[codegen] Test and document codegen under different layouts (#3085)

* [codegen] Test and document codegen under different layouts

This improves the robustness of our suite to different code paths, and
illustrates in our documentation the relationship between type
complexity and codegen complexity.

gherrit-pr-id: G39e297d5b891f5115ef9607b090c5fbe95427d68

* [ci] Change zizmor persona from 'pedantic' to 'regular'

* [ci] Temporarily disable `zizmor` job

---------

Co-authored-by: Joshua Liebow-Feeser <joshlf@users.noreply.github.com>

Author: jswrenn

.github/workflows/ci.yml

@@ -988,6 +988,8 @@ jobs:
   zizmor:
     runs-on: ubuntu-latest
     name: zizmor
+    # FIXME(#3093): Re-enable
+    if: ${{ false }}
     steps:
       - uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # v6.0.1
         with:
@@ -1001,7 +1003,7 @@ jobs:
           # (without manually configuring a ruleset) is to fail the job, which
           # the action only does if advanced-security is false.
           advanced-security: false
-          persona: pedantic
+          persona: regular
 
   # Used to signal to branch protections that all other jobs have succeeded.
   all-jobs-succeed:

Cargo.lock

@@ -15,7 +15,7 @@
 [package]
 edition = "2021"
 name = "zerocopy"
-version = "0.8.41-alpha"
+version = "0.8.41"
 authors = [
     "Joshua Liebow-Feeser <joshlf@google.com>",
     "Jack Wrenn <jswrenn@amazon.com>",
@@ -112,13 +112,13 @@ __internal_use_only_features_that_work_on_stable = [
 ]
 
 [dependencies]
-zerocopy-derive = { version = "=0.8.41-alpha", path = "zerocopy-derive", optional = true }
+zerocopy-derive = { version = "=0.8.41", path = "zerocopy-derive", optional = true }
 
 # The "associated proc macro pattern" ensures that the versions of zerocopy and
 # zerocopy-derive remain equal, even if the 'derive' feature isn't used.
 # See: https://github.com/matklad/macro-dep-test
 [target.'cfg(any())'.dependencies]
-zerocopy-derive = { version = "=0.8.41-alpha", path = "zerocopy-derive" }
+zerocopy-derive = { version = "=0.8.41", path = "zerocopy-derive" }
 
 [dev-dependencies]
 # FIXME(#381) Remove this dependency once we have our own layout gadgets.
@@ -129,4 +129,4 @@ rustversion = "1.0"
 static_assertions = "1.1"
 testutil = { path = "testutil" }
 # In tests, unlike in production, zerocopy-derive is not optional
-zerocopy-derive = { version = "=0.8.41-alpha", path = "zerocopy-derive" }
+zerocopy-derive = { version = "=0.8.41", path = "zerocopy-derive" }

Cargo.toml

@@ -0,0 +1,23 @@
+use zerocopy_derive::*;
+
+// The only valid value of this type are the bytes `0xC0C0`.
+#[derive(TryFromBytes, KnownLayout, Immutable)]
+#[repr(u16)]
+pub enum C0C0 {
+    _XC0C0 = 0xC0C0,
+}
+
+#[derive(FromBytes, KnownLayout, Immutable)]
+#[repr(C, align(2))]
+pub struct Packet<Magic> {
+    magic_number: Magic,
+    mug_size: u8,
+    temperature: u8,
+    marshmallows: [u8; 2],
+}
+
+/// A packet begining with the magic number `0xC0C0`.
+pub type CocoPacket = Packet<C0C0>;
+
+/// A packet beginning with any two initialized bytes.
+pub type LocoPacket = Packet<[u8; 2]>;

benches/formats/coco.rs benches/formats/coco_dynamic_padding.rsbenches/formats/coco.rs renamed to benches/formats/coco_dynamic_padding.rs

@@ -0,0 +1,7 @@
+#[path = "formats/coco_static_size.rs"]
+mod format;
+
+#[unsafe(no_mangle)]
+fn bench_read_from_bytes_static_size(source: &[u8]) -> Option<format::LocoPacket> {
+    zerocopy::FromBytes::read_from_bytes(source).ok()
+}

benches/formats/coco_no_padding.rs benches/formats/coco_dynamic_size.rsbenches/formats/coco_no_padding.rs renamed to benches/formats/coco_dynamic_size.rs

