ile 'description' to name the repository.</subtitle>
<link rel='alternate' type='text/html' href='http://git.arm.linux.org.uk/cgit/git.armlinux.org.uk/linux-net-next.git/'/>
<entry>
<title>bpf: Add bpf_read_branch_records() helper</title>
<updated>2020-02-19T22:37:36+00:00</updated>
<author>
<name>Daniel Xu</name>
<email>dxu@dxuuu.xyz</email>
</author>
<published>2020-02-18T03:04:31+00:00</published>
<link rel='alternate' type='text/html' href='http://git.arm.linux.org.uk/cgit/git.armlinux.org.uk/linux-net-next.git/commit/?id=fff7b64355eac6e29b50229ad1512315bc04b44e'/>
<id>fff7b64355eac6e29b50229ad1512315bc04b44e</id>
<content type='text'>
Branch records are a CPU feature that can be configured to record
certain branches that are taken during code execution. This data is
particularly interesting for profile guided optimizations. perf has had
branch record support for a while but the data collection can be a bit
coarse grained.

We (Facebook) have seen in experiments that associating metadata with
branch records can improve results (after postprocessing). We generally
use bpf_probe_read_*() to get metadata out of userspace. That's why bpf
support for branch records is useful.

Aside from this particular use case, having branch data available to bpf
progs can be useful to get stack traces out of userspace applications
that omit frame pointers.

Signed-off-by: Daniel Xu &lt;dxu@dxuuu.xyz&gt;
Signed-off-by: Alexei Starovoitov &lt;ast@kernel.org&gt;
Acked-by: Andrii Nakryiko &lt;andriin@fb.com&gt;
Link: https://lore.kernel.org/bpf/20200218030432.4600-2-dxu@dxuuu.xyz
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
Branch records are a CPU feature that can be configured to record
certain branches that are taken during code execution. This data is
particularly interesting for profile guided optimizations. perf has had
branch record support for a while but the data collection can be a bit
coarse grained.

We (Facebook) have seen in experiments that associating metadata with
branch records can improve results (after postprocessing). We generally
use bpf_probe_read_*() to get metadata out of userspace. That's why bpf
support for branch records is useful.

Aside from this particular use case, having branch data available to bpf
progs can be useful to get stack traces out of userspace applications
that omit frame pointers.

Signed-off-by: Daniel Xu &lt;dxu@dxuuu.xyz&gt;
Signed-off-by: Alexei Starovoitov &lt;ast@kernel.org&gt;
Acked-by: Andrii Nakryiko &lt;andriin@fb.com&gt;
Link: https://lore.kernel.org/bpf/20200218030432.4600-2-dxu@dxuuu.xyz
</pre>
</div>
</content>
</entry>
<entry>
<title>bpf: Allow bpf_perf_event_read_value in all BPF programs</title>
<updated>2020-02-18T15:08:27+00:00</updated>
<author>
<name>Song Liu</name>
<email>songliubraving@fb.com</email>
</author>
<published>2020-02-14T23:41:46+00:00</published>
<link rel='alternate' type='text/html' href='http://git.arm.linux.org.uk/cgit/git.armlinux.org.uk/linux-net-next.git/commit/?id=b80b033bedae68dae8fc703ab8a69811ce678f2e'/>
<id>b80b033bedae68dae8fc703ab8a69811ce678f2e</id>
<content type='text'>
bpf_perf_event_read_value() is NMI safe. Enable it for all BPF programs.
This can be used in fentry/fexit to profile BPF program and individual
kernel function with hardware counters.

Signed-off-by: Song Liu &lt;songliubraving@fb.com&gt;
Signed-off-by: Daniel Borkmann &lt;daniel@iogearbox.net&gt;
Link: https://lore.kernel.org/bpf/20200214234146.2910011-1-songliubraving@fb.com
</content>
<content type='xhtml'>
<div xmlns='http://www.w3.org/1999/xhtml'>
<pre>
bpf_perf_event_read_value() is NMI safe. Enable it for all BPF programs.
This can be used in fentry/fexit to profile BPF program and individual
kernel function with hardware counters.

Signed-off-by: Song Liu &lt;songliubraving@fb.com&gt;
Signed-off-by: Daniel Borkmann &lt;daniel@iogearbox.net&gt;
Link: https://lore.kernel.org/bpf/20200214234146.2910011-1-songliubraving@fb.com
</pre>
</div>
</content>
</entry>