<feed xmlns='http://www.w3.org/2005/Atom'>
<title>kernel/linux.git/tools/testing/selftests/rseq/rseq.c, branch v6.3.6</title>
<subtitle>Linux kernel stable tree (mirror)</subtitle>
<id>https://git.radix-linux.su/kernel/linux.git/atom?h=v6.3.6</id>
<link rel='self' href='https://git.radix-linux.su/kernel/linux.git/atom?h=v6.3.6'/>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/'/>
<updated>2022-12-27T11:52:11+00:00</updated>
<entry>
<title>selftests/rseq: Implement rseq numa node id field selftest</title>
<updated>2022-12-27T11:52:11+00:00</updated>
<author>
<name>Mathieu Desnoyers</name>
<email>mathieu.desnoyers@efficios.com</email>
</author>
<published>2022-11-22T20:39:08+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=99babd04b25054717d21840298b0b46046b42cd9'/>
<id>urn:sha1:99babd04b25054717d21840298b0b46046b42cd9</id>
<content type='text'>
Test the NUMA node id extension rseq field. Compare it against the value
returned by the getcpu(2) system call while pinned on a specific core.

Signed-off-by: Mathieu Desnoyers &lt;mathieu.desnoyers@efficios.com&gt;
Signed-off-by: Peter Zijlstra (Intel) &lt;peterz@infradead.org&gt;
Link: https://lore.kernel.org/r/20221122203932.231377-7-mathieu.desnoyers@efficios.com
</content>
</entry>
<entry>
<title>selftests/rseq: Use ELF auxiliary vector for extensible rseq</title>
<updated>2022-12-27T11:52:11+00:00</updated>
<author>
<name>Mathieu Desnoyers</name>
<email>mathieu.desnoyers@efficios.com</email>
</author>
<published>2022-11-22T20:39:07+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=03f5c0272d1b59343144e199becc911dae52c37e'/>
<id>urn:sha1:03f5c0272d1b59343144e199becc911dae52c37e</id>
<content type='text'>
Use the ELF auxiliary vector AT_RSEQ_FEATURE_SIZE to detect the RSEQ
features supported by the kernel.

Signed-off-by: Mathieu Desnoyers &lt;mathieu.desnoyers@efficios.com&gt;
Signed-off-by: Peter Zijlstra (Intel) &lt;peterz@infradead.org&gt;
Link: https://lore.kernel.org/r/20221122203932.231377-6-mathieu.desnoyers@efficios.com
</content>
</entry>
<entry>
<title>selftests/rseq: Fix: Fail thread registration when CONFIG_RSEQ=n</title>
<updated>2022-12-27T11:52:10+00:00</updated>
<author>
<name>Mathieu Desnoyers</name>
<email>mathieu.desnoyers@efficios.com</email>
</author>
<published>2022-11-22T20:39:03+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=4842dadfc66f627083ec46c4e9a426e805c765f3'/>
<id>urn:sha1:4842dadfc66f627083ec46c4e9a426e805c765f3</id>
<content type='text'>
When linking the selftests against a libc which does not handle rseq
registration (before 2.35),  rseq thread registration silently succeed
even with CONFIG_RSEQ=n because it erroneously thinks that libc is
handling rseq registration.

This is caused by setting the rseq ownership flag only after the
rseq_available() check. It should rather be set before the
rseq_available() check.

Set the rseq_size to 0 (error value) immediately after the
rseq_available() check fails rather than in the thread registration
functions.

Signed-off-by: Mathieu Desnoyers &lt;mathieu.desnoyers@efficios.com&gt;
Signed-off-by: Peter Zijlstra (Intel) &lt;peterz@infradead.org&gt;
Link: https://lore.kernel.org/r/20221122203932.231377-2-mathieu.desnoyers@efficios.com
</content>
</entry>
<entry>
<title>selftests/rseq: check if libc rseq support is registered</title>
<updated>2022-06-28T07:08:28+00:00</updated>
<author>
<name>Michael Jeanson</name>
<email>mjeanson@efficios.com</email>
</author>
<published>2022-06-14T15:48:30+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=d1a997ba4c1bf65497d956aea90de42a6398f73a'/>
<id>urn:sha1:d1a997ba4c1bf65497d956aea90de42a6398f73a</id>
<content type='text'>
When checking for libc rseq support in the library constructor, don't
only depend on the symbols presence, check that the registration was
completed.

This targets a scenario where the libc has rseq support but it is not
wired for the current architecture in 'bits/rseq.h', we want to fallback
to our internal registration mechanism.

