Hi Yuanchu,
On Wed, 4 May 2022 18:29:08 +0000 Yuanchu Xie <[email protected]> wrote:
> The test case added in commit db7a347b26fe ("mm/damon/dbgfs:
> use '__GFP_NOWARN' for user-specified size buffer allocation")
> intentionally writes and reads with a large count to cause
> allocation failure and check for kernel warnings. We suppress
> the compiler warnings for these calls as they work as intended.
>
> Signed-off-by: Yuanchu Xie <[email protected]>
> ---
It would be a good practice to mention the changes from the previous version of
this patch here[1].
[1] https://docs.kernel.org/process/submitting-patches.html#the-canonical-patch-format
> tools/testing/selftests/damon/huge_count_read_write.c | 2 ++
> 1 file changed, 2 insertions(+)
>
> diff --git a/tools/testing/selftests/damon/huge_count_read_write.c b/tools/testing/selftests/damon/huge_count_read_write.c
> index ad7a6b4cf338..91bd80c75cd9 100644
> --- a/tools/testing/selftests/damon/huge_count_read_write.c
> +++ b/tools/testing/selftests/damon/huge_count_read_write.c
> @@ -2,6 +2,8 @@
> /*
> * Author: SeongJae Park <[email protected]>
> */
> +#pragma GCC diagnostic ignored "-Wstringop-overflow"
> +#pragma GCC diagnostic ignored "-Wstringop-overread"
I agree that this must be the cleaner way than v2. But, I get below warning
after applying this:
$ sudo make -C tools/testing/selftests/damon run_tests
make: Entering directory '/home/sjpark/linux/tools/testing/selftests/damon'
gcc huge_count_read_write.c -o /home/sjpark/linux/tools/testing/selftests/damon/huge_count_read_write
huge_count_read_write.c:6:32: warning: unknown option after ‘#pragma GCC diagnostic’ kind [-Wpragmas]
6 | #pragma GCC diagnostic ignored "-Wstringop-overread"
| ^~~~~~~~~~~~~~~~~~~~~
My gcc version is:
$ gcc --version
gcc (Ubuntu 9.4.0-1ubuntu1~20.04.1) 9.4.0
Thanks,
SJ
>
> #include <fcntl.h>
> #include <stdlib.h>
> --
> 2.36.0.464.gb9c8b46e94-goog
>
Hi SeongJae,
On Wed, May 4, 2022 at 11:45 AM SeongJae Park <[email protected]> wrote:
>
> Hi Yuanchu,
>
> On Wed, 4 May 2022 18:29:08 +0000 Yuanchu Xie <[email protected]> wrote:
>
> > The test case added in commit db7a347b26fe ("mm/damon/dbgfs:
> > use '__GFP_NOWARN' for user-specified size buffer allocation")
> > intentionally writes and reads with a large count to cause
> > allocation failure and check for kernel warnings. We suppress
> > the compiler warnings for these calls as they work as intended.
> >
> > Signed-off-by: Yuanchu Xie <[email protected]>
> > ---
>
> It would be a good practice to mention the changes from the previous version of
> this patch here[1].
>
> [1] https://docs.kernel.org/process/submitting-patches.html#the-canonical-patch-format
>
Thank you, I missed this when trying to figure out how to add
additional comments for a revision.
> > tools/testing/selftests/damon/huge_count_read_write.c | 2 ++
> > 1 file changed, 2 insertions(+)
> >
> > diff --git a/tools/testing/selftests/damon/huge_count_read_write.c b/tools/testing/selftests/damon/huge_count_read_write.c
> > index ad7a6b4cf338..91bd80c75cd9 100644
> > --- a/tools/testing/selftests/damon/huge_count_read_write.c
> > +++ b/tools/testing/selftests/damon/huge_count_read_write.c
> > @@ -2,6 +2,8 @@
> > /*
> > * Author: SeongJae Park <[email protected]>
> > */
> > +#pragma GCC diagnostic ignored "-Wstringop-overflow"
> > +#pragma GCC diagnostic ignored "-Wstringop-overread"
>
> I agree that this must be the cleaner way than v2. But, I get below warning
> after applying this:
>
> $ sudo make -C tools/testing/selftests/damon run_tests
> make: Entering directory '/home/sjpark/linux/tools/testing/selftests/damon'
> gcc huge_count_read_write.c -o /home/sjpark/linux/tools/testing/selftests/damon/huge_count_read_write
> huge_count_read_write.c:6:32: warning: unknown option after ‘#pragma GCC diagnostic’ kind [-Wpragmas]
> 6 | #pragma GCC diagnostic ignored "-Wstringop-overread"
> | ^~~~~~~~~~~~~~~~~~~~~
>
> My gcc version is:
>
> $ gcc --version
> gcc (Ubuntu 9.4.0-1ubuntu1~20.04.1) 9.4.0
I see, I'm running
$ gcc --version
gcc (Debian 11.2.0-16+build1) 11.2.0
I believe this is a new warning for gcc-11 [1], and somewhat
unfortunate that it results in a warning for gcc-9.4. I'm not sure
what the preference is here.
