From: Arnaldo Carvalho de Melo <acme@kernel.org>
To: Alan Maguire <alan.maguire@oracle.com>
Cc: Jiri Olsa <jolsa@kernel.org>,
Clark Williams <williams@redhat.com>,
dwarves@vger.kernel.org,
Arnaldo Carvalho de Melo <acme@redhat.com>
Subject: Re: [PATCH 3/3] dwarf_loader: Allow forcing the merge of CUs for solving inter CU tag references
Date: Thu, 18 Jun 2026 10:57:30 -0300 [thread overview]
Message-ID: <ajP5Sqj19vwNzayw@x1> (raw)
In-Reply-To: <9bc30199-be51-4825-82de-4ac28a1f9e97@oracle.com>
On Mon, Mar 30, 2026 at 09:58:18AM +0100, Alan Maguire wrote:
> On 23/03/2026 21:15, Arnaldo Carvalho de Melo wrote:
> > From: Arnaldo Carvalho de Melo <acme@redhat.com>
> >
> > The Linux perf tool now includes some Rust code that then gets linked
> > into perf and comes with its DWARF that has tags referencing tags in
> > different CUs, and as the current DWARF loading algorithm uses
> > parallelization and recodes the big DWARF types (DWARF_off, usually
> > 64-bit) into smaller ones as a step into converting to CTF (initially)
> > and later BTF, the resolution fails.
> >
> > There is a case whe this inter CU happens, LTO builds, and so there is
> > an alternative algorithm for that case, that serializes DWARF CU loading
> > and merges all the CUs into just one meta/mega-CU, which then has all
> > the types and thus doesn't have a problem with inter CU references, as
> > the recoding into smaller ids is done only after all CUs are loaded.
> >
> > So while we don't refactor the loading in a way that allows for inter CU
> > while allowing parallelization, maybe by doing the recoding just at the
> > end of parallel loading, add minimal code to force this CU merging for
> > experimentation in such cases, getting back the regression test
> > prettify_perf.data.sh to work, making it force CU merging.
> >
> > $ pahole ~/bin/perf > unmerged.txt
> > <Suppress lots of warnings when recoding DWARF types.>
> > $ pahole --force_cu_merging ~/bin/perf > merged.txt
> > $
> >
>
> Is there then a case for adding this as a pahole flag automatically if we are doing
> an LTO build? If so, it might make sense to rework this into a btf_feature since they
I see, the build asks for it since it knows it is doing an LTO build, so
instead of asking for --force_cu_merging, it adds the btf_features
request, that if unknown, doesn't cause any problems, the
--force_cu_merging cause problems, I'll add the force_cu_merging btf
feature.
At some point we can rename --btf_features to --features, while leaving
--btf_features to not force users to change their scripts, but allowing
new usage to use the more fitting --features name.
- Arnaldo
> have a better compatibility story; if --btf_features=force_cu_merging is unknown, pahole
> encoding will continue. I realize it's not strictly a BTF feature but given that
> defining it as such will reduce pahole compatibility pain it might be worth doing it
> that way.
>
> > With the current set of Rust types that are representable with the
> > pahole data structures and then pretty printed as if they were C we see
> > 12 differences:
> >
> > $ diff -u unmerged.txt merged.txt | grep ^@@ | wc -l
> > 12
> > $ diff -u unmerged.txt merged.txt | wc -l
> > 198
> >
> > Of this kind, due to some types not being resolved as tags are
> > referencing tags in other CUs.
> >
> > $ diff -u unmerged.txt merged.txt | head
> > --- unmerged.txt 2026-03-23 17:56:54.971785023 -0300
> > +++ merged.txt 2026-03-23 17:56:59.826872178 -0300
> > @@ -9643,10 +9643,11 @@
> > u64 __0 __attribute__((__aligned__(8))); /* 0 8 */
> > struct Abbreviation __1 __attribute__((__aligned__(8))); /* 8 112 */
> >
> > - /* XXX last struct has 5 bytes of padding */
> > + /* XXX last struct has 16 bytes of padding, 1 hole */
> >
> > /* size: 120, cachelines: 2, members: 2 */
> > $
> >
> > Now the pretty printing perf.data test case passes:
> >
> > ⬢ [acme@toolbx tests]$ ./prettify_perf.data.sh
> > Pretty printing of files using DWARF type information.
