdx, dl
 lea rdi, [rbp-0x38]
 mov esi, r14d
 mov rax, 0xD1FFAB1E ; code for System.Decimal+DecCalc:OverflowUnscale(byref,int,ubyte):int
 call [rax]System.Decimal+DecCalc:OverflowUnscale(byref,int,ubyte):int
 mov r14d, eax
- jmp G_M21333_IG33
+ jmp G_M21333_IG35
 						;; size=38 bbWeight=0.50 PerfScore 4.75
+G_M21333_IG23:
+ mov eax, dword ptr [rbp-0x30]
+ xor edx, edx
+ div edx:eax, r12d
+ mov dword ptr [rbp-0x30], eax
+						;; size=11 bbWeight=0.50 PerfScore 13.62
+G_M21333_IG24:
+ mov eax, dword ptr [rbp-0x34]
+ div edx:eax, r12d
+ mov dword ptr [rbp-0x34], eax
+						;; size=9 bbWeight=0.50 PerfScore 13.50
 G_M21333_IG25:
+ mov eax, dword ptr [rbp-0x38]
+ div edx:eax, r12d
+ mov dword ptr [rbp-0x38], eax
+ mov eax, edx
+						;; size=11 bbWeight=0.50 PerfScore 13.62
+G_M21333_IG26:
+ mov dword ptr [rbp-0x40], eax
+ test eax, eax
+ jne SHORT G_M21333_IG28
+						;; size=7 bbWeight=4 PerfScore 9.00
+G_M21333_IG27:
+ test r14d, r14d
+ jge G_M21333_IG35
+ mov r8d, r14d
+ neg r8d
+ cmp r8d, 9
+ jge G_M21333_IG31
+ jmp G_M21333_IG32
+						;; size=30 bbWeight=2 PerfScore 10.00
+G_M21333_IG28:
 mov r13d, 1
 cmp r14d, 28
- je SHORT G_M21333_IG26
+ je SHORT G_M21333_IG29
 lea rdi, [rbp-0x38]
 mov esi, r14d
 mov rcx, 0xD1FFAB1E ; code for System.Decimal+DecCalc:SearchScale(byref,int):int
 call [rcx]System.Decimal+DecCalc:SearchScale(byref,int):int
- test eax, eax
- jne SHORT G_M21333_IG30
-						;; size=35 bbWeight=2 PerfScore 13.50
-G_M21333_IG26:
+ mov r8d, eax
+ test r8d, r8d
+ jne SHORT G_M21333_IG32
+						;; size=39 bbWeight=2 PerfScore 14.00
+G_M21333_IG29:
 mov eax, dword ptr [rbp-0x40]
 lea ecx, [rax+rax]
 cmp ecx, eax
- jb SHORT G_M21333_IG27
+ jb SHORT G_M21333_IG30
 cmp ecx, r12d
- jb G_M21333_IG33
- ja SHORT G_M21333_IG27
+ jb G_M21333_IG35
+ ja SHORT G_M21333_IG30
 test byte ptr [rbp-0x38], 1
- je G_M21333_IG33
+ je G_M21333_IG35
 						;; size=31 bbWeight=0.50 PerfScore 4.00
-G_M21333_IG27:
+G_M21333_IG30:
 mov rdi, qword ptr [rbp-0x38]
 inc rdi
 mov qword ptr [rbp-0x38], rdi
 test rdi, rdi
- jne G_M21333_IG33
+ jne G_M21333_IG35
 mov edi, dword ptr [rbp-0x30]
 inc edi
 mov dword ptr [rbp-0x30], edi
 test edi, edi
- jne G_M21333_IG33
+ jne G_M21333_IG35
 lea rdi, [rbp-0x38]
 mov esi, r14d
 mov edx, 1
 mov rax, 0xD1FFAB1E ; code for System.Decimal+DecCalc:OverflowUnscale(byref,int,ubyte):int
 call [rax]System.Decimal+DecCalc:OverflowUnscale(byref,int,ubyte):int
 mov r14d, eax
- jmp G_M21333_IG33
+ jmp G_M21333_IG35
 						;; size=68 bbWeight=0.50 PerfScore 6.75
-G_M21333_IG28:
- mov ecx, 9
-						;; size=5 bbWeight=2 PerfScore 0.50
-G_M21333_IG29:
- mov edi, ecx
- mov eax, edi
-						;; size=4 bbWeight=2 PerfScore 1.00
-G_M21333_IG30:
- cmp eax, 10
- jae G_M21333_IG46
- mov edi, eax
- mov rsi, 0xD1FFAB1E ; static handle
- mov edi, dword ptr [rsi+4*rdi]
- mov ecx, edi
- add r14d, eax
- lea rdi, [rbp-0x38]
- mov dword ptr [rbp-0x3C], ecx
- mov esi, ecx
- mov rax, 0xD1FFAB1E ; code for System.Decimal+DecCalc:IncreaseScale(byref,uint):uint
- call [rax]System.Decimal+DecCalc:IncreaseScale(byref,uint):uint
- test eax, eax
- jne G_M21333_IG44
- mov edi, dword ptr [rbp-0x40]
- mov esi, dword ptr [rbp-0x3C]
- imul rdi, rsi
- mov esi, r12d
- mov rax, rdi
- xor edx, edx
- div rdx:rax, rsi
- mov edx, eax
- imul edx, r12d
- sub edi, edx
- mov edx, edi
- mov edi, eax
- add rdi, qword ptr [rbp-0x38]
- mov qword ptr [rbp-0x38], rdi
- mov esi, eax
- cmp rdi, rsi
- mov eax, edx
- jae G_M21333_IG04
-						;; size=112 bbWeight=4 PerfScore 336.00
 G_M21333_IG31:
- mov edx, dword ptr [rbp-0x30]
- inc edx
+ mov r8d, 9
+						;; size=6 bbWeight=2 PerfScore 0.50
+G_M21333_IG32:
+ cmp r8d, 10
+ jae G_M21333_IG48
+ mov edx, r8d
+ mov rdi, 0xD1FFAB1E ; static handle
+ mov edx, dword ptr [rdi+4*rdx]
+ mov dword ptr [rbp-0x3C], edx
+ add r14d, r8d
+ mov rdx, qword ptr [rbp-0x38]
+ mov edi, dword ptr [rbp-0x3C]
+ lea rsi, [rbp-0x68]
+ mulx rdx, rcx, rdi
+ mov qword ptr [rsi], rcx
+ mov rsi, qword ptr [rbp-0x68]
+ mov qword ptr [rbp-0x38], rsi
+ mov esi, dword ptr [rbp-0x30]
+ imul rsi, rdi
+ add rdx, rsi
 mov dword ptr [rbp-0x30], edx
+ shr rdx, 32
 test edx, edx
- jne G_M21333_IG04
+ jne G_M21333_IG46
+ mov eax, dword ptr [rbp-0x40]
+ imul rax, rdi
+ mov rdx, rax
+ shr rdx, 32
+ div edx:eax, r12d
+ mov edi, edx
+ mov esi, eax
+ add rsi, qword ptr [rbp-0x38]
+ mov qword ptr [rbp-0x38], rsi
+ cmp rsi, rax
+ mov eax, edi
+ jae G_M21333_IG26
+						;; size=124 bbWeight=4 PerfScore 213.00
+G_M21333_IG33:
+ mov edi, dword ptr [rbp-0x30]
+ inc edi
+ mov dword ptr [rbp-0x30], edi
+ test edi, edi
+ jne G_M21333_IG26
 						;; size=16 bbWeight=2 PerfScore 7.00
-G_M21333_IG32:
- test eax, eax
+G_M21333_IG34:
+ test edx, edx
 setne dl
 movzx rdx, dl
 lea rdi, [rbp-0x38]
 mov esi, r14d
 mov rax, 0xD1FFAB1E ; code for System.Decimal+DecCalc:OverflowUnscale(byref,int,ubyte):int
 call [rax]System.Decimal+DecCalc:OverflowUnscale(byref,int,ubyte):int
 mov r14d, eax
 						;; size=30 bbWeight=0.50 PerfScore 2.88
-G_M21333_IG33:
+G_M21333_IG35:
 test r13d, r13d
- je SHORT G_M21333_IG37
+ je SHORT G_M21333_IG39
 						;; size=5 bbWeight=1 PerfScore 1.25
-G_M21333_IG34:
+G_M21333_IG36:
 mov ecx, dword ptr [rbp-0x38]
 mov rdi, qword ptr [rbp-0x34]
- jmp SHORT G_M21333_IG36
+ jmp SHORT G_M21333_IG38
 						;; size=9 bbWeight=0.50 PerfScore 2.00
-G_M21333_IG35:
+G_M21333_IG37:
 mov rdi, rsi
 mov ecx, edx
 add r14d, -8
 						;; size=9 bbWeight=4 PerfScore 3.00
-G_M21333_IG36:
+G_M21333_IG38:
 test cl, cl
- jne SHORT G_M21333_IG38
+ jne SHORT G_M21333_IG40
 cmp r14d, 8
- jl SHORT G_M21333_IG38
+ jl SHORT G_M21333_IG40
 mov rdx, 0xD1FFAB1E
 mov rax, rdi
 mul rdx:rax, rdx
 mov rsi, rdx
 shr rsi, 26
 imul rax, rsi, 0xD1FFAB1E
 mov rdx, rdi
 sub rdx, rax
 shl rdx, 32
 mov eax, ecx
 add rdx, rax
 mov r8, 0xD1FFAB1E
 mov rax, rdx
 mul rdx:rax, r8
 shr rdx, 26
 imul eax, edx, 0xD1FFAB1E
 cmp ecx, eax
- jne SHORT G_M21333_IG38
- jmp SHORT G_M21333_IG35
+ jne SHORT G_M21333_IG40
+ jmp SHORT G_M21333_IG37
 						;; size=87 bbWeight=4 PerfScore 78.00
-G_M21333_IG37:
+G_M21333_IG39:
 mov rcx, qword ptr [rbp-0x38]
 mov qword ptr [rbx+0x08], rcx
 mov edi, dword ptr [rbp-0x30]
 mov dword ptr [rbx+0x04], edi
- jmp G_M21333_IG42
+ jmp G_M21333_IG44
 						;; size=19 bbWeight=0.50 PerfScore 3.00
-G_M21333_IG38:
+G_M21333_IG40:
 test cl, 15
- jne SHORT G_M21333_IG39
+ jne SHORT G_M21333_IG41
 cmp r14d, 4
- jl SHORT G_M21333_IG39
+ jl SHORT G_M21333_IG41
 mov rdx, 0xD1FFAB1E
 mov rax, rdi
 mul rdx:rax, rdx
 mov rsi, rdx
 shr rsi, 11
 imul rax, rsi, 0x2710
 mov rdx, rdi
 sub rdx, rax
 shl rdx, 32
 mov eax, ecx
 add rdx, rax
 mov r8, 0xD1FFAB1E
 mov rax, rdx
 mul rdx:rax, r8
 shr rdx, 11
 imul eax, edx, 0x2710
 cmp ecx, eax
- jne SHORT G_M21333_IG39
+ jne SHORT G_M21333_IG41
 mov rdi, rsi
 mov ecx, edx
 add r14d, -4
 						;; size=95 bbWeight=0.50 PerfScore 9.12
-G_M21333_IG39:
+G_M21333_IG41:
 test cl, 3
- jne SHORT G_M21333_IG40
+ jne SHORT G_M21333_IG42
 cmp r14d, 2
- jl SHORT G_M21333_IG40
+ jl SHORT G_M21333_IG42
 mov rdx, 0xD1FFAB1E
 mov rax, rdi
 shr rax, 2
 mul rdx:rax, rdx
 mov rsi, rdx
 shr rsi, 2
 imul rax, rsi, 100
 mov rdx, rdi
 sub rdx, rax
 shl rdx, 32
 mov eax, ecx
 add rax, rdx
 mov rdx, 0xD1FFAB1E
 shr rax, 2
 mul rdx:rax, rdx
 shr rdx, 2
 imul eax, edx, 100
 cmp ecx, eax
- jne SHORT G_M21333_IG40
+ jne SHORT G_M21333_IG42
 mov rdi, rsi
 mov ecx, edx
 add r14d, -2
 						;; size=94 bbWeight=0.50 PerfScore 9.50
-G_M21333_IG40:
+G_M21333_IG42:
 test cl, 1
- jne SHORT G_M21333_IG41
+ jne SHORT G_M21333_IG43
 test r14d, r14d
- jle SHORT G_M21333_IG41
+ jle SHORT G_M21333_IG43
 mov rdx, 0xD1FFAB1E
 mov rax, rdi
 mul rdx:rax, rdx
 mov rsi, rdx
 shr rsi, 3
 lea rax, [rsi+4*rsi]
 add rax, rax
 mov rdx, rdi
 sub rdx, rax
 shl rdx, 32
 mov eax, ecx
 add rdx, rax
 mov r8, 0xD1FFAB1E
 mov rax, rdx
 mul rdx:rax, r8
 shr rdx, 3
 lea eax, [rdx+4*rdx]
 add eax, eax
 cmp ecx, eax
- jne SHORT G_M21333_IG41
+ jne SHORT G_M21333_IG43
 mov rdi, rsi
 mov ecx, edx
 dec r14d
 						;; size=92 bbWeight=0.50 PerfScore 7.88
-G_M21333_IG41:
+G_M21333_IG43:
 mov dword ptr [rbx+0x08], ecx
 mov dword ptr [rbx+0x0C], edi
 shr rdi, 32
 mov dword ptr [rbx+0x04], edi
 						;; size=13 bbWeight=0.50 PerfScore 1.75
-G_M21333_IG42:
+G_M21333_IG44:
 mov eax, dword ptr [rbx]
 xor eax, dword ptr [r15]
 and eax, 0xD1FFAB1E
 mov ecx, r14d
 shl ecx, 16
 or eax, ecx
 mov dword ptr [rbx], eax
 						;; size=20 bbWeight=1 PerfScore 7.25
-G_M21333_IG43:
- add rsp, 56
+G_M21333_IG45:
+ add rsp, 88
 pop rbx
 pop r12
 pop r13
 pop r14
 pop r15
 pop rbp
 ret 
 						;; size=15 bbWeight=1 PerfScore 4.25
-G_M21333_IG44:
+G_M21333_IG46:
 mov rax, 0xD1FFAB1E ; code for System.SR:get_Overflow_Decimal():System.String
 call [rax]System.SR:get_Overflow_Decimal():System.String
 mov rdi, rax
 mov rax, 0xD1FFAB1E ; code for System.Number:ThrowOverflowException(System.String)
 call [rax]System.Number:ThrowOverflowException(System.String)
 int3 
 						;; size=28 bbWeight=0 PerfScore 0.00
-G_M21333_IG45:
+G_M21333_IG47:
 mov rdi, 0xD1FFAB1E ; System.DivideByZeroException
 call CORINFO_HELP_NEWSFAST
 mov rbx, rax
 mov rdi, rbx
 mov rax, 0xD1FFAB1E ; code for System.DivideByZeroException:.ctor():this
 call [rax]System.DivideByZeroException:.ctor():this
 mov rdi, rbx
 call CORINFO_HELP_THROW
 int3 
 						;; size=42 bbWeight=0 PerfScore 0.00
-G_M21333_IG46:
+G_M21333_IG48:
 call CORINFO_HELP_RNGCHKFAIL
 int3 
 						;; size=6 bbWeight=0 PerfScore 0.00
 
