From: Michel Dagenais via lttng-dev <lttng-dev@lists.lttng.org> To: Steven Rostedt <rostedt@goodmis.org> Cc: ahmadkhorrami <ahmadkhorrami@ut.ac.ir>, linux-trace-users-owner@vger.kernel.org, linux-trace-users <linux-trace-users@vger.kernel.org>, lttng-dev <lttng-dev@lists.lttng.org>, Namhyung Kim <namhyung@kernel.org> Subject: Re: Capturing User-Level Function Calls/Returns Date: Wed, 15 Jul 2020 21:06:01 -0400 (EDT) [thread overview] Message-ID: <489547987.230950.1594861561764.JavaMail.zimbra@polymtl.ca> (raw) In-Reply-To: <20200715174858.4698803c@oasis.local.home> > Without recompiling, how would that be implemented? As you mentioned, this is possible when "jump patching" 5 bytes instructions. Fast tracepoints in GDB and in kprobe do it. Kprobe goes further and patches sequences of instructions (because the target instruction is less than 5 bytes) if there is no incoming branch into the middle of the sequence. You can go even further, for instance using 3 bytes jumps to a trampoline installed in alignment nops. If you combine different strategies like this, you can eventually reach almost 100% success rate for "jump patching" tracepoints. This gets quite hairy though. However, the short story is that there is currently no tool as far as I know that does that easily and reliably in user space. https://onlinelibrary.wiley.com/doi/abs/10.1002/spe.2746 https://dl.acm.org/doi/pdf/10.1145/3062341.3062344 If you can afford a more invasive tool, that requires a lot of memory and stops your application for quite some time, you can look at approaches like dyninst that decompile the binary, insert instrumentation code and reassemble the code. https://dyninst.org/ > You would need to insert a jump on top of code, and still be able to > preserve that code. What a trap does, is to insert a int3, that will > trap into the kernel, it would then emulate the code that the int3 was > on, and also call some code that can trace the current state. > > To do it in user land, you would need to find way to replace the code > at the location you want to trace, with a jump to the tracing > infrastructure, that will also be able to emulate the code that the > jump was inserted on top of. As on x86, that jump will need to be 5 > bytes long (covering 5 bytes of text to emulate), where as a int3 is a > single byte. > > Thus, you either recompile and insert nops where you want to place your > jumps, or you trap using int3 that can do the work from within the > kernel. > > -- Steve > _______________________________________________ > lttng-dev mailing list > lttng-dev@lists.lttng.org > https://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
WARNING: multiple messages have this Message-ID (diff)
From: Michel Dagenais via lttng-dev <lttng-dev@lists.lttng.org> To: Steven Rostedt <rostedt@goodmis.org> Cc: ahmadkhorrami <ahmadkhorrami@ut.ac.ir>, linux-trace-users-owner@vger.kernel.org, linux-trace-users <linux-trace-users@vger.kernel.org>, lttng-dev <lttng-dev@lists.lttng.org>, Namhyung Kim <namhyung@kernel.org> Subject: Re: [lttng-dev] Capturing User-Level Function Calls/Returns Date: Wed, 15 Jul 2020 21:06:01 -0400 (EDT) [thread overview] Message-ID: <489547987.230950.1594861561764.JavaMail.zimbra@polymtl.ca> (raw) Message-ID: <20200716010601.lV_wZOyYg_dk-JxBGFZgbszWls3xuqbLkHlqhGbu6qM@z> (raw) In-Reply-To: <20200715174858.4698803c@oasis.local.home> > Without recompiling, how would that be implemented? As you mentioned, this is possible when "jump patching" 5 bytes instructions. Fast tracepoints in GDB and in kprobe do it. Kprobe goes further and patches sequences of instructions (because the target instruction is less than 5 bytes) if there is no incoming branch into