From b7e02d0ae4639fb7d83e05774d549dd29089e919 Mon Sep 17 00:00:00 2001
From: Joe Thornber <ejt@redhat.com>
Date: Thu, 9 Jul 2020 13:52:53 +0100
Subject: [PATCH] [thin_metadata_pack] FIx bug where pack would try and read
 too much.

---
 Cargo.lock           |  7 +++++++
 Cargo.toml           |  1 +
 src/pack/toplevel.rs | 21 +++++++++------------
 3 files changed, 17 insertions(+), 12 deletions(-)

diff --git a/Cargo.lock b/Cargo.lock
index 0ca6d78..25d5f5b 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -21,6 +21,11 @@ dependencies = [
  "winapi 0.3.8 (registry+https://github.com/rust-lang/crates.io-index)",
 ]
 
+[[package]]
+name = "anyhow"
+version = "1.0.31"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+
 [[package]]
 name = "arrayvec"
 version = "0.4.12"
@@ -380,6 +385,7 @@ dependencies = [
 name = "thinp"
 version = "0.1.0"
 dependencies = [
+ "anyhow 1.0.31 (registry+https://github.com/rust-lang/crates.io-index)",
  "byteorder 1.3.4 (registry+https://github.com/rust-lang/crates.io-index)",
  "clap 2.33.1 (registry+https://github.com/rust-lang/crates.io-index)",
  "crc32c 0.4.0 (registry+https://github.com/rust-lang/crates.io-index)",
@@ -456,6 +462,7 @@ source = "registry+https://github.com/rust-lang/crates.io-index"
 "checksum adler32 1.0.4 (registry+https://github.com/rust-lang/crates.io-index)" = "5d2e7343e7fc9de883d1b0341e0b13970f764c14101234857d2ddafa1cb1cac2"
 "checksum aho-corasick 0.7.10 (registry+https://github.com/rust-lang/crates.io-index)" = "8716408b8bc624ed7f65d223ddb9ac2d044c0547b6fa4b0d554f3a9540496ada"
 "checksum ansi_term 0.11.0 (registry+https://github.com/rust-lang/crates.io-index)" = "ee49baf6cb617b853aa8d93bf420db2383fab46d314482ca2803b40d5fde979b"
+"checksum anyhow 1.0.31 (registry+https://github.com/rust-lang/crates.io-index)" = "85bb70cc08ec97ca5450e6eba421deeea5f172c0fc61f78b5357b2a8e8be195f"
 "checksum arrayvec 0.4.12 (registry+https://github.com/rust-lang/crates.io-index)" = "cd9fd44efafa8690358b7408d253adf110036b88f55672a933f01d616ad9b1b9"
 "checksum atty 0.2.14 (registry+https://github.com/rust-lang/crates.io-index)" = "d9b39be18770d11421cdb1b9947a45dd3f37e93092cbf377614828a319d5fee8"
 "checksum autocfg 1.0.0 (registry+https://github.com/rust-lang/crates.io-index)" = "f8aac770f1885fd7e387acedd76065302551364496e46b3dd00860b2f8359b9d"
diff --git a/Cargo.toml b/Cargo.toml
index e1e1c7c..dadedba 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -6,6 +6,7 @@ edition = "2018"
 license = "GPL3"
 
 [dependencies]
+anyhow = "1.0"
 byteorder = "1.3"
 clap = "2.33"
 crc32c = "0.4"
diff --git a/src/pack/toplevel.rs b/src/pack/toplevel.rs
index bd1757d..76f9bf6 100644
--- a/src/pack/toplevel.rs
+++ b/src/pack/toplevel.rs
@@ -1,3 +1,4 @@
+use anyhow::Result;
 use byteorder::{LittleEndian, ReadBytesExt, WriteBytesExt};
 
 use flate2::{read::ZlibDecoder, write::ZlibEncoder, Compression};
@@ -34,11 +35,6 @@ fn shuffle<T>(v: &mut Vec<T>) {
     v.shuffle(&mut rng);
 }
 
-// FIXME: move to a utils module
-fn div_up(n: u64, d: u64) -> u64 {
-    (n + d - 1) / d
-}
-
 // Each thread processes multiple contiguous runs of blocks, called
 // chunks.  Chunks are shuffled so each thread gets chunks spread
 // across the dev in case there are large regions that don't contain
@@ -47,12 +43,17 @@ fn mk_chunk_vecs(nr_blocks: u64, nr_jobs: u64) -> Vec<Vec<(u64, u64)>> {
     use std::cmp::{max, min};
 
     let chunk_size = min(4 * 1024u64, max(128u64, nr_blocks / (nr_jobs * 64)));
-    let nr_chunks = div_up(nr_blocks, chunk_size);
+    let nr_chunks = nr_blocks / chunk_size;
     let mut chunks = Vec::with_capacity(nr_chunks as usize);
     for i in 0..nr_chunks {
         chunks.push((i * chunk_size, (i + 1) * chunk_size));
     }
 
+    // there may be a smaller chunk at the back of the file.
+    if nr_chunks * chunk_size < nr_blocks {
+        chunks.push((nr_chunks * chunk_size, nr_blocks));
+    }
+
     shuffle(&mut chunks);
 
     let mut vs = Vec::with_capacity(nr_jobs as usize);
@@ -104,11 +105,7 @@ pub fn pack(input_file: &str, output_file: &str) -> Result<(), Box<dyn Error>> {
     Ok(())
 }
 
-fn crunch<R, W>(
-    input: Arc<Mutex<R>>,
-    output: Arc<Mutex<W>>,
-    ranges: Vec<(u64, u64)>,
-) -> io::Result<()>
+fn crunch<R, W>(input: Arc<Mutex<R>>, output: Arc<Mutex<W>>, ranges: Vec<(u64, u64)>) -> Result<()>
 where
     R: Read + Seek,
     W: Write,
@@ -256,7 +253,7 @@ fn pack_block<W: Write>(w: &mut W, kind: BT, buf: &[u8]) {
         BT::NODE => check(&pack_btree_node(w, buf)),
         BT::INDEX => check(&pack_index(w, buf)),
         BT::BITMAP => check(&pack_bitmap(w, buf)),
-        BT::UNKNOWN => {panic!("asked to pack an unknown block type")}
+        BT::UNKNOWN => panic!("asked to pack an unknown block type"),
     }
 }