Optimize bounds checking by validating metadata against maximum

Introduces `KnownLayout::is_valid_metadata` which produces `true` iff
the given metadata can describe a valid allocation of `Self` by
comparing the given metadata against the theoretical maximum for
`Self`'s layout. The theoretical metadata can (and is, in practice)
computed at compile time.

We insert this check in the two critical places that bounds-check user-
provided metadata: `new_box` and `validate_cast_and_convert_metadata`.

For sized and simple dynamically sized types, this does not meaningfully
impact codegen, as the optimizer was already able to deduce the maximum
valid metadata. However, for dynamically padded types, this permits the
compiler to use unchecked arithmetic, resulting in MCA cycle count
reductions of as much as 44%.

Makes progress towards #3079.

gherrit-pr-id: Gcf66958135e905f7cd7d2fac87d9f881e5e5185f

Author: jswrenn

benches/new_box_zeroed_with_elems_dynamic_padding.x86-64

@@ -3,32 +3,20 @@ bench_new_box_zeroed_with_elems_dynamic_padding:
 	push rbx
 	push rax
 	mov rbx, rdi
-	mov ecx, 3
-	mov rax, rdi
-	mul rcx
-	jo .LBB5_6
-	mov r14, rax
-	cmp rax, -10
-	ja .LBB5_6
-	lea rax, [r14 + 9]
-	not r14d
-	and r14d, 3
-	add r14, rax
-	setb al
-	movabs rcx, 9223372036854775803
-	cmp r14, rcx
-	seta cl
-	or cl, al
-	je .LBB5_4
-.LBB5_6:
-	xor eax, eax
-	jmp .LBB5_5
-.LBB5_4:
+	movabs rax, 3074457345618258598
+	cmp rdi, rax
+	ja .LBB5_1
+	lea r14, [rbx + 2*rbx]
+	or r14, 3
+	add r14, 9
 	call qword ptr [rip + __rustc::__rust_no_alloc_shim_is_unstable_v2@GOTPCREL]
 	mov esi, 4
 	mov rdi, r14
 	call qword ptr [rip + __rustc::__rust_alloc_zeroed@GOTPCREL]
-.LBB5_5:
+	jmp .LBB5_3
+.LBB5_1:
+	xor eax, eax
+.LBB5_3:
 	mov rdx, rbx
 	add rsp, 8
 	pop rbx

benches/new_box_zeroed_with_elems_dynamic_padding.x86-64.mca

@@ -1,12 +1,12 @@
 Iterations:        100
-Instructions:      3200
-Total Cycles:      2989
-Total uOps:        4300
+Instructions:      2100
+Total Cycles:      2990
+Total uOps:        3000
 
 Dispatch Width:    4
-uOps Per Cycle:    1.44
-IPC:               1.07
-Block RThroughput: 10.8
+uOps Per Cycle:    1.00
+IPC:               0.70
+Block RThroughput: 7.5
 
 
 Instruction Info:
@@ -22,29 +22,18 @@ Instruction Info:
  2      5     1.00           *            push	rbx
  2      5     1.00           *            push	rax
  1      1     0.33                        mov	rbx, rdi
- 1      1     0.33                        mov	ecx, 3
- 1      1     0.33                        mov	rax, rdi
- 2      4     1.00                        mul	rcx
- 1      1     1.00                        jo	.LBB5_6
- 1      1     0.33                        mov	r14, rax
- 1      1     0.33                        cmp	rax, -10
- 1      1     1.00                        ja	.LBB5_6
- 1      1     0.50                        lea	rax, [r14 + 9]
- 1      1     0.33                        not	r14d
- 1      1     0.33                        and	r14d, 3
- 1      1     0.33                        add	r14, rax
- 1      1     0.50                        setb	al
- 1      1     0.33                        movabs	rcx, 9223372036854775803
- 1      1     0.33                        cmp	r14, rcx
- 2      2     1.00                        seta	cl
- 1      1     0.33                        or	cl, al
- 1      1     1.00                        je	.LBB5_4
- 1      0     0.25                        xor	eax, eax
- 1      1     1.00                        jmp	.LBB5_5
+ 1      1     0.33                        movabs	rax, 3074457345618258598
+ 1      1     0.33                        cmp	rdi, rax
+ 1      1     1.00                        ja	.LBB5_1
+ 1      1     0.50                        lea	r14, [rbx + 2*rbx]
+ 1      1     0.33                        or	r14, 3
+ 1      1     0.33                        add	r14, 9
  4      7     1.00    *                   call	qword ptr [rip + __rustc::__rust_no_alloc_shim_is_unstable_v2@GOTPCREL]
  1      1     0.33                        mov	esi, 4
  1      1     0.33                        mov	rdi, r14
  4      7     1.00    *                   call	qword ptr [rip + __rustc::__rust_alloc_zeroed@GOTPCREL]
+ 1      1     1.00                        jmp	.LBB5_3
+ 1      0     0.25                        xor	eax, eax
  1      1     0.33                        mov	rdx, rbx
  1      1     0.33                        add	rsp, 8
  1      6     0.50    *                   pop	rbx
@@ -65,39 +54,28 @@ Resources:
 