-; Total bytes of code 1825, prolog size 25, PerfScore 900.50, instruction count 489, allocated bytes for code 1825 (MethodHash=142cacaa) for method System.Decimal+DecCalc:VarDecDiv(byref,byref) (FullOpts)
+; Total bytes of code 1920, prolog size 28, PerfScore 902.50, instruction count 513, allocated bytes for code 1920 (MethodHash=142cacaa) for method System.Decimal+DecCalc:VarDecDiv(byref,byref) (FullOpts)

4 (13.79 % of base) - System.Decimal+DecCalc:IncreaseScale64(byref,uint)

 ; Assembly listing for method System.Decimal+DecCalc:IncreaseScale64(byref,uint) (FullOpts)
 ; Emitting BLENDED_CODE for X64 with AVX512 - Unix
 ; FullOpts code
 ; optimized code
 ; rsp based frame
 ; partially interruptible
 ; No PGO data
-; 0 inlinees with PGO data; 3 single block inlinees; 0 inlinees without PGO data
+; 0 inlinees with PGO data; 3 single block inlinees; 1 inlinees without PGO data
 ; Final local variable assignments
 ;
-; V00 arg0 [V00,T00] ( 6, 6 ) byref -> rdi single-def
+; V00 arg0 [V00,T00] ( 5, 5 ) byref -> rdi single-def
 ; V01 arg1 [V01,T01] ( 3, 3 ) int -> rsi single-def
-; V02 loc0 [V02,T02] ( 3, 3 ) long -> rax 
+; V02 loc0 [V02,T03] ( 2, 2 ) long -> rcx ld-addr-op
 ;# V03 OutArgs [V03 ] ( 1, 1 ) struct ( 0) [rsp+0x00] do-not-enreg[XS] addr-exposed "OutgoingArgSpace"
-;* V04 tmp1 [V04 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
-;* V05 tmp2 [V05 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
-; V06 cse0 [V06,T03] ( 3, 3 ) long -> rcx "CSE #01: aggressive"
+;* V04 tmp1 [V04 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+;* V05 tmp2 [V05 ] ( 0, 0 ) long -> zero-ref "Inline return value spill temp"
+;* V06 tmp3 [V06 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+; V07 tmp4 [V07 ] ( 2, 2 ) long -> [rsp+0x00] do-not-enreg[X] addr-exposed ld-addr-op "Inline ldloca(s) first use temp"
+; V08 tmp5 [V08,T02] ( 2, 4 ) long -> rax "impAppendStmt"
+;* V09 tmp6 [V09 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
 ;
-; Lcl frame size = 0
+; Lcl frame size = 8
 
 G_M64539_IG01:
-						;; size=0 bbWeight=1 PerfScore 0.00
+ push rax
+						;; size=1 bbWeight=1 PerfScore 1.00
 G_M64539_IG02:
- mov eax, dword ptr [rdi]
- mov ecx, esi
- imul rax, rcx
- mov dword ptr [rdi], eax
- mov edx, dword ptr [rdi+0x04]
- imul rcx, rdx
- shr rax, 32
- add rax, rcx
- mov qword ptr [rdi+0x04], rax
-						;; size=28 bbWeight=1 PerfScore 11.00
+ mov rdx, qword ptr [rdi]
+ mov eax, esi
+ lea rcx, [rsp]
+ mulx rax, rsi, rax
+ mov qword ptr [rcx], rsi
+ mov rcx, qword ptr [rsp]
+ mov dword ptr [rdi+0x08], eax
+ mov qword ptr [rdi], rcx
+						;; size=27 bbWeight=1 PerfScore 9.75
 G_M64539_IG03:
+ add rsp, 8
 ret 
-						;; size=1 bbWeight=1 PerfScore 1.00
+						;; size=5 bbWeight=1 PerfScore 1.25
 
-; Total bytes of code 29, prolog size 0, PerfScore 12.00, instruction count 10, allocated bytes for code 29 (MethodHash=beae03e4) for method System.Decimal+DecCalc:IncreaseScale64(byref,uint) (FullOpts)
+; Total bytes of code 33, prolog size 1, PerfScore 12.00, instruction count 11, allocated bytes for code 33 (MethodHash=beae03e4) for method System.Decimal+DecCalc:IncreaseScale64(byref,uint) (FullOpts)

[MihuBot]

Top method improvements

-158 (-19.73 % of base) - System.Decimal+DecCalc:VarDecMul(byref,byref)