@@ -0,0 +1,15 @@
+bench_read_from_bytes_static_size:
+	mov rcx, rsi
+	cmp rsi, 6
+	jne .LBB5_2
+	mov eax, dword ptr [rdi]
+	movzx ecx, word ptr [rdi + 4]
+	shl rcx, 32
+	or rcx, rax
+.LBB5_2:
+	shl rcx, 16
+	inc rcx
+	xor eax, eax
+	cmp rsi, 6
+	cmove rax, rcx
+	ret

benches/formats/coco_static_size.rs

@@ -0,0 +1,65 @@
+Iterations:        100
+Instructions:      1300
+Total Cycles:      377
+Total uOps:        1400
+
+Dispatch Width:    4
+uOps Per Cycle:    3.71
+IPC:               3.45
+Block RThroughput: 3.5
+
+
+Instruction Info:
+[1]: #uOps
+[2]: Latency
+[3]: RThroughput
+[4]: MayLoad
+[5]: MayStore
+[6]: HasSideEffects (U)
+
+[1]    [2]    [3]    [4]    [5]    [6]    Instructions:
+ 1      1     0.33                        mov	rcx, rsi
+ 1      1     0.33                        cmp	rsi, 6
+ 1      1     1.00                        jne	.LBB5_2
+ 1      5     0.50    *                   mov	eax, dword ptr [rdi]
+ 1      5     0.50    *                   movzx	ecx, word ptr [rdi + 4]
+ 1      1     0.50                        shl	rcx, 32
+ 1      1     0.33                        or	rcx, rax
+ 1      1     0.50                        shl	rcx, 16
+ 1      1     0.33                        inc	rcx
+ 1      0     0.25                        xor	eax, eax
+ 1      1     0.33                        cmp	rsi, 6
+ 2      2     0.67                        cmove	rax, rcx
+ 1      1     1.00                  U     ret
+
+
+Resources:
+[0]   - SBDivider
+[1]   - SBFPDivider
+[2]   - SBPort0
+[3]   - SBPort1
+[4]   - SBPort4
+[5]   - SBPort5
+[6.0] - SBPort23
+[6.1] - SBPort23
+
+
+Resource pressure per iteration:
+[0]    [1]    [2]    [3]    [4]    [5]    [6.0]  [6.1]  
+ -      -     3.66   3.67    -     3.67   1.00   1.00   
+
+Resource pressure by instruction:
+[0]    [1]    [2]    [3]    [4]    [5]    [6.0]  [6.1]  Instructions:
+ -      -     0.63   0.36    -     0.01    -      -     mov	rcx, rsi
+ -      -     0.05   0.05    -     0.90    -      -     cmp	rsi, 6
+ -      -      -      -      -     1.00    -      -     jne	.LBB5_2
+ -      -      -      -      -      -      -     1.00   mov	eax, dword ptr [rdi]
+ -      -      -      -      -      -     1.00    -     movzx	ecx, word ptr [rdi + 4]
+ -      -     0.97    -      -     0.03    -      -     shl	rcx, 32
+ -      -     0.02   0.35    -     0.63    -      -     or	rcx, rax
+ -      -     0.98    -      -     0.02    -      -     shl	rcx, 16
+ -      -      -     0.98    -     0.02    -      -     inc	rcx
+ -      -      -      -      -      -      -      -     xor	eax, eax
+ -      -     0.03   0.93    -     0.04    -      -     cmp	rsi, 6
+ -      -     0.98   1.00    -     0.02    -      -     cmove	rax, rcx
+ -      -      -      -      -     1.00    -      -     ret

benches/read_from_bytes.rs

@@ -0,0 +1,10 @@
+#[path = "formats/coco_static_size.rs"]
+mod format;
+
+#[unsafe(no_mangle)]
+fn bench_read_from_prefix_static_size(source: &[u8]) -> Option<format::LocoPacket> {
+    match zerocopy::FromBytes::read_from_prefix(source) {
+        Ok((packet, _rest)) => Some(packet),
+        _ => None,
+    }
+}