From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from firstgate.proxmox.com (firstgate.proxmox.com [212.224.123.68]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits)) (No client certificate requested) by lists.proxmox.com (Postfix) with ESMTPS id 6ED36695B8 for ; Fri, 12 Feb 2021 15:44:36 +0100 (CET) Received: from firstgate.proxmox.com (localhost [127.0.0.1]) by firstgate.proxmox.com (Proxmox) with ESMTP id 6CF4B2810C for ; Fri, 12 Feb 2021 15:44:36 +0100 (CET) Received: from proxmox-new.maurer-it.com (proxmox-new.maurer-it.com [212.186.127.180]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits)) (No client certificate requested) by firstgate.proxmox.com (Proxmox) with ESMTPS id 380BF280FA for ; Fri, 12 Feb 2021 15:44:35 +0100 (CET) Received: from proxmox-new.maurer-it.com (localhost.localdomain [127.0.0.1]) by proxmox-new.maurer-it.com (Proxmox) with ESMTP id 0310C4624F for ; Fri, 12 Feb 2021 15:44:35 +0100 (CET) From: Dominik Csapak To: pbs-devel@lists.proxmox.com Date: Fri, 12 Feb 2021 15:44:32 +0100 Message-Id: <20210212144433.30668-3-d.csapak@proxmox.com> X-Mailer: git-send-email 2.20.1 In-Reply-To: <20210212144433.30668-1-d.csapak@proxmox.com> References: <20210212144433.30668-1-d.csapak@proxmox.com> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit X-SPAM-LEVEL: Spam detection results: 0 AWL 0.225 Adjusted score from AWL reputation of From: address KAM_DMARC_STATUS 0.01 Test Rule for DKIM or SPF Failure with Strict Alignment RCVD_IN_DNSWL_MED -2.3 Sender listed at https://www.dnswl.org/, medium trust SPF_HELO_NONE 0.001 SPF: HELO does not publish an SPF Record SPF_PASS -0.001 SPF: sender matches SPF record URIBL_BLOCKED 0.001 ADMINISTRATOR NOTICE: The query to URIBL was blocked. See http://wiki.apache.org/spamassassin/DnsBlocklists#dnsbl-block for more information. [mod.rs] Subject: [pbs-devel] [PATCH proxmox v4 2/2] proxmox: add sparse_copy(_async) to tools::io X-BeenThere: pbs-devel@lists.proxmox.com X-Mailman-Version: 2.1.29 Precedence: list List-Id: Proxmox Backup Server development discussion List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 12 Feb 2021 14:44:36 -0000 this is able to seek the target instead of writing zeroes, which generates sparse files where supported also add tests for it Signed-off-by: Dominik Csapak --- proxmox/src/tools/io/mod.rs | 320 ++++++++++++++++++++++++++++++++++++ 1 file changed, 320 insertions(+) diff --git a/proxmox/src/tools/io/mod.rs b/proxmox/src/tools/io/mod.rs index 2e92ebb..b935921 100644 --- a/proxmox/src/tools/io/mod.rs +++ b/proxmox/src/tools/io/mod.rs @@ -3,8 +3,328 @@ //! The [`ReadExt`] trait provides additional operations for handling byte buffers for types //! implementing [`Read`](std::io::Read). +use std::io::{self, ErrorKind, Read, Seek, SeekFrom, Write}; + mod read; pub use read::*; mod write; pub use write::*; + +fn buffer_is_zero(buf: &[u8]) -> bool { + !buf.chunks(128) + .map(|aa| aa.iter().fold(0, |a, b| a | b) != 0) + .any(|a| a) +} + +/// Result of a sparse copy call +/// contains the amount of written/seeked bytes +/// and if the last operation was a seek +pub struct SparseCopyResult { + pub written: u64, + pub seeked_last: bool, +} + +/// copy similar to io::copy, but seeks the target when encountering +/// zero bytes instead of writing them +/// +/// Example use: +/// ``` +/// # use std::io; +/// # use proxmox::tools::io::sparse_copy; +/// fn code(mut reader: R, mut writer: W) -> io::Result<()> +/// where +/// R: io::Read, +/// W: io::Write + io::Seek, +/// { +/// let res = sparse_copy(&mut reader, &mut writer)?; +/// +/// println!