[1] https://gcc.gnu.org/onlinedocs/gcc-10.3.0/gcc/Warning-Options.html
Thanks,
Yuanchu
SeongJae,
Do you have a preference on how this should be handled?
Thanks,
Yuanchu
On Wed, May 4, 2022 at 6:12 PM Yuanchu Xie <[email protected]> wrote:
>
> Hi SeongJae,
>
> On Wed, May 4, 2022 at 11:45 AM SeongJae Park <[email protected]> wrote:
> >
> > Hi Yuanchu,
> >
> > On Wed, 4 May 2022 18:29:08 +0000 Yuanchu Xie <[email protected]> wrote:
> >
> > > The test case added in commit db7a347b26fe ("mm/damon/dbgfs:
> > > use '__GFP_NOWARN' for user-specified size buffer allocation")
> > > intentionally writes and reads with a large count to cause
> > > allocation failure and check for kernel warnings. We suppress
> > > the compiler warnings for these calls as they work as intended.
> > >
> > > Signed-off-by: Yuanchu Xie <[email protected]>
> > > ---
> >
> > It would be a good practice to mention the changes from the previous version of
> > this patch here[1].
> >
> > [1] https://docs.kernel.org/process/submitting-patches.html#the-canonical-patch-format
> >
> Thank you, I missed this when trying to figure out how to add
> additional comments for a revision.
>
> > > tools/testing/selftests/damon/huge_count_read_write.c | 2 ++
> > > 1 file changed, 2 insertions(+)
> > >
> > > diff --git a/tools/testing/selftests/damon/huge_count_read_write.c b/tools/testing/selftests/damon/huge_count_read_write.c
> > > index ad7a6b4cf338..91bd80c75cd9 100644
> > > --- a/tools/testing/selftests/damon/huge_count_read_write.c
> > > +++ b/tools/testing/selftests/damon/huge_count_read_write.c
> > > @@ -2,6 +2,8 @@
> > > /*
> > > * Author: SeongJae Park <[email protected]>
> > > */
> > > +#pragma GCC diagnostic ignored "-Wstringop-overflow"
> > > +#pragma GCC diagnostic ignored "-Wstringop-overread"
> >
> > I agree that this must be the cleaner way than v2. But, I get below warning
> > after applying this:
> >
> > $ sudo make -C tools/testing/selftests/damon run_tests
> > make: Entering directory '/home/sjpark/linux/tools/testing/selftests/damon'
> > gcc huge_count_read_write.c -o /home/sjpark/linux/tools/testing/selftests/damon/huge_count_read_write
> > huge_count_read_write.c:6:32: warning: unknown option after ‘#pragma GCC diagnostic’ kind [-Wpragmas]
> > 6 | #pragma GCC diagnostic ignored "-Wstringop-overread"
> > | ^~~~~~~~~~~~~~~~~~~~~
> >
> > My gcc version is:
> >
> > $ gcc --version
> > gcc (Ubuntu 9.4.0-1ubuntu1~20.04.1) 9.4.0
> I see, I'm running
>
> $ gcc --version
> gcc (Debian 11.2.0-16+build1) 11.2.0
>
> I believe this is a new warning for gcc-11 [1], and somewhat
> unfortunate that it results in a warning for gcc-9.4. I'm not sure
> what the preference is here.
> [1] https://gcc.gnu.org/onlinedocs/gcc-10.3.0/gcc/Warning-Options.html
>
> Thanks,
> Yuanchu
Hi Yuanchu,
On Mon, 16 May 2022 21:07:25 -0400 Yuanchu Xie <[email protected]> wrote:
> SeongJae,
>
> Do you have a preference on how this should be handled?
Sorry for late response. I was thinking you were asking Shuah's opinion. I
have no strong opinion but the approach you made in v2 looks slightly better
for me.
Thanks,
SJ
>
> Thanks,
> Yuanchu
>
> On Wed, May 4, 2022 at 6:12 PM Yuanchu Xie <[email protected]> wrote:
> >
> > Hi SeongJae,
> >
> > On Wed, May 4, 2022 at 11:45 AM SeongJae Park <[email protected]> wrote:
> > >
> > > Hi Yuanchu,
> > >
> > > On Wed, 4 May 2022 18:29:08 +0000 Yuanchu Xie <[email protected]> wrote:
> > >
> > > > The test case added in commit db7a347b26fe ("mm/damon/dbgfs:
> > > > use '__GFP_NOWARN' for user-specified size buffer allocation")
> > > > intentionally writes and reads with a large count to cause
> > > > allocation failure and check for kernel warnings. We suppress
> > > > the compiler warnings for these calls as they work as intended.
> > > >
> > > > Signed-off-by: Yuanchu Xie <[email protected]>
> > > > ---
> > >
> > > It would be a good practice to mention the changes from the previous version of
> > > this patch here[1].
> > >
> > > [1] https://docs.kernel.org/process/submitting-patches.html#the-canonical-patch-format
> > >
> > Thank you, I missed this when trying to figure out how to add
> > additional comments for a revision.