 Resource pressure per iteration:
 [0]    [1]    [2]    [3]    [4]    [5]    [6.0]  [6.1]  
- -      -     8.99   8.98   5.00   10.03  4.49   4.51   
+ -      -     4.49   4.50   5.00   6.01   4.50   4.50   
 
 Resource pressure by instruction:
 [0]    [1]    [2]    [3]    [4]    [5]    [6.0]  [6.1]  Instructions:
- -      -      -      -     1.00    -     0.49   0.51   push	r14
- -      -      -      -     1.00    -     0.51   0.49   push	rbx
- -      -      -      -     1.00    -     0.49   0.51   push	rax
- -      -     0.95   0.04    -     0.01    -      -     mov	rbx, rdi
- -      -      -     0.97    -     0.03    -      -     mov	ecx, 3
- -      -     0.02   0.02    -     0.96    -      -     mov	rax, rdi
- -      -     1.00   1.00    -      -      -      -     mul	rcx
- -      -      -      -      -     1.00    -      -     jo	.LBB5_6
- -      -     0.02   0.97    -     0.01    -      -     mov	r14, rax
- -      -     0.97   0.03    -      -      -      -     cmp	rax, -10
- -      -      -      -      -     1.00    -      -     ja	.LBB5_6
- -      -     0.99   0.01    -      -      -      -     lea	rax, [r14 + 9]
- -      -     0.01   0.99    -      -      -      -     not	r14d
- -      -     0.97   0.03    -      -      -      -     and	r14d, 3
- -      -     0.01   0.98    -     0.01    -      -     add	r14, rax
- -      -     1.00    -      -      -      -      -     setb	al
- -      -     0.02    -      -     0.98    -      -     movabs	rcx, 9223372036854775803
- -      -      -     0.97    -     0.03    -      -     cmp	r14, rcx
- -      -     2.00    -      -      -      -      -     seta	cl
- -      -     0.03   0.03    -     0.94    -      -     or	cl, al
- -      -      -      -      -     1.00    -      -     je	.LBB5_4
+ -      -      -      -     1.00    -     0.50   0.50   push	r14
+ -      -      -      -     1.00    -     0.50   0.50   push	rbx
+ -      -      -      -     1.00    -     0.50   0.50   push	rax
+ -      -     0.49   0.50    -     0.01    -      -     mov	rbx, rdi
+ -      -     0.50   0.50    -      -      -      -     movabs	rax, 3074457345618258598
+ -      -     0.50   0.50    -      -      -      -     cmp	rdi, rax
+ -      -      -      -      -     1.00    -      -     ja	.LBB5_1
+ -      -     0.50   0.50    -      -      -      -     lea	r14, [rbx + 2*rbx]
+ -      -     0.50   0.50    -      -      -      -     or	r14, 3
+ -      -     0.50    -      -     0.50    -      -     add	r14, 9
+ -      -      -      -     1.00   1.00   1.00   1.00   call	qword ptr [rip + __rustc::__rust_no_alloc_shim_is_unstable_v2@GOTPCREL]
+ -      -      -     0.50    -     0.50    -      -     mov	esi, 4
+ -      -     0.50   0.50    -      -      -      -     mov	rdi, r14
+ -      -      -      -     1.00   1.00   1.00   1.00   call	qword ptr [rip + __rustc::__rust_alloc_zeroed@GOTPCREL]
+ -      -      -      -      -     1.00    -      -     jmp	.LBB5_3
  -      -      -      -      -      -      -      -     xor	eax, eax
- -      -      -      -      -     1.00    -      -     jmp	.LBB5_5
- -      -      -      -     1.00   1.00   1.02   0.98   call	qword ptr [rip + __rustc::__rust_no_alloc_shim_is_unstable_v2@GOTPCREL]
- -      -     0.03   0.97    -      -      -      -     mov	esi, 4
- -      -     0.96   0.01    -     0.03    -      -     mov	rdi, r14
- -      -      -      -     1.00   1.00   0.98   1.02   call	qword ptr [rip + __rustc::__rust_alloc_zeroed@GOTPCREL]
- -      -      -     0.97    -     0.03    -      -     mov	rdx, rbx
- -      -     0.01   0.99    -      -      -      -     add	rsp, 8
+ -      -     0.51   0.49    -      -      -      -     mov	rdx, rbx
+ -      -     0.49   0.51    -      -      -      -     add	rsp, 8
  -      -      -      -      -      -     0.50   0.50   pop	rbx
  -      -      -      -      -      -     0.50   0.50   pop	r14
  -      -      -      -      -     1.00    -      -     ret

benches/new_box_zeroed_with_elems_dynamic_size.x86-64

@@ -3,17 +3,17 @@ bench_new_box_zeroed_with_elems_dynamic_size:
 	push rbx
 	push rax
 	mov rbx, rdi
-	movabs rax, 4611686018427387900
+	movabs rax, 4611686018427387901
 	cmp rdi, rax
-	jbe .LBB5_2
-	xor eax, eax
-	jmp .LBB5_3
-.LBB5_2:
+	ja .LBB5_1
 	lea r14, [2*rbx + 4]
 	call qword ptr [rip + __rustc::__rust_no_alloc_shim_is_unstable_v2@GOTPCREL]
 	mov esi, 2
 	mov rdi, r14
 	call qword ptr [rip + __rustc::__rust_alloc_zeroed@GOTPCREL]
+	jmp .LBB5_3
+.LBB5_1:
+	xor eax, eax
 .LBB5_3:
 	mov rdx, rbx
 	add rsp, 8

benches/new_box_zeroed_with_elems_dynamic_size.x86-64.mca

@@ -22,16 +22,16 @@ Instruction Info:
  2      5     1.00           *            push	rbx
  2      5     1.00           *            push	rax
  1      1     0.33                        mov	rbx, rdi
- 1      1     0.33                        movabs	rax, 4611686018427387900
+ 1      1     0.33                        movabs	rax, 4611686018427387901
  1      1     0.33                        cmp	rdi, rax
- 1      1     1.00                        jbe	.LBB5_2
- 1      0     0.25                        xor	eax, eax
- 1      1     1.00                        jmp	.LBB5_3
+ 1      1     1.00                        ja	.LBB5_1
  1      1     0.50                        lea	r14, [2*rbx + 4]
  4      7     1.00    *                   call	qword ptr [rip + __rustc::__rust_no_alloc_shim_is_unstable_v2@GOTPCREL]
  1      1     0.33                        mov	esi, 2
  1      1     0.33                        mov	rdi, r14
  4      7     1.00    *                   call	qword ptr [rip + __rustc::__rust_alloc_zeroed@GOTPCREL]
+ 1      1     1.00                        jmp	.LBB5_3
+ 1      0     0.25                        xor	eax, eax
  1      1     0.33                        mov	rdx, rbx
  1      1     0.33                        add	rsp, 8
  1      6     0.50    *                   pop	rbx
@@ -52,26 +52,26 @@ Resources:
 
 Resource pressure per iteration:
 [0]    [1]    [2]    [3]    [4]    [5]    [6.0]  [6.1]  
- -      -     3.97   3.97   5.00   5.06   4.50   4.50   
+ -      -     3.97   3.98   5.00   5.05   4.50   4.50   
 
 Resource pressure by instruction:
 [0]    [1]    [2]    [3]    [4]    [5]    [6.0]  [6.1]  Instructions:
  -      -      -      -     1.00    -     0.50   0.50   push	r14
  -      -      -      -     1.00    -     0.50   0.50   push	rbx
  -      -      -      -     1.00    -     0.50   0.50   push	rax
- -      -     0.94   0.05    -     0.01    -      -     mov	rbx, rdi
- -      -     0.05   0.95    -      -      -      -     movabs	rax, 4611686018427387900
- -      -     0.95    -      -     0.05    -      -     cmp	rdi, rax
- -      -      -      -      -     1.00    -      -     jbe	.LBB5_2
- -      -      -      -      -      -      -      -     xor	eax, eax
- -      -      -      -      -     1.00    -      -     jmp	.LBB5_3
- -      -      -     1.00    -      -      -      -     lea	r14, [2*rbx + 4]
+ -      -     0.05   0.94    -     0.01    -      -     mov	rbx, rdi
+ -      -     0.94   0.06    -      -      -      -     movabs	rax, 4611686018427387901
+ -      -     0.06   0.94    -      -      -      -     cmp	rdi, rax
+ -      -      -      -      -     1.00    -      -     ja	.LBB5_1
+ -      -     0.94   0.06    -      -      -      -     lea	r14, [2*rbx + 4]
  -      -      -      -     1.00   1.00   1.00   1.00   call	qword ptr [rip + __rustc::__rust_no_alloc_shim_is_unstable_v2@GOTPCREL]
- -      -     0.06   0.94    -      -      -      -     mov	esi, 2
- -      -     0.94   0.06    -      -      -      -     mov	rdi, r14
+ -      -     0.98   0.02    -      -      -      -     mov	esi, 2
+ -      -     0.02   0.94    -     0.04    -      -     mov	rdi, r14
  -      -      -      -     1.00   1.00   1.00   1.00   call	qword ptr [rip + __rustc::__rust_alloc_zeroed@GOTPCREL]
- -      -     0.05   0.95    -      -      -      -     mov	rdx, rbx
- -      -     0.98   0.02    -      -      -      -     add	rsp, 8
+ -      -      -      -      -     1.00    -      -     jmp	.LBB5_3
+ -      -      -      -      -      -      -      -     xor	eax, eax
+ -      -     0.94   0.06    -      -      -      -     mov	rdx, rbx
+ -      -     0.04   0.96    -      -      -      -     add	rsp, 8
  -      -      -      -      -      -     0.50   0.50   pop	rbx
  -      -      -      -      -      -     0.50   0.50   pop	r14
  -      -      -      -      -     1.00    -      -     ret

benches/new_vec_zeroed.x86-64

@@ -1,44 +1,40 @@
 bench_new_vec_zeroed:
+	mov rax, rdi
+	movabs rcx, 1537228672809129301
+	cmp rsi, rcx
+	ja .LBB5_5
+	test rsi, rsi
+	je .LBB5_2
 	push r15
 	push r14
-	push r12
 	push rbx
-	push rax
-	mov rbx, rdi
-	movabs r12, 9223372036854775805
-	mov ecx, 6
-	mov rax, rsi
-	mul rcx
-	jo .LBB5_6
-	cmp rax, r12
-	jbe .LBB5_2
-.LBB5_6:
-	add r12, 3
-	mov qword ptr [rbx], r12
-	jmp .LBB5_7
-.LBB5_2:
-	mov r14, rsi
-	test rax, rax
-	je .LBB5_3
-	mov r15, rax
+	lea rcx, [rsi + rsi]
+	lea rbx, [rcx + 2*rcx]
+	mov r14, rax
+	mov r15, rsi
 	call qword ptr [rip + __rustc::__rust_no_alloc_shim_is_unstable_v2@GOTPCREL]
 	mov esi, 2
-	mov rdi, r15
+	mov rdi, rbx
 	call qword ptr [rip + __rustc::__rust_alloc_zeroed@GOTPCREL]
-	test rax, rax
-	jne .LBB5_5
-	jmp .LBB5_6
-.LBB5_3:
-	mov eax, 2
-.LBB5_5:
-	mov qword ptr [rbx], r14
-	mov qword ptr [rbx + 8], rax
-	mov qword ptr [rbx + 16], r14
-.LBB5_7:
-	mov rax, rbx
-	add rsp, 8
+	mov rsi, r15
+	mov rcx, rax
+	mov rax, r14
+	test rcx, rcx
 	pop rbx
-	pop r12
 	pop r14
 	pop r15
+	je .LBB5_5
+	mov qword ptr [rax], rsi
+	mov qword ptr [rax + 8], rcx
+	mov qword ptr [rax + 16], rsi
+	ret
+.LBB5_5:
+	movabs rcx, -9223372036854775808
+	mov qword ptr [rax], rcx
+	ret
+.LBB5_2:
+	mov ecx, 2
+	mov qword ptr [rax], rsi
+	mov qword ptr [rax + 8], rcx
+	mov qword ptr [rax + 16], rsi
 	ret