gen_init_cpio: attempt copy_file_range for file data

The copy_file_range syscall can improve copy performance by cloning
extents between cpio archive source and destination files.
Existing read / write based copy logic is retained for fallback in case
the copy_file_range syscall is unsupported or unavailable due to
cross-filesystem EXDEV, etc.

Clone or reflink, as opposed to copy, of source file extents into the
output cpio archive may (e.g. on Btrfs and XFS) require alignment of the
output to the filesystem block size. This could be achieved by inserting
padding entries into the cpio archive manifest.

Signed-off-by: David Disseldorp <ddiss@suse.de>
Reviewed-by: Nicolas Schier <nsc@kernel.org>
Link: https://lore.kernel.org/r/20250819032607.28727-4-ddiss@suse.de
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
This commit is contained in:
David Disseldorp 2025-08-19 13:05:46 +10:00 committed by Nathan Chancellor
parent ae18b94099
commit 97169cd6d9
No known key found for this signature in database
GPG Key ID: 1D6B269171C01A96
1 changed files with 12 additions and 1 deletions

View File

@ -354,6 +354,7 @@ static int cpio_mkfile(const char *name, const char *location,
int namesize;
unsigned int i;
uint32_t csum = 0;
ssize_t this_read;
mode |= S_IFREG;
@ -429,9 +430,19 @@ static int cpio_mkfile(const char *name, const char *location,
push_pad(padlen(offset, 4)) < 0)
goto error;
if (size) {
this_read = copy_file_range(file, NULL, outfd, NULL, size, 0);
if (this_read > 0) {
if (this_read > size)
goto error;
offset += this_read;
size -= this_read;
}
/* short or failed copy falls back to read/write... */
}
while (size) {
unsigned char filebuf[65536];
ssize_t this_read;
size_t this_size = MIN(size, sizeof(filebuf));
this_read = read(file, filebuf, this_size);