linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: peterz@infradead.org
To: Jin Yao <yao.jin@linux.intel.com>
Cc: mingo@redhat.com, oleg@redhat.com, acme@kernel.org,
	jolsa@kernel.org, Linux-kernel@vger.kernel.org,
	ak@linux.intel.com, kan.liang@intel.com, yao.jin@intel.com,
	alexander.shishkin@linux.intel.com, mark.rutland@arm.com
Subject: Re: [PATCH v1 2/2] perf/core: Fake regs for leaked kernel samples
Date: Tue, 4 Aug 2020 13:49:00 +0200	[thread overview]
Message-ID: <20200804114900.GI2657@hirez.programming.kicks-ass.net> (raw)
In-Reply-To: <20200731025617.16243-2-yao.jin@linux.intel.com>

On Fri, Jul 31, 2020 at 10:56:17AM +0800, Jin Yao wrote:
> @@ -6973,7 +6973,8 @@ static struct perf_callchain_entry __empty_callchain = { .nr = 0, };
>  struct perf_callchain_entry *
>  perf_callchain(struct perf_event *event, struct pt_regs *regs)
>  {
> -	bool kernel = !event->attr.exclude_callchain_kernel;
> +	bool kernel = !event->attr.exclude_callchain_kernel &&
> +		      !event->attr.exclude_kernel;

This seems weird; how can we get there. Also it seems to me that if you
have !exclude_callchain_kernel you already have permission for kernel
bits, so who cares.

>  	bool user   = !event->attr.exclude_callchain_user;
>  	/* Disallow cross-task user callchains. */
>  	bool crosstask = event->ctx->task && event->ctx->task != current;
> @@ -6988,12 +6989,39 @@ perf_callchain(struct perf_event *event, struct pt_regs *regs)
>  	return callchain ?: &__empty_callchain;
>  }
>  
> +static struct pt_regs *leak_check(struct perf_event_header *header,
> +				  struct perf_event *event,
> +				  struct pt_regs *regs)
> +{
> +	struct pt_regs *regs_fake = NULL;
> +
> +	if (event->attr.exclude_kernel && !user_mode(regs)) {
> +		if (!(current->flags & PF_KTHREAD)) {
> +			regs_fake = task_pt_regs(current);
> +			if (!user_mode(regs_fake)) {
> +				regs_fake = NULL;
> +				instruction_pointer_set(regs, -1L);
> +			}
> +		} else
> +			instruction_pointer_set(regs, -1L);

That violates coding style, also:

		if (!(current->flags & PF_KTHREAD)) {
			regs_fake = task_pt_regs(current);
			if (!user_mode(regs_fake)) /* is this not a BUG?  */
				regs_fake = NULL;
		}

		if (!regs_fake)
			instruction_pointer_set(regs, -1L);

Seems simpler to me.

> +		if ((header->misc & PERF_RECORD_MISC_CPUMODE_MASK) ==
> +		     PERF_RECORD_MISC_KERNEL) {
> +			header->misc &= ~PERF_RECORD_MISC_CPUMODE_MASK;
> +			header->misc |= PERF_RECORD_MISC_USER;
> +		}

Why the conditional? At this point it had better be unconditionally
user, no?

		headers->misc &= ~PERF_RECORD_MISC_CPUMODE_MASK;
		headers->misc |= PERF_RECORD_MISC_USER;

> +	}
> +
> +	return regs_fake;
> +}

  reply	other threads:[~2020-08-04 11:54 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-07-31  2:56 [PATCH v1 1/2] Missing instruction_pointer_set() instances Jin Yao
2020-07-31  2:56 ` [PATCH v1 2/2] perf/core: Fake regs for leaked kernel samples Jin Yao
2020-08-04 11:49   ` peterz [this message]
2020-08-05  2:15     ` Jin, Yao
2020-08-05 12:44       ` peterz
2020-08-05 12:57         ` peterz
2020-08-06  2:26         ` Jin, Yao
2020-08-06  9:18           ` peterz
2020-08-06  9:24             ` peterz
2020-08-07  5:32               ` Jin, Yao
2020-08-06 11:00             ` peterz
2020-08-07  6:24               ` Jin, Yao
2020-08-07  9:02                 ` peterz
2020-08-10  2:03                   ` Jin, Yao
2020-08-07  5:23             ` Jin, Yao
2020-08-11  7:50           ` Jin, Yao
2020-08-11  7:59             ` Peter Zijlstra
2020-08-11  8:31               ` Jin, Yao
2020-08-11  8:45                 ` Peter Zijlstra
2020-08-12  3:52                   ` Jin, Yao
2020-08-12  7:25                     ` Like Xu
2020-08-04 11:31 ` [PATCH v1 1/2] Missing instruction_pointer_set() instances peterz
2020-08-05  0:26   ` Jin, Yao
2020-08-04 21:31 ` Max Filippov

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=20200804114900.GI2657@hirez.programming.kicks-ass.net \
    --to=peterz@infradead.org \
    --cc=Linux-kernel@vger.kernel.org \
    --cc=acme@kernel.org \
    --cc=ak@linux.intel.com \
    --cc=alexander.shishkin@linux.intel.com \
    --cc=jolsa@kernel.org \
    --cc=kan.liang@intel.com \
    --cc=mark.rutland@arm.com \
    --cc=mingo@redhat.com \
    --cc=oleg@redhat.com \
    --cc=yao.jin@intel.com \
    --cc=yao.jin@linux.intel.com \
    /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).