Signed-off-by: Michael Jeanson &lt;mjeanson@efficios.com&gt;
Signed-off-by: Peter Zijlstra (Intel) &lt;peterz@infradead.org&gt;
Reviewed-by: Mathieu Desnoyers &lt;mathieu.desnoyers@efficios.com&gt;
Link: https://lore.kernel.org/r/20220614154830.1367382-4-mjeanson@efficios.com
</content>
</entry>
<entry>
<title>selftests/rseq: Change type of rseq_offset to ptrdiff_t</title>
<updated>2022-02-11T22:30:08+00:00</updated>
<author>
<name>Mathieu Desnoyers</name>
<email>mathieu.desnoyers@efficios.com</email>
</author>
<published>2022-02-03T15:05:32+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=889c5d60fbcf332c8b6ab7054d45f2768914a375'/>
<id>urn:sha1:889c5d60fbcf332c8b6ab7054d45f2768914a375</id>
<content type='text'>
Just before the 2.35 release of glibc, the __rseq_offset userspace ABI
was changed from int to ptrdiff_t.

Adapt to this change in the kernel selftests.

Signed-off-by: Mathieu Desnoyers &lt;mathieu.desnoyers@efficios.com&gt;
Signed-off-by: Peter Zijlstra (Intel) &lt;peterz@infradead.org&gt;
Link: https://sourceware.org/pipermail/libc-alpha/2022-February/136024.html
</content>
</entry>
<entry>
<title>selftests/rseq: Uplift rseq selftests for compatibility with glibc-2.35</title>
<updated>2022-02-02T12:11:35+00:00</updated>
<author>
<name>Mathieu Desnoyers</name>
<email>mathieu.desnoyers@efficios.com</email>
</author>
<published>2022-01-24T17:12:45+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=233e667e1ae3e348686bd9dd0172e62a09d852e1'/>
<id>urn:sha1:233e667e1ae3e348686bd9dd0172e62a09d852e1</id>
<content type='text'>
glibc-2.35 (upcoming release date 2022-02-01) exposes the rseq per-thread
data in the TCB, accessible at an offset from the thread pointer, rather
than through an actual Thread-Local Storage (TLS) variable, as the
Linux kernel selftests initially expected.

The __rseq_abi TLS and glibc-2.35's ABI for per-thread data cannot
actively coexist in a process, because the kernel supports only a single
rseq registration per thread.

Here is the scheme introduced to ensure selftests can work both with an
older glibc and with glibc-2.35+:

- librseq exposes its own "rseq_offset, rseq_size, rseq_flags" ABI.

- librseq queries for glibc rseq ABI (__rseq_offset, __rseq_size,
  __rseq_flags) using dlsym() in a librseq library constructor. If those
  are found, copy their values into rseq_offset, rseq_size, and
  rseq_flags.

- Else, if those glibc symbols are not found, handle rseq registration
  from librseq and use its own IE-model TLS to implement the rseq ABI
  per-thread storage.

Signed-off-by: Mathieu Desnoyers &lt;mathieu.desnoyers@efficios.com&gt;
Signed-off-by: Peter Zijlstra (Intel) &lt;peterz@infradead.org&gt;
Link: https://lkml.kernel.org/r/20220124171253.22072-8-mathieu.desnoyers@efficios.com
</content>
</entry>
<entry>
<title>selftests/rseq: Remove volatile from __rseq_abi</title>
<updated>2022-02-02T12:11:34+00:00</updated>
<author>
<name>Mathieu Desnoyers</name>
<email>mathieu.desnoyers@efficios.com</email>
</author>
<published>2022-01-24T17:12:42+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=94b80a19ebfe347a01301d750040a61c38200e2b'/>
<id>urn:sha1:94b80a19ebfe347a01301d750040a61c38200e2b</id>
<content type='text'>
This is done in preparation for the selftest uplift to become compatible
with glibc-2.35.

All accesses to the __rseq_abi fields are volatile, but remove the
volatile from the TLS variable declaration, otherwise we are stuck with
volatile for the upcoming rseq_get_abi() helper.

