process_vm_readv, process_vm_writev should be available on android

[rupansh]

process_vm_readv and process_vm_writev generally return -EPERM(-1) when used in android, however the syscalls work fine when running under privileged shell.
The following was tested(with modified nix) on Android 9/10 running Linux Kernel 4.9 on a SDM 845 phone:

fn read_mem_real(pid: i32, addr: usize, vsize: usize) -> Result<Vec<u8>, String> {
    let mut res = vec![0; vsize];
    let riovec: [RemoteIoVec; 1] = [RemoteIoVec {base: addr, len: vsize}];

    match process_vm_readv(Pid::from_raw(pid), &[IoVec::from_mut_slice(&mut res)], &riovec) {
        Err(e) => {
            log::info!("ERR {}", e);
            return Err(e.to_string());
        },
        Ok(_) => {
            return Ok(res);
        }
    }
}

This will work fine when run under su shell.

[rust-highfive]

Thanks for the pull request, and welcome! The Rust team is excited to review your changes, and you should hear from @JohnTitor (or someone else) soon.

If any changes to this PR are deemed necessary, please add them as extra commits. This ensures that the reviewer can see what has changed since they last reviewed the code. Due to the way GitHub handles out-of-date commits, this should also make it reasonably obvious what issues have or haven't been addressed. Large or tricky changes may require several passes of review and changes.

Please see the contribution instructions for more information.

[joshtriplett]

Closing and reopening to re-run CI.

[rupansh]

@joshtriplett I am not sure why its failing. I used android-ndk's aarch64-linux-android27-clang

[joshtriplett]

On Fri, Oct 02, 2020 at 07:46:05PM -0700, Rupansh wrote: @joshtriplett I am not sure why its failing. I used android-ndk's aarch64-linux-android27-clang

Since it succeeds on other Android targets, and fails on aarch64, is it possible that the Android SDK for aarch64 doesn't have that syscall in its headers yet?

[rupansh]

On Fri, Oct 02, 2020 at 07:46:05PM -0700, Rupansh wrote: @joshtriplett I am not sure why its failing. I used android-ndk's aarch64-linux-android27-clang
Since it succeeds on other Android targets, and fails on aarch64, is it possible that the Android SDK for aarch64 doesn't have that syscall in its headers yet?

sysroot/usr/include/sys/uio.h has it in my case for android27. It was added in a later version. android21(Lollipop, the TC used by CI) is pretty old now!
Edit: it was added with android23
sysroot/usr/include/sys/uio.h

#if __ANDROID_API__ >= 23
ssize_t process_vm_readv(pid_t __pid, const struct iovec* __local_iov, unsigned long __local_iov_count, const struct iovec* __remote_iov, unsigned long __remote_iov_count, unsigned long __flags) __INTRODUCED_IN(23)
ssize_t process_vm_writev(pid_t __pid, const struct iovec* __local_iov, unsigned long __local_iov_count, const struct iovec* __remote_iov, unsigned long __remote_iov_count, unsigned long __flags) __INTRODUCED_IN(23)
#endif /* __ANDROID_API__ >= 23 */

[JohnTitor]

The failure on AArch64 should be related to #1765. Could you skip the test for it like https://github.com/rust-lang/libc/pull/1867/files#diff-aedce4e8d381f53dae77653609e3174da7ee2a184308251ad33f7a47bebabe14R1593-R1608?

[JohnTitor]

Thanks!
@bors r+

[bors]

📌 Commit 0121c5c has been approved by JohnTitor

[bors]

⌛ Testing commit 0121c5c with merge 2d044d9...

[bors]

☀️ Test successful - checks-actions, checks-cirrus-freebsd-10, checks-cirrus-freebsd-11, checks-cirrus-freebsd-12
Approved by: JohnTitor
Pushing 2d044d9 to master...