From: Peter Zijlstra <email@example.com> To: Adrian Hunter <firstname.lastname@example.org> Cc: Andi Kleen <email@example.com>, Nadav Amit <firstname.lastname@example.org>, Ingo Molnar <email@example.com>, Andy Lutomirski <firstname.lastname@example.org>, Josh Poimboeuf <email@example.com>, Edward Cree <firstname.lastname@example.org>, "H . Peter Anvin" <email@example.com>, Thomas Gleixner <firstname.lastname@example.org>, LKML <email@example.com>, Nadav Amit <firstname.lastname@example.org>, X86 ML <email@example.com>, Paolo Abeni <firstname.lastname@example.org>, Borislav Petkov <email@example.com>, David Woodhouse <firstname.lastname@example.org>, Alexander Shishkin <email@example.com>, firstname.lastname@example.org Subject: Re: [RFC v2 0/6] x86: dynamic indirect branch promotion Date: Tue, 8 Jan 2019 10:25:59 +0100 Message-ID: <20190108092559.GA6808@hirez.programming.kicks-ass.net> (raw) In-Reply-To: <email@example.com> On Tue, Jan 08, 2019 at 09:47:18AM +0200, Adrian Hunter wrote: > On 7/01/19 6:32 PM, Peter Zijlstra wrote: > > On Thu, Jan 03, 2019 at 02:18:15PM -0800, Andi Kleen wrote: > >> Nadav Amit <firstname.lastname@example.org> writes: > >>> > >>> - Do we use periodic learning or not? Josh suggested to reconfigure the > >>> branches whenever a new target is found. However, I do not know at > >>> this time how to do learning efficiently, without making learning much > >>> more expensive. > >> > >> FWIW frequent patching will likely completely break perf Processor Trace > >> decoding, which needs a somewhat stable kernel text image to decode the > >> traces generated by the CPU. Right now it relies on kcore dumped after > >> the trace usually being stable because jumplabel changes happen only > >> infrequently. But if you start patching frequently this assumption will > >> break. > >> > >> You would either need a way to turn this off, or provide > >> updates for every change to the trace, so that the decoder can > >> keep track. > > > > I'm thining it would be entirely possible to create and feed text_poke > > events into the regular (!aux) buffer which can be timestamp correlated > > to the PT data. > > To rebuild kernel text from such events would require a starting point. > What is the starting point? The problem with kcore is that people can > deconfig it without realising it is needed to enable the tracing of kernel > self-modifying code. It would be nice if it was all tied together, so that > if someone selects the ability to trace kernel self-modifying code, then all > the bits needed are also selected. Perhaps we should expose another ELF > image that contains only kernel executable code, and take the opportunity to > put the symbols in it also. Meh; you always need a magic combo of CONFIG symbols to make stuff work. We don't even have a CONFIG symbol for PT, so if you really care you should probably start there. If you want symbols; what stops us from exposing kallsyms in kcore as is? > Also what about BPF jitted code? Will it always fit in an event? I was > thinking of trying to add a way to prevent temporarily the unload of modules > or jitted code, which would be a good-enough solution for now. We're working on BPF and kallsym events, those should, esp. when combined with kcore, allow you to extract the actual instructions. We don't have module events, but I suppose the kallsym events should cover module loading (a new module results in lots of new symbols after all). With all that there is still a race in that nothing blocks module-unload while we're still harvesting the information, not sure how/if we want to cure that.