> > Test ./prettify_perf.data.sh passed
> > ⬢ [acme@toolbx tests]$
> >
> > Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
> > ---
> > dwarf_loader.c | 2 +-
> > dwarves.h | 1 +
> > man-pages/pahole.1 | 12 ++++++++++++
> > pahole.c | 8 ++++++++
> > tests/prettify_perf.data.sh | 4 ++--
> > 5 files changed, 24 insertions(+), 3 deletions(-)
> >
> > diff --git a/dwarf_loader.c b/dwarf_loader.c
> > index b5a92160ecf82f74..de2e9b70c32f85de 100644
> > --- a/dwarf_loader.c
> > +++ b/dwarf_loader.c
> > @@ -3967,7 +3967,7 @@ static int cus__load_module(struct cus *cus, struct conf_load *conf,
> > }
> > }
> >
> > - if (cus__merging_cu(dw, elf)) {
> > + if (conf->force_cu_merging || cus__merging_cu(dw, elf)) {
> > res = cus__merge_and_process_cu(cus, conf, mod, dw, elf, filename,
> > build_id, build_id_len,
> > type_cu ? &type_dcu : NULL);
> > diff --git a/dwarves.h b/dwarves.h
> > index 95d84b8ce3a6e95d..7887af93693ebad5 100644
> > --- a/dwarves.h
> > +++ b/dwarves.h
> > @@ -102,6 +102,7 @@ struct conf_load {
> > bool btf_gen_distilled_base;
> > bool btf_attributes;
> > bool true_signature;
> > + bool force_cu_merging;
> > uint8_t hashtable_bits;
> > uint8_t max_hashtable_bits;
> > uint16_t kabi_prefix_len;
> > diff --git a/man-pages/pahole.1 b/man-pages/pahole.1
> > index 90a8f4566de621d3..39bb53816f4fac9f 100644
> > --- a/man-pages/pahole.1
> > +++ b/man-pages/pahole.1
> > @@ -515,6 +515,18 @@ This is useful for scripts where it provides a way to ask for that exclusion
> > for pahole and pfunct, no need to use --lang_exclude in all calls to those
> > tools, just set that environment variable.
> >
> > +.TP
> > +.B \-\-force_cu_merging
> > +Force merging all CUs into one. Use when there are references across CUs.
> > +
> > +This happens in some LTO cases and was observed with Rust CUs, where types
> > +of tags (function parameters, abstract origins for inlines, etc) reference
> > +types in another CU.
> > +
> > +For LTO this is being autodetected and the merging of cus is done
> > +automatically, but for the Rust case, and maybe others this is needed with the
> > +current DWARF loading algorithm.
> > +
> > .TP
> > .B \-y, \-\-prefix_filter=PREFIX
> > Include PREFIXed classes.
