* Expected warning maybe-uninitialized does not appear using g++13.2.0?
@ 2023-12-20 19:16 Eric Batchelor
2023-12-22 3:43 ` David Malcolm
0 siblings, 1 reply; 3+ messages in thread
From: Eric Batchelor @ 2023-12-20 19:16 UTC (permalink / raw)
To: gcc
Hello, I unintentionally stumbled upon some strange behaviour that
occurred due to a typo.
I reproduced the behaviour where an object (std::string in my case) can
be passed to a function by reference, uninitialized, WITHOUT a compiler
warning.
Changing the code to pass the object by value DOES emit the warning.
I don't think the compiled code is incorrect, it segfaults presumably
due to uninitialized members.
I understand there may seldom be a reason to use uninitialized objects,
so "don't do that," but as I said this was unintentional and it seems
that it should have generated a warning, which have saved some
head-scratching.
Code to reproduce:
#include <string>
std::string f(std::string &s) {
s.append("x");
return s;
}
int main() {
std::string a = f(a);
}
Compile and run (no warning):
$ g++ -o uninit_obj uninit_obj.cpp -std=c++23 -Wall -Wpedantic -Wextra
&& ./uninit_obj
Segmentation fault (core dumped)
No difference whether using -O0 (or 1 2 3)
If I change the function to pass by value, std::string f(std::string s),
and rerun, I get the expected compiler warning:
$ g++ -o uninit_obj uninit_obj.cpp -std=c++23 -Wall -Wpedantic -Wextra
&& ./uninit_obj
uninit_obj.cpp: In function 'int main()':
uninit_obj.cpp:7:22: warning: 'a' may be used uninitialized
[-Wmaybe-uninitialized]
7 | std::string a = f(a);
[...]
terminate called after throwing an instance of 'std::bad_alloc'
what(): std::bad_alloc
Aborted (core dumped)
Output from g++ -v:
Using built-in specs.
COLLECT_GCC=g++
COLLECT_LTO_WRAPPER=/usr/local/gcc13/libexec/gcc/x86_64-pc-linux-gnu/13.2.0/lto-wrapper
Target: x86_64-pc-linux-gnu
Configured with: ../gcc-13.2.0/configure --disable-multilib
--enable-languages=c,c++ --prefix=/usr/local/gcc13 --program-suffix=-13
--enable-libstdcxx-backtrace=yes
Thread model: posix
Supported LTO compression algorithms: zlib
gcc version 13.2.0 (GCC)
Thanks
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: Expected warning maybe-uninitialized does not appear using g++13.2.0?
2023-12-20 19:16 Expected warning maybe-uninitialized does not appear using g++13.2.0? Eric Batchelor
@ 2023-12-22 3:43 ` David Malcolm
2023-12-22 4:45 ` Marc Glisse
0 siblings, 1 reply; 3+ messages in thread
From: David Malcolm @ 2023-12-22 3:43 UTC (permalink / raw)
To: Eric Batchelor, gcc
On Wed, 2023-12-20 at 11:16 -0800, Eric Batchelor wrote:
> Hello, I unintentionally stumbled upon some strange behaviour that
> occurred due to a typo.
> I reproduced the behaviour where an object (std::string in my case)
> can
> be passed to a function by reference, uninitialized, WITHOUT a
> compiler
> warning.
> Changing the code to pass the object by value DOES emit the warning.
> I don't think the compiled code is incorrect, it segfaults presumably
> due to uninitialized members.
> I understand there may seldom be a reason to use uninitialized
> objects,
> so "don't do that," but as I said this was unintentional and it seems
> that it should have generated a warning, which have saved some
> head-scratching.
>
> Code to reproduce:
>
> #include <string>
> std::string f(std::string &s) {
> s.append("x");
> return s;
> }
> int main() {
> std::string a = f(a);
> }
>
> Compile and run (no warning):
>
> $ g++ -o uninit_obj uninit_obj.cpp -std=c++23 -Wall -Wpedantic -
> Wextra
> && ./uninit_obj
> Segmentation fault (core dumped)
>
> No difference whether using -O0 (or 1 2 3)
As I understand it, -Wmaybe-uninitialized is purely intraprocedural
i.e. it works within each individual function, without considering the
interactions *between* functions.
