https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110827
--- Comment #20 from Iain Sandoe ---
(In reply to Michael Duggan from comment #19)
> I submitted a patch.
Yes a saw that - thanks, it does need other people to review it though,
> I also recently thought of another potential solution.
> We co
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110827
--- Comment #19 from Michael Duggan ---
I submitted a patch. I also recently thought of another potential solution.
We could patch the code that outputs the function information for the graph
file, changing it to *not* mark the actor function
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110827
--- Comment #18 from Iain Sandoe ---
(In reply to Michael Duggan from comment #17)
> Gcov doesn't have access to the function declaration, only to the
> information that has been output to the count and graph files. Neither of
> those contain a
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110827
--- Comment #17 from Michael Duggan ---
Gcov doesn't have access to the function declaration, only to the information
that has been output to the count and graph files. Neither of those contain
any information (as far as I can tell) that would
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110827
--- Comment #16 from Iain Sandoe ---
(In reply to Michael Duggan from comment #15)
> Would it be possible for one of these to make it into version 15?
I'd prefer the change that teaches gcov about coroutines, I fear the change to
the status of
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110827
--- Comment #15 from Michael Duggan ---
Would it be possible for one of these to make it into version 15?
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110827
--- Comment #14 from Michael Duggan ---
So, I applied the following patch instead, with identical results. Pro:
doesn't change the artificiality status of the function. Con: more complex
code.
1 file changed, 34 insertions(+), 2 deletions(-)
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110827
--- Comment #13 from Iain Sandoe ---
I would say that perhaps (noting comment #9) it would be useful to explore
teaching gcov that it should not ignore the actor function (it is identifiable
that a function is a coroutine component)
Part of the
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110827
--- Comment #12 from Michael Duggan ---
I finally got around to this experiment today, after more than a year. I
changed this line in `coro_build_actor_or_destory_function` from:
DECL_ARTIFICIAL (fn) = true;
to
DECL_ARTIFICIAL (fn) = !act
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110827
--- Comment #11 from Iain Sandoe ---
(In reply to Michael Duggan from comment #9)
> More data:
> The coroutine actor is marked as artificial in
> coro_build_actor_or_destroy_function. As a result, it is completely ignored
> by gcov. In gcov's
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110827
--- Comment #10 from Michael Duggan ---
To sum up what I have figured out, C++ transforms the coroutine "function" into
a trio of functions: a ramp function, an actor function, and a destruction
function. The ramp function acts as the actual fu
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110827
--- Comment #9 from Michael Duggan ---
More data:
The coroutine actor is marked as artificial in
coro_build_actor_or_destroy_function. As a result, it is completely ignored by
gcov. In gcov's process_all_functions function, artificial function
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110827
--- Comment #8 from Michael Duggan ---
Using the better test case, I have determined that the coroutine _is_ being
instrumented with gcov counters. When disassembled, the output contains the
following in the bar() actor function:
Dump of assem
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110827
Michael Duggan changed:
What|Removed |Added
Attachment #55648|0 |1
is obsolete|
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110827
--- Comment #6 from Iain Sandoe ---
(In reply to Richard Biener from comment #5)
> (In reply to Iain Sandoe from comment #2)
> > (In reply to Richard Biener from comment #1)
> > > I'm seeing all code properly instrumented. The coverage I see is
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110827
--- Comment #5 from Richard Biener ---
(In reply to Iain Sandoe from comment #2)
> (In reply to Richard Biener from comment #1)
> > I'm seeing all code properly instrumented. The coverage I see is
>
> > 1: 27:task foo() {
> >
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110827
--- Comment #4 from Michael Duggan ---
I should be more explicit. The `std::cout` line in the example is just a
placeholder for "does some work here," and this example is specifically the
simplest version of a coroutine I could come up with tha
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110827
--- Comment #3 from Michael Duggan ---
(In reply to Richard Biener from comment #1)
> I'm seeing all code properly instrumented. The coverage I see is
>
> -:1:#include
> -:2:#include
> -:3:
> -:
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110827
--- Comment #2 from Iain Sandoe ---
(In reply to Richard Biener from comment #1)
> I'm seeing all code properly instrumented. The coverage I see is
> 1: 27:task foo() {
> -: 28: std::cout << "Running..." << std::endl;
>
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=110827
Richard Biener changed:
What|Removed |Added
Last reconfirmed||2023-07-27
Status|UNCONFIR
20 matches
Mail list logo
