Re: [Linaro-mm-sig] Re: [PATCH 2/2] procfs: Add 'path' to /proc/<pid>/fdinfo/

From: Kalesh Singh
Date: Wed Jun 01 2022 - 23:31:27 EST

On Wed, Jun 1, 2022 at 8:02 AM Christian König
<ckoenig.leichtzumerken@xxxxxxxxx> wrote:
>
> Am 01.06.22 um 00:48 schrieb Stephen Brennan:
> > Kalesh Singh <kaleshsingh@xxxxxxxxxx> writes:
> >> On Tue, May 31, 2022 at 3:07 PM Stephen Brennan
> >> <stephen.s.brennan@xxxxxxxxxx> wrote:
> >>> On 5/31/22 14:25, Kalesh Singh wrote:
> >>>> In order to identify the type of memory a process has pinned through
> >>>> its open fds, add the file path to fdinfo output. This allows
> >>>> identifying memory types based on common prefixes. e.g. "/memfd...",
> >>>> "/dmabuf...", "/dev/ashmem...".
> >>>>
> >>>> Access to /proc/<pid>/fdinfo is governed by PTRACE_MODE_READ_FSCREDS
> >>>> the same as /proc/<pid>/maps which also exposes the file path of
> >>>> mappings; so the security permissions for accessing path is consistent
> >>>> with that of /proc/<pid>/maps.
> >>> Hi Kalesh,
> >> Hi Stephen,
> >>
> >> Thanks for taking a look.
> >>
> >>> I think I see the value in the size field, but I'm curious about path,
> >>> which is available via readlink /proc/<pid>/fd/<n>, since those are
> >>> symlinks to the file themselves.
> >> This could work if we are root, but the file permissions wouldn't
> >> allow us to do the readlink on other processes otherwise. We want to
> >> be able to capture the system state in production environments from
> >> some trusted process with ptrace read capability.
> > Interesting, thanks for explaining. It seems weird to have a duplicate
> > interface for the same information but such is life.
>
> Yeah, the size change is really straight forward but for this one I'm
> not 100% sure either.

The 2 concerns I think are:
1. Fun characters in the path names
2. If exposing the path is appropriate to begin with.

One way I think we can address both is to only expose the path for
anon inodes. Then we have well-known path formats and we don't expose
much about which files a process is accessing since these aren't real
paths.

+ if (is_anon_inode(inode)) {
+ seq_puts(m, "path:\t");
+ seq_file_path(m, file, "\n");
+ seq_putc(m, '\n');
+ }

Interested to hear thoughts on it.

>
> Probably best to ping some core fs developer before going further with it.

linux-fsdevel is cc'd here. Adding Al Vrio as well. Please let me know
if there are other parties I should include.

>
> BTW: Any preferred branch to push this upstream? If not I can take it
> through drm-misc-next.

No other dependencies for this, so drm-misc-next is good.

