linux-perf-users.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Namhyung Kim <namhyung@kernel.org>
To: Zecheng Li <zecheng@google.com>
Cc: Peter Zijlstra <peterz@infradead.org>,
	Ingo Molnar <mingo@redhat.com>,
	Arnaldo Carvalho de Melo <acme@kernel.org>,
	Mark Rutland <mark.rutland@arm.com>,
	Alexander Shishkin <alexander.shishkin@linux.intel.com>,
	Jiri Olsa <jolsa@kernel.org>, Ian Rogers <irogers@google.com>,
	Adrian Hunter <adrian.hunter@intel.com>,
	"Liang, Kan" <kan.liang@linux.intel.com>,
	Masami Hiramatsu <mhiramat@kernel.org>,
	Zecheng Li <zli94@ncsu.edu>, Xu Liu <xliuprof@google.com>,
	linux-perf-users@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH v1 5/6] perf dwarf-aux: Find pointer type to a type
Date: Sat, 26 Jul 2025 17:24:37 -0700	[thread overview]
Message-ID: <aIVxxfiHH0JTQkpr@google.com> (raw)
In-Reply-To: <20250725202809.1230085-6-zecheng@google.com>

On Fri, Jul 25, 2025 at 08:28:08PM +0000, Zecheng Li wrote:
> Implement die_find_pointer_to_type that searches for the pointer type to
> a given type in the CU. There's no guarantee that a pointer DIE exists
> for every possible base type. Compilers only generate DIEs for types
> actually used or defined.

Please describe why it's necessary first.  What happens if it fails to
fine one?

Thanks,
Namhyung

> 
> The performance could be improved by adding a cache for the pointer
> types. Currently its impact on the annotation time for vmlinux is low.
> 
> Signed-off-by: Zecheng Li <zecheng@google.com>
> ---
>  tools/perf/util/dwarf-aux.c | 69 +++++++++++++++++++++++++++++++++++++
>  tools/perf/util/dwarf-aux.h |  4 +++
>  2 files changed, 73 insertions(+)
> 
> diff --git a/tools/perf/util/dwarf-aux.c b/tools/perf/util/dwarf-aux.c
> index 4039dbd2b8c0..49d509839a85 100644
> --- a/tools/perf/util/dwarf-aux.c
> +++ b/tools/perf/util/dwarf-aux.c
> @@ -2121,3 +2121,72 @@ Dwarf_Die *die_deref_ptr_type(Dwarf_Die *ptr_die, int offset,
>  
>  	return die_get_member_type(&type_die, offset, die_mem);
>  }
> +
> +struct find_pointer_type_data {
> +	/* DIE offset of the type we want to point to */
> +	Dwarf_Off target_type_offset;
> +	Dwarf_Die *found_die;
> +};
> +
> +static int __die_find_pointer_to_type_cb(Dwarf_Die *die_mem, void *arg)
> +{
> +	struct find_pointer_type_data *data = arg;
> +	Dwarf_Attribute type_attr;
> +	Dwarf_Die type_die;
> +	Dwarf_Off ref_type_offset;
> +
> +	if (dwarf_tag(die_mem) != DW_TAG_pointer_type)
> +		return DIE_FIND_CB_CONTINUE;
> +
> +	if (!dwarf_attr(die_mem, DW_AT_type, &type_attr))
> +		return DIE_FIND_CB_SIBLING;
> +
> +	/* Get the DIE this pointer points to */
> +	if (!dwarf_formref_die(&type_attr, &type_die))
> +		return DIE_FIND_CB_SIBLING;
> +
> +	ref_type_offset = dwarf_dieoffset(&type_die);
> +
> +	if (ref_type_offset != 0 && ref_type_offset == data->target_type_offset) {
> +		/* This die_mem is a pointer to the target type */
> +		if (data->found_die)
> +			*data->found_die = *die_mem;
> +		return DIE_FIND_CB_END;
> +	}
> +
> +	return DIE_FIND_CB_SIBLING;
> +}
> +
> +/**
> + * die_find_pointer_to_type - Find a DIE for a pointer to a given type
> + * @cu_die: The compilation unit to search within.
> + * @target_type: The DIE of the type you want to find a pointer to.
> + * @result_die: Buffer to store the found DW_TAG_pointer_type DIE.
> + *
> + * Scans the children of the @cu_die for a DW_TAG_pointer_type DIE
> + * whose DW_AT_type attribute references the @target_type.
> + *
> + * Return: @result_die if found, NULL otherwise.
> + */
> +Dwarf_Die *die_find_pointer_to_type(Dwarf_Die *cu_die, Dwarf_Die *target_type,
> +				   Dwarf_Die *result_die)
> +{
> +	struct find_pointer_type_data data;
> +	Dwarf_Die search_mem;
> +
> +	if (!cu_die || !target_type || !result_die)
> +		return NULL;
> +
> +	data.target_type_offset = dwarf_dieoffset(target_type);
> +	if (data.target_type_offset == 0) {
> +		pr_debug("Target type DIE has no offset\n");
> +		return NULL;
> +	}
> +	data.found_die = result_die;
> +
> +	if (die_find_child(cu_die, __die_find_pointer_to_type_cb, &data, &search_mem))
> +		return result_die;
> +
> +	return NULL;
> +}
> +
> diff --git a/tools/perf/util/dwarf-aux.h b/tools/perf/util/dwarf-aux.h
> index 892c8c5c23fc..7e1336ff276e 100644
> --- a/tools/perf/util/dwarf-aux.h
> +++ b/tools/perf/util/dwarf-aux.h
> @@ -156,6 +156,10 @@ Dwarf_Die *die_get_member_type(Dwarf_Die *type_die, int offset, Dwarf_Die *die_m
>  /* Return type info where the pointer and offset point to */
>  Dwarf_Die *die_deref_ptr_type(Dwarf_Die *ptr_die, int offset, Dwarf_Die *die_mem);
>  
> +/* Find a DIE for a pointer to a given type */
> +Dwarf_Die *die_find_pointer_to_type(Dwarf_Die *cu_die, Dwarf_Die *target_type,
> +				   Dwarf_Die *result_die);
> +
>  /* Get byte offset range of given variable DIE */
>  int die_get_var_range(Dwarf_Die *sp_die, Dwarf_Die *vr_die, struct strbuf *buf);
>  
> -- 
> 2.50.1.470.g6ba607880d-goog
> 

