squashfs: use -no-hardlinks for reproducible squashfs images (#114454)

the nix store may contain hardlinks: derivations may output them
directly, or users may be using store optimization which automatically
hardlinks identical files in the nix store.

The presence of these links are intended to be a 'transparent'
optimization. However, when creating a squashfs image, the image
will be different depending on whether hard links were present
on the filesystem, leading to reproducibility problems.

By passing '-no-hardlinks' to mksquashfs the files are stored
as duplicates in the squashfs image. Since squashfs has support
for duplicate files this does not lead to a larger image.

For more details see
https://github.com/NixOS/nixpkgs/issues/114331

3 files changed, 81 insertions, 1 deletions

diff --git a/nixos/lib/make-squashfs.nix b/nixos/lib/make-squashfs.nix
index ee76c9c5bf24..8690c42e7ac9 100644
--- a/nixos/lib/make-squashfs.nix
+++ b/nixos/lib/make-squashfs.nix
@@ -23,6 +23,6 @@ stdenv.mkDerivation {
 
       # Generate the squashfs image.
       mksquashfs nix-path-registration $(cat $closureInfo/store-paths) $out \
-        -keep-as-directory -all-root -b 1048576 -comp ${comp}
+        -no-hardlinks -keep-as-directory -all-root -b 1048576 -comp ${comp}
     '';
 }
diff --git a/pkgs/tools/filesystems/squashfs/0001-Mksquashfs-add-no-hardlinks-option.patch b/pkgs/tools/filesystems/squashfs/0001-Mksquashfs-add-no-hardlinks-option.patch
new file mode 100644
index 000000000000..0d6804a647b2
--- /dev/null
+++ b/pkgs/tools/filesystems/squashfs/0001-Mksquashfs-add-no-hardlinks-option.patch
@@ -0,0 +1,76 @@
+From d925c9a11ee2e88ac8aac03f51892746f2bcf8cd Mon Sep 17 00:00:00 2001
+From: Phillip Lougher <phillip@squashfs.org.uk>
+Date: Thu, 25 Feb 2021 23:12:10 +0000
+Subject: [PATCH] Mksquashfs: add -no-hardlinks option
+
+Normally Mksquashfs will detect hardlinks (multiple files with the
+same inode) and hardlink them in the Squashfs image.
+
+But often hardlinks are used in the original filesystem
+to save space, when files are discovered to be duplicate.
+In this special case the only reason the files are
+hardlinked is to save space, and where the filesystem
+doesn't handle duplicate files (different inode, same
+data).
+
+Squashfs does handle duplicate files, and so add
+an option to ignore hardlinks and instead
+store them as duplicates.
+
+Signed-off-by: Phillip Lougher <phillip@squashfs.org.uk>
+---
+ squashfs-tools/mksquashfs.c | 16 +++++++++++-----
+ 1 file changed, 11 insertions(+), 5 deletions(-)
+
+diff --git a/squashfs-tools/mksquashfs.c b/squashfs-tools/mksquashfs.c
+index a45b77f..d4dc359 100644
+--- a/squashfs-tools/mksquashfs.c
++++ b/squashfs-tools/mksquashfs.c
+@@ -312,6 +312,9 @@ struct dir_info *root_dir;
+ FILE *log_fd;
+ int logging=FALSE;
+ 
++/* Should Mksquashfs detect hardlinked files? */
++int no_hardlinks = FALSE;
++
+ static char *read_from_disk(long long start, unsigned int avail_bytes);
+ void add_old_root_entry(char *name, squashfs_inode inode, int inode_number,
+ 	int type);
+@@ -3093,11 +3096,11 @@ struct inode_info *lookup_inode3(struct stat *buf, int pseudo, int id,
+ 
+ 	/*
+ 	 * Look-up inode in hash table, if it already exists we have a
+-	 * hard-link, so increment the nlink count and return it.
+-	 * Don't do the look-up for directories because we don't hard-link
+-	 * directories.
++	 * hardlink, so increment the nlink count and return it.
++	 * Don't do the look-up for directories because Unix/Linux doesn't
++	 * allow hard-links to directories.
+ 	 */
+-	if ((buf->st_mode & S_IFMT) != S_IFDIR) {
++	if ((buf->st_mode & S_IFMT) != S_IFDIR && !no_hardlinks) {
+ 		for(inode = inode_info[ino_hash]; inode; inode = inode->next) {
+ 			if(memcmp(buf, &inode->buf, sizeof(struct stat)) == 0) {
+ 				inode->nlink ++;
+@@ -5447,7 +5450,9 @@ int main(int argc, char *argv[])
+ 		comp = lookup_compressor(COMP_DEFAULT);
+ 
+ 	for(i = source + 2; i < argc; i++) {
+-		if(strcmp(argv[i], "-mkfs-time") == 0 ||
++		if(strcmp(argv[i], "-no-hardlinks") == 0)
++			no_hardlinks = TRUE;
++		else if(strcmp(argv[i], "-mkfs-time") == 0 ||
+ 				strcmp(argv[i], "-fstime") == 0) {
+ 			if((++i == argc) || !parse_num_unsigned(argv[i], &mkfs_time)) {
+ 				ERROR("%s: %s missing or invalid time value\n", argv[0], argv[i - 1]);
+@@ -5893,6 +5898,7 @@ printOptions:
+ 				"files larger than block size\n");
+ 			ERROR("-no-duplicates\t\tdo not perform duplicate "
+ 				"checking\n");
++			ERROR("-no-hardlinks\t\tdo not hardlink files, instead store duplicates\n");
+ 			ERROR("-all-root\t\tmake all files owned by root\n");
+ 			ERROR("-root-mode <mode>\tset root directory permissions to octal <mode>\n");
+ 			ERROR("-force-uid <uid>\tset all file uids to <uid>\n");
+-- 
+2.30.0
+
diff --git a/pkgs/tools/filesystems/squashfs/default.nix b/pkgs/tools/filesystems/squashfs/default.nix
index c833cfe10ada..7e55bcf6e0e4 100644
--- a/pkgs/tools/filesystems/squashfs/default.nix
+++ b/pkgs/tools/filesystems/squashfs/default.nix
@@ -19,6 +19,10 @@ stdenv.mkDerivation {
     # This patch adds an option to pad filesystems (increasing size) in
     # exchange for better chunking / binary diff calculation.
     ./4k-align.patch
+    # Add -no-hardlinks option. This is a rebased version of
+    # c37bb4da4a5fa8c1cf114237ba364692dd522262, can be removed
+    # when upgrading to the next version after 4.4
+    ./0001-Mksquashfs-add-no-hardlinks-option.patch
   ] ++ lib.optional stdenv.isDarwin ./darwin.patch;
 
   buildInputs = [ zlib xz zstd lz4 ];