zkvm: add partial std support

Co-authored-by: Frank Laub <[email protected]>
Co-authored-by: nils <[email protected]>
Co-authored-by: Victor Graf <[email protected]>
Co-authored-by: weikengchen <[email protected]>

library/std/build.rs

@@ -35,6 +35,7 @@ fn main() {
         || target.contains("hurd")
         || target.contains("uefi")
         || target.contains("teeos")
+        || target.contains("zkvm")
         // See src/bootstrap/synthetic_targets.rs
         || env::var("RUSTC_BOOTSTRAP_SYNTHETIC_TARGET").is_ok()
     {

library/std/src/sys/pal/common/alloc.rs

@@ -16,7 +16,7 @@ use crate::ptr;
     target_arch = "sparc",
     target_arch = "wasm32",
     target_arch = "hexagon",
-    all(target_arch = "riscv32", not(target_os = "espidf")),
+    all(target_arch = "riscv32", not(any(target_os = "espidf", target_os = "zkvm"))),
     all(target_arch = "xtensa", not(target_os = "espidf")),
 ))]
 pub const MIN_ALIGN: usize = 8;
@@ -32,11 +32,11 @@ pub const MIN_ALIGN: usize = 8;
     target_arch = "wasm64",
 ))]
 pub const MIN_ALIGN: usize = 16;
-// The allocator on the esp-idf platform guarantees 4 byte alignment.
-#[cfg(any(
-    all(target_arch = "riscv32", target_os = "espidf"),
+// The allocator on the esp-idf and zkvm platforms guarantee 4 byte alignment.
+#[cfg(all(any(
+    all(target_arch = "riscv32", any(target_os = "espidf", target_os = "zkvm")),
     all(target_arch = "xtensa", target_os = "espidf"),
-))]
+)))]
 pub const MIN_ALIGN: usize = 4;
 
 pub unsafe fn realloc_fallback(

library/std/src/sys/pal/mod.rs

@@ -55,6 +55,9 @@ cfg_if::cfg_if! {
     } else if #[cfg(target_os = "teeos")] {
         mod teeos;
         pub use self::teeos::*;
+    } else if #[cfg(target_os = "zkvm")] {
+        mod zkvm;
+        pub use self::zkvm::*;
     } else {
         mod unsupported;
         pub use self::unsupported::*;

library/std/src/sys/pal/zkvm/abi.rs

@@ -0,0 +1,55 @@
+//! ABI definitions for symbols exported by risc0-zkvm-platform.
+
+// Included here so we don't have to depend on risc0-zkvm-platform.
+//
+// FIXME: Should we move this to the "libc" crate?  It seems like other
+// architectures put a lot of this kind of stuff there.  But there's
+// currently no risc0 fork of the libc crate, so we'd either have to
+// fork it or upstream it.
+
+#![allow(dead_code)]
+pub const DIGEST_WORDS: usize = 8;
+
+/// Standard IO file descriptors for use with sys_read and sys_write.
+pub mod fileno {
+    pub const STDIN: u32 = 0;
+    pub const STDOUT: u32 = 1;
+    pub const STDERR: u32 = 2;
+    pub const JOURNAL: u32 = 3;
+}
+
+extern "C" {
+    // Wrappers around syscalls provided by risc0-zkvm-platform:
+    pub fn sys_halt();
+    pub fn sys_output(output_id: u32, output_value: u32);
+    pub fn sys_sha_compress(
+        out_state: *mut [u32; DIGEST_WORDS],
+        in_state: *const [u32; DIGEST_WORDS],
+        block1_ptr: *const [u32; DIGEST_WORDS],
+        block2_ptr: *const [u32; DIGEST_WORDS],
+    );
+    pub fn sys_sha_buffer(
+        out_state: *mut [u32; DIGEST_WORDS],
+        in_state: *const [u32; DIGEST_WORDS],
+        buf: *const u8,
+        count: u32,
+    );
+    pub fn sys_rand(recv_buf: *mut u32, words: usize);
+    pub fn sys_panic(msg_ptr: *const u8, len: usize) -> !;
+    pub fn sys_log(msg_ptr: *const u8, len: usize);
+    pub fn sys_cycle_count() -> usize;
+    pub fn sys_read(fd: u32, recv_buf: *mut u8, nrequested: usize) -> usize;
+    pub fn sys_write(fd: u32, write_buf: *const u8, nbytes: usize);
+    pub fn sys_getenv(
+        recv_buf: *mut u32,
+        words: usize,
+        varname: *const u8,
+        varname_len: usize,
+    ) -> usize;
+    pub fn sys_argc() -> usize;
+    pub fn sys_argv(out_words: *mut u32, out_nwords: usize, arg_index: usize) -> usize;
+
+    // Allocate memory from global HEAP.
+    pub fn sys_alloc_words(nwords: usize) -> *mut u32;
+    pub fn sys_alloc_aligned(nwords: usize, align: usize) -> *mut u8;
+}

