Re: [PATCH v2] perf,x86: avoid missing caller address in stack traces captured in uprobe

From: Josh Poimboeuf
Date: Wed Jul 03 2024 - 18:41:17 EST


On Wed, Jul 03, 2024 at 01:23:39PM -0700, Andrii Nakryiko wrote:
> > > [0] https://docs.google.com/presentation/d/1k10-HtK7pP5CMMa86dDCdLW55fHOut4co3Zs5akk0t4
> >
> > I don't seem to have permission to open it.
> >
>
> Argh, sorry, it's under my corporate account which doesn't allow
> others to view it. Try this, I "published" it, let me know if that
> still doesn't work:
>
> [0] https://docs.google.com/presentation/d/e/2PACX-1vRgL3UPbkrznwtNPKn-sSjvan7tFeMqOrIyZAFSSEPYiWG20JGSP80jBmZqGwqMuBGVmv9vyLU4KRTx/pub

The new link doesn't work either :-)

> > > Few questions, while we are at it. Does it mean that
> > > perf_callchain_user() will support working from sleepable context and
> > > will wait for data to be paged in? Is anyone already working on this?
> > > Any pointers?
> >
> > I had a prototype here:
> >
> > https://lkml.kernel.org/lkml/cover.1699487758.git.jpoimboe@xxxxxxxxxx
> >
> > Hopefully I can get started on v2 soon.
>
> Ok, so you are going to work on this. Please cc me on future revisions
> then. Thanks!

Will do!

--
Josh