search for: renameat2

...libc/klibc.git;a=commit;h=ebdc262bd8a4d650c58de48f67e6b08aeb953a8f Author: Ben Hutchings <ben at decadent.org.uk> AuthorDate: Mon, 16 Jul 2018 18:24:08 +0100 Committer: Ben Hutchings <ben at decadent.org.uk> CommitDate: Fri, 18 Jan 2019 03:10:14 +0000 [klibc] rename, renameat: Use renameat2() system call New architectures only define the renameat2() system call, which was added in Linux 3.15. Define rename() and renameat() as wrappers for it if necessary. Link: https://www.zytor.com/pipermail/klibc/2018-July/003996.html Signed-off-by: Ben Hutchings <ben at decadent.org.uk> -...

New architectures only define the renameat2() system call, which was added in Linux 3.15. Define rename() and renameat() as wrappers for it if necessary. Signed-off-by: Ben Hutchings <ben at decadent.org.uk> --- --- a/usr/klibc/Kbuild +++ b/usr/klibc/Kbuild @@ -59,7 +59,8 @@ klib-y += vsnprintf.o snprintf.o vsprint inet/inet_ntoa...

...DCWD, "destination", AT_EMPTY_PATH) Without special permissions and have instead to depend on proc: linkat(AT_FDCWD, "/proc/self/fd/<num>", AT_FDCWD, "destination", AT_SYMLINK_FOLLOW) Another annoyance is that linkat will not replace the destination and renameat2 doesn't support AT_EMPTY_PATH. The result is that we have to use unlink+linkat and loop. On windows there is FILE_FLAG_DELETE_ON_CLOSE, but there seems to be no way to cancel it. If a file is created with it I can rename it, but it is still deleted in the end. I couldn't find any support...

On Tue, May 7, 2019 at 9:05 AM Tomas Kalibera <tomas.kalibera at gmail.com> wrote: > > Thanks for the report. According to my reading, this use of "mv" is ok > and the renameat2() call which the invocation of "mv" leads to is also > ok and allowed by POSIX in this context. It could only fail with EEXIST > if the target directory (path/pkg) was not empty. So far I've not been > able to reproduce but we could fall back to copy like on Windows. Thanks...

...= 34002 ioctl(0, TCGETS, {B38400 opost isig icanon echo ...}) = 0 stat("path", {st_mode=S_IFDIR|0755, st_size=2, ...}) = 0 lstat("path/to/pkg", {st_mode=S_IFDIR|0755, st_size=0, ...}) = 0 lstat("path/pkg", {st_mode=S_IFDIR|0755, st_size=0, ...}) = 0 renameat2(AT_FDCWD, "path/to/pkg", AT_FDCWD, "path/pkg", 0) = -1 EEXIST (File exists) open("/usr/share/locale/locale.alias", O_RDONLY|O_CLOEXEC) = 3 fstat(3, {st_mode=S_IFREG|0644, st_size=2502, ...}) = 0 mmap(NULL, 4096, PROT_READ|PROT_WRITE, MAP_PRIVATE|MAP_ANONYMOUS, -1, 0) =...

Thanks for the report.? According to my reading, this use of "mv" is ok and the renameat2() call which the invocation of "mv" leads to is also ok and allowed by POSIX in this context. It could only fail with EEXIST if the target directory (path/pkg) was not empty. So far I've not been able to reproduce but we could fall back to copy like on Windows. Best Tomas On 5/5...

On 5/7/19 6:18 PM, Henrik Bengtsson wrote: > On Tue, May 7, 2019 at 9:05 AM Tomas Kalibera <tomas.kalibera at gmail.com> wrote: >> Thanks for the report. According to my reading, this use of "mv" is ok >> and the renameat2() call which the invocation of "mv" leads to is also >> ok and allowed by POSIX in this context. It could only fail with EEXIST >> if the target directory (path/pkg) was not empty. So far I've not been >> able to reproduce but we could fall back to copy like on Windo...

...nk: Add -f option [klibc] mount: Implement -o defaults [klibc] Install headers with consistent mode [klibc] dash: mkbuiltins: Fix sort order harder [klibc] gzip: Fix silent fallback to decompression [klibc] run-init: Add dry-run mode [klibc] rename, renameat: Use renameat2() system call [klibc] Add RISC-V (RV64) port [klibc] x86_64: Reduce ld max-page-size option again [klibc] Never clean files in quilt status directory [klibc] x86_64: Use -Ttext-segment to avoid address collision [klibc] i386: Use -Ttext-segment to avoid address collisi...

...--------------------------------------- Alfonso Sanchez-beato (1): [klibc] Add support for reboot syscall argument Baptiste Jonglez (1): [klibc] nfsmount: support nfsvers= and vers= options Ben Hutchings (6): [klibc] run-init: Add dry-run mode [klibc] rename, renameat: Use renameat2() system call [klibc] Add RISC-V (RV64) port [klibc] x86_64: Reduce ld max-page-size option again [klibc] Never clean files in quilt status directory [klibc] x86_64: Use -Ttext-segment to avoid address collision Benjamin Drung (2): [klibc] ipconfig: Implement classles...