mail archive of the barebox mailing list
 help / color / mirror / Atom feed
From: Ahmad Fatoum <a.fatoum@pengutronix.de>
To: Sascha Hauer <s.hauer@pengutronix.de>,
	BAREBOX <barebox@lists.infradead.org>
Cc: "Claude Sonnet 4.5" <noreply@anthropic.com>
Subject: Re: [PATCH v2 03/21] elf: add dynamic relocation support
Date: Tue, 6 Jan 2026 14:51:07 +0100	[thread overview]
Message-ID: <14aa93ff-c0e8-4604-9afb-1c74a570ee6d@pengutronix.de> (raw)
In-Reply-To: <20260106-pbl-load-elf-v2-3-487bc760f045@pengutronix.de>

Hi,

On 1/6/26 1:53 PM, Sascha Hauer wrote:
> Add support for applying dynamic relocations to ELF binaries. This allows
> loading ET_DYN (position-independent) binaries and ET_EXEC binaries at
> custom load addresses.
> 
> Key changes:
> - Add elf_image.reloc_offset to track offset between vaddr and load address
> - Implement elf_compute_load_offset() to calculate relocation offset
> - Add elf_set_load_address() API to specify custom load address
> - Implement elf_find_dynamic_segment() to locate PT_DYNAMIC
> - Add elf_relocate() to apply relocations
> - Provide weak default elf_apply_relocations() stub for unsupported architectures
> - Add ELF dynamic section accessors
> 
> The relocation offset type is unsigned long to properly handle pointer
> arithmetic and avoid casting issues.
> 
> Architecture-specific implementations should override the weak
> elf_apply_relocations() function to handle their relocation types.
> 
> Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
> Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

Wrong way round ^

Reviewed-by: Ahmad Fatoum <a.fatoum@pengutronix.de>

But some nitpicks and a question below.

> +static void *elf_get_dest(struct elf_image *elf, void *phdr)

How about elf_phdr_relocated_paddr as a more descriptive name?

> +	if (elf->load_address) {
> +		elf->base_load_addr = elf->load_address;
> +	} else if (elf->type == ET_EXEC) {
> +		elf->base_load_addr = NULL;
> +	} else {
> +		elf->base_load_addr = (void *)(phys_addr_t)min_paddr;
> +	}

Curly braces can be removed.

> +	if (elf->type == ET_EXEC && !elf->load_address) {
> +		elf->reloc_offset = 0;
> +	} else {
> +		elf->reloc_offset = ((unsigned long)elf->base_load_addr - min_vaddr);
> +	}

Ditto.


> +void elf_set_load_address(struct elf_image *elf, void *addr)
> +{
> +	elf->load_address = addr;
> +}
> +
> +static void *elf_find_dynamic_segment(struct elf_image *elf)
> +{
> +	void *buf = elf->hdr_buf;
> +	void *phdr = buf + elf_hdr_e_phoff(elf, buf);
> +	int i;
> +
> +	for (i = 0; i < elf_hdr_e_phnum(elf, buf); i++) {
> +		if (elf_phdr_p_type(elf, phdr) == PT_DYNAMIC) {
> +			u64 offset = elf_phdr_p_offset(elf, phdr);
> +
> +			/* If loaded from file, PT_DYNAMIC might not be in hdr_buf */
> +			if (elf->filename)
> +				return elf_get_dest(elf, phdr);
> +			else
> +				/* Binary in memory, use offset */
> +				return elf->hdr_buf + offset;

Does elf_get_dest(elf, phdr) not compute the same address that we would
get here?

> +	/* Check that we found exactly one relocation type */
> +	if (rel && rela) {
> +		pr_err("ELF has both REL and RELA relocations\n");
> +		return -EINVAL;
> +	}

Future work could be checking these things at compile time and skipping
the checks in PBL.

> diff --git a/lib/Makefile b/lib/Makefile
> index 6d259dd94e163336d7fdab38c7b74b301aabc5c5..da2c8ffe1dbf512a901295c89494e0837f31a0d9 100644
> --- a/lib/Makefile
> +++ b/lib/Makefile
> @@ -24,6 +24,7 @@ obj-y			+= readkey.o
>  obj-y			+= kfifo.o
>  obj-y			+= libbb.o
>  obj-y			+= libgen.o
> +obj-y			+= elf_reloc.o

Why is this not obj-pbl-y?