library/std/src/sys/pal/zkvm/alloc.rs

@@ -0,0 +1,15 @@
+use super::abi;
+use crate::alloc::{GlobalAlloc, Layout, System};
+
+#[stable(feature = "alloc_system_type", since = "1.28.0")]
+unsafe impl GlobalAlloc for System {
+    #[inline]
+    unsafe fn alloc(&self, layout: Layout) -> *mut u8 {
+        abi::sys_alloc_aligned(layout.size(), layout.align())
+    }
+
+    #[inline]
+    unsafe fn dealloc(&self, _ptr: *mut u8, _layout: Layout) {
+        // this allocator never deallocates memory
+    }
+}

library/std/src/sys/pal/zkvm/args.rs

@@ -0,0 +1,80 @@
+use super::{abi, WORD_SIZE};
+use crate::ffi::OsString;
+use crate::fmt;
+use crate::sys_common::FromInner;
+
+pub struct Args {
+    i_forward: usize,
+    i_back: usize,
+    count: usize,
+}
+
+pub fn args() -> Args {
+    let count = unsafe { abi::sys_argc() };
+    Args { i_forward: 0, i_back: 0, count }
+}
+
+impl Args {
+    /// Use sys_argv to get the arg at the requested index. Does not check that i is less than argc
+    /// and will not return if the index is out of bounds.
+    fn argv(i: usize) -> OsString {
+        let arg_len = unsafe { abi::sys_argv(crate::ptr::null_mut(), 0, i) };
+
+        let arg_len_words = (arg_len + WORD_SIZE - 1) / WORD_SIZE;
+        let words = unsafe { abi::sys_alloc_words(arg_len_words) };
+
+        let arg_len2 = unsafe { abi::sys_argv(words, arg_len_words, i) };
+        debug_assert_eq!(arg_len, arg_len2);
+
+        // Convert to OsString.
+        //
+        // FIXME: We can probably get rid of the extra copy here if we
+        // reimplement "os_str" instead of just using the generic unix
+        // "os_str".
+        let arg_bytes: &[u8] =
+            unsafe { crate::slice::from_raw_parts(words.cast() as *const u8, arg_len) };
+        OsString::from_inner(super::os_str::Buf { inner: arg_bytes.to_vec() })
+    }
+}
+
+impl fmt::Debug for Args {
+    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
+        f.debug_list().finish()
+    }
+}
+
+impl Iterator for Args {
+    type Item = OsString;
+
+    fn next(&mut self) -> Option<OsString> {
+        if self.i_forward >= self.count - self.i_back {
+            None
+        } else {
+            let arg = Self::argv(self.i_forward);
+            self.i_forward += 1;
+            Some(arg)
+        }
+    }
+
+    fn size_hint(&self) -> (usize, Option<usize>) {
+        (self.count, Some(self.count))
+    }
+}
+
+impl ExactSizeIterator for Args {
+    fn len(&self) -> usize {
+        self.count
+    }
+}
+
+impl DoubleEndedIterator for Args {
+    fn next_back(&mut self) -> Option<OsString> {
+        if self.i_back >= self.count - self.i_forward {
+            None
+        } else {
+            let arg = Self::argv(self.count - 1 - self.i_back);
+            self.i_back += 1;
+            Some(arg)
+        }
+    }
+}

