linux-kernel - Re: [PATCH] x86: fix oops caused by old EFI info on kexec boot

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <aTQXWdVdxG6CgQoD@gmail.com>
Date: Sat, 6 Dec 2025 12:45:29 +0100
From: Ingo Molnar <mingo@...nel.org>
To: "H. Peter Anvin" <hpa@...or.com>
Cc: James Le Cuirot <chewi@...too.org>, x86@...nel.org,
	linux-kernel@...r.kernel.org, Thomas Gleixner <tglx@...utronix.de>,
	Ingo Molnar <mingo@...hat.com>, Borislav Petkov <bp@...en8.de>,
	Dave Hansen <dave.hansen@...ux.intel.com>,
	Ard Biesheuvel <ardb@...nel.org>
Subject: Re: [PATCH] x86: fix oops caused by old EFI info on kexec boot

* H. Peter Anvin <hpa@...or.com> wrote:

> --- a/arch/x86/include/uapi/asm/bootparam.h
> +++ b/arch/x86/include/uapi/asm/bootparam.h
> @@ -26,6 +26,28 @@
>  #define XLF_5LEVEL_ENABLED		(1<<6)
>  #define XLF_MEM_ENCRYPTION		(1<<7)
>
> +/* bootloader ID */
> +#define BOOTLOADER_LILO		0x00
> +#define BOOTLOADER_LOADLIN	0x01
> +/* 0x02 used by kernel internal boot sector - long since obsolete */
> +#define BOOTLOADER_SYSLINUX	0x03
> +#define BOOTLOADER_IPXE		0x04
> +#define BOOTLOADER_ELILO	0x05
> +/* 0x06 unknown user */
> +#define BOOTLOADER_GRUB		0x07
> +#define BOOTLOADER_UBOOT	0x08
> +#define BOOTLOADER_XEN		0x09
> +#define BOOTLOADER_GUJIN	0x0a
> +#define BOOTLOADER_QEMU		0x0b
> +#define BOOTLOADER_ARCTURUS	0x0c	/* "Arcturus Networks uCbootloader" */
> +#define BOOTLOADER_KEXEC_TOOLS	0x0d
> +#define BOOTLOADER_EXTENDED_ID	0x0e	/* Escape into the 0x10+ space */
> +#define BOOTLOADER_MISSING_ID	0x0f	/* Bootloader unknown */
> +#define BOOTLOADER_EXTID_BASE	0x10	/* Error: ext_loader_type never set */
> +#define BOOTLOADER_MINIMAL	0x11	/* "Minimal Linux Bootloader" */
> +#define BOOTLOADER_OVMF		0x12
> +#define BOOTLOADER_BAREBOX	0x13
> +
>  #ifndef __ASSEMBLER__
>
>  #include <linux/types.h>
> @@ -162,6 