From: Kees Cook <kees@kernel.org>
To: Svetlana Parfenova <svetlana.parfenova@syntacore.com>
Cc: linux-fsdevel@vger.kernel.org, linux-mm@kvack.org,
linux-kernel@vger.kernel.org, viro@zeniv.linux.org.uk,
brauner@kernel.org, jack@suse.cz, akpm@linux-foundation.org,
david@redhat.com, lorenzo.stoakes@oracle.com,
Liam.Howlett@oracle.com, vbabka@suse.cz, rppt@kernel.org,
surenb@google.com, mhocko@suse.com
Subject: Re: [RFC RESEND v2] binfmt_elf: preserve original ELF e_flags for core dumps
Date: Mon, 25 Aug 2025 10:17:07 -0700 [thread overview]
Message-ID: <202508251009.CB5EB2E304@keescook> (raw)
In-Reply-To: <20250811095328.256869-1-svetlana.parfenova@syntacore.com>
On Mon, Aug 11, 2025 at 03:53:28PM +0600, Svetlana Parfenova wrote:
> Some architectures, such as RISC-V, use the ELF e_flags field to encode
> ABI-specific information (e.g., ISA extensions, fpu support). Debuggers
> like GDB rely on these flags in core dumps to correctly interpret
> optional register sets. If the flags are missing or incorrect, GDB may
> warn and ignore valid data, for example:
>
> warning: Unexpected size of section '.reg2/213' in core file.
>
> This can prevent access to fpu or other architecture-specific registers
> even when they were dumped.
>
> Save the e_flags field during ELF binary loading (in load_elf_binary())
> into the mm_struct, and later retrieve it during core dump generation
> (in fill_note_info()). A new macro ELF_CORE_USE_PROCESS_EFLAGS allows
> architectures to enable this behavior - currently just RISC-V.
>
> Signed-off-by: Svetlana Parfenova <svetlana.parfenova@syntacore.com>
> ---
> Changes in v2:
> - Remove usage of Kconfig option.
> - Add an architecture-optional macro to set process e_flags. Enabled
> by defining ELF_CORE_USE_PROCESS_EFLAGS. Defaults to no-op if not
> used.
>
> arch/riscv/include/asm/elf.h | 1 +
> fs/binfmt_elf.c | 34 ++++++++++++++++++++++++++++------
> include/linux/mm_types.h | 3 +++
> 3 files changed, 32 insertions(+), 6 deletions(-)
>
> diff --git a/arch/riscv/include/asm/elf.h b/arch/riscv/include/asm/elf.h
> index c7aea7886d22..5d9f0ac851ee 100644
> --- a/arch/riscv/include/asm/elf.h
> +++ b/arch/riscv/include/asm/elf.h
> @@ -20,6 +20,7 @@
> * These are used to set parameters in the core dumps.
> */
> #define ELF_ARCH EM_RISCV
> +#define ELF_CORE_USE_PROCESS_EFLAGS
Let's move this to the per-arch Kconfig instead, that way we can use it
in other places. Maybe call in CONFIG_ARCH_HAS_ELF_CORE_EFLAGS?
>
> #ifndef ELF_CLASS
> #ifdef CONFIG_64BIT
> diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c
> index caeddccaa1fe..e52b1e077218 100644
> --- a/fs/binfmt_elf.c
> +++ b/fs/binfmt_elf.c
> @@ -66,6 +66,14 @@
> #define elf_check_fdpic(ex) false
> #endif
>
> +#ifdef ELF_CORE_USE_PROCESS_EFLAGS
> +#define elf_coredump_get_process_eflags(dump_task, e_flags) \
> + (*(e_flags) = (dump_task)->mm->saved_e_flags)
> +#else
> +#define elf_coredump_get_process_eflags(dump_task, e_flags) \
> + do { (void)(dump_task); (void)(e_flags); } while (0)
> +#endif
Let's make specific set/get helpers here, instead.