Cheers,
Ahmad

-- 
Pengutronix e.K.                  |                             |
Steuerwalder Str. 21              | http://www.pengutronix.de/  |
31137 Hildesheim, Germany         | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686  | Fax:   +49-5121-206917-5555 |




  reply	other threads:[~2026-01-06 13:51 UTC|newest]

Thread overview: 46+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-01-06 12:53 [PATCH v2 00/21] PBL: Add PBL ELF loading support with dynamic relocations Sascha Hauer
2026-01-06 12:53 ` [PATCH v2 01/21] elf: only accept images matching the native ELF_CLASS Sascha Hauer
2026-01-06 12:58   ` Ahmad Fatoum
2026-01-06 12:53 ` [PATCH v2 02/21] elf: build for PBL as well Sascha Hauer
2026-01-06 13:26   ` Ahmad Fatoum
2026-01-06 12:53 ` [PATCH v2 03/21] elf: add dynamic relocation support Sascha Hauer
2026-01-06 13:51   ` Ahmad Fatoum [this message]
2026-01-06 12:53 ` [PATCH v2 04/21] ARM: implement elf_apply_relocations() for ELF " Sascha Hauer
2026-01-06 13:07   ` Ahmad Fatoum
2026-01-06 14:25     ` Ahmad Fatoum
2026-01-06 12:53 ` [PATCH v2 05/21] riscv: define generic relocate_image Sascha Hauer
2026-01-06 13:10   ` Ahmad Fatoum
2026-01-06 13:11     ` Sascha Hauer
2026-01-06 12:53 ` [PATCH v2 06/21] riscv: implement elf_apply_relocations() for ELF relocation support Sascha Hauer
2026-01-06 13:11   ` Ahmad Fatoum
2026-01-06 12:53 ` [PATCH v2 07/21] elf: implement elf_load_inplace() Sascha Hauer
2026-01-06 13:53   ` Ahmad Fatoum
2026-01-06 12:53 ` [PATCH v2 08/21] elf: create elf_open_binary_into() Sascha Hauer
2026-01-06 13:55   ` Ahmad Fatoum
2026-01-06 12:53 ` [PATCH v2 09/21] Makefile: add barebox.elf build target Sascha Hauer
2026-01-06 13:13   ` Ahmad Fatoum
2026-01-06 12:53 ` [PATCH v2 10/21] PBL: allow to link ELF image into PBL Sascha Hauer
2026-01-06 13:18   ` Ahmad Fatoum
2026-01-06 12:53 ` [PATCH v2 11/21] mmu: add MAP_CACHED_RO mapping type Sascha Hauer
2026-01-06 13:14   ` Ahmad Fatoum
2026-01-06 12:53 ` [PATCH v2 12/21] mmu: introduce pbl_remap_range() Sascha Hauer
2026-01-06 13:15   ` Ahmad Fatoum
2026-01-06 12:53 ` [PATCH v2 13/21] ARM: use relative jumps in exception table Sascha Hauer
2026-01-06 13:57   ` Ahmad Fatoum
2026-01-06 12:53 ` [PATCH v2 14/21] ARM: exceptions: make in-binary exception table const Sascha Hauer
2026-01-06 14:00   ` Ahmad Fatoum
2026-01-06 12:53 ` [PATCH v2 15/21] ARM: linker script: create separate PT_LOAD segments for text, rodata, and data Sascha Hauer
2026-01-06 14:05   ` Ahmad Fatoum
2026-01-06 12:53 ` [PATCH v2 16/21] ARM: link ELF image into PBL Sascha Hauer
2026-01-06 14:06   ` Ahmad Fatoum
2026-01-06 12:53 ` [PATCH v2 17/21] ARM: PBL: setup MMU with proper permissions from ELF segments Sascha Hauer
2026-01-06 14:10   ` Ahmad Fatoum
2026-01-06 12:53 ` [PATCH v2 18/21] riscv: link ELF image into PBL Sascha Hauer
2026-01-06 14:11   ` Ahmad Fatoum
2026-01-06 12:53 ` [PATCH v2 19/21] riscv: linker script: create separate PT_LOAD segments for text, rodata, and data Sascha Hauer
2026-01-06 14:12   ` Ahmad Fatoum
2026-01-08 15:25     ` Sascha Hauer
2026-01-06 12:53 ` [PATCH v2 20/21] riscv: Allwinner D1: Drop M-Mode Sascha Hauer
2026-01-06 13:11   ` Ahmad Fatoum
2026-01-06 12:53 ` [PATCH v2 21/21] riscv: add ELF segment-based memory protection with MMU Sascha Hauer
2026-01-06 14:20   ` Ahmad Fatoum

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=14aa93ff-c0e8-4604-9afb-1c74a570ee6d@pengutronix.de \
    --to=a.fatoum@pengutronix.de \
    --cc=barebox@lists.infradead.org \
    --cc=noreply@anthropic.com \
    --cc=s.hauer@pengutronix.de \
    /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