FWIW, -fanalyzer does attempt to model interprocedural interactions,
but doesn't yet work properly on C++ code. For your example, it
happens to generate some warnings, but the wording is really vague;
see: https://godbolt.org/z/a1q7xYMjb
and it might well be getting other things wrong (as I said, it doesn't
yet properly work on C++).
Dave
>
> If I change the function to pass by value, std::string f(std::string
> s),
> and rerun, I get the expected compiler warning:
>
> $ g++ -o uninit_obj uninit_obj.cpp -std=c++23 -Wall -Wpedantic -
> Wextra
> && ./uninit_obj
> uninit_obj.cpp: In function 'int main()':
> uninit_obj.cpp:7:22: warning: 'a' may be used uninitialized
> [-Wmaybe-uninitialized]
> 7 | std::string a = f(a);
> [...]
> terminate called after throwing an instance of 'std::bad_alloc'
> what(): std::bad_alloc
> Aborted (core dumped)
>
> Output from g++ -v:
>
> Using built-in specs.
> COLLECT_GCC=g++
> COLLECT_LTO_WRAPPER=/usr/local/gcc13/libexec/gcc/x86_64-pc-linux-
> gnu/13.2.0/lto-wrapper
> Target: x86_64-pc-linux-gnu
> Configured with: ../gcc-13.2.0/configure --disable-multilib
> --enable-languages=c,c++ --prefix=/usr/local/gcc13 --program-suffix=-
> 13
> --enable-libstdcxx-backtrace=yes
> Thread model: posix
> Supported LTO compression algorithms: zlib
> gcc version 13.2.0 (GCC)
>
> Thanks
>
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: Expected warning maybe-uninitialized does not appear using g++13.2.0?
2023-12-22 3:43 ` David Malcolm
@ 2023-12-22 4:45 ` Marc Glisse
0 siblings, 0 replies; 3+ messages in thread
From: Marc Glisse @ 2023-12-22 4:45 UTC (permalink / raw)
To: David Malcolm; +Cc: Eric Batchelor, gcc
On Thu, 21 Dec 2023, David Malcolm via Gcc wrote:
> On Wed, 2023-12-20 at 11:16 -0800, Eric Batchelor wrote:
>> Hello, I unintentionally stumbled upon some strange behaviour that
>> occurred due to a typo.
>> I reproduced the behaviour where an object (std::string in my case)
>> can
>> be passed to a function by reference, uninitialized, WITHOUT a
>> compiler
>> warning.
>> Changing the code to pass the object by value DOES emit the warning.
>> I don't think the compiled code is incorrect, it segfaults presumably
>> due to uninitialized members.
>> I understand there may seldom be a reason to use uninitialized
>> objects,
>> so "don't do that," but as I said this was unintentional and it seems
>> that it should have generated a warning, which have saved some
>> head-scratching.
>>
>> Code to reproduce:
>>
>> #include <string>
>> std::string f(std::string &s) {
>> s.append("x");
>> return s;
>> }
>> int main() {
>> std::string a = f(a);
>> }
>>
>> Compile and run (no warning):
>>
>> $ g++ -o uninit_obj uninit_obj.cpp -std=c++23 -Wall -Wpedantic -
>> Wextra
>> && ./uninit_obj
>> Segmentation fault (core dumped)
>>
>> No difference whether using -O0 (or 1 2 3)
>
> As I understand it, -Wmaybe-uninitialized is purely intraprocedural
> i.e. it works within each individual function, without considering the
> interactions *between* functions.
If you compile
#include <string>
static std::string f(std::string &s) {
s.append("x");
return s;
}
void g() {
std::string a = f(a);
}
with -O3, by the time we get to the uninit pass, function g starts with
void g ()
{
size_type __dnew;
struct string a;
[...]
<bb 2> [local count: 1073741824]:
_26 = a._M_string_length;
if (_26 == 4611686018427387903)
which should not require any interprocedural logic.
--
Marc Glisse
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2023-12-22 4:45 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-12-20 19:16 Expected warning maybe-uninitialized does not appear using g++13.2.0? Eric Batchelor
2023-12-22 3:43 ` David Malcolm
2023-12-22 4:45 ` Marc Glisse
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).