Re: [PATCH v20 0/5] Introducing trace buffer mapping by user-space

2024-04-16 Thread Mike Rapoport
(added linux-mm) On Wed, Apr 10, 2024 at 01:56:12PM -0400, Steven Rostedt wrote: > > Hi Andrew, et.al. > > Linus said it's a hard requirement that this code gets an Acked-by (or > Reviewed-by) from the memory sub-maintainers before he will accept it. > He was upset that we faulted in pages one

Re: [PATCH v20 0/5] Introducing trace buffer mapping by user-space

2024-04-10 Thread Steven Rostedt
Hi Andrew, et.al. Linus said it's a hard requirement that this code gets an Acked-by (or Reviewed-by) from the memory sub-maintainers before he will accept it. He was upset that we faulted in pages one at a time instead of mapping it in one go:

[PATCH v20 0/5] Introducing trace buffer mapping by user-space

2024-04-06 Thread Vincent Donnefort
The tracing ring-buffers can be stored on disk or sent to network without any copy via splice. However the later doesn't allow real time processing of the traces. A solution is to give userspace direct access to the ring-buffer pages via a mapping. An application can now become a consumer of the