 ; Assembly listing for method System.Decimal+DecCalc:VarDecMul(byref,byref) (FullOpts)
 ; Emitting BLENDED_CODE for X64 with AVX512 - Unix
 ; FullOpts code
 ; optimized code
 ; rbp based frame
 ; fully interruptible
 ; No PGO data
-; 0 inlinees with PGO data; 69 single block inlinees; 0 inlinees without PGO data
+; 0 inlinees with PGO data; 52 single block inlinees; 5 inlinees without PGO data
 ; Final local variable assignments
 ;
-; V00 arg0 [V00,T02] ( 29, 17 ) byref -> rbx single-def
-; V01 arg1 [V01,T03] ( 25, 14 ) byref -> r15 single-def
-; V02 loc0 [V02,T06] ( 13, 7 ) int -> registers 
-; V03 loc1 [V03,T00] ( 47, 23.50) long -> registers 
-; V04 loc2 [V04,T01] ( 11, 19.50) int -> rsi 
-; V05 loc3 [V05 ] ( 17, 12 ) struct (24) [rbp-0x28] do-not-enreg[XSF] addr-exposed ld-addr-op <System.Decimal+DecCalc+Buf24>
-;* V06 loc4 [V06,T16] ( 0, 0 ) long -> zero-ref 
-; V07 loc5 [V07,T08] ( 6, 3 ) long -> registers 
-; V08 loc6 [V08,T07] ( 7, 3.50) long -> rsi 
+; V00 arg0 [V00,T01] ( 22, 13.50) byref -> rbx single-def
+; V01 arg1 [V01,T02] ( 17, 10 ) byref -> r15 single-def
+; V02 loc0 [V02,T04] ( 13, 7 ) int -> [rbp-0x14] 
+; V03 loc1 [V03,T03] ( 16, 8 ) long -> registers ld-addr-op
+; V04 loc2 [V04,T00] ( 11, 19.50) int -> rsi 
+; V05 loc3 [V05 ] ( 13, 10 ) struct (24) [rbp-0x30] do-not-enreg[XSF] addr-exposed ld-addr-op <System.Decimal+DecCalc+Buf24>
+;* V06 loc4 [V06,T23] ( 0, 0 ) long -> zero-ref 
+; V07 loc5 [V07,T07] ( 6, 3 ) long -> registers 
+; V08 loc6 [V08,T06] ( 7, 3.50) long -> rsi 
 ;* V09 loc7 [V09 ] ( 0, 0 ) long -> zero-ref 
 ;* V10 loc8 [V10 ] ( 0, 0 ) struct (16) zero-ref ld-addr-op <System.ReadOnlySpan`1[ulong]>
-; V11 loc9 [V11,T04] ( 17, 8.50) long -> rdi 
-; V12 loc10 [V12,T05] ( 14, 7 ) int -> rsi 
-;# V13 OutArgs [V13 ] ( 1, 1 ) struct ( 0) [rsp+0x00] do-not-enreg[XS] addr-exposed "OutgoingArgSpace"
-;* V14 tmp1 [V14 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
-;* V15 tmp2 [V15 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
-;* V16 tmp3 [V16 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
-;* V17 tmp4 [V17 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
-;* V18 tmp5 [V18 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
-;* V19 tmp6 [V19 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
-;* V20 tmp7 [V20 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
-;* V21 tmp8 [V21 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
-;* V22 tmp9 [V22 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
-;* V23 tmp10 [V23 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
-;* V24 tmp11 [V24 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
-;* V25 tmp12 [V25 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
-;* V26 tmp13 [V26 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
-;* V27 tmp14 [V27 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
-;* V28 tmp15 [V28 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
-;* V29 tmp16 [V29 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
-;* V30 tmp17 [V30 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
-;* V31 tmp18 [V31 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
-;* V32 tmp19 [V32 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
-;* V33 tmp20 [V33 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
-;* V34 tmp21 [V34 ] ( 0, 0 ) struct (16) zero-ref "dup spill" <System.ValueTuple`2[ulong,ulong]>
-;* V35 tmp22 [V35 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
-;* V36 tmp23 [V36 ] ( 0, 0 ) struct (16) zero-ref "ReadOnlySpan<T> for CreateSpan<T>" <System.ReadOnlySpan`1[ulong]>
-; V37 tmp24 [V37,T09] ( 5, 2.50) long -> rax "Inline stloc first use temp"
-;* V38 tmp25 [V38 ] ( 0, 0 ) struct (16) zero-ref ld-addr-op "NewObj constructor temp" <System.ValueTuple`2[ulong,ulong]>
-;* V39 tmp26 [V39 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
-;* V40 tmp27 [V40 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
-;* V41 tmp28 [V41 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
-;* V42 tmp29 [V42 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
-;* V43 tmp30 [V43 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
-;* V44 tmp31 [V44 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
-;* V45 tmp32 [V45 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
-;* V46 tmp33 [V46 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
-;* V47 tmp34 [V47 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
-;* V48 tmp35 [V48 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
-;* V49 tmp36 [V49 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
-;* V50 tmp37 [V50 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
-;* V51 tmp38 [V51 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
-;* V52 tmp39 [V52 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
-;* V53 tmp40 [V53 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
-;* V54 tmp41 [V54 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
-; V55 tmp42 [V55,T10] ( 2, 2 ) long -> rax "Inlining Arg"
-;* V56 tmp43 [V56 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
-;* V57 tmp44 [V57 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
-;* V58 tmp45 [V58 ] ( 0, 0 ) byref -> zero-ref single-def "field V10._reference (fldOffset=0x0)" P-INDEP
-;* V59 tmp46 [V59 ] ( 0, 0 ) int -> zero-ref "field V10._length (fldOffset=0x8)" P-INDEP
-;* V60 tmp47 [V60 ] ( 0, 0 ) long -> zero-ref "field V34.Item1 (fldOffset=0x0)" P-INDEP
-;* V61 tmp48 [V61 ] ( 0, 0 ) long -> zero-ref "field V34.Item2 (fldOffset=0x8)" P-INDEP
-;* V62 tmp49 [V62 ] ( 0, 0 ) byref -> zero-ref single-def "field V36._reference (fldOffset=0x0)" P-INDEP
-;* V63 tmp50 [V63 ] ( 0, 0 ) int -> zero-ref "field V36._length (fldOffset=0x8)" P-INDEP
-;* V64 tmp51 [V64 ] ( 0, 0 ) long -> zero-ref "field V38.Item1 (fldOffset=0x0)" P-INDEP
-; V65 tmp52 [V65,T11] ( 3, 1.50) long -> rdi "field V38.Item2 (fldOffset=0x8)" P-INDEP
-; V66 cse0 [V66,T12] ( 3, 1.50) int -> rcx "CSE #06: moderate"
-; V67 cse1 [V67,T13] ( 3, 1.50) int -> rsi "CSE #11: moderate"
-; V68 cse2 [V68,T14] ( 3, 1.50) int -> rdi "CSE #13: moderate"
-; V69 cse3 [V69,T15] ( 3, 1.50) int -> rcx "CSE #14: moderate"
+; V11 loc9 [V11,T21] ( 2, 1 ) long -> rsi ld-addr-op
+; V12 loc10 [V12,T22] ( 2, 1 ) long -> rdi ld-addr-op
+; V13 loc11 [V13,T08] ( 6, 3 ) long -> rax 
+; V14 loc12 [V14,T05] ( 12, 6 ) long -> rsi 
+;# V15 OutArgs [V15 ] ( 1, 1 ) struct ( 0) [rsp+0x00] do-not-enreg[XS] addr-exposed "OutgoingArgSpace"
+;* V16 tmp1 [V16 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
+;* V17 tmp2 [V17 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
+;* V18 tmp3 [V18 ] ( 0, 0 ) long -> zero-ref "impAppendStmt"
+;* V19 tmp4 [V19 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
+;* V20 tmp5 [V20 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
+;* V21 tmp6 [V21 ] ( 0, 0 ) long -> zero-ref "impAppendStmt"
+;* V22 tmp7 [V22 ] ( 0, 0 ) long -> zero-ref "impAppendStmt"
+;* V23 tmp8 [V23 ] ( 0, 0 ) long -> zero-ref "impAppendStmt"
+;* V24 tmp9 [V24 ] ( 0, 0 ) long -> zero-ref "impAppendStmt"
+;* V25 tmp10 [V25 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
+;* V26 tmp11 [V26 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
+;* V27 tmp12 [V27 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
+;* V28 tmp13 [V28 ] ( 0, 0 ) long -> zero-ref "impAppendStmt"
+;* V29 tmp14 [V29 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
+;* V30 tmp15 [V30 ] ( 0, 0 ) int -> zero-ref "impAppendStmt"
+;* V31 tmp16 [V31 ] ( 0, 0 ) struct (16) zero-ref "dup spill" <System.ValueTuple`2[ulong,ulong]>
+;* V32 tmp17 [V32 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
+;* V33 tmp18 [V33 ] ( 0, 0 ) struct (16) zero-ref "ReadOnlySpan<T> for CreateSpan<T>" <System.ReadOnlySpan`1[ulong]>
+; V34 tmp19 [V34,T12] ( 5, 2.50) long -> rax "Inline stloc first use temp"
+;* V35 tmp20 [V35 ] ( 0, 0 ) struct (16) zero-ref ld-addr-op "NewObj constructor temp" <System.ValueTuple`2[ulong,ulong]>
+;* V36 tmp21 [V36 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+;* V37 tmp22 [V37 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+;* V38 tmp23 [V38 ] ( 0, 0 ) long -> zero-ref "Inline return value spill temp"
+;* V39 tmp24 [V39 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+; V40 tmp25 [V40 ] ( 2, 1 ) long -> [rbp-0x38] do-not-enreg[X] addr-exposed ld-addr-op "Inline ldloca(s) first use temp"
+; V41 tmp26 [V41,T09] ( 3, 3 ) long -> rdi "impAppendStmt"
+;* V42 tmp27 [V42 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+;* V43 tmp28 [V43 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
+;* V44 tmp29 [V44 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+;* V45 tmp30 [V45 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
+;* V46 tmp31 [V46 ] ( 0, 0 ) long -> zero-ref "Inline return value spill temp"
+;* V47 tmp32 [V47 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+; V48 tmp33 [V48 ] ( 2, 1 ) long -> [rbp-0x40] do-not-enreg[X] addr-exposed ld-addr-op "Inline ldloca(s) first use temp"
+; V49 tmp34 [V49,T10] ( 3, 3 ) long -> rdx "impAppendStmt"
+;* V50 tmp35 [V50 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+;* V51 tmp36 [V51 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
+;* V52 tmp37 [V52 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+;* V53 tmp38 [V53 ] ( 0, 0 ) long -> zero-ref "Inline return value spill temp"
+;* V54 tmp39 [V54 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+; V55 tmp40 [V55 ] ( 2, 1 ) long -> [rbp-0x48] do-not-enreg[X] addr-exposed ld-addr-op "Inline ldloca(s) first use temp"
+; V56 tmp41 [V56,T11] ( 3, 3 ) long -> rax "impAppendStmt"
+;* V57 tmp42 [V57 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+;* V58 tmp43 [V58 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
+;* V59 tmp44 [V59 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
+;* V60 tmp45 [V60 ] ( 0, 0 ) long -> zero-ref "Inline return value spill temp"
+;* V61 tmp46 [V61 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+; V62 tmp47 [V62 ] ( 2, 1 ) long -> [rbp-0x50] do-not-enreg[X] addr-exposed ld-addr-op "Inline ldloca(s) first use temp"
+; V63 tmp48 [V63,T13] ( 2, 2 ) long -> rdx "impAppendStmt"
+;* V64 tmp49 [V64 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
+;* V65 tmp50 [V65 ] ( 0, 0 ) long -> zero-ref "Inline return value spill temp"
+;* V66 tmp51 [V66 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+; V67 tmp52 [V67 ] ( 2, 1 ) long -> [rbp-0x58] do-not-enreg[X] addr-exposed ld-addr-op "Inline ldloca(s) first use temp"
+; V68 tmp53 [V68,T14] ( 2, 2 ) long -> rdx "impAppendStmt"
+;* V69 tmp54 [V69 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+;* V70 tmp55 [V70 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
+;* V71 tmp56 [V71 ] ( 0, 0 ) byref -> zero-ref single-def "field V10._reference (fldOffset=0x0)" P-INDEP
+;* V72 tmp57 [V72 ] ( 0, 0 ) int -> zero-ref "field V10._length (fldOffset=0x8)" P-INDEP
+;* V73 tmp58 [V73 ] ( 0, 0 ) long -> zero-ref "field V31.Item1 (fldOffset=0x0)" P-INDEP
+;* V74 tmp59 [V74 ] ( 0, 0 ) long -> zero-ref "field V31.Item2 (fldOffset=0x8)" P-INDEP
+;* V75 tmp60 [V75 ] ( 0, 0 ) byref -> zero-ref single-def "field V33._reference (fldOffset=0x0)" P-INDEP
+;* V76 tmp61 [V76 ] ( 0, 0 ) int -> zero-ref "field V33._length (fldOffset=0x8)" P-INDEP
+;* V77 tmp62 [V77 ] ( 0, 0 ) long -> zero-ref "field V35.Item1 (fldOffset=0x0)" P-INDEP
+; V78 tmp63 [V78,T15] ( 3, 1.50) long -> rdi "field V35.Item2 (fldOffset=0x8)" P-INDEP
+; V79 cse0 [V79,T16] ( 3, 1.50) long -> rdi "CSE #14: moderate"
+; V80 cse1 [V80,T17] ( 3, 1.50) int -> rdi "CSE #13: moderate"
+; V81 cse2 [V81,T18] ( 3, 1.50) int -> rsi "CSE #03: moderate"
+; V82 cse3 [V82,T19] ( 3, 1.50) int -> rdi "CSE #09: moderate"
+; V83 cse4 [V83,T20] ( 3, 1.50) int -> rdx "CSE #12: moderate"
 ;
-; Lcl frame size = 32
+; Lcl frame size = 80
 
