From: Steven Rostedt <rostedt@goodmis.org>
To: Tatsuyuki Ishi <ishitatsuyuki@gmail.com>
Cc: Peter Zijlstra <peterz@infradead.org>,
Indu Bhagat <indu.bhagat@oracle.com>,
linux-toolchains@vger.kernel.org, daandemeyer@meta.com,
andrii@kernel.org, kris.van.hees@oracle.com,
elena.zannoni@oracle.com, nick.alcock@oracle.com
Subject: Re: [POC 5/5] x86_64: invoke SFrame based stack tracer for user space
Date: Wed, 13 Mar 2024 10:52:28 -0400 [thread overview]
Message-ID: <20240313105228.2a6bbd2f@gandalf.local.home> (raw)
In-Reply-To: <4CDF4EB4-C8AF-43DB-88AB-DB4BB76995F2@gmail.com>
On Wed, 13 Mar 2024 22:37:26 +0800
Tatsuyuki Ishi <ishitatsuyuki@gmail.com> wrote:
>
> Hi, just discovered this idea while looking for options to implement
> in-kernel unwinding, and in our case, for Wine applications where
> recompiling existing proprietary applications with frame pointers is not
> feasible.
>
> I looked a bit more into how deferring work from NMI could work, but it
> looks like this might not be easy for now. For normal IRQs, we can call
> task_work_add and irqentry_exit will handle it for us. But task works are
> not triggered for NMI, and that’s likely because unmasking NMI requires
> IRET on x86, so if the NMI comes in during user mode, then we need to
> forge a kernel IRET frame to do non-NMI work.
>
> Does the forging stuff sound feasible? Suggestions are also welcome.
The simple answer is to call irq_work. That will trigger an interrupt when
irqs are enabled again (returning from NMI and just before entering user
space).
The return from the irq_work can do the proper ptrace path.
-- Steve
next prev parent reply other threads:[~2024-03-13 14:50 UTC|newest]
Thread overview: 32+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-05-01 20:04 [POC 0/5] SFrame based stack tracer for user space in the kernel Indu Bhagat
2023-05-01 20:04 ` [POC 1/5] Kconfig: x86: Add new config options for userspace unwinder Indu Bhagat
2023-05-01 20:04 ` [POC 2/5] task_struct : add additional member for sframe state Indu Bhagat
2023-05-01 20:04 ` [POC 3/5] sframe: add new SFrame library Indu Bhagat
2023-05-01 22:40 ` Steven Rostedt
2023-05-02 5:07 ` Indu Bhagat
2023-05-02 8:46 ` Peter Zijlstra
2023-05-02 9:09 ` Peter Zijlstra
2023-05-02 9:20 ` Peter Zijlstra
2023-05-02 9:28 ` Peter Zijlstra
2023-05-02 9:30 ` Peter Zijlstra
2023-05-03 6:03 ` Indu Bhagat
2023-05-02 10:31 ` Peter Zijlstra
2023-05-02 10:41 ` Peter Zijlstra
2023-05-02 15:22 ` Steven Rostedt
2023-05-01 20:04 ` [POC 4/5] sframe: add an SFrame format stack tracer Indu Bhagat
2023-05-01 23:00 ` Steven Rostedt
2023-05-02 6:16 ` Indu Bhagat
2023-05-02 8:53 ` Peter Zijlstra
2023-05-02 9:04 ` Peter Zijlstra
2023-05-01 20:04 ` [POC 5/5] x86_64: invoke SFrame based stack tracer for user space Indu Bhagat
2023-05-01 23:11 ` Steven Rostedt
2023-05-02 10:53 ` Peter Zijlstra
2023-05-02 15:27 ` Steven Rostedt
2023-05-16 17:25 ` Andrii Nakryiko
2023-05-16 17:38 ` Steven Rostedt
2023-05-16 17:51 ` Andrii Nakryiko
2024-03-13 14:37 ` Tatsuyuki Ishi
2024-03-13 14:52 ` Steven Rostedt [this message]
2024-03-13 14:58 ` Tatsuyuki Ishi
2024-03-13 15:04 ` Steven Rostedt
2023-05-01 22:15 ` [POC 0/5] SFrame based stack tracer for user space in the kernel Steven Rostedt
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=20240313105228.2a6bbd2f@gandalf.local.home \
--to=rostedt@goodmis.org \
--cc=andrii@kernel.org \
--cc=daandemeyer@meta.com \
--cc=elena.zannoni@oracle.com \
--cc=indu.bhagat@oracle.com \
--cc=ishitatsuyuki@gmail.com \
--cc=kris.van.hees@oracle.com \
--cc=linux-toolchains@vger.kernel.org \
--cc=nick.alcock@oracle.com \
--cc=peterz@infradead.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
Be 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).