("last part was seeked: {}", res.seeked_last); +/// println!("written: {}", res.written); +/// +/// Ok(()) +/// } +/// ``` +pub fn sparse_copy( + reader: &mut R, + writer: &mut W, +) -> Result { + let mut buf = crate::tools::byte_buffer::ByteBuffer::with_capacity(4096); + let mut written = 0; + let mut seek_amount: i64 = 0; + let mut seeked_last = false; + loop { + buf.clear(); + let len = match buf.read_from(reader) { + Ok(len) => len, + Err(ref e) if e.kind() == ErrorKind::Interrupted => continue, + Err(e) => return Err(e), + }; + + if len > 0 && buffer_is_zero(&buf[..]) { + seek_amount += len as i64; + continue; + } + + if seek_amount > 0 { + writer.seek(SeekFrom::Current(seek_amount))?; + written += seek_amount as u64; + seek_amount = 0; + seeked_last = true; + } + + if len > 0 { + writer.write_all(&buf[..])?; + written += len as u64; + seeked_last = false; + } else { + return Ok(SparseCopyResult { + written, + seeked_last, + }); + } + } +} + +#[cfg(feature = "tokio")] +use tokio::io::{AsyncRead, AsyncSeek, AsyncSeekExt, AsyncWrite, AsyncWriteExt}; + +#[cfg(feature = "tokio")] +/// copy similar to tokio::io::copy, but seeks the target when encountering +/// zero bytes instead of writing them +/// +/// Example: +/// ```no_run +/// # use std::io; +/// # use tokio::io::{AsyncRead, AsyncWrite, AsyncSeek}; +/// # use proxmox::tools::io::sparse_copy_async; +/// async fn code(mut reader: R, mut writer: W) -> io::Result<()> +/// where +/// R: AsyncRead + Unpin, +/// W: AsyncWrite + AsyncSeek + Unpin, +/// { +/// let res = sparse_copy_async(&mut reader, &mut writer).await?; +/// +/// println!("last part was seeked: {}", res.seeked_last); +/// println!("written: {}", res.written); +/// +/// Ok(()) +/// } +/// ``` +pub async fn sparse_copy_async( + reader: &mut R, + writer: &mut W, +) -> Result +where + R: AsyncRead + Unpin, + W: AsyncWrite + AsyncSeek + Unpin, +{ + let mut buf = crate::tools::byte_buffer::ByteBuffer::with_capacity(4096); + let mut written = 0; + let mut seek_amount: i64 = 0; + let mut seeked_last = false; + loop { + buf.clear(); + let len = match buf.read_from_async(reader).await { + Ok(len) => len, + Err(ref e) if e.kind() == ErrorKind::Interrupted => continue, + Err(e) => return Err(e), + }; + + if len > 0 && buffer_is_zero(&buf[..]) { + seek_amount += len as i64; + continue; + } + + if seek_amount > 0 { + writer.seek(SeekFrom::Current(seek_amount)).await?; + written += seek_amount as u64; + seek_amount = 0; + seeked_last = true; + } + + if len > 0 { + writer.write_all(&buf[..]).await?; + written += len as u64; + seeked_last = false; + } else { + return Ok(SparseCopyResult { + written, + seeked_last, + }); + } + } +} + +#[cfg(test)] +mod test { + use std::io::Cursor; + use std::pin::Pin; + use std::task::{Context, Poll}; + + use tokio::io::{AsyncRead, AsyncSeek, AsyncWrite, ReadBuf}; + + use crate::tools::io::{sparse_copy, sparse_copy_async}; + + const LEN: usize = 10000; + + #[test] + fn test_sparse_copy() { + // test sparse + let mut test_data = Vec::new(); + for _ in 0..LEN / 2 { + test_data.push(1u8); + } + for _ in 0..LEN / 2 { + test_data.push(0u8); + } + let mut test_data = Cursor::new(test_data); + let mut result_data = Cursor::new(vec![0; LEN]); + + let result = + sparse_copy(&mut test_data, &mut result_data).expect("error during sparse copy"); + assert_eq!