 G_M55226_IG01:
 push rbp
 push r15
 push rbx
- sub rsp, 32
- lea rbp, [rsp+0x30]
+ sub rsp, 80
+ lea rbp, [rsp+0x60]
 mov rbx, rdi
 mov r15, rsi
 						;; size=19 bbWeight=1 PerfScore 4.25
 G_M55226_IG02:
 mov eax, dword ptr [rbx]
 add eax, dword ptr [r15]
 shr eax, 16
 movzx rdx, al
 mov eax, dword ptr [rbx+0x04]
 or eax, dword ptr [rbx+0x0C]
 jne G_M55226_IG08
 						;; size=23 bbWeight=1 PerfScore 11.75
 G_M55226_IG03:
 mov eax, dword ptr [r15+0x04]
 or eax, dword ptr [r15+0x0C]
- jne G_M55226_IG16
+ jne G_M55226_IG11
 mov edi, dword ptr [rbx+0x08]
 mov esi, dword ptr [r15+0x08]
 imul rdi, rsi
 cmp edx, 28
 jle SHORT G_M55226_IG06
 cmp edx, 47
- jg G_M55226_IG22
+ jg G_M55226_IG21
 add edx, -29
 cmp edx, 19
- jae G_M55226_IG24
+ jae G_M55226_IG23
 mov eax, edx
 mov rdx, 0xD1FFAB1E ; static handle
 mov rsi, qword ptr [rdx+8*rax]
 mov rax, rdi
 xor edx, edx
 div rdx:rax, rsi
 mov rdx, rax
 imul rdx, rsi
 sub rdi, rdx
 mov rdx, rax
 shr rsi, 1
 cmp rdi, rsi
 jb SHORT G_M55226_IG05
 ja SHORT G_M55226_IG04
 test al, 1
 je SHORT G_M55226_IG05
 						;; size=102 bbWeight=0.50 PerfScore 43.38
 G_M55226_IG04:
 lea rdx, [rax+0x01]
 						;; size=4 bbWeight=0.50 PerfScore 0.25
 G_M55226_IG05:
 mov eax, 28
 mov rdi, rdx
 mov edx, eax
 						;; size=10 bbWeight=0.50 PerfScore 0.38
 G_M55226_IG06:
 mov qword ptr [rbx+0x08], rdi
 mov edi, dword ptr [r15]
 xor edi, dword ptr [rbx]
 and edi, 0xD1FFAB1E
- mov esi, edx
- shl esi, 16
- or edi, esi
- mov dword ptr [rbx], edi
-						;; size=24 bbWeight=0.50 PerfScore 4.12
+ shl edx, 16
+ or edx, edi
+ mov dword ptr [rbx], edx
+						;; size=22 bbWeight=0.50 PerfScore 4.00
 G_M55226_IG07:
- add rsp, 32
+ add rsp, 80
 pop rbx
 pop r15
 pop rbp
 ret 
 						;; size=9 bbWeight=0.50 PerfScore 1.38
 G_M55226_IG08:
 mov edi, dword ptr [r15+0x04]
 or edi, dword ptr [r15+0x0C]
- jne G_M55226_IG13
+ jne SHORT G_M55226_IG09
+ mov dword ptr [rbp-0x14], edx
+ mov rdx, qword ptr [rbx+0x08]
 mov edi, dword ptr [r15+0x08]
- mov esi, dword ptr [rbx+0x08]
- imul rdi, rsi
- mov dword ptr [rbp-0x28], edi
- mov esi, dword ptr [r15+0x08]
- mov eax, dword ptr [rbx+0x0C]
- imul rsi, rax
- shr rdi, 32
- add rdi, rsi
- mov dword ptr [rbp-0x24], edi
- shr rdi, 32
- mov esi, dword ptr [rbx+0x04]
- test esi, esi
- je G_M55226_IG18
+ lea rsi, [rbp-0x40]
+ mulx rdx, rax, rdi
+ mov qword ptr [rsi], rax
+ mov rdi, qword ptr [rbp-0x40]
+ mov rax, rdx
+ mov qword ptr [rbp-0x30], rdi
+ mov edi, dword ptr [rbx+0x04]
+ test edi, edi
+ je G_M55226_IG16
 mov eax, dword ptr [r15+0x08]
- imul rsi, rax
- add rsi, rdi
- mov rdi, rsi
- mov esi, 0xD1FFAB1E
- cmp rdi, rsi
- jbe G_M55226_IG18
-						;; size=92 bbWeight=0.50 PerfScore 15.25
+ imul rax, rdi
+ add rax, rdx
+ mov edx, 0xD1FFAB1E
+ cmp rax, rdx
+ ja G_M55226_IG12
+ jmp G_M55226_IG16
+						;; size=85 bbWeight=0.50 PerfScore 14.38
 G_M55226_IG09:
- mov qword ptr [rbp-0x20], rdi
+ mov dword ptr [rbp-0x14], edx
+ mov rdx, qword ptr [rbx+0x08]
+ mov rax, qword ptr [r15+0x08]
+ lea rdi, [rbp-0x48]
+ mulx rax, rsi, rax
+ mov qword ptr [rdi], rsi
+ mov rdx, qword ptr [rbp-0x48]
+ mov qword ptr [rbp-0x30], rdx
+ mov edx, dword ptr [rbx+0x04]
+ mov edi, dword ptr [r15+0x04]
+ mov esi, edx
+ or esi, edi
+ je SHORT G_M55226_IG10
+ mov esi, edx
+ imul rsi, rdi
+ mov rdx, qword ptr [rbx+0x08]
+ lea rcx, [rbp-0x50]
+ mulx rdx, r8, rdi
+ mov qword ptr [rcx], r8
+ mov rdi, qword ptr [rbp-0x50]
+ add rsi, rdx
+ add rax, rdi
+ lea rdx, [rsi+0x01]
+ cmp rax, rdi
+ cmovb rsi, rdx
+ mov rdx, qword ptr [r15+0x08]
+ mov edi, dword ptr [rbx+0x04]
+ lea rcx, [rbp-0x58]
+ mulx rdx, r8, rdi
+ mov qword ptr [rcx], r8
+ mov rdi, qword ptr [rbp-0x58]
+ add rsi, rdx
+ add rax, rdi
+ lea rdx, [rsi+0x01]
+ cmp rax, rdi
+ cmovb rsi, rdx
+ mov qword ptr [rbp-0x28], rax
+ mov qword ptr [rbp-0x20], rsi
+ mov esi, 5
+ jmp G_M55226_IG17
+ align [0 bytes for IG18]
+						;; size=145 bbWeight=0.50 PerfScore 21.75
+G_M55226_IG10:
+ mov qword ptr [rbp-0x28], rax
+ mov esi, 3
+ jmp G_M55226_IG17
+						;; size=14 bbWeight=0.50 PerfScore 1.62
+G_M55226_IG11:
+ mov dword ptr [rbp-0x14], edx
+ mov rdx, qword ptr [r15+0x08]
+ mov edi, dword ptr [rbx+0x08]
+ lea rsi, [rbp-0x38]
+ mulx rdi, rax, rdi
+ mov qword ptr [rsi], rax
+ mov rsi, qword ptr [rbp-0x38]
+ mov rax, rdi
+ mov qword ptr [rbp-0x30], rsi
+ mov esi, dword ptr [r15+0x04]
+ test esi, esi
+ je SHORT G_M55226_IG16
+ mov eax, dword ptr [rbx+0x08]
+ imul rax, rsi
+ add rax, rdi
+ mov edi, 0xD1FFAB1E
+ cmp rax, rdi
+ jbe SHORT G_M55226_IG16
+						;; size=61 bbWeight=0.50 PerfScore 10.38
+G_M55226_IG12:
+ mov qword ptr [rbp-0x28], rax
 mov esi, 3
 						;; size=9 bbWeight=0.50 PerfScore 0.62
-G_M55226_IG10:
- lea rdi, [rbp-0x28]
+G_M55226_IG13:
+ lea rdi, [rbp-0x30]
+ mov edx, dword ptr [rbp-0x14]
 mov rax, 0xD1FFAB1E ; code for System.Decimal+DecCalc:ScaleResult(ulong,uint,int):int