next prev parent reply index Thread overview: 42+ messages / expand[flat|nested] mbox.gz Atom feed top 2018-12-31 7:21 Nadav Amit 2018-12-31 7:21 ` [RFC v2 1/6] x86: introduce kernel restartable sequence Nadav Amit 2018-12-31 20:08 ` Andy Lutomirski 2018-12-31 21:12 ` Nadav Amit 2019-01-03 22:21 ` Andi Kleen 2019-01-03 22:29 ` Nadav Amit 2019-01-03 22:48 ` Andi Kleen 2019-01-03 22:52 ` Nadav Amit 2019-01-03 23:40 ` Andi Kleen 2019-01-03 23:56 ` Nadav Amit 2019-01-04 0:34 ` hpa 2018-12-31 7:21 ` [RFC v2 2/6] objtool: ignore instructions Nadav Amit 2018-12-31 7:21 ` [RFC v2 3/6] x86: patch indirect branch promotion Nadav Amit 2018-12-31 7:21 ` [RFC v2 4/6] x86: interface for accessing indirect branch locations Nadav Amit 2018-12-31 7:21 ` [RFC v2 5/6] x86: learning and patching indirect branch targets Nadav Amit 2018-12-31 20:05 ` Andy Lutomirski 2018-12-31 21:07 ` Nadav Amit 2018-12-31 7:21 ` [RFC v2 6/6] x86: outline optpoline Nadav Amit 2018-12-31 19:51 ` [RFC v2 0/6] x86: dynamic indirect branch promotion Andy Lutomirski 2018-12-31 19:53 ` Nadav Amit 2019-01-03 18:10 ` Josh Poimboeuf 2019-01-03 18:30 ` Nadav Amit 2019-01-03 20:31 ` Josh Poimboeuf 2019-01-03 22:18 ` Andi Kleen 2019-01-07 16:32 ` Peter Zijlstra 2019-01-08 7:47 ` Adrian Hunter 2019-01-08 9:25 ` Peter Zijlstra [this message] 2019-01-08 10:01 ` Adrian Hunter 2019-01-08 10:10 ` Peter Zijlstra 2019-01-08 17:27 ` Andi Kleen 2019-01-08 18:28 ` Nadav Amit 2019-01-08 19:01 ` Peter Zijlstra 2019-01-08 20:47 ` Nadav Amit 2019-01-08 20:53 ` Andi Kleen 2019-01-09 10:35 ` Peter Zijlstra 2019-08-29 8:23 ` Tracing text poke / kernel self-modifying code (Was: Re: [RFC v2 0/6] x86: dynamic indirect branch promotion) Adrian Hunter 2019-08-29 8:53 ` Peter Zijlstra 2019-08-29 9:40 ` Adrian Hunter 2019-08-29 11:46 ` Peter Zijlstra 2019-09-12 7:00 ` Adrian Hunter 2019-09-12 12:17 ` hpa 2019-01-08 18:57 ` [RFC v2 0/6] x86: dynamic indirect branch promotion Peter Zijlstra
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=20190108092559.GA6808@hirez.programming.kicks-ass.net \ --email@example.com \ --firstname.lastname@example.org \ --email@example.com \ --firstname.lastname@example.org \ --email@example.com \ --firstname.lastname@example.org \ --email@example.com \ --firstname.lastname@example.org \ --email@example.com \ --firstname.lastname@example.org \ --email@example.com \ --firstname.lastname@example.org \ --email@example.com \ --firstname.lastname@example.org \ --email@example.com \ --firstname.lastname@example.org \ --email@example.com \ --firstname.lastname@example.org \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: link
LKML Archive on lore.kernel.org Archives are clonable: git clone --mirror https://lore.kernel.org/lkml/0 lkml/git/0.git git clone --mirror https://lore.kernel.org/lkml/1 lkml/git/1.git git clone --mirror https://lore.kernel.org/lkml/2 lkml/git/2.git git clone --mirror https://lore.kernel.org/lkml/3 lkml/git/3.git git clone --mirror https://lore.kernel.org/lkml/4 lkml/git/4.git git clone --mirror https://lore.kernel.org/lkml/5 lkml/git/5.git git clone --mirror https://lore.kernel.org/lkml/6 lkml/git/6.git git clone --mirror https://lore.kernel.org/lkml/7 lkml/git/7.git git clone --mirror https://lore.kernel.org/lkml/8 lkml/git/8.git # If you have public-inbox 1.1+ installed, you may # initialize and index your mirror using the following commands: public-inbox-init -V2 lkml lkml/ https://lore.kernel.org/lkml \ email@example.com public-inbox-index lkml Example config snippet for mirrors Newsgroup available over NNTP: nntp://nntp.lore.kernel.org/org.kernel.vger.linux-kernel AGPL code for this site: git clone https://public-inbox.org/public-inbox.git