(result.written, LEN as u64); + assert_eq!(result.seeked_last, true); + for i in 0..LEN { + if i < LEN / 2 { + assert_eq!(result_data.get_ref()[i], 1); + } else { + assert_eq!(result_data.get_ref()[i], 0); + } + } + + // test non sparse + let mut test_data = Cursor::new(vec![1; LEN]); + let mut result_data = Cursor::new(vec![0; LEN]); + + let result = + sparse_copy(&mut test_data, &mut result_data).expect("error during sparse copy"); + assert_eq!(result.written, LEN as u64); + assert_eq!(result.seeked_last, false); + for i in 0..LEN { + assert_eq!(result_data.get_ref()[i], 1); + } + } + + struct DummyAsyncReader { + inner: R, + } + + struct DummyAsyncWriter { + inner: W, + seek_pos: u64, + } + + impl AsyncRead for DummyAsyncReader { + fn poll_read( + self: Pin<&mut Self>, + _cx: &mut Context<'_>, + buf: &mut ReadBuf<'_>, + ) -> Poll> { + let this = Pin::get_mut(self); + let mut read_buf = buf.initialize_unfilled(); + match this.inner.read(&mut read_buf) { + Ok(len) => { + buf.advance(len); + Poll::Ready(Ok(())) + } + Err(err) => Poll::Ready(Err(err)), + } + } + } + + impl AsyncWrite for DummyAsyncWriter { + fn poll_write( + self: Pin<&mut Self>, + _cx: &mut Context<'_>, + buf: &[u8], + ) -> Poll> { + let this = Pin::get_mut(self); + match this.inner.write(buf) { + Ok(len) => Poll::Ready(Ok(len)), + Err(err) => Poll::Ready(Err(err)), + } + } + + fn poll_flush(self: Pin<&mut Self>, _cx: &mut Context<'_>) -> Poll> { + Poll::Ready(Ok(())) + } + + fn poll_shutdown(self: Pin<&mut Self>, _cx: &mut Context<'_>) -> Poll> { + Poll::Ready(Ok(())) + } + } + + impl AsyncSeek for DummyAsyncWriter { + fn start_seek(self: Pin<&mut Self>, position: std::io::SeekFrom) -> std::io::Result<()> { + let this = Pin::get_mut(self); + this.seek_pos = this.inner.seek(position)?; + Ok(()) + } + + fn poll_complete( + self: Pin<&mut Self>, + _cx: &mut Context<'_>, + ) -> Poll> { + let this = Pin::get_mut(self); + Poll::Ready(Ok(this.seek_pos)) + } + } + + #[test] + fn test_sparse_copy_async() { + let fut = async { + // test sparse + let mut test_data = Vec::new(); + for _ in 0..LEN / 2 { + test_data.push(1u8); + } + for _ in 0..LEN / 2 { + test_data.push(0u8); + } + let mut test_data = DummyAsyncReader { + inner: Cursor::new(test_data), + }; + let mut result_data = DummyAsyncWriter { + inner: Cursor::new(vec![0; LEN]), + seek_pos: 0, + }; + + let result = sparse_copy_async(&mut test_data, &mut result_data) + .await + .expect("error during sparse copy"); + + assert_eq!(result.written, LEN as u64); + assert_eq!(result.seeked_last, true); + for i in 0..LEN { + if i < LEN / 2 { + assert_eq!(result_data.inner.get_ref()[i], 1); + } else { + assert_eq!(result_data.inner.get_ref()[i], 0); + } + } + + // test non sparse + let mut test_data = DummyAsyncReader { + inner: Cursor::new(vec![1; LEN]), + }; + let mut result_data = DummyAsyncWriter { + inner: Cursor::new(vec![0; LEN]), + seek_pos: 0, + }; + + let result = sparse_copy_async(&mut test_data, &mut result_data) + .await + .expect("error during sparse copy"); + + assert_eq!(result.written, LEN as u64); + assert_eq!(result.seeked_last, false); + for i in 0..LEN { + assert_eq!(result_data.inner.get_ref()[i], 1); + } + Ok(()) + }; + + crate::tools::poll_once::poll_result_once(fut).expect("ok") + } +} -- 2.20.1