From: Thomas Perale via buildroot <buildroot@buildroot.org>
To: Florian Larysch <fl@n621.de>
Cc: Thomas Perale <thomas.perale@mind.be>, buildroot@buildroot.org
Subject: Re: [Buildroot] [PATCH 2025.02.x 1/1] package/pahole: fix build on modern toolchains
Date: Fri, 24 Jul 2026 16:14:09 +0200 [thread overview]
Message-ID: <20260724141410.477344-1-thomas.perale@mind.be> (raw)
In-Reply-To: <20260714153242.20947-1-fl@n621.de>
In reply of:
> When building pahole using a modern GCC (e.g. when building
> host-pahole), discarded-qualifiers warnings are produced. In builds that
> don't set CMAKE_BUILD_TYPE=Release, -Werror gets set and causes a build
> failure.
>
> Fix this by backporting the applicable parts of an upstream patch and
> fixing a remaining issue that implicitly got refactored away upstream
> already.
>
> Signed-off-by: Florian Larysch <fl@n621.de>
Applied to 2025.02.x. Thanks
> ---
> ...scarded-qualifiers-for-strchr-strstr.patch | 134 ++++++++++++++++++
> ...reserve-const-ness-of-bsearch-result.patch | 38 +++++
> 2 files changed, 172 insertions(+)
> create mode 100644 package/pahole/0002-pahole-Fix-discarded-qualifiers-for-strchr-strstr.patch
> create mode 100644 package/pahole/0003-pahole-preserve-const-ness-of-bsearch-result.patch
>
> diff --git a/package/pahole/0002-pahole-Fix-discarded-qualifiers-for-strchr-strstr.patch b/package/pahole/0002-pahole-Fix-discarded-qualifiers-for-strchr-strstr.patch
> new file mode 100644
> index 0000000000..fc0bcc4f7c
> --- /dev/null
> +++ b/package/pahole/0002-pahole-Fix-discarded-qualifiers-for-strchr-strstr.patch
> @@ -0,0 +1,134 @@
> +From 5eeb534e02183bcaf2b86f71e0e96efa2d2ce3d0 Mon Sep 17 00:00:00 2001
> +From: Damien Hourtoulle <damien.hourtoulle@gmail.com>
> +Date: Wed, 11 Mar 2026 15:09:39 +0100
> +Subject: [PATCH] pahole: Fix discarded-qualifiers for strchr/strstr.
> +
> +Glibc 2.43 added C23 const-preserving overloads :
> +https://sourceware.org/glibc/wiki/Release/2.43.
> +
> +For the function prototype__new, fix local variable declaration to use
> +the correct const char* type instead of char*, removing the need to
> +discard the const qualifier.
> +
> +Also fix a pre-existing bug in type__find_type_enum: when iterating
> +over a '+'-concatenated list of enum names, after advancing to the next
> +segment the new separator was not nullified, causing "second+third" to
> +be passed to cu__find_enumeration_by_name instead of "second". Add a
> +null check before writing to sep_mutable to avoid a null dereference
> +when no further '+' is present.
> +
> +Signed-off-by: Damien Hourtoulle <damien.hourtoulle@gmail.com>
> +Signed-off-by: Alan Maguire <alan.maguire@oracle.com>
> +Link: https://lore.kernel.org/dwarves/20260311140938.81459-2-d.hourtoulle@geoide.fr/
> +
> +Upstream: https://git.kernel.org/pub/scm/devel/pahole/pahole.git/commit/?id=ba31cfa08ffcfe6ff5757f325de5b03ca157cb78
> +[Florian: Backport of the applicable bits to v1.27]
> +Signed-off-by: Florian Larysch <fl@n621.de>
> +---
> + pahole.c | 35 ++++++++++++++++++-----------------
> + 1 file changed, 18 insertions(+), 17 deletions(-)
> +
> +diff --git a/pahole.c b/pahole.c
> +index 954498d..3dcb525 100644
> +--- a/pahole.c
> ++++ b/pahole.c
> +@@ -2505,7 +2505,7 @@ static int64_t type_instance__int_value(struct type_instance *instance, const ch
> + int byte_offset = 0;
> +
> + if (!member) {
> +- char *sep = strchr(member_name_orig, '.');
> ++ const char *sep = strchr(member_name_orig, '.');
> +
> + if (!sep)
> + return -1;
> +@@ -2518,8 +2518,8 @@ static int64_t type_instance__int_value(struct type_instance *instance, const ch
> + char *member_name = member_name_alloc;
> + struct type *type = instance->type;
> +
> +- sep = member_name_alloc + (sep - member_name_orig);
> +- *sep = 0;
> ++ char *sep_mutable = member_name_alloc + (sep - member_name_orig); // sep mutable for the copy
> ++ *sep_mutable = 0;
> +
> + while (1) {
> + member = type__find_member_by_name(type, member_name);
> +@@ -2532,9 +2532,9 @@ out_free_member_name:
> + type = tag__type(cu__type(cu, member->tag.type));
> + if (type == NULL)
> + goto out_free_member_name;
> +- member_name = sep + 1;
> +- sep = strchr(member_name, '.');
> +- if (!sep)
> ++ member_name = sep_mutable + 1;
> ++ sep_mutable = strchr(member_name, '.');
> ++ if (!sep_mutable)
> + break;
> +
> + }
