From: Florian Schmaus <florian.schmaus@codasip.com>
To: Paul Walmsley <pjw@kernel.org>,
Palmer Dabbelt <palmer@dabbelt.com>,
Albert Ou <aou@eecs.berkeley.edu>,
Alexandre Ghiti <alex@ghiti.fr>
Cc: Andrew Jones <ajones@ventanamicro.com>,
Samuel Holland <samuel.holland@sifive.com>,
Miaoqian Lin <linmq006@gmail.com>,
Sean Chang <seanwascoding@gmail.com>,
linux-riscv@lists.infradead.or, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] riscv: module: Use scope-based resource management for scratch buf
Date: Thu, 27 Aug 2026 13:21:45 +0200 [thread overview]
Message-ID: <babc7c19-8b33-425b-a872-1569e036c722@codasip.com> (raw)
In-Reply-To: <20260512063359.713142-1-florian.schmaus@codasip.com>
On 12/05/2026 08.33, Florian Schmaus wrote:
> The scratch buffer in module_frob_arch_sections() is currently managed
> manually, requiring explicit calls to kvfree().
>
> Convert it to use scope-based resource management via the
> __free(kvfree) attribute. This automatically guarantees that the
> buffer is freed when it goes out of scope, preventing potential memory
> leaks. Most notably, it simplifies the kvrealloc() error path by
> removing the need to manually free the original buffer before
> returning -ENOMEM.
>
> Signed-off-by: Florian Schmaus <florian.schmaus@codasip.com>
> ---
> arch/riscv/kernel/module-sections.c | 8 +++-----
> 1 file changed, 3 insertions(+), 5 deletions(-)
>
> diff --git a/arch/riscv/kernel/module-sections.c b/arch/riscv/kernel/module-sections.c
> index b3b11b7f7ed9..ed3be8be377e 100644
> --- a/arch/riscv/kernel/module-sections.c
> +++ b/arch/riscv/kernel/module-sections.c
> @@ -116,7 +116,7 @@ int module_frob_arch_sections(Elf_Ehdr *ehdr, Elf_Shdr *sechdrs,
> size_t num_scratch_relas = 0;
> unsigned int num_plts = 0;
> unsigned int num_gots = 0;
> - Elf_Rela *scratch = NULL;
> + Elf_Rela *scratch __free(kvfree) = NULL;
> Elf_Rela *new_scratch;
> size_t scratch_size = 0;
> int i;
> @@ -168,10 +168,9 @@ int module_frob_arch_sections(Elf_Ehdr *ehdr, Elf_Shdr *sechdrs,
> if (scratch_size_needed > scratch_size) {
> scratch_size = scratch_size_needed;
> new_scratch = kvrealloc(scratch, scratch_size, GFP_KERNEL);
> - if (!new_scratch) {
> - kvfree(scratch);
> + if (!new_scratch)
> return -ENOMEM;
> - }
> +
> scratch = new_scratch;
> }
>
> @@ -184,7 +183,6 @@ int module_frob_arch_sections(Elf_Ehdr *ehdr, Elf_Shdr *sechdrs,
> /* sort the accumulated PLT/GOT relocations so duplicates are adjacent */
> sort(scratch, num_scratch_relas, sizeof(*scratch), cmp_rela, NULL);
> count_max_entries(scratch, num_scratch_relas, &num_plts, &num_gots);
> - kvfree(scratch);
> }
>
> mod->arch.plt.shdr->sh_type = SHT_NOBITS;
Friendly ping. Have the RISC-V maintainer had a chance to look at the
patch? It's a minor cleanup.
Thanks!
Florian
prev parent reply other threads:[~2026-08-27 11:21 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-05-12 6:33 Florian Schmaus
2026-08-27 11:21 ` Florian Schmaus [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=babc7c19-8b33-425b-a872-1569e036c722@codasip.com \
--to=florian.schmaus@codasip.com \
--cc=ajones@ventanamicro.com \
--cc=alex@ghiti.fr \
--cc=aou@eecs.berkeley.edu \
--cc=linmq006@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-riscv@lists.infradead.or \
--cc=palmer@dabbelt.com \
--cc=pjw@kernel.org \
--cc=samuel.holland@sifive.com \
--cc=seanwascoding@gmail.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
all inboxes | Powered by JetHome®