[PATCH 0/3] Fix the O_PATH support for FIFOs
Thu Jan 23 17:13:00 GMT 2020
On 1/23/2020 11:31 AM, Ken Brown wrote:
> Commit aa55d22c, "Cygwin: honor the O_PATH flag when opening a FIFO",
> fixed a hang but otherwise didn't accomplish the purpose of the O_PATH
> flag as stated in the Linux man page for open(2):
> Obtain a file descriptor that can be used for two purposes: to
> indicate a location in the filesystem tree and to perform
> operations that act purely at the file descriptor level. The
> file itself is not opened, and other file operations (e.g.,
> read(2), write(2), fchmod(2), fchown(2), fgetxattr(2),
> ioctl(2), mmap(2)) fail with the error EBADF.
> [The man page goes on to describe operations that *can* be
> performed: close(2), fchdir(2), fstat(2),....]
> Opening a file or directory with the O_PATH flag requires no
> permissions on the object itself (but does require execute
> permission on the directories in the path prefix).
> The first problem in the current implementation is that if open(2) is
> called on a FIFO, fhandler_base::device_access_denied is called and
> tries to open the FIFO with read access, which isn't supposed to be
> required. This is fixed by the first patch in this series.
> The second patch makes fhandler_fifo::open call fhandler_base::open_fs
> if O_PATH is set, so that we actually obtain a handle that can be used
> for the purposes stated above.
> The third page tweaks fhandler_fifo::fcntl and fhandler_fifo::dup so
> that they work with O_PATH.
> In a followup email I'll provide the program I used to test this
Test program attached.
$ gcc -o o_path_fifo_test o_path_fifo_test.c
The following calls should fail with EBADF:
The following calls should succeed:
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
More information about the Cygwin-patches