> +@@ -2972,7 +2972,7 @@ static struct prototype *prototype__new(const char *expression)
> +
> + strcpy(prototype->name, expression);
> +
> +- const char *name = prototype->name;
> ++ char *name = prototype->name;
> +
> + prototype->nr_args = 0;
> +
> +@@ -2986,10 +2986,9 @@ static struct prototype *prototype__new(const char *expression)
> + if (args_close == NULL)
> + goto out_no_closing_parens;
> +
> ++ *args_open++ = *args_close = '\0';
> + char *args = args_open;
> +
> +- *args++ = *args_close = '\0';
> +-
> + while (isspace(*args))
> + ++args;
> +
> +@@ -3136,7 +3135,7 @@ static int type__find_type_enum(struct type *type, struct cu *cu, const char *ty
> + return type__add_type_enum(type, te, cu);
> +
> + // Now look at a 'virtual enum', i.e. the concatenation of multiple enums
> +- char *sep = strchr(type_enum, '+');
> ++ const char *sep = strchr(type_enum, '+');
> +
> + if (!sep)
> + return -1;
> +@@ -3148,13 +3147,13 @@ static int type__find_type_enum(struct type *type, struct cu *cu, const char *ty
> +
> + int ret = -1;
> +
> +- sep = type_enums + (sep - type_enum);
> ++ char *sep_mutable = type_enums + (sep - type_enum);
> ++ char *cur = type_enums;
> +
> +- type_enum = type_enums;
> +- *sep = '\0';
> ++ *sep_mutable = '\0';
> +
> + while (1) {
> +- te = cu__find_enumeration_by_name(cu, type_enum, NULL);
> ++ te = cu__find_enumeration_by_name(cu, cur, NULL);
> +
> + if (!te)
> + goto out;
> +@@ -3163,10 +3162,12 @@ static int type__find_type_enum(struct type *type, struct cu *cu, const char *ty
> + if (ret)
> + goto out;
> +
> +- if (sep == NULL)
> ++ if (sep_mutable == NULL)
> + break;
> +- type_enum = sep + 1;
> +- sep = strchr(type_enum, '+');
> ++ cur = sep_mutable + 1;
> ++ sep_mutable = strchr(cur, '+');
> ++ if (sep_mutable)
> ++ *sep_mutable = '\0';
> + }
> +
> + ret = 0;
> +--
> +2.55.0
> +
> diff --git a/package/pahole/0003-pahole-preserve-const-ness-of-bsearch-result.patch b/package/pahole/0003-pahole-preserve-const-ness-of-bsearch-result.patch
> new file mode 100644
> index 0000000000..7a4cddf5d1
> --- /dev/null
> +++ b/package/pahole/0003-pahole-preserve-const-ness-of-bsearch-result.patch
> @@ -0,0 +1,38 @@
> +From f510bf5d3f3e3405710311ceb4a3539267218bde Mon Sep 17 00:00:00 2001
> +From: Florian Larysch <larysch@fixme.gmbh>
> +Date: Tue, 14 Jul 2026 16:03:12 +0200
> +Subject: [PATCH] pahole: preserve const-ness of bsearch result
> +MIME-Version: 1.0
> +Content-Type: text/plain; charset=UTF-8
> +Content-Transfer-Encoding: 8bit
> +
> +Fixes:
> +
> +pahole/btf_encoder.c: In function ‘btf_encoder__tag_kfunc’:
> +pahole/btf_encoder.c:1526:16: error: assignment discards ‘const’ qualifier from pointer target type [-Werror=discarded-qualifiers]
> + 1526 | target = bsearch(&key, base, cnt, sizeof(key), btf_func_cmp);
> + | ^
> +
> +Upstream: N/A (implicitly fixed as part of a refactor in commit c567717
> +("btf_encoder: Refactor btf_encoder__tag_kfuncs()"), released in v1.30)
> +Signed-off-by: Florian Larysch <fl@n621.de>
> +---
> + btf_encoder.c | 2 +-
> + 1 file changed, 1 insertion(+), 1 deletion(-)
> +
> +diff --git a/btf_encoder.c b/btf_encoder.c
> +index c2df2bc..4118028 100644
> +--- a/btf_encoder.c
> ++++ b/btf_encoder.c
> +@@ -1516,7 +1516,7 @@ static int btf_encoder__tag_kfunc(struct btf_encoder *encoder, struct gobuffer *
> + {
> + struct btf_func key = { .name = kfunc };
> + struct btf *btf = encoder->btf;
> +- struct btf_func *target;
> ++ const struct btf_func *target;
> + const void *base;
> + unsigned int cnt;
> + int err = -1;
> +--
> +2.55.0
> +
> --
> 2.55.0
>
> _______________________________________________
> buildroot mailing list
> buildroot@buildroot.org
> https://lists.buildroot.org/mailman/listinfo/buildroot
_______________________________________________
buildroot mailing list
buildroot@buildroot.org
https://lists.buildroot.org/mailman/listinfo/buildroot
prev parent reply other threads:[~2026-07-24 14:14 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-07-14 15:32 [Buildroot] [PATCH 2025.02.x 1/1] package/pahole: fix build on modern toolchains Florian Larysch
2026-07-24 14:14 ` Thomas Perale via buildroot [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=20260724141410.477344-1-thomas.perale@mind.be \
--to=buildroot@buildroot.org \
--cc=fl@n621.de \
--cc=thomas.perale@mind.be \
/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