static inline
u32 coredump_get_mm_eflags(struct mm_struct *mm, u32 flags)
{
#ifdef CONFIG_ARCH_HAS_ELF_CORE_EFLAGS
flags = mm->saved_e_flags;
#else
return flags;
}
static inline
void coredump_set_mm_eflags(struct mm_struct *mm, u32 flags)
{
#ifdef CONFIG_ARCH_HAS_ELF_CORE_EFLAGS
mm->saved_e_flags = flags;
#endif
}
> +
> static int load_elf_binary(struct linux_binprm *bprm);
>
> /*
> @@ -1290,6 +1298,9 @@ static int load_elf_binary(struct linux_binprm *bprm)
> mm->end_data = end_data;
> mm->start_stack = bprm->p;
>
> + /* stash e_flags for use in core dumps */
> + mm->saved_e_flags = elf_ex->e_flags;
Then this is:
coredump_set_mm_eflags(mm, elf_ex->e_flags);
> +
> /**
> * DOC: "brk" handling
> *
> @@ -1804,6 +1815,8 @@ static int fill_note_info(struct elfhdr *elf, int phdrs,
> struct elf_thread_core_info *t;
> struct elf_prpsinfo *psinfo;
> struct core_thread *ct;
> + u16 machine;
> + u32 flags;
>
> psinfo = kmalloc(sizeof(*psinfo), GFP_KERNEL);
> if (!psinfo)
> @@ -1831,17 +1844,26 @@ static int fill_note_info(struct elfhdr *elf, int phdrs,
> return 0;
> }
>
> - /*
> - * Initialize the ELF file header.
> - */
> - fill_elf_header(elf, phdrs,
> - view->e_machine, view->e_flags);
> + machine = view->e_machine;
> + flags = view->e_flags;
> #else
> view = NULL;
> info->thread_notes = 2;
> - fill_elf_header(elf, phdrs, ELF_ARCH, ELF_CORE_EFLAGS);
> + machine = ELF_ARCH;
> + flags = ELF_CORE_EFLAGS;
> #endif
>
> + /*
> + * Override ELF e_flags with value taken from process,
> + * if arch wants to.
> + */
> + elf_coredump_get_process_eflags(dump_task, &flags);
And this is:
flags = coredump_get_mm_eflags(dump_task->mm, flags);
> +
> + /*
> + * Initialize the ELF file header.
> + */
> + fill_elf_header(elf, phdrs, machine, flags);
> +
> /*
> * Allocate a structure for each thread.
> */
> diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
> index d6b91e8a66d6..e46f554f8d91 100644
> --- a/include/linux/mm_types.h
> +++ b/include/linux/mm_types.h
> @@ -1098,6 +1098,9 @@ struct mm_struct {
>
> unsigned long saved_auxv[AT_VECTOR_SIZE]; /* for /proc/PID/auxv */
>
And then add:
#ifdef CONFIG_ARCH_HAS_ELF_CORE_EFLAGS
> + /* the ABI-related flags from the ELF header. Used for core dump */
> + unsigned long saved_e_flags;
#endif
around this part
> +
> struct percpu_counter rss_stat[NR_MM_COUNTERS];
>
> struct linux_binfmt *binfmt;
> --
> 2.50.1
>
--
Kees Cook
next prev parent reply other threads:[~2025-08-25 17:17 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-08-06 16:18 [RFC RESEND] binfmt_elf: preserve original ELF e_flags in " Svetlana Parfenova
2025-08-06 18:57 ` Kees Cook
2025-08-07 13:13 ` Svetlana Parfenova
2025-08-07 21:14 ` Kees Cook
2025-08-08 15:54 ` Svetlana Parfenova
2025-08-08 21:54 ` Kees Cook
2025-08-11 9:53 ` [RFC RESEND v2] binfmt_elf: preserve original ELF e_flags for " Svetlana Parfenova
2025-08-25 17:17 ` Kees Cook [this message]
2025-09-01 13:58 ` Svetlana Parfenova
2025-09-01 13:53 ` [RFC RESEND v3] " Svetlana Parfenova
2025-09-04 3:52 ` Kees Cook
2025-09-04 3:52 ` Kees Cook
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=202508251009.CB5EB2E304@keescook \
--to=kees@kernel.org \
--cc=Liam.Howlett@oracle.com \
--cc=akpm@linux-foundation.org \
--cc=brauner@kernel.org \
--cc=david@redhat.com \
--cc=jack@suse.cz \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=lorenzo.stoakes@oracle.com \
--cc=mhocko@suse.com \
--cc=rppt@kernel.org \
--cc=surenb@google.com \
--cc=svetlana.parfenova@syntacore.com \
--cc=vbabka@suse.cz \
--cc=viro@zeniv.linux.org.uk \
/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