  reply	other threads:[~2025-07-27  0:24 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-07-25 20:28 [PATCH v1 0/6] perf tools: Some improvements on data type profiler Zecheng Li
2025-07-25 20:28 ` [PATCH v1 1/6] perf dwarf-aux: Use signed comparison in match_var_offset Zecheng Li
2025-07-26  0:58   ` Ian Rogers
2025-07-26 23:56     ` Namhyung Kim
2025-07-25 20:28 ` [PATCH v1 2/6] perf dwarf-aux: More accurate variable type match for breg Zecheng Li
2025-07-27  0:14   ` Namhyung Kim
2025-07-25 20:28 ` [PATCH v1 3/6] perf dwarf-aux: Better type matching for stack variables Zecheng Li
2025-07-26  1:17   ` Ian Rogers
2025-07-27  0:22     ` Namhyung Kim
2025-07-25 20:28 ` [PATCH v1 4/6] perf annotate: Skip annotating data types to lea instructions Zecheng Li
2025-07-26  1:19   ` Ian Rogers
2025-07-25 20:28 ` [PATCH v1 5/6] perf dwarf-aux: Find pointer type to a type Zecheng Li
2025-07-27  0:24   ` Namhyung Kim [this message]
2025-07-25 20:28 ` [PATCH v1 6/6] perf annotate: Track arithmetic instructions on pointers Zecheng Li
2025-07-27  0:32   ` Namhyung Kim

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=aIVxxfiHH0JTQkpr@google.com \
    --to=namhyung@kernel.org \
    --cc=acme@kernel.org \
    --cc=adrian.hunter@intel.com \
    --cc=alexander.shishkin@linux.intel.com \
    --cc=irogers@google.com \
    --cc=jolsa@kernel.org \
    --cc=kan.liang@linux.intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-perf-users@vger.kernel.org \
    --cc=mark.rutland@arm.com \
    --cc=mhiramat@kernel.org \
    --cc=mingo@redhat.com \
    --cc=peterz@infradead.org \
    --cc=xliuprof@google.com \
    --cc=zecheng@google.com \
    --cc=zli94@ncsu.edu \
    /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).