Libguestfs - Feb 2023 - [libnbd PATCH v3 09/29] lib/utils: introduce async-signal-safe execvpe()

On Tue, Feb 21, 2023 at 07:07:38PM +0100, Laszlo Ersek
wrote:
> On 2/21/23 18:06, Eric Blake wrote:
> > On Tue, Feb 21, 2023 at 12:55:49PM +0100, Laszlo Ersek wrote:
> >> On 2/15/23 23:28, Eric Blake wrote:
> >>> On Wed, Feb 15, 2023 at 03:11:38PM +0100, Laszlo Ersek wrote:
> >>>> execvp() [01] is powerful:
> >>>
> >>> You KNOW this is going to be a juicy commit message when the
first
> >>> line contains a 2-digit footnote ;)
> >>
> >> Haha, yes :) Clearly I originally started the commit message with
> >> single-digit footnotes, but I ended up needing another digit :)
> > 
> > Trimming my responses to just your open questions...
> 
> Sorry, I should have been trimming too.
No worries.  Some email clients make it easier to gloss over quoted
text than others, but I tend to trim to try and focus the reader's
attention on where I am actually adding to the conversation.  At the
other extreme, sometimes I trim too much context for someone jumping
into the conversation late - in open source this is fine because there
are public archives, but in other contexts it can have negative
consequences.
> 
> >>> [1] And this simplifies if empty PATH can imply searching in
".".
> >>
> >> If we ever get practical problems with PATH="",
let's update the code
> >> then, incrementally.
> > 
> > Indeed.  No need to fret about a corner case that no one will hit; or
> > if they can prove that it matters, it's an easy incremental patch
at
> > that time.
> 
> Well, given Daniel's comments meanwhile, it seems like the original
> execvp() is something we shouldn't fret about. :/
glibc marks execvp() and exevpe() as 'MT-Safe env', which means it
does not modify 'environ' and presumably does not use 'malloc'. 
If
the only reason POSIX marks execvp() as thread-unsafe is because of
its interaction with getenv() for PATH and therefore unsafe to exec a
child in one thread while another is calling putenv(), then it is no
worse than our use of getenv() without locking on the grounds that no
sane app will be doing setenv() after spawning threads.

But you've gone to a lot of work on this series; I'm still in favor of
including your work, even if decide our premise behind it is weaker
than intended.  As I understood it, the premise is that actively
avoiding as many non-async-safe functions as possible between fork()
and exec() is always a good idea to avoid the deadlock created when a
multithreaded process fork()s in one thread while another thread holds
a mutex on the non-async-safe resource (in the child process, the
other thread no longer exists and therefore can never release the
resource).  Knowing WHICH resources are liable to be locked in other
threads makes it easier to reason about which generically
non-async-safe functions can be used if we make other limiting
restrictions (such as glibc's 'MT-Safe env' designation), but that
requires more thought than blindly avoiding all non-async-safe
functions.
> 
> > Again agreed - any user desperate enough to pass in an atypical binary
> > name or put atypical relative directory names in PATH gets what they
> > deserve; as long as we don't think our code misbehaving on argv[0]
of
> > "+s" can be abused as a security hole, then not worrying
about the
> > corner case in our code is just fine for this patch, and an
> > incremental patch on top if we can think of why it is
> > security-sensitive after all.
> 
> Can you please elaborate on "+s"? (I'd like to understand
your point
> regardless of this patch, too.)
https://www.austingroupbugs.net/view.php?id=1440 is where it came up
in the Austin Group.  Basically, having system("+s") be _required_ to
invoke ["/path/to/sh", "-c", "+s"] is risky -
"+s" is ambiguous
between being the name of a real script and being a shell option
(POSIX already warns that naming an executable with a leading "-" is
unwise because of potential conflicts with options, but 'sh' has the
special rules that + as well as - can introduce options for historical
reasons).  Calling ["/path/to/sh", "-c", "--",
"+s"] is unambiguous
for a POSIX sh that understands "--" as the end-of-options marker, but
not all historical sh did that.  It is also possible to use
["/path/to/sh", "-c", " +s"] (note the added
leading space, which is
ignored by the shell), but that requires space for injecting the
leading space, and how do you do that concisely while still
maintaining async-safety?
> 
> >> How does "//binary" differ from "/binary", in
pathname resolution?
> > 
> > POSIX intentionally says that file names beginning with leading
"//"
> > but not "///" have implementation-defined semantics. 
Linux's
> > definition is that "//" is a synonym for "/", so
it makes no
> > difference there.  But on Cygwin, "/machine" is a file (or
directory)
> > living in the root directory, while "//machine/share" is a
network
> > share access path that locates the shared resource "share"
on
> > network-addressable "machine" (think Windows \\MACHINE\SHARE
syntax).
> > Native windows does not let you access bare \\ or \\MACHINE, but
> > Cygwin has further generalized those so that "ls //" shows
you a list
> > of all machines currently advertising shares in your local subnet (do
> > not try it on a large enterprise subnet - it can take a LOONG time).
> > Cygwin also defaults to having "/cygdrive/c" be a synonym to
Windows
> > C:\, but has an option for you to re-spell it as "//c" (that
is,
> > expose all of your drive letters the same as remote-access machines by
> > setting the cygdrive prefix to empty instead of "cygdrive").
> 
> Ah, found it:
> 
>
https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap04.html#tag_04_13
> 
> """
> If a pathname begins with two successive <slash> characters, the
first
> component following the leading <slash> characters may be interpreted
in
> an implementation-defined manner, although more than two leading
<slash>
> characters shall be treated as a single <slash> character.
> """
> 
> (... I wanted to say that "there's no replacement for reading
POSIX in
> full", but now I'm wondering if reading POSIX *at all* makes
sense...)
Alas, this rings too close to home...  Standards are only useful if
they are likely to be followed, which in turn is harder if they are
hard to read.
> 
> > And thanks again for a detailed commit message and code comments for
> > something that would otherwise be hard to maintain if we didn't
have
> > strong rationale for why we are doing it.
> 
> ... That rationale may just have fallen away.
It certainly got weakened.
> 
> Laszlo
> 
-- 
Eric Blake, Principal Software Engineer
Red Hat, Inc.           +1-919-301-3266
Virtualization:  qemu.org | libvirt.org