library/std/src/sys/pal/zkvm/env.rs

@@ -0,0 +1,9 @@
+pub mod os {
+    pub const FAMILY: &str = "";
+    pub const OS: &str = "";
+    pub const DLL_PREFIX: &str = "";
+    pub const DLL_SUFFIX: &str = ".elf";
+    pub const DLL_EXTENSION: &str = "elf";
+    pub const EXE_SUFFIX: &str = ".elf";
+    pub const EXE_EXTENSION: &str = "elf";
+}

library/std/src/sys/pal/zkvm/mod.rs

@@ -0,0 +1,93 @@
+//! System bindings for the risc0 zkvm platform
+//!
+//! This module contains the facade (aka platform-specific) implementations of
+//! OS level functionality for zkvm.
+//!
+//! This is all super highly experimental and not actually intended for
+//! wide/production use yet, it's still all in the experimental category. This
+//! will likely change over time.
+
+const WORD_SIZE: usize = core::mem::size_of::<u32>();
+
+pub mod alloc;
+#[path = "../zkvm/args.rs"]
+pub mod args;
+#[path = "../unix/cmath.rs"]
+pub mod cmath;
+pub mod env;
+#[path = "../unsupported/fs.rs"]
+pub mod fs;
+#[path = "../unsupported/io.rs"]
+pub mod io;
+#[path = "../unsupported/net.rs"]
+pub mod net;
+#[path = "../unsupported/once.rs"]
+pub mod once;
+pub mod os;
+#[path = "../unix/os_str.rs"]
+pub mod os_str;
+#[path = "../unix/path.rs"]
+pub mod path;
+#[path = "../unsupported/pipe.rs"]
+pub mod pipe;
+#[path = "../unsupported/process.rs"]
+pub mod process;
+pub mod stdio;
+pub mod thread_local_key;
+#[path = "../unsupported/time.rs"]
+pub mod time;
+
+#[path = "../unsupported/locks/mod.rs"]
+pub mod locks;
+#[path = "../unsupported/thread.rs"]
+pub mod thread;
+
+#[path = "../unsupported/thread_parking.rs"]
+pub mod thread_parking;
+
+mod abi;
+
+use crate::io as std_io;
+
+pub mod memchr {
+    pub use core::slice::memchr::{memchr, memrchr};
+}
+
+// SAFETY: must be called only once during runtime initialization.
+// NOTE: this is not guaranteed to run, for example when Rust code is called externally.
+pub unsafe fn init(_argc: isize, _argv: *const *const u8, _sigpipe: u8) {}
+
+// SAFETY: must be called only once during runtime cleanup.
+// NOTE: this is not guaranteed to run, for example when the program aborts.
+pub unsafe fn cleanup() {}
+
+pub fn unsupported<T>() -> std_io::Result<T> {
+    Err(unsupported_err())
+}
+
+pub fn unsupported_err() -> std_io::Error {
+    std_io::const_io_error!(
+        std_io::ErrorKind::Unsupported,
+        "operation not supported on this platform",
+    )
+}
+
+pub fn is_interrupted(_code: i32) -> bool {
+    false
+}
+
+pub fn decode_error_kind(_code: i32) -> crate::io::ErrorKind {
+    crate::io::ErrorKind::Uncategorized
+}
+
+pub fn abort_internal() -> ! {
+    core::intrinsics::abort();
+}
+
+pub fn hashmap_random_keys() -> (u64, u64) {
+    let mut buf = [0u32; 4];
+    unsafe {
+        abi::sys_rand(buf.as_mut_ptr(), 4);
+    };
+    ((buf[0] as u64) << 32 + buf[1] as u64, (buf[2] as u64) << 32 + buf[3] as u64)
+}