Thanks,
Kalesh

>
> Regards,
> Christian.
>
> >
> >>> File paths can contain fun characters like newlines or colons, which
> >>> could make parsing out filenames in this text file... fun. How would your
> >>> userspace parsing logic handle "/home/stephen/filename\nsize:\t4096"? The
> >>> readlink(2) API makes that easy already.
> >> I think since we have escaped the "\n" (seq_file_path(m, file, "\n")),
> > I really should have read through that function before commenting,
> > thanks for teaching me something new :)
> >
> > Stephen
> >
> >> then user space might parse this line like:
> >>
> >> if (strncmp(line, "path:\t", 6) == 0)
> >> char* path = line + 6;
> >>
> >>
> >> Thanks,
> >> Kalesh
> >>
> >>> Is the goal avoiding races (e.g. file descriptor 3 is closed and reopened
> >>> to a different path between reading fdinfo and stating the fd)?
> >>>
> >>> Stephen
> >>>
> >>>> Signed-off-by: Kalesh Singh <kaleshsingh@xxxxxxxxxx>
> >>>> ---
> >>>>
> >>>> Changes from rfc:
> >>>> - Split adding 'size' and 'path' into a separate patches, per Christian
> >>>> - Fix indentation (use tabs) in documentaion, per Randy
> >>>>
> >>>> Documentation/filesystems/proc.rst | 14 ++++++++++++--
> >>>> fs/proc/fd.c | 4 ++++
> >>>> 2 files changed, 16 insertions(+), 2 deletions(-)
> >>>>
> >>>> diff --git a/Documentation/filesystems/proc.rst b/Documentation/filesystems/proc.rst
> >>>> index 779c05528e87..591f12d30d97 100644
> >>>> --- a/Documentation/filesystems/proc.rst
> >>>> +++ b/Documentation/filesystems/proc.rst
> >>>> @@ -1886,14 +1886,16 @@ if precise results are needed.
> >>>> 3.8 /proc/<pid>/fdinfo/<fd> - Information about opened file
> >>>> ---------------------------------------------------------------
> >>>> This file provides information associated with an opened file. The regular
> >>>> -files have at least five fields -- 'pos', 'flags', 'mnt_id', 'ino', and 'size'.
> >>>> +files have at least six fields -- 'pos', 'flags', 'mnt_id', 'ino', 'size',
> >>>> +and 'path'.
> >>>>
> >>>> The 'pos' represents the current offset of the opened file in decimal
> >>>> form [see lseek(2) for details], 'flags' denotes the octal O_xxx mask the
> >>>> file has been created with [see open(2) for details] and 'mnt_id' represents
> >>>> mount ID of the file system containing the opened file [see 3.5
> >>>> /proc/<pid>/mountinfo for details]. 'ino' represents the inode number of
> >>>> -the file, and 'size' represents the size of the file in bytes.
> >>>> +the file, 'size' represents the size of the file in bytes, and 'path'
> >>>> +represents the file path.
> >>>>
> >>>> A typical output is::
> >>>>
> >>>> @@ -1902,6 +1904,7 @@ A typical output is::
> >>>> mnt_id: 19
> >>>> ino: 63107
> >>>> size: 0
> >>>> + path: /dev/null
> >>>>
> >>>> All locks associated with a file descriptor are shown in its fdinfo too::
> >>>>
> >>>> @@ -1920,6 +1923,7 @@ Eventfd files
> >>>> mnt_id: 9
> >>>> ino: 63107
> >>>> size: 0
> >>>> + path: anon_inode:[eventfd]
> >>>> eventfd-count: 5a
> >>>>
> >>>> where 'eventfd-count' is hex value of a counter.
> >>>> @@ -1934,6 +1938,7 @@ Signalfd files
> >>>> mnt_id: 9
> >>>> ino: 63107
> >>>> size: 0
> >>>> + path: anon_inode:[signalfd]
> >>>> sigmask: 0000000000000200
> >>>>
> >>>> where 'sigmask' is hex value of the signal mask associated
> >>>> @@ -1949,6 +1954,7 @@ Epoll files
> >>>> mnt_id: 9
> >>>> ino: 63107
> >>>> size: 0
> >>>> + path: anon_inode:[eventpoll]
> >>>> tfd: 5 events: 1d data: ffffffffffffffff pos:0 ino:61af sdev:7
> >>>>
> >>>> where 'tfd' is a target file descriptor number in decimal form,
> >>>> @@ -1968,6 +1974,7 @@ For inotify files the format is the following::
> >>>> mnt_id: 9
> >>>> ino: 63107
> >>>> size: 0
> >>>> + path: anon_inode:inotify
> >>>> inotify wd:3 ino:9e7e sdev:800013 mask:800afce ignored_mask:0 fhandle-bytes:8 fhandle-type:1 f_handle:7e9e0000640d1b6d
> >>>>
> >>>> where 'wd' is a watch descriptor in decimal form, i.e. a target file
> >>>> @@ -1992,6 +1999,7 @@ For fanotify files the format is::
> >>>> mnt_id: 9
> >>>> ino: 63107
> >>>> size: 0
> >>>> + path: anon_inode:[fanotify]
> >>>> fanotify flags:10 event-flags:0
> >>>> fanotify mnt_id:12 mflags:40 mask:38 ignored_mask:40000003
> >>>> fanotify ino:4f969 sdev:800013 mflags:0 mask:3b ignored_mask:40000000 fhandle-bytes:8 fhandle-type:1 f_handle:69f90400c275b5b4
> >>>> @@ -2018,6 +2026,7 @@ Timerfd files
> >>>> mnt_id: 9
> >>>> ino: 63107
> >>>> size: 0
> >>>> + path: anon_inode:[timerfd]
> >>>> clockid: 0
> >>>> ticks: 0
> >>>> settime flags: 01
> >>>> @@ -2042,6 +2051,7 @@ DMA Buffer files
> >>>> mnt_id: 9
> >>>> ino: 63107
> >>>> size: 32768
> >>>> + path: /dmabuf:
> >>>> count: 2
> >>>> exp_name: system-heap
> >>>>
> >>>> diff --git a/fs/proc/fd.c b/fs/proc/fd.c
> >>>> index 464bc3f55759..8889a8ba09d4 100644
> >>>> --- a/fs/proc/fd.c
> >>>> +++ b/fs/proc/fd.c
> >>>> @@ -60,6 +60,10 @@ static int seq_show(struct seq_file *m, void *v)
> >>>> seq_printf(m, "ino:\t%lu\n", file_inode(file)->i_ino);
> >>>> seq_printf(m, "size:\t%lli\n", (long long)file_inode(file)->i_size);
> >>>>
> >>>> + seq_puts(m, "path:\t");
> >>>> + seq_file_path(m, file, "\n");
> >>>> + seq_putc(m, '\n');
> >>>> +
> >>>> /* show_fd_locks() never deferences files so a stale value is safe */
> >>>> show_fd_locks(m, file, files);
> >>>> if (seq_has_overflowed(m))
> >>> --
> >>> To unsubscribe from this group and stop receiving emails from it, send an email to kernel-team+unsubscribe@xxxxxxxxxxx.
> >>>
> > _______________________________________________
> > Linaro-mm-sig mailing list -- linaro-mm-sig@xxxxxxxxxxxxxxxx
> > To unsubscribe send an email to linaro-mm-sig-leave@xxxxxxxxxxxxxxxx
>