[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20220422110903.GW2731@worktop.programming.kicks-ass.net>
Date: Fri, 22 Apr 2022 13:09:03 +0200
From: Peter Zijlstra <peterz@...radead.org>
To: Delyan Kratunov <delyank@...com>
Cc: "valentin.schneider@....com" <valentin.schneider@....com>,
"bigeasy@...utronix.de" <bigeasy@...utronix.de>,
"dietmar.eggemann@....com" <dietmar.eggemann@....com>,
"keescook@...omium.org" <keescook@...omium.org>,
"andrii@...nel.org" <andrii@...nel.org>,
"vincent.guittot@...aro.org" <vincent.guittot@...aro.org>,
"akpm@...ux-foundation.org" <akpm@...ux-foundation.org>,
"mingo@...nel.org" <mingo@...nel.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"rdunlap@...radead.org" <rdunlap@...radead.org>,
"rostedt@...dmis.org" <rostedt@...dmis.org>,
"Kenta.Tada@...y.com" <Kenta.Tada@...y.com>,
"adharmap@...cinc.com" <adharmap@...cinc.com>,
"tglx@...utronix.de" <tglx@...utronix.de>,
"bristot@...hat.com" <bristot@...hat.com>,
"ebiederm@...ssion.com" <ebiederm@...ssion.com>,
"ast@...nel.org" <ast@...nel.org>,
"legion@...nel.org" <legion@...nel.org>,
"ed.tsai@...iatek.com" <ed.tsai@...iatek.com>,
"u.kleine-koenig@...gutronix.de" <u.kleine-koenig@...gutronix.de>,
"juri.lelli@...hat.com" <juri.lelli@...hat.com>,
"x86@...nel.org" <x86@...nel.org>
Subject: Re: [PATCH] sched/tracing: append prev_state to tp args instead
On Thu, Apr 21, 2022 at 10:12:25PM +0000, Delyan Kratunov wrote:
> Hi folks,
>
> While working on bpf tooling, we noticed that the sched_switch tracepoint
> signature recently changed in an incompatible manner. This affects the
> runqslower tools in the kernel tree, as well as multiple libbpf tools in iovisor/bcc.
>
> It would be a fair amount of churn to fix all these tools, not to mention any
> non-public tools people may be using.
So on the one hand, too bad, deal with it. None of this is ABI.
> If you are open to it, here's a
> description and a patch that moves the new argument to the end,
> so existing tools can continue working without change (the new argument
> just won't be extracted in existing programs):
And on the other hand; those users need to be fixed anyway, right?
Accessing prev->__state is equally broken.
Yes, the proposed hack is cute, but I have very little sympathy for any
of this. These are in-kernel interfaces, they change, there must not be
any impediment to change.
If bpf wants to ride on them, it needs to suffer the pain of doing so.
Powered by blists - more mailing lists