[PATCH v5 1/3] Introduce -l/--syscall-limit option

[Sahil Siddiq]

Hi,

On Saturday, 1 April 2023 03:12:12 IST Dmitry V. Levin wrote:
> On Sat, Apr 01, 2023 at 02:59:03AM +0530, Sahil Siddiq wrote:
> > On Friday, 31 March 2023 17:03:20 IST Dmitry V. Levin wrote:
> > > On Fri, Mar 31, 2023 at 12:14:13PM +0530, Sahil Siddiq wrote:
> > > > On Friday, 31 March 2023 03:17:40 IST Kossi GLOKPOR wrote:
> > > > > Hi,
> > > > > What do you think about using the --daemonize options ? For example
> > > > > with --daemonize=grandchild, the shell will be the parent of the
> > > > > tracee, and detaching & stopping strace will have no incidence on
> > > > > it.
> > > > > When we set an auto stop condition, we assume that strace can have
> > > > > to
> > > > > stop while the tracee is still running, and enabling tracer
> > > > > daemonization seems like a good way to prevent any issues with the
> > > > > tracee when strace stops.
> > > > > 
> > > > > Regards,
> > > > > Kossi
> > > > 
> > > > Thank you for the suggestion. I have tested this out and it gives the
> > > > results that I was looking for. We could let "--syscall-limit" be
> > > > enabled
> > > > only when using it with the -D or -p options and disable it otherwise.
> > > 
> > > To be honest, although using --syscall-limit along with --daemonize
> > > makes
> > > perfect sense, I'm not sure it is necessary to make this combination, or
> > > combination of --syscall-limit with -p, mandatory.
> > 
> > Right, logically --syscall-limit should not be tied to the -D or -p
> > options. But I was thinking of using it as a hack to work around the
> > problem that arises after the tracees are detached. The detached tracee
> > interferes with the foreground process in the terminal session which was
> > initially used to run strace.
> strace --daemonize has its own side effects.  For example, compared to a
> regular strace invocation, it's harder to find out whether strace
> --daemonize has finished writing its output.  That's why I'm reluctant to
> force --daemonize option without giving user an option to turn it off.

I wouldn't mind working on introducing --daemonize=none or a similar option
if that's alright and if it solves the problem. I might have to comb through the
relevant parts of the code to understand how it works before I can get into the
actual implementation.

Regards,
Sahil