> >
> > > > tools/testing/selftests/damon/huge_count_read_write.c | 2 ++
> > > > 1 file changed, 2 insertions(+)
> > > >
> > > > diff --git a/tools/testing/selftests/damon/huge_count_read_write.c b/tools/testing/selftests/damon/huge_count_read_write.c
> > > > index ad7a6b4cf338..91bd80c75cd9 100644
> > > > --- a/tools/testing/selftests/damon/huge_count_read_write.c
> > > > +++ b/tools/testing/selftests/damon/huge_count_read_write.c
> > > > @@ -2,6 +2,8 @@
> > > > /*
> > > > * Author: SeongJae Park <[email protected]>
> > > > */
> > > > +#pragma GCC diagnostic ignored "-Wstringop-overflow"
> > > > +#pragma GCC diagnostic ignored "-Wstringop-overread"
> > >
> > > I agree that this must be the cleaner way than v2. But, I get below warning
> > > after applying this:
> > >
> > > $ sudo make -C tools/testing/selftests/damon run_tests
> > > make: Entering directory '/home/sjpark/linux/tools/testing/selftests/damon'
> > > gcc huge_count_read_write.c -o /home/sjpark/linux/tools/testing/selftests/damon/huge_count_read_write
> > > huge_count_read_write.c:6:32: warning: unknown option after ‘#pragma GCC diagnostic’ kind [-Wpragmas]
> > > 6 | #pragma GCC diagnostic ignored "-Wstringop-overread"
> > > | ^~~~~~~~~~~~~~~~~~~~~
> > >
> > > My gcc version is:
> > >
> > > $ gcc --version
> > > gcc (Ubuntu 9.4.0-1ubuntu1~20.04.1) 9.4.0
> > I see, I'm running
> >
> > $ gcc --version
> > gcc (Debian 11.2.0-16+build1) 11.2.0
> >
> > I believe this is a new warning for gcc-11 [1], and somewhat
> > unfortunate that it results in a warning for gcc-9.4. I'm not sure
> > what the preference is here.
> > [1] https://gcc.gnu.org/onlinedocs/gcc-10.3.0/gcc/Warning-Options.html
> >
> > Thanks,
> > Yuanchu
>
Hi Shuah,
On Wed, May 4, 2022 at 11:45 AM SeongJae Park <[email protected]> wrote:
>
> Hi Yuanchu,
>
> On Wed, 4 May 2022 18:29:08 +0000 Yuanchu Xie <[email protected]> wrote:
>
> > The test case added in commit db7a347b26fe ("mm/damon/dbgfs:
> > use '__GFP_NOWARN' for user-specified size buffer allocation")
> > intentionally writes and reads with a large count to cause
> > allocation failure and check for kernel warnings. We suppress
> > the compiler warnings for these calls as they work as intended.
> >
> > Signed-off-by: Yuanchu Xie <[email protected]>
> > ---
>
> It would be a good practice to mention the changes from the previous version of
> this patch here[1].
>
> [1] https://docs.kernel.org/process/submitting-patches.html#the-canonical-patch-format
>
> > tools/testing/selftests/damon/huge_count_read_write.c | 2 ++
> > 1 file changed, 2 insertions(+)
> >
> > diff --git a/tools/testing/selftests/damon/huge_count_read_write.c b/tools/testing/selftests/damon/huge_count_read_write.c
> > index ad7a6b4cf338..91bd80c75cd9 100644
> > --- a/tools/testing/selftests/damon/huge_count_read_write.c
> > +++ b/tools/testing/selftests/damon/huge_count_read_write.c
> > @@ -2,6 +2,8 @@
> > /*
> > * Author: SeongJae Park <[email protected]>
> > */
> > +#pragma GCC diagnostic ignored "-Wstringop-overflow"
> > +#pragma GCC diagnostic ignored "-Wstringop-overread"
>
> I agree that this must be the cleaner way than v2. But, I get below warning
> after applying this:
>
> $ sudo make -C tools/testing/selftests/damon run_tests
> make: Entering directory '/home/sjpark/linux/tools/testing/selftests/damon'
> gcc huge_count_read_write.c -o /home/sjpark/linux/tools/testing/selftests/damon/huge_count_read_write
> huge_count_read_write.c:6:32: warning: unknown option after ‘#pragma GCC diagnostic’ kind [-Wpragmas]
> 6 | #pragma GCC diagnostic ignored "-Wstringop-overread"
> | ^~~~~~~~~~~~~~~~~~~~~
>
> My gcc version is:
>
> $ gcc --version
> gcc (Ubuntu 9.4.0-1ubuntu1~20.04.1) 9.4.0
$ gcc --version
gcc (Debian 11.2.0-16+build1) 11.2.0
I believe this is a new warning for gcc-11 [1], and somewhat unfortunate that
it results in a warning for gcc-9.4. So a patch that would resolve the test
warnings for gcc 11 would introduce a new warning for gcc 9, and vice versa.
What's the preferred solution here?
[1] https://gcc.gnu.org/onlinedocs/gcc-10.3.0/gcc/Warning-Options.html
Thanks,
Yuanchu