 call [rax]System.Decimal+DecCalc:ScaleResult(ulong,uint,int):int
 mov edx, eax
-						;; size=18 bbWeight=0.50 PerfScore 2.00
-G_M55226_IG11:
- mov rax, qword ptr [rbp-0x28]
+						;; size=21 bbWeight=0.50 PerfScore 2.50
+G_M55226_IG14:
+ mov rax, qword ptr [rbp-0x30]
 mov qword ptr [rbx+0x08], rax
- mov eax, dword ptr [rbp-0x20]
+ mov eax, dword ptr [rbp-0x28]
 mov dword ptr [rbx+0x04], eax
 mov eax, dword ptr [r15]
 xor eax, dword ptr [rbx]
 and eax, 0xD1FFAB1E
 shl edx, 16
 or eax, edx
 mov dword ptr [rbx], eax
 						;; size=31 bbWeight=0.50 PerfScore 5.50
-G_M55226_IG12:
- add rsp, 32
+G_M55226_IG15:
+ add rsp, 80
 pop rbx
 pop r15
 pop rbp
 ret 
 						;; size=9 bbWeight=0.50 PerfScore 1.38
-G_M55226_IG13:
- mov edi, dword ptr [rbx+0x08]
- mov eax, dword ptr [r15+0x08]
- imul rdi, rax
- mov dword ptr [rbp-0x28], edi
- mov eax, dword ptr [rbx+0x08]
- mov ecx, dword ptr [r15+0x0C]
- imul rax, rcx
- shr rdi, 32
- add rdi, rax
- mov eax, dword ptr [rbx+0x0C]
- mov ecx, dword ptr [r15+0x08]
- imul rax, rcx
- add rax, rdi
- mov dword ptr [rbp-0x24], eax
- mov rcx, rax
- shr rcx, 32
- mov rsi, rax
- shr rsi, 32
- mov r8, 0xD1FFAB1E
- or rsi, r8
- cmp rax, rdi
- mov rdi, rsi
- cmovae rdi, rcx
- mov eax, dword ptr [rbx+0x0C]
- mov ecx, dword ptr [r15+0x0C]
- imul rax, rcx
- add rax, rdi
- mov ecx, dword ptr [rbx+0x04]
- mov edi, dword ptr [r15+0x04]
- mov esi, ecx
- or esi, edi
- je G_M55226_IG15
- mov esi, dword ptr [rbx+0x08]
- imul rdi, rsi
- add rax, rdi
- xor esi, esi
- mov r8d, 1
- cmp rax, rdi
- cmovb esi, r8d
- mov edi, ecx
- mov ecx, dword ptr [r15+0x08]
- imul rdi, rcx
- add rax, rdi
- mov dword ptr [rbp-0x20], eax
- lea ecx, [rsi+0x01]
- cmp rax, rdi
- cmovb esi, ecx
- mov edi, esi
- shl rdi, 32
- shr rax, 32
- or rdi, rax
- mov eax, dword ptr [rbx+0x0C]
- mov esi, dword ptr [r15+0x04]
- imul rax, rsi
- add rax, rdi
- xor esi, esi
- cmp rax, rdi
- cmovb esi, r8d
- mov edi, dword ptr [rbx+0x04]
- mov ecx, dword ptr [r15+0x0C]
- imul rdi, rcx
- add rax, rdi
- mov dword ptr [rbp-0x1C], eax
- lea ecx, [rsi+0x01]
- cmp rax, rdi
- cmovb esi, ecx
- mov ecx, dword ptr [rbx+0x04]
- mov edi, dword ptr [r15+0x04]
- imul rcx, rdi
- mov edi, esi
- shl rdi, 32
- shr rax, 32
- or rax, rdi
-						;; size=253 bbWeight=0.50 PerfScore 35.75
-G_M55226_IG14:
- add rax, rcx
- mov qword ptr [rbp-0x18], rax
- mov esi, 5
- jmp SHORT G_M55226_IG19
- align [6 bytes for IG20]
-						;; size=20 bbWeight=0.50 PerfScore 1.75
-G_M55226_IG15:
- mov qword ptr [rbp-0x20], rax
- mov esi, 3
- jmp SHORT G_M55226_IG19
-						;; size=11 bbWeight=0.50 PerfScore 1.62
 G_M55226_IG16:
- mov eax, dword ptr [rbx+0x08]
- mov esi, dword ptr [r15+0x08]
- imul rax, rsi
 mov dword ptr [rbp-0x28], eax
- mov ecx, dword ptr [rbx+0x08]
- mov edi, dword ptr [r15+0x0C]
- imul rcx, rdi
- shr rax, 32
- add rax, rcx
- mov dword ptr [rbp-0x24], eax
- shr rax, 32
- mov ecx, dword ptr [r15+0x04]
- test ecx, ecx
- je SHORT G_M55226_IG17
- mov edi, dword ptr [rbx+0x08]
- imul rcx, rdi
- add rcx, rax
- mov rax, rcx
- mov ecx, 0xD1FFAB1E
- cmp rax, rcx
- mov rdi, rax
- jbe SHORT G_M55226_IG18
- jmp G_M55226_IG09
-						;; size=78 bbWeight=0.50 PerfScore 13.38
-G_M55226_IG17:
- mov rdi, rax
-						;; size=3 bbWeight=0.25 PerfScore 0.06
-G_M55226_IG18:
- mov dword ptr [rbp-0x20], edi
 mov esi, 2
 						;; size=8 bbWeight=0.50 PerfScore 0.62
-G_M55226_IG19:
- lea rax, [rbp-0x28]
- movsxd rcx, esi
+G_M55226_IG17:
+ lea rax, [rbp-0x30]
+ mov ecx, esi
 cmp dword ptr [rax+4*rcx], 0
- jne SHORT G_M55226_IG21
-						;; size=13 bbWeight=0.50 PerfScore 2.38
-G_M55226_IG20:
+ jne SHORT G_M55226_IG19
+						;; size=12 bbWeight=0.50 PerfScore 2.38
+G_M55226_IG18:
 test esi, esi
- je SHORT G_M55226_IG22
+ je SHORT G_M55226_IG21
 dec esi
- lea rax, [rbp-0x28]
- movsxd rcx, esi
+ lea rax, [rbp-0x30]
+ mov ecx, esi
 cmp dword ptr [rax+4*rcx], 0
- je SHORT G_M55226_IG20
-						;; size=19 bbWeight=4 PerfScore 25.00
-G_M55226_IG21:
+ je SHORT G_M55226_IG18
+						;; size=18 bbWeight=4 PerfScore 25.00
+G_M55226_IG19:
 cmp esi, 2
- ja G_M55226_IG10
+ ja SHORT G_M55226_IG13
+ mov edx, dword ptr [rbp-0x14]
 cmp edx, 28
- jle G_M55226_IG11
- jmp G_M55226_IG10
-						;; size=23 bbWeight=0.50 PerfScore 2.25
-G_M55226_IG22:
+ jle SHORT G_M55226_IG14
+						;; size=13 bbWeight=0.50 PerfScore 1.75
+G_M55226_IG20:
+ mov dword ptr [rbp-0x14], edx
+ jmp SHORT G_M55226_IG13
+						;; size=5 bbWeight=0.25 PerfScore 0.75
+G_M55226_IG21:
 vxorps xmm0, xmm0, xmm0
 vmovdqu xmmword ptr [rbx], xmm0
 						;; size=8 bbWeight=0.50 PerfScore 1.17
-G_M55226_IG23:
- add rsp, 32
+G_M55226_IG22:
+ add rsp, 80
 pop rbx
 pop r15
 pop rbp
 ret 
 						;; size=9 bbWeight=0.50 PerfScore 1.38
-G_M55226_IG24:
+G_M55226_IG23:
 call CORINFO_HELP_RNGCHKFAIL
 int3 
 						;; size=6 bbWeight=0 PerfScore 0.00
 
