vec macro into_boxed codegen regression

[RustyYato]

This is a regression from stable 1.42.0 to the latest beta & nightly

found because of this: https://internals.rust-lang.org/t/more-efficient-boxed-slice-creation/12271

pub fn bar2(n: usize) -> Box<[u32]> {
    vec![0; n].into_boxed_slice()
}

I think the regression happened because of a missed optimization to inline alloc::raw_vec::RawVec<T,A>::allocate_in::{{closure}}, so it generates unnecessary panic handlers

stable

alloc::raw_vec::RawVec<T,A>::allocate_in::{{closure}}:
        push    rax
        call    qword ptr [rip + alloc::raw_vec::capacity_overflow@GOTPCREL]
        ud2

example::bar2:
        push    r14
        push    rbx
        push    rax
        mov     ecx, 4
        mov     rax, rdi
        mul     rcx
        jo      .LBB1_5
        mov     rbx, rdi
        mov     r14, rax
        test    rax, rax
        je      .LBB1_4
        mov     esi, 4
        mov     rdi, r14
        call    qword ptr [rip + __rust_alloc_zeroed@GOTPCREL]
        test    rax, rax
        je      .LBB1_6
        mov     rcx, rax
.LBB1_4:
        mov     rax, rcx
        mov     rdx, rbx
        add     rsp, 8
        pop     rbx
        pop     r14
        ret
.LBB1_5:
        call    alloc::raw_vec::RawVec<T,A>::allocate_in::{{closure}}
        ud2
.LBB1_6:
        mov     esi, 4
        mov     rdi, r14
        call    qword ptr [rip + alloc::alloc::handle_alloc_error@GOTPCREL]
        ud2

nightly

core::ptr::drop_in_place:
        mov     rsi, qword ptr [rdi + 8]
        test    rsi, rsi
        je      .LBB0_2
        shl     rsi, 2
        test    rsi, rsi
        je      .LBB0_2
        mov     rdi, qword ptr [rdi]
        mov     edx, 4
        jmp     qword ptr [rip + __rust_dealloc@GOTPCREL]
.LBB0_2:
        ret

alloc::raw_vec::RawVec<T,A>::allocate_in::{{closure}}:
        push    rax
        call    qword ptr [rip + alloc::raw_vec::capacity_overflow@GOTPCREL]
        ud2

alloc::raw_vec::RawVec<T,A>::allocate_in::{{closure}}:
        push    rax
        mov     rax, qword ptr [rdi]
        mov     rsi, qword ptr [rdi + 8]
        mov     rdi, rax
        call    qword ptr [rip + alloc::alloc::handle_alloc_error@GOTPCREL]
        ud2

example::bar2:
        push    r15
        push    r14
        push    rbx
        sub     rsp, 32
        mov     ecx, 4
        xor     esi, esi
        mov     rax, rdi
        mul     rcx
        mov     rbx, rax
        setno   al
        jo      .LBB3_18
        mov     r14, rdi
        mov     sil, al
        shl     rsi, 2
        mov     qword ptr [rsp + 8], rbx
        mov     qword ptr [rsp + 16], rsi
        test    rbx, rbx
        je      .LBB3_3
        mov     rdi, rbx
        call    qword ptr [rip + __rust_alloc_zeroed@GOTPCREL]
        mov     rsi, rax
.LBB3_3:
        test    rsi, rsi
        je      .LBB3_19
        mov     rax, rbx
        shr     rax, 2
        mov     qword ptr [rsp + 8], rsi
        mov     qword ptr [rsp + 16], rax
        mov     qword ptr [rsp + 24], r14
        cmp     rax, r14
        je      .LBB3_15
        jb      .LBB3_10
        test    rbx, rbx
        je      .LBB3_15
        lea     r15, [4*r14]
        cmp     rbx, r15
        je      .LBB3_14
        mov     edx, 4
        mov     rdi, rsi
        mov     rsi, rbx
        test    r15, r15
        je      .LBB3_9
        mov     rcx, r15
        call    qword ptr [rip + __rust_realloc@GOTPCREL]
        test    rax, rax
        je      .LBB3_16
        mov     rsi, rax
        mov     rbx, r15
        jmp     .LBB3_14
.LBB3_9:
        call    qword ptr [rip + __rust_dealloc@GOTPCREL]
        xor     ebx, ebx
        mov     esi, 4
.LBB3_14:
        mov     qword ptr [rsp + 8], rsi
        shr     rbx, 2
        mov     qword ptr [rsp + 16], rbx
.LBB3_15:
        mov     rax, rsi
        mov     rdx, r14
        add     rsp, 32
        pop     rbx
        pop     r14
        pop     r15
        ret
.LBB3_18:
        call    alloc::raw_vec::RawVec<T,A>::allocate_in::{{closure}}
        ud2
.LBB3_19:
        lea     rdi, [rsp + 8]
        call    alloc::raw_vec::RawVec<T,A>::allocate_in::{{closure}}
        ud2
.LBB3_10:
        lea     rdi, [rip + .L__unnamed_1]
        lea     rdx, [rip + .L__unnamed_2]
        mov     esi, 36
        call    qword ptr [rip + core::panicking::panic@GOTPCREL]
        ud2
.LBB3_16:
        mov     esi, 4
        mov     rdi, r15
        call    qword ptr [rip + alloc::alloc::handle_alloc_error@GOTPCREL]
        ud2
        mov     rbx, rax
        lea     rdi, [rsp + 8]
        call    core::ptr::drop_in_place
        mov     rdi, rbx
        call    _Unwind_Resume@PLT
        ud2

.L__unnamed_1:
        .ascii  "Tried to shrink to a larger capacity"

.L__unnamed_3:
        .ascii  "/rustc/f05a5240440b3eaef1684a7965860fab40301947/src/libcore/macros/mod.rs"

.L__unnamed_2:
        .quad   .L__unnamed_3
        .asciz  "I\000\000\000\000\000\000\000\n\000\000\000\t\000\000"

[spastorino]

Assigning P-high as discussed as part of the Prioritization Working Group process and removing I-prioritize.

[spastorino]

Changing the priority to P-medium as we've discussed as part of the Prioritization Working Group process, if there's evidence that this regress badly some real world code and someone can prove where and how we can switch this back to P-high.

[spastorino]

Would be nice to identify where this has regressed ...

@rustbot ping cleanup

[rustbot]

Hey Cleanup Crew ICE-breakers! This bug has been identified as a good
"Cleanup ICE-breaking candidate". In case it's useful, here are some
instructions for tackling these sorts of bugs. Maybe take a look?
Thanks! <3

cc @AminArria @chrissimpkins @contrun @DutchGhost @elshize @ethanboxx @h-michael @HallerPatrick @hdhoang @hellow554 @imtsuki @jakevossen5 @kanru @KarlK90 @LeSeulArtichaut @MAdrianMattocks @matheus-consoli @mental32 @nmccarty @Noah-Kennedy @pard68 @PeytonT @pierreN @Redblueflame @RobbieClarken @RobertoSnap @robjtede @SarthakSingh31 @senden9 @shekohex @sinato @spastorino @turboladen @woshilapin @yerke