Signed-off-by: Mathieu Desnoyers &lt;mathieu.desnoyers@efficios.com&gt;
Signed-off-by: Peter Zijlstra (Intel) &lt;peterz@infradead.org&gt;
Link: https://lkml.kernel.org/r/20220124171253.22072-5-mathieu.desnoyers@efficios.com
</content>
</entry>
<entry>
<title>selftests/rseq: introduce own copy of rseq uapi header</title>
<updated>2022-02-02T12:11:33+00:00</updated>
<author>
<name>Mathieu Desnoyers</name>
<email>mathieu.desnoyers@efficios.com</email>
</author>
<published>2022-01-24T17:12:39+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=5c105d55a9dc9e01535116ccfc26e703168a574f'/>
<id>urn:sha1:5c105d55a9dc9e01535116ccfc26e703168a574f</id>
<content type='text'>
The Linux kernel rseq uapi header has a broken layout for the
rseq_cs.ptr field on 32-bit little endian architectures. The entire
rseq_cs.ptr field is planned for removal, leaving only the 64-bit
rseq_cs.ptr64 field available.

Both glibc and librseq use their own copy of the Linux kernel uapi
header, where they introduce proper union fields to access to the 32-bit
low order bits of the rseq_cs pointer on 32-bit architectures.

Introduce a copy of the Linux kernel uapi headers in the Linux kernel
selftests.

Signed-off-by: Mathieu Desnoyers &lt;mathieu.desnoyers@efficios.com&gt;
Signed-off-by: Peter Zijlstra (Intel) &lt;peterz@infradead.org&gt;
Link: https://lkml.kernel.org/r/20220124171253.22072-2-mathieu.desnoyers@efficios.com
</content>
</entry>
<entry>
<title>selftests/rseq: remove ARRAY_SIZE define from individual tests</title>
<updated>2021-12-11T00:51:03+00:00</updated>
<author>
<name>Shuah Khan</name>
<email>skhan@linuxfoundation.org</email>
</author>
<published>2021-12-09T23:19:09+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=07ad4f7629d4802ff0d962b0ac23ea6445964e2a'/>
<id>urn:sha1:07ad4f7629d4802ff0d962b0ac23ea6445964e2a</id>
<content type='text'>
ARRAY_SIZE is defined in several selftests. Remove definitions from
individual test files and include header file for the define instead.
ARRAY_SIZE define is added in a separate patch to prepare for this
change.

Remove ARRAY_SIZE from rseq tests and pickup the one defined in
kselftest.h.

Signed-off-by: Shuah Khan &lt;skhan@linuxfoundation.org&gt;
</content>
</entry>
<entry>
<title>rseq/selftests: Use __rseq_handled symbol to coexist with glibc</title>
<updated>2019-05-07T21:31:46+00:00</updated>
<author>
<name>Mathieu Desnoyers</name>
<email>mathieu.desnoyers@efficios.com</email>
</author>
<published>2019-04-29T15:27:55+00:00</published>
<link rel='alternate' type='text/html' href='https://git.radix-linux.su/kernel/linux.git/commit/?id=5b0c308a0565a94d2e1070cbf287197b676faaaf'/>
<id>urn:sha1:5b0c308a0565a94d2e1070cbf287197b676faaaf</id>
<content type='text'>
In order to integrate rseq into user-space applications, expose a
__rseq_handled symbol so many rseq users can be linked into the same
application (e.g. librseq and glibc).

The __rseq_refcount TLS variable is static to the librseq library. It
ensures that rseq syscall registration/unregistration happens only for
the most early/late caller to rseq_{,un}register_current_thread for each
thread, thus ensuring that rseq is registered across the lifetime of all
rseq users for a given thread.

Signed-off-by: Mathieu Desnoyers &lt;mathieu.desnoyers@efficios.com&gt;
CC: Shuah Khan &lt;shuah@kernel.org&gt;
CC: Carlos O'Donell &lt;carlos@redhat.com&gt;
CC: Florian Weimer &lt;fweimer@redhat.com&gt;
CC: Joseph Myers &lt;joseph@codesourcery.com&gt;
CC: Szabolcs Nagy &lt;szabolcs.nagy@arm.com&gt;
CC: Thomas Gleixner &lt;tglx@linutronix.de&gt;
CC: Ben Maurer &lt;bmaurer@fb.com&gt;
CC: Peter Zijlstra &lt;peterz@infradead.org&gt;
CC: "Paul E. McKenney" &lt;paulmck@linux.vnet.ibm.com&gt;
CC: Boqun Feng &lt;boqun.feng@gmail.com&gt;
CC: Will Deacon &lt;will.deacon@arm.com&gt;
CC: Dave Watson &lt;davejwatson@fb.com&gt;
CC: Paul Turner &lt;pjt@google.com&gt;
CC: linux-api@vger.kernel.org
Signed-off-by: Shuah Khan &lt;skhan@linuxfoundation.org&gt;
</content>
</entry>
</feed>