On 2/21/23 20:32, Eric Blake wrote:
> On Tue, Feb 21, 2023 at 07:07:38PM +0100, Laszlo Ersek wrote:
>> Well, given Daniel's comments meanwhile, it seems like the original
>> execvp() is something we shouldn't fret about. :/
> 
> glibc marks execvp() and exevpe() as 'MT-Safe env', which means it
> does not modify 'environ' and presumably does not use
'malloc'.  If
> the only reason POSIX marks execvp() as thread-unsafe is because of
> its interaction with getenv() for PATH and therefore unsafe to exec a
> child in one thread while another is calling putenv(), then it is no
> worse than our use of getenv() without locking on the grounds that no
> sane app will be doing setenv() after spawning threads.
> 
> But you've gone to a lot of work on this series; I'm still in favor
of
> including your work, even if decide our premise behind it is weaker
> than intended.  As I understood it, the premise is that actively
> avoiding as many non-async-safe functions as possible between fork()
> and exec() is always a good idea to avoid the deadlock created when a
> multithreaded process fork()s in one thread while another thread holds
> a mutex on the non-async-safe resource (in the child process, the
> other thread no longer exists and therefore can never release the
> resource).  Knowing WHICH resources are liable to be locked in other
> threads makes it easier to reason about which generically
> non-async-safe functions can be used if we make other limiting
> restrictions (such as glibc's 'MT-Safe env' designation), but
that
> requires more thought than blindly avoiding all non-async-safe
> functions.
It's killing me to see my work turn out as waste, but I don't want it to
be included (and to present maintenance burden) just for "saving" the
work. I kind of expected it to be unwelcome (due to it being opinionated and not
fixing acute symptoms); I didn't expect it to be lost from the specification
/ premise up. This early paragraph in the commit message is torpedoed:

    However, execvp() is not async-signal-safe [03], and so we shouldn't
call
    it in a child process forked from a multi-threaded parent process [04],
    which the libnbd client application may well be.

That "and so" implication is busted because the glibc manual (from
GNU, not the Linux manual pages!) *implies* that fork(), as opposed to _Fork(),
does not restrict the child process to AS-Safe functions.

Anyway, I think your characterization is right. My approach was, "avoid
calling anything that's not explicitly async-signal-safe". A laxer
approach is "avoid calling anything that might interfere with particular
resources".

Unfortunately, this laxer approach, while it may require some "further
caution" in case we want to call further functions in the affected context,
quite summarily obviates this particular execvp() replacement.

Plus, regarding said "further caution" in general, I'm getting the
impression (from RHBZ#906468) that glibc's intent is actually the opposite
-- i.e., the intent seems to be that any particular API should opt out of,
rather than opt in to, usability after fork().

Considering how much work is still needed to address the review feedback thus
far (and the necessary v4 reviews!), I'm seriously torn if we should just
drop the whole thing. (And there's a really sore lesson for me in this,
regarding cleanups for standards conformance.)
>> Can you please elaborate on "+s"? (I'd like to understand
your point
>> regardless of this patch, too.)
> 
> https://www.austingroupbugs.net/view.php?id=1440 is where it came up
> in the Austin Group.  Basically, having system("+s") be
_required_ to
> invoke ["/path/to/sh", "-c", "+s"] is risky -
"+s" is ambiguous
> between being the name of a real script and being a shell option
So what does the "+s" option do specifically?

"-s" makes the shell read commands from stdin, but "+s"
would only apply to such an "-s" option that were taken by the
"set" special builtin -- it would *disable* that set-option (while
"set -s" would enable it). However, I don't see *any*
"-s" under "set", so I don't know what "set
+s" would do.
> (POSIX already warns that naming an executable with a leading "-"
is
> unwise because of potential conflicts with options, but 'sh' has
the
> special rules that + as well as - can introduce options for historical
> reasons).
So is "+s" just a synonym for "-s", meaning the shell will
read commands from stdin?
> Calling ["/path/to/sh", "-c", "--",
"+s"] is unambiguous
> for a POSIX sh that understands "--" as the end-of-options
marker, but
> not all historical sh did that.  It is also possible to use
> ["/path/to/sh", "-c", " +s"] (note the added
leading space, which is
> ignored by the shell),
ignored why?
> but that requires space for injecting the
> leading space, and how do you do that concisely while still
> maintaining async-safety?
(right, not doing it, just learning about +s)
>> (... I wanted to say that "there's no replacement for reading
POSIX in
>> full", but now I'm wondering if reading POSIX *at all* makes
sense...)
> 
> Alas, this rings too close to home...  Standards are only useful if
> they are likely to be followed, which in turn is harder if they are
> hard to read.
"Hard to read" is a problem, but not the core issue IMO here. The
problem is that for the sake of various technological advances, POSIX is
intentionally abandoned, and that leaves us with *zero* common documentation
that might uniformly cover even a small handful of modern OSes. And we have no
alternative to individually testing out every "standard" API on each
platform of interest. I have no problem with prctl() being Linux-specific;
I'm very much irritated by fork() relaxing its standard restrictions without
proper documentation. It's a recipe for wasting work.

Anyway I'll log out now because I'm hardly coherent.

Laszlo