the middle of the sequence. You can go even further, for instance using 3 bytes jumps to a trampoline installed in alignment nops. If you combine different strategies like this, you can eventually reach almost 100% success rate for "jump patching" tracepoints. This gets quite hairy though. However, the short story is that there is currently no tool as far as I know that does that easily and reliably in user space. https://onlinelibrary.wiley.com/doi/abs/10.1002/spe.2746 https://dl.acm.org/doi/pdf/10.1145/3062341.3062344 If you can afford a more invasive tool, that requires a lot of memory and stops your application for quite some time, you can look at approaches like dyninst that decompile the binary, insert instrumentation code and reassemble the code. https://dyninst.org/ > You would need to insert a jump on top of code, and still be able to > preserve that code. What a trap does, is to insert a int3, that will > trap into the kernel, it would then emulate the code that the int3 was > on, and also call some code that can trace the current state. > > To do it in user land, you would need to find way to replace the code > at the location you want to trace, with a jump to the tracing > infrastructure, that will also be able to emulate the code that the > jump was inserted on top of. As on x86, that jump will need to be 5 > bytes long (covering 5 bytes of text to emulate), where as a int3 is a > single byte. > > Thus, you either recompile and insert nops where you want to place your > jumps, or you trap using int3 that can do the work from within the > kernel. > > -- Steve > _______________________________________________ > lttng-dev mailing list > lttng-dev@lists.lttng.org > https://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev _______________________________________________ lttng-dev mailing list lttng-dev@lists.lttng.org https://lists.lttng.org/cgi-bin/mailman/listinfo/lttng-dev
next prev parent reply other threads:[~2020-07-16 1:06 UTC|newest] Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top [not found] <e9c7400ff0075f3beba2863c4432a905@ut.ac.ir> 2020-07-15 18:28 ` Capturing User-Level Function Calls/Returns Steven Rostedt 2020-07-15 18:28 ` [lttng-dev] " Steven Rostedt via lttng-dev 2020-07-15 18:45 ` Mathieu Desnoyers 2020-07-15 18:45 ` [lttng-dev] " Mathieu Desnoyers via lttng-dev 2020-07-15 21:39 ` ahmadkhorrami 2020-07-15 21:39 ` [lttng-dev] " ahmadkhorrami via lttng-dev 2020-07-15 21:48 ` Steven Rostedt 2020-07-15 21:48 ` [lttng-dev] " Steven Rostedt via lttng-dev 2020-07-15 22:25 ` ahmadkhorrami 2020-07-15 22:25 ` [lttng-dev] " ahmadkhorrami via lttng-dev 2020-07-16 1:06 ` Michel Dagenais via lttng-dev [this message] 2020-07-16 1:06 ` Michel Dagenais via lttng-dev 2020-07-16 1:49 ` Frank Ch. Eigler 2020-07-16 1:49 ` Frank Ch. Eigler via lttng-dev 2020-07-16 16:26 ` ahmadkhorrami 2020-07-16 16:26 ` ahmadkhorrami via lttng-dev 2020-07-16 16:20 ` ahmadkhorrami 2020-07-16 16:20 ` ahmadkhorrami via lttng-dev 2020-07-16 16:34 ` ahmadkhorrami 2020-07-16 16:34 ` ahmadkhorrami via lttng-dev 2020-07-16 1:04 ` Namhyung Kim 2020-07-16 1:04 ` [lttng-dev] " Namhyung Kim via lttng-dev 2020-07-16 16:07 ` ahmadkhorrami 2020-07-16 16:07 ` [lttng-dev] " ahmadkhorrami via lttng-dev
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=489547987.230950.1594861561764.JavaMail.zimbra@polymtl.ca \ --to=lttng-dev@lists.lttng.org \ --cc=ahmadkhorrami@ut.ac.ir \ --cc=linux-trace-users-owner@vger.kernel.org \ --cc=linux-trace-users@vger.kernel.org \ --cc=michel.dagenais@polymtl.ca \ --cc=namhyung@kernel.org \ --cc=rostedt@goodmis.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: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).