From a14fa38a0de97173cf86476ded9d5bd7f7060305 Mon Sep 17 00:00:00 2001
From: Marcus Holland-Moritz <github@mhxnet.de>
Date: Sun, 6 Nov 2022 10:52:02 +0100
Subject: [PATCH] Reintroduce --num-scanner-workers

---
 doc/mkdwarfs.md  | 17 ++++++++++++-----
 src/mkdwarfs.cpp | 13 ++++++++++---
 2 files changed, 22 insertions(+), 8 deletions(-)

diff --git a/doc/mkdwarfs.md b/doc/mkdwarfs.md
index 2df9da4a..e4d4fd8b 100644
--- a/doc/mkdwarfs.md
+++ b/doc/mkdwarfs.md
@@ -66,16 +66,23 @@ Most other options are concerned with compression tuning:
 - `-N`, `--num-workers=`*value*:
   Number of worker threads used for building the filesystem. This defaults
   to the number of processors available on your system. Use this option if
-  you want to limit the resources used by `mkdwarfs`.
-  This option affects both the scanning phase and the compression phase.
+  you want to limit the resources used by `mkdwarfs` or to optimize build
+  speed. This option affects only the compression phase.
+  In the compression phase, the worker threads are used to compress the
+  individual filesystem blocks in the background. Ordering, segmenting
+  and block building are, again, single-threaded and run independently.
+
+- `--num-scanner-workers=`*value*:
+  Number of worker threads used for building the filesystem. This defaults
+  to the number of processors available on your system. Use this option if
+  you want to limit the resources used by `mkdwarfs` or to optimize build
+  speed. This option affects only the scanning phase. By default, the same
+  value is used as for `--num-workers`.
   In the scanning phase, the worker threads are used to scan files in the
   background as they are discovered. File scanning includes checksumming
   for de-duplication as well as (optionally) checksumming for similarity
   computation, depending on the `--order` option. File discovery itself
   is single-threaded and runs independently from the scanning threads.
-  In the compression phase, the worker threads are used to compress the
-  individual filesystem blocks in the background. Ordering, segmenting
-  and block building are, again, single-threaded and run independently.
 
 - `-B`, `--max-lookback-blocks=`*value*:
   Specify how many of the most recent blocks to scan for duplicate segments.
diff --git a/src/mkdwarfs.cpp b/src/mkdwarfs.cpp
index 44b79fe8..70a14543 100644
--- a/src/mkdwarfs.cpp
+++ b/src/mkdwarfs.cpp
@@ -374,7 +374,7 @@ int mkdwarfs(int argc, char** argv) {
       time_resolution, order, progress_mode, recompress_opts, pack_metadata,
       file_hash_algo, debug_filter;
   std::vector<std::string> filter;
-  size_t num_workers;
+  size_t num_workers, num_scanner_workers;
   bool no_progress = false, remove_header = false, no_section_index = false,
        force_overwrite = false;
   unsigned level;
@@ -421,7 +421,10 @@ int mkdwarfs(int argc, char** argv) {
         "block size bits (size = 2^arg bits)")
     ("num-workers,N",
         po::value<size_t>(&num_workers)->default_value(num_cpu),
-        "number of scanner/writer worker threads")
+        "number of writer (compression) worker threads")
+    ("num-scanner-workers",
+        po::value<size_t>(&num_scanner_workers),
+        "number of scanner (hashing) worker threads")
     ("max-lookback-blocks,B",
         po::value<size_t>(&cfg.max_active_blocks)->default_value(1),
         "how many blocks to scan for segments")
@@ -715,8 +718,12 @@ int mkdwarfs(int argc, char** argv) {
 
   size_t mem_limit = parse_size_with_unit(memory_limit);
 
+  if (!vm.count("num-scanner-workers")) {
+    num_scanner_workers = num_workers;
+  }
+
   worker_group wg_compress("compress", num_workers);
-  worker_group wg_scanner("scanner", num_workers);
+  worker_group wg_scanner("scanner", num_scanner_workers);
 
   if (vm.count("debug-filter")) {
     if (auto it = debug_filter_modes.find(debug_filter);