-; Total bytes of code 801, prolog size 19, PerfScore 175.60, instruction count 237, allocated bytes for code 801 (MethodHash=34b92845) for method System.Decimal+DecCalc:VarDecMul(byref,byref) (FullOpts)
+; Total bytes of code 643, prolog size 19, PerfScore 156.54, instruction count 194, allocated bytes for code 643 (MethodHash=34b92845) for method System.Decimal+DecCalc:VarDecMul(byref,byref) (FullOpts)

-142 (-88.75 % of base) - System.Decimal+DecCalc:Div96By64(byref,ulong):uint

 ; Assembly listing for method System.Decimal+DecCalc:Div96By64(byref,ulong):uint (FullOpts)
 ; Emitting BLENDED_CODE for X64 with AVX512 - Unix
 ; FullOpts code
 ; optimized code
 ; rbp based frame
-; fully interruptible
+; partially interruptible
 ; No PGO data
-; 0 inlinees with PGO data; 9 single block inlinees; 0 inlinees without PGO data
+; 0 inlinees with PGO data; 2 single block inlinees; 0 inlinees without PGO data
 ; Final local variable assignments
 ;
-; V00 arg0 [V00,T03] ( 9, 6 ) byref -> rdi single-def
-; V01 arg1 [V01,T01] ( 12, 21 ) long -> rsi single-def
-; V02 loc0 [V02,T00] ( 16, 29 ) long -> r8 
-; V03 loc1 [V03,T05] ( 3, 2.50) int -> rcx 
-; V04 loc2 [V04,T02] ( 9, 18.50) int -> rcx 
-; V05 loc3 [V05,T06] ( 4, 2 ) int -> r8 
-; V06 loc4 [V06,T07] ( 4, 2 ) long -> r9 
-; V07 loc5 [V07,T08] ( 3, 1.50) long -> rax 
-;* V08 loc6 [V08 ] ( 0, 0 ) byref -> zero-ref single-def
-;* V09 loc7 [V09 ] ( 0, 0 ) struct (16) zero-ref <System.ValueTuple`2[ulong,ulong]>
-;# V10 OutArgs [V10 ] ( 1, 1 ) struct ( 0) [rsp+0x00] do-not-enreg[XS] addr-exposed "OutgoingArgSpace"
-; V11 tmp1 [V11,T09] ( 3, 1.50) long -> rax "Inline stloc first use temp"
-;* V12 tmp2 [V12 ] ( 0, 0 ) struct (16) zero-ref ld-addr-op "NewObj constructor temp" <System.ValueTuple`2[ulong,ulong]>
-;* V13 tmp3 [V13 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
-;* V14 tmp4 [V14 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
-;* V15 tmp5 [V15 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
-; V16 tmp6 [V16,T04] ( 3, 3 ) int -> rax "Single return block return value"
-;* V17 tmp7 [V17 ] ( 0, 0 ) long -> zero-ref "field V09.Item1 (fldOffset=0x0)" P-INDEP
-;* V18 tmp8 [V18 ] ( 0, 0 ) long -> zero-ref "field V09.Item2 (fldOffset=0x8)" P-INDEP
-;* V19 tmp9 [V19 ] ( 0, 0 ) long -> zero-ref "field V12.Item1 (fldOffset=0x0)" P-INDEP
-; V20 tmp10 [V20,T11] ( 2, 1 ) long -> r8 "field V12.Item2 (fldOffset=0x8)" P-INDEP
-; V21 cse0 [V21,T10] ( 3, 1.50) long -> rcx "CSE #02: moderate"
+; V00 arg0 [V00,T00] ( 5, 5 ) byref -> rdi single-def
+; V01 arg1 [V01,T01] ( 3, 3 ) long -> rsi single-def
+;* V02 loc0 [V02 ] ( 0, 0 ) long -> zero-ref 
+;* V03 loc1 [V03 ] ( 0, 0 ) int -> zero-ref 
+;* V04 loc2 [V04 ] ( 0, 0 ) int -> zero-ref 
+;* V05 loc3 [V05 ] ( 0, 0 ) int -> zero-ref 
+;* V06 loc4 [V06 ] ( 0, 0 ) long -> zero-ref 
+;* V07 loc5 [V07 ] ( 0, 0 ) int -> zero-ref 
+;* V08 loc6 [V08 ] ( 0, 0 ) long -> zero-ref 
+;* V09 loc7 [V09 ] ( 0, 0 ) byref -> zero-ref single-def
+;* V10 loc8 [V10 ] ( 0, 0 ) struct (16) zero-ref multireg-ret <System.ValueTuple`2[ulong,ulong]>
+;# V11 OutArgs [V11 ] ( 1, 1 ) struct ( 0) [rsp+0x00] do-not-enreg[XS] addr-exposed "OutgoingArgSpace"
+;* V12 tmp1 [V12 ] ( 0, 0 ) struct (16) zero-ref do-not-enreg[SBR] multireg-ret "Return value temp for multireg return" <System.ValueTuple`2[ulong,ulong]>
+;* V13 tmp2 [V13 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+; V14 tmp3 [V14,T02] ( 2, 2 ) long -> rax "field V10.Item1 (fldOffset=0x0)" P-INDEP
+; V15 tmp4 [V15,T03] ( 2, 2 ) long -> rdx "field V10.Item2 (fldOffset=0x8)" P-INDEP
+;* V16 tmp5 [V16 ] ( 0, 0 ) long -> zero-ref "field V12.Item1 (fldOffset=0x0)" P-DEP
+;* V17 tmp6 [V17 ] ( 0, 0 ) long -> zero-ref "field V12.Item2 (fldOffset=0x8)" P-DEP
 ;
 ; Lcl frame size = 0
 
 G_M1421_IG01:
 push rbp
 mov rbp, rsp
 						;; size=4 bbWeight=1 PerfScore 1.25
 G_M1421_IG02:
- mov ecx, dword ptr [rdi+0x08]
- test ecx, ecx
- jne SHORT G_M1421_IG05
-						;; size=7 bbWeight=1 PerfScore 3.25
-G_M1421_IG03:
- mov r8, qword ptr [rdi]
- cmp r8, rsi
- jb SHORT G_M1421_IG08
- mov rax, r8
- xor edx, edx
+ mov rax, qword ptr [rdi]
+ mov edx, dword ptr [rdi+0x08]
 div rdx:rax, rsi
- imul rsi, rax
- sub r8, rsi
- mov qword ptr [rdi], r8
-						;; size=26 bbWeight=0.50 PerfScore 34.00
-G_M1421_IG04:
- pop rbp
- ret 
-						;; size=2 bbWeight=0.50 PerfScore 0.75
-G_M1421_IG05:
- mov r8, rsi
- shr r8, 32
- cmp ecx, r8d
- jb SHORT G_M1421_IG07
- mov r8, qword ptr [rdi]
- mov rax, rsi
- shl rax, 32
- sub r8, rax
- xor ecx, ecx
- align [0 bytes for IG06]
-						;; size=27 bbWeight=0.50 PerfScore 2.62
-G_M1421_IG06:
- dec ecx
- add r8, rsi
- cmp r8, rsi
- jae SHORT G_M1421_IG06
- jmp SHORT G_M1421_IG12
-						;; size=12 bbWeight=4 PerfScore 15.00
-G_M1421_IG07:
- mov r9, qword ptr [rdi+0x04]
- mov ecx, r8d
- cmp r9, rcx
- jae SHORT G_M1421_IG10
-						;; size=12 bbWeight=0.50 PerfScore 1.75
-G_M1421_IG08:
- xor eax, eax
-						;; size=2 bbWeight=0.50 PerfScore 0.12
-G_M1421_IG09:
+ mov qword ptr [rdi], rdx
+						;; size=12 bbWeight=1 PerfScore 66.00
+G_M1421_IG03:
 pop rbp
 ret 
-						;; size=2 bbWeight=0.50 PerfScore 0.75
-G_M1421_IG10:
- mov rax, r9
- xor edx, edx
- div rdx:rax, rcx
- mov ecx, eax
- imul r8d, ecx
- mov eax, r8d
- sub r9, rax
- shl r9, 32
- mov r8d, dword ptr [rdi]
- or r8, r9
- mov eax, ecx
- mov edx, esi
- imul rax, rdx
- sub r8, rax
- not rax
- cmp rax, r8
- jae SHORT G_M1421_IG12
- align [0 bytes for IG11]
-						;; size=49 bbWeight=0.50 PerfScore 35.62
-G_M1421_IG11:
- dec ecx
- add r8, rsi
- cmp r8, rsi
- jae SHORT G_M1421_IG11
-						;; size=10 bbWeight=4 PerfScore 7.00
-G_M1421_IG12:
- mov qword ptr [rdi], r8
- mov eax, ecx
- jmp SHORT G_M1421_IG04
-						;; size=7 bbWeight=0.50 PerfScore 1.62
+						;; size=2 bbWeight=1 PerfScore 1.50
 
-; Total bytes of code 160, prolog size 4, PerfScore 103.75, instruction count 63, allocated bytes for code 163 (MethodHash=4595fa72) for method System.Decimal+DecCalc:Div96By64(byref,ulong):uint (FullOpts)
+; Total bytes of code 18, prolog size 4, PerfScore 68.75, instruction count 8, allocated bytes for code 18 (MethodHash=4595fa72) for method System.Decimal+DecCalc:Div96By64(byref,ulong):uint (FullOpts)