> > diff --git a/pahole.c b/pahole.c
> > index e4bfb69de56ada59..05e61b61dddad8ea 100644
> > --- a/pahole.c
> > +++ b/pahole.c
> > @@ -1153,6 +1153,7 @@ ARGP_PROGRAM_VERSION_HOOK_DEF = dwarves_print_version;
> > #define ARG_padding 348
> > #define ARGP_with_embedded_flexible_array 349
> > #define ARGP_btf_attributes 350
> > +#define ARGP_force_cu_merging 351
> >
> > /* --btf_features=feature1[,feature2,..] allows us to specify
> > * a list of requested BTF features or "default" to enable all default
> > @@ -1818,6 +1819,11 @@ static const struct argp_option pahole__options[] = {
> > .key = ARGP_btf_attributes,
> > .doc = "Allow generation of attributes in BTF. Attributes are the type tags and decl tags with the kind_flag set to 1.",
> > },
> > + {
> > + .name = "force_cu_merging",
> > + .key = ARGP_force_cu_merging,
> > + .doc = "Force merging all CUs into one. Use when there are references across CUs.",
> > + },
> > {
> > .name = NULL,
> > }
> > @@ -2014,6 +2020,8 @@ static error_t pahole__options_parser(int key, char *arg,
> > parse_btf_features(arg, true); break;
> > case ARGP_btf_attributes:
> > conf_load.btf_attributes = true; break;
> > + case ARGP_force_cu_merging:
> > + conf_load.force_cu_merging = true; break;
> > default:
> > return ARGP_ERR_UNKNOWN;
> > }
> > diff --git a/tests/prettify_perf.data.sh b/tests/prettify_perf.data.sh
> > index 1fae95154d710aae..3b903e32da24b489 100755
> > --- a/tests/prettify_perf.data.sh
> > +++ b/tests/prettify_perf.data.sh
> > @@ -25,7 +25,7 @@ fi
> > perf_lacks_type_info() {
> > local type_keyword=$1
> > local type_name=$2
> > - if ! pahole -C $type_name $perf | grep -q "^$type_keyword $type_name {"; then
> > + if ! pahole --force_cu_merging -C $type_name $perf | grep -q "^$type_keyword $type_name {"; then
> > info_log "skip: $perf doesn't have '$type_keyword $type_name' type info"
> > test_skip
> > fi
> > @@ -41,7 +41,7 @@ $perf record --quiet -o $perf_data sleep 0.00001
> >
> > number_of_filtered_perf_record_metadata() {
> > local metadata_record=$1
> > - local count=$(pahole -F dwarf -V $perf --header=perf_file_header --seek_bytes '$header.data.offset' --size_bytes='$header.data.size' -C "perf_event_header(sizeof,type,type_enum=perf_event_type+perf_user_event_type,filter=type==PERF_RECORD_$metadata_record)" --prettify $perf_data | grep ".type = PERF_RECORD_$metadata_record," | wc -l)
> > + local count=$(pahole --force_cu_merging -F dwarf -V $perf --header=perf_file_header --seek_bytes '$header.data.offset' --size_bytes='$header.data.size' -C "perf_event_header(sizeof,type,type_enum=perf_event_type+perf_user_event_type,filter=type==PERF_RECORD_$metadata_record)" --prettify $perf_data | grep ".type = PERF_RECORD_$metadata_record," | wc -l)
> > echo "$count"
> > }
> >
prev parent reply other threads:[~2026-06-18 13:57 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-03-23 21:15 [PATCH 0/3] Initial support for some Rust tags + way to ask for CU merging at load time Arnaldo Carvalho de Melo
2026-03-23 21:15 ` [PATCH 1/3] dwarf_loader: Initial support for DW_TAG_variant_part Arnaldo Carvalho de Melo
2026-04-08 14:05 ` Alan Maguire
2026-04-08 17:36 ` Arnaldo Carvalho de Melo
2026-06-18 13:51 ` Arnaldo Carvalho de Melo
2026-03-23 21:15 ` [PATCH 2/3] dwarf_loader: Initial support for DW_TAG_subprogram in DW_TAG_enumeration Arnaldo Carvalho de Melo
2026-03-30 9:05 ` Alan Maguire
2026-03-30 22:39 ` Arnaldo Carvalho de Melo
2026-03-23 21:15 ` [PATCH 3/3] dwarf_loader: Allow forcing the merge of CUs for solving inter CU tag references Arnaldo Carvalho de Melo
2026-03-30 8:58 ` Alan Maguire
2026-06-18 13:57 ` Arnaldo Carvalho de Melo [this message]
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=ajP5Sqj19vwNzayw@x1 \
--to=acme@kernel.org \
--cc=acme@redhat.com \
--cc=alan.maguire@oracle.com \
--cc=dwarves@vger.kernel.org \
--cc=jolsa@kernel.org \
--cc=williams@redhat.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