|
|
Пишет LWN.net (![[info]](http://lj.rossia.org/img/syndicated.gif){kind=small} syn_lwnheadline)@ 2019-05-14 17:18:00 |
 |
|
 |
|
|
|
|
|
|
 |
|
 |
An eBPF overview, part 5: Tracing user processes (Collabora blog)
The fifth
and final article in Adrian Ratiu's series
on eBPF delves into userspace tracing. "In our previous parts we focused on tracing the Linux kernel, for which the eBPF-based projects are, in our humble opinion, the most safe, widely available and useful methods (eBPF is fully upstreamed in Linux, guarantees a stable ABI, comes enabled by default in almost all distributions and integrates with all other tracing mechanisms). It has really become a no-brainer choice for kernel work. However, up until now, talking in-depth about userspace tracing was deliberately avoided because it merits special treatment, hence this full part 5 article devoted to it."
The fifth
and final article in Adrian Ratiu's series
on eBPF delves into userspace tracing. "In our previous parts we focused on tracing the Linux kernel, for which the eBPF-based projects are, in our humble opinion, the most safe, widely available and useful methods (eBPF is fully upstreamed in Linux, guarantees a stable ABI, comes enabled by default in almost all distributions and integrates with all other tracing mechanisms). It has really become a no-brainer choice for kernel work. However, up until now, talking in-depth about userspace tracing was deliberately avoided because it merits special treatment, hence this full part 5 article devoted to it."