-40 (-39.22 % of base) - System.Decimal+DecCalc:Div96By32(byref,uint):uint

 ; Assembly listing for method System.Decimal+DecCalc:Div96By32(byref,uint):uint (FullOpts)
 ; Emitting BLENDED_CODE for X64 with AVX512 - Unix
 ; FullOpts code
 ; optimized code
 ; rbp based frame
 ; partially interruptible
 ; No PGO data
-; 0 inlinees with PGO data; 4 single block inlinees; 0 inlinees without PGO data
 ; Final local variable assignments
 ;
-; V00 arg0 [V00,T00] ( 9, 6 ) byref -> rdi single-def
-; V01 arg1 [V01,T01] ( 6, 4 ) int -> rsi single-def
-; V02 loc0 [V02,T02] ( 11, 5.50) long -> rcx 
-; V03 loc1 [V03,T03] ( 6, 3 ) long -> rax 
-; V04 loc2 [V04,T05] ( 3, 1.50) int -> rax 
-;# V05 OutArgs [V05 ] ( 1, 1 ) struct ( 0) [rsp+0x00] do-not-enreg[XS] addr-exposed "OutgoingArgSpace"
-; V06 cse0 [V06,T04] ( 6, 3 ) long -> r8 multi-def "CSE #01: aggressive"
+; V00 arg0 [V00,T00] ( 11, 8 ) byref -> rdi single-def
+; V01 arg1 [V01,T01] ( 6, 4.50) int -> rsi single-def
+; V02 loc0 [V02,T02] ( 6, 4 ) int -> rdx 
+;* V03 loc1 [V03 ] ( 0, 0 ) struct ( 8) zero-ref multireg-ret <System.ValueTuple`2[uint,uint]>
+;* V04 loc2 [V04 ] ( 0, 0 ) long -> zero-ref 
+;* V05 loc3 [V05 ] ( 0, 0 ) long -> zero-ref 
+;* V06 loc4 [V06 ] ( 0, 0 ) struct (16) zero-ref <System.ValueTuple`2[ulong,ulong]>
+;# V07 OutArgs [V07 ] ( 1, 1 ) struct ( 0) [rsp+0x00] do-not-enreg[XS] addr-exposed "OutgoingArgSpace"
+;* V08 tmp1 [V08 ] ( 0, 0 ) struct ( 8) zero-ref do-not-enreg[SR] multireg-ret "Return value temp for multireg return" <System.ValueTuple`2[uint,uint]>
+; V09 tmp2 [V09,T05] ( 2, 2 ) byref -> rcx single-def "impAppendStmt"
+;* V10 tmp3 [V10 ] ( 0, 0 ) struct ( 8) zero-ref do-not-enreg[SR] multireg-ret "Return value temp for multireg return" <System.ValueTuple`2[uint,uint]>
+; V11 tmp4 [V11,T06] ( 2, 2 ) byref -> rcx single-def "impAppendStmt"
+;* V12 tmp5 [V12 ] ( 0, 0 ) struct ( 8) zero-ref do-not-enreg[SR] multireg-ret "Return value temp for multireg return" <System.ValueTuple`2[uint,uint]>
+;* V13 tmp6 [V13 ] ( 0, 0 ) byref -> zero-ref single-def "impAppendStmt"
+; V14 tmp7 [V14,T03] ( 6, 4 ) int -> rax "field V03.Item1 (fldOffset=0x0)" P-INDEP
+; V15 tmp8 [V15,T04] ( 6, 4 ) int -> rdx "field V03.Item2 (fldOffset=0x4)" P-INDEP
+;* V16 tmp9 [V16 ] ( 0, 0 ) long -> zero-ref "field V06.Item1 (fldOffset=0x0)" P-INDEP
+;* V17 tmp10 [V17 ] ( 0, 0 ) long -> zero-ref "field V06.Item2 (fldOffset=0x8)" P-INDEP
+; V18 cse0 [V18,T07] ( 3, 1.50) int -> rax "CSE #01: moderate"
 ;
 ; Lcl frame size = 0
 
 G_M12183_IG01:
 push rbp
 mov rbp, rsp
 						;; size=4 bbWeight=1 PerfScore 1.25
 G_M12183_IG02:
+ xor edx, edx
 cmp dword ptr [rdi+0x08], 0
- je SHORT G_M12183_IG06
-						;; size=6 bbWeight=1 PerfScore 4.00
+ jne SHORT G_M12183_IG04
+						;; size=8 bbWeight=1 PerfScore 4.25
 G_M12183_IG03:
- mov rcx, qword ptr [rdi+0x04]
- mov r8d, esi
- mov rax, rcx
- xor edx, edx
- div rdx:rax, r8
- mov qword ptr [rdi+0x04], rax
- imul eax, esi
- sub rcx, rax
- shl rcx, 32
- mov eax, dword ptr [rdi]
- or rcx, rax
- jne SHORT G_M12183_IG08
-						;; size=36 bbWeight=0.50 PerfScore 35.38
-G_M12183_IG04:
+ mov eax, dword ptr [rdi+0x04]
+ cmp eax, esi
+ jae SHORT G_M12183_IG05
+ mov edx, eax
 xor eax, eax
-						;; size=2 bbWeight=0.50 PerfScore 0.12
+ mov dword ptr [rdi+0x04], eax
+ jmp SHORT G_M12183_IG06
+						;; size=16 bbWeight=0.50 PerfScore 3.38
+G_M12183_IG04:
+ lea rcx, bword ptr [rdi+0x08]
+ mov eax, dword ptr [rdi+0x08]
+ xor edx, edx
+ div edx:eax, esi
+ mov dword ptr [rcx], eax
+						;; size=13 bbWeight=0.50 PerfScore 14.38
 G_M12183_IG05:
- pop rbp
- ret 
-						;; size=2 bbWeight=0.50 PerfScore 0.75
+ lea rcx, bword ptr [rdi+0x04]
+ mov eax, dword ptr [rdi+0x04]
+ div edx:eax, esi
+ mov dword ptr [rcx], eax
+						;; size=11 bbWeight=0.50 PerfScore 14.25
 G_M12183_IG06:
- mov rcx, qword ptr [rdi]
- test rcx, rcx
- je SHORT G_M12183_IG04
- mov r8d, esi
- mov rax, rcx
- xor edx, edx
- div rdx:rax, r8
- mov qword ptr [rdi], rax
- imul r8d, eax
- mov eax, ecx
- sub eax, r8d
-						;; size=31 bbWeight=0.50 PerfScore 34.25
-G_M12183_IG07:
- pop rbp
- ret 
-						;; size=2 bbWeight=0.50 PerfScore 0.75
-G_M12183_IG08:
- mov rax, rcx
- xor edx, edx
- div rdx:rax, r8
+ mov eax, dword ptr [rdi]
+ div edx:eax, esi
 mov dword ptr [rdi], eax
- imul esi, eax
- mov eax, ecx
- sub eax, esi
-						;; size=17 bbWeight=0.50 PerfScore 32.50
-G_M12183_IG09:
+ mov eax, edx
+						;; size=8 bbWeight=1 PerfScore 28.25
+G_M12183_IG07:
 pop rbp
 ret 
-						;; size=2 bbWeight=0.50 PerfScore 0.75
+						;; size=2 bbWeight=1 PerfScore 1.50
 
-; Total bytes of code 102, prolog size 4, PerfScore 109.75, instruction count 41, allocated bytes for code 102 (MethodHash=a751d068) for method System.Decimal+DecCalc:Div96By32(byref,uint):uint (FullOpts)
+; Total bytes of code 62, prolog size 4, PerfScore 67.25, instruction count 27, allocated bytes for code 62 (MethodHash=a751d068) for method System.Decimal+DecCalc:Div96By32(byref,uint):uint (FullOpts)

-26 (-11.21 % of base) - System.Decimal+DecCalc:VarDecCmpSub(byref,byref):int

 ; Assembly listing for method System.Decimal+DecCalc:VarDecCmpSub(byref,byref):int (FullOpts)
 ; Emitting BLENDED_CODE for X64 with AVX512 - Unix
 ; FullOpts code
 ; optimized code
 ; rbp based frame
 ; fully interruptible
 ; No PGO data
-; 0 inlinees with PGO data; 8 single block inlinees; 0 inlinees without PGO data
+; 0 inlinees with PGO data; 7 single block inlinees; 1 inlinees without PGO data
 ; Final local variable assignments
 ;
 ; V00 arg0 [V00,T08] ( 5, 5 ) byref -> rdi single-def
 ; V01 arg1 [V01,T09] ( 5, 5 ) byref -> rsi single-def
-; V02 loc0 [V02,T12] ( 7, 4.50) int -> rcx 
-; V03 loc1 [V03,T11] ( 8, 5.25) int -> rax 
-; V04 loc2 [V04,T04] ( 8, 15 ) long -> registers 
+; V02 loc0 [V02,T12] ( 7, 4.50) int -> rax 
+; V03 loc1 [V03,T11] ( 8, 5.25) int -> rdx 
+; V04 loc2 [V04,T05] ( 7, 11 ) long -> registers ld-addr-op
 ; V05 loc3 [V05,T06] ( 7, 11 ) int -> registers 
 ; V06 loc4 [V06,T13] ( 4, 2.50) long -> registers 
 ; V07 loc5 [V07,T14] ( 4, 2.50) int -> registers 
-; V08 loc6 [V08,T17] ( 3, 1.50) int -> rax 
-; V09 loc7 [V09,T18] ( 3, 1.50) long -> rax 
+; V08 loc6 [V08,T17] ( 3, 1.50) int -> rcx 
+; V09 loc7 [V09,T18] ( 3, 1.50) long -> rcx 
 ;* V10 loc8 [V10 ] ( 0, 0 ) int -> zero-ref 
-; V11 loc9 [V11,T05] ( 3, 12 ) long -> r10 
-; V12 loc10 [V12,T00] ( 6, 24 ) long -> rdx 
-;* V13 loc11 [V13 ] ( 0, 0 ) struct (16) zero-ref ld-addr-op <System.ReadOnlySpan`1[uint]>
-;# V14 OutArgs [V14 ] ( 1, 1 ) struct ( 0) [rsp+0x00] do-not-enreg[XS] addr-exposed "OutgoingArgSpace"
-; V15 tmp1 [V15,T10] ( 3, 6 ) int -> rax "dup spill"
-; V16 tmp2 [V16,T07] ( 3, 8 ) int -> rax 
-; V17 tmp3 [V17,T03] ( 2, 16 ) int -> r9 "dup spill"
-; V18 tmp4 [V18,T15] ( 2, 2 ) long -> rdx "impSpillLclRefs"
-; V19 tmp5 [V19,T16] ( 2, 2 ) int -> rdi "impSpillLclRefs"
-;* V20 tmp6 [V20 ] ( 0, 0 ) struct (16) zero-ref "ReadOnlySpan<T> for CreateSpan<T>" <System.ReadOnlySpan`1[uint]>
-;* V21 tmp7 [V21 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
-;* V22 tmp8 [V22 ] ( 0, 0 ) int -> zero-ref "Inlining Arg"
-;* V23 tmp9 [V23 ] ( 0, 0 ) byref -> zero-ref "field V13._reference (fldOffset=0x0)" P-INDEP
-;* V24 tmp10 [V24 ] ( 0, 0 ) int -> zero-ref "field V13._length (fldOffset=0x8)" P-INDEP
-;* V25 tmp11 [V25 ] ( 0, 0 ) byref -> zero-ref "field V20._reference (fldOffset=0x0)" P-INDEP
-;* V26 tmp12 [V26 ] ( 0, 0 ) int -> zero-ref "field V20._length (fldOffset=0x8)" P-INDEP
-; V27 cse0 [V27,T02] ( 4, 16 ) long -> rax "CSE #01: aggressive"
-; V28 rat0 [V28,T01] ( 7, 20.75) long -> r9 "Widened IV V03"
+; V11 loc9 [V11,T03] ( 3, 12 ) long -> rdi 
+;* V12 loc10 [V12 ] ( 0, 0 ) struct (16) zero-ref ld-addr-op <System.ReadOnlySpan`1[uint]>
+;# V13 OutArgs [V13 ] ( 1, 1 ) struct ( 0) [rsp+0x00] do-not-enreg[XS] addr-exposed "OutgoingArgSpace"
+; V14 tmp1 [V14,T10] ( 3, 6 ) int -> rdx "dup spill"
+; V15 tmp2 [V15,T07] ( 3, 8 ) int -> rdx 
+; V16 tmp3 [V16,T01] ( 2, 16 ) int -> r9 "dup spill"
+; V17 tmp4 [V17,T15] ( 2, 2 ) long -> rcx "impAppendStmt"
+; V18 tmp5 [V18,T16] ( 2, 2 ) int -> rdi "impSpillLclRefs"
+;* V19 tmp6 [V19 ] ( 0, 0 ) struct (16) zero-ref "ReadOnlySpan<T> for CreateSpan<T>" <System.ReadOnlySpan`1[uint]>
+;* V20 tmp7 [V20 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+;* V21 tmp8 [V21 ] ( 0, 0 ) long -> zero-ref "Inline return value spill temp"
+;* V22 tmp9 [V22 ] ( 0, 0 ) long -> zero-ref "Inlining Arg"
+; V23 tmp10 [V23 ] ( 2, 8 ) long -> [rbp-0x10] do-not-enreg[X] addr-exposed ld-addr-op "Inline ldloca(s) first use temp"
+; V24 tmp11 [V24,T02] ( 2, 16 ) long -> rcx "impAppendStmt"
+;* V25 tmp12 [V25 ] ( 0, 0 ) byref -> zero-ref "field V12._reference (fldOffset=0x0)" P-INDEP
+;* V26 tmp13 [V26 ] ( 0, 0 ) int -> zero-ref "field V12._length (fldOffset=0x8)" P-INDEP
+;* V27 tmp14 [V27 ] ( 0, 0 ) byref -> zero-ref "field V19._reference (fldOffset=0x0)" P-INDEP
+;* V28 tmp15 [V28 ] ( 0, 0 ) int -> zero-ref "field V19._length (fldOffset=0x8)" P-INDEP
+; V29 cse0 [V29,T04] ( 3, 12 ) long -> r10 "CSE #01: aggressive"
+; V30 rat0 [V30,T00] ( 7, 20.75) long -> r9 "Widened IV V03"
 ;
-; Lcl frame size = 0
+; Lcl frame size = 8
 
 G_M9821_IG01:
 push rbp
- mov rbp, rsp
-						;; size=4 bbWeight=1 PerfScore 1.25
+ push rbx
+ push rax
+ lea rbp, [rsp+0x10]
+						;; size=8 bbWeight=1 PerfScore 3.50
 G_M9821_IG02:
- mov eax, dword ptr [rsi]
- mov ecx, eax
- sar ecx, 31
- or ecx, 1
- sub eax, dword ptr [rdi]
- mov rdx, qword ptr [rdi+0x08]
+ mov edx, dword ptr [rsi]
+ mov eax, edx
+ sar eax, 31
+ or eax, 1
+ sub edx, dword ptr [rdi]
+ mov rcx, qword ptr [rdi+0x08]
 mov edi, dword ptr [rdi+0x04]
 mov r8, qword ptr [rsi+0x08]
 mov esi, dword ptr [rsi+0x04]
- test eax, eax
+ test edx, edx
 je SHORT G_M9821_IG07
 						;; size=30 bbWeight=1 PerfScore 15.25
 G_M9821_IG03:
- sar eax, 16
+ sar edx, 16
 jns SHORT G_M9821_IG04
+ neg edx
 neg eax
- neg ecx
- mov r9, rdx
- mov rdx, r8
+ mov r9, rcx
+ mov rcx, r8
 mov r8, r9
 mov r9d, esi
 mov esi, edi
 mov edi, r9d
 						;; size=26 bbWeight=0.50 PerfScore 1.75
 G_M9821_IG04:
- mov r9d, eax
+ mov r9d, edx
 align [0 bytes for IG05]
 						;; size=3 bbWeight=0.75 PerfScore 0.19
 G_M9821_IG05:
 cmp r9d, 9
 jge SHORT G_M9821_IG08
 						;; size=6 bbWeight=4 PerfScore 5.00
 G_M9821_IG06:
 cmp r9d, 10
- jae G_M9821_IG18
- mov rax, 0xD1FFAB1E ; static handle
- mov eax, dword ptr [rax+4*r9]
+ jae SHORT G_M9821_IG17
+ mov rdx, 0xD1FFAB1E ; static handle
+ mov edx, dword ptr [rdx+4*r9]
 jmp SHORT G_M9821_IG09
-						;; size=26 bbWeight=2 PerfScore 11.00
+						;; size=22 bbWeight=2 PerfScore 11.00
 G_M9821_IG07:
- mov r11, rdx
- jmp SHORT G_M9821_IG13
+ mov rdx, rcx
+ jmp SHORT G_M9821_IG12
 						;; size=5 bbWeight=0.50 PerfScore 1.12
 G_M9821_IG08:
- mov eax, 0xD1FFAB1E
+ mov edx, 0xD1FFAB1E
 						;; size=5 bbWeight=2 PerfScore 0.50
 G_M9821_IG09:
 mov r10d, edx
- mov eax, eax
- imul r10, rax
- shr rdx, 32
- mov edx, edx
- imul rdx, rax
- mov r11, r10
- shr r11, 32
- add rdx, r11
- mov r10d, r10d
- mov r11, rdx
- shl r11, 32
- add r11, r10
+ lea r11, [rbp-0x10]
+ mov rdx, rcx
+ mulx rcx, rbx, r10
+ mov qword ptr [r11], rbx
+ mov rdx, qword ptr [rbp-0x10]
 mov edi, edi
- imul rax, rdi
- shr rdx, 32
- add rdx, rax
- mov eax, 0xD1FFAB1E
- cmp rdx, rax
- jbe SHORT G_M9821_IG12
-						;; size=65 bbWeight=4 PerfScore 48.00
+ imul rdi, r10
+ add rdi, rcx
+ mov ecx, 0xD1FFAB1E
+ cmp rdi, rcx
+ jbe SHORT G_M9821_IG11
+						;; size=41 bbWeight=4 PerfScore 40.00
 G_M9821_IG10:
- mov eax, ecx
-						;; size=2 bbWeight=1 PerfScore 0.25
-G_M9821_IG11:
+ add rsp, 8
+ pop rbx
 pop rbp
 ret 
-						;; size=2 bbWeight=1 PerfScore 1.50
-G_M9821_IG12:
- mov edi, edx
+						;; size=7 bbWeight=1 PerfScore 2.25
+G_M9821_IG11:
 add r9d, -9
 test r9d, r9d
- jg SHORT G_M9821_IG15
-						;; size=11 bbWeight=4 PerfScore 7.00
-G_M9821_IG13:
- mov eax, edi
- sub eax, esi
- je SHORT G_M9821_IG16
- cmp eax, edi
+ jg SHORT G_M9821_IG14
+						;; size=9 bbWeight=4 PerfScore 6.00
+G_M9821_IG12:
+ mov ecx, edi
+ sub ecx, esi
+ je SHORT G_M9821_IG15
+ cmp ecx, edi
 jbe SHORT G_M9821_IG10
 						;; size=10 bbWeight=0.50 PerfScore 1.38
-G_M9821_IG14:
- neg ecx
+G_M9821_IG13:
+ neg eax
 jmp SHORT G_M9821_IG10
 						;; size=4 bbWeight=0.50 PerfScore 1.12
+G_M9821_IG14:
+ mov rcx, rdx
+ jmp SHORT G_M9821_IG05
+						;; size=5 bbWeight=2 PerfScore 4.50
 G_M9821_IG15:
- mov rdx, r11
- jmp G_M9821_IG05
-						;; size=8 bbWeight=2 PerfScore 4.50
-G_M9821_IG16:
- mov rax, r11
- sub rax, r8
- jne SHORT G_M9821_IG17
- xor ecx, ecx
+ mov rcx, rdx
+ sub rcx, r8
+ jne SHORT G_M9821_IG16
+ xor eax, eax
 jmp SHORT G_M9821_IG10
 						;; size=12 bbWeight=0.50 PerfScore 1.88
-G_M9821_IG17:
- cmp rax, r11
+G_M9821_IG16:
+ cmp rcx, rdx
 jbe SHORT G_M9821_IG10
- jmp SHORT G_M9821_IG14
+ jmp SHORT G_M9821_IG13
 						;; size=7 bbWeight=0.50 PerfScore 1.62
-G_M9821_IG18:
+G_M9821_IG17:
 call CORINFO_HELP_RNGCHKFAIL
 int3 
 						;; size=6 bbWeight=0 PerfScore 0.00
 
-; Total bytes of code 232, prolog size 4, PerfScore 103.31, instruction count 81, allocated bytes for code 232 (MethodHash=418ad9a2) for method System.Decimal+DecCalc:VarDecCmpSub(byref,byref):int (FullOpts)
+; Total bytes of code 206, prolog size 8, PerfScore 97.06, instruction count 75, allocated bytes for code 206 (MethodHash=418ad9a2) for method System.Decimal+DecCalc:VarDecCmpSub(byref,byref):int (FullOpts)

Larger list of diffs: https://gist.github.com/MihuBot/6fc131614f94bf73d7408685d0ed9b4a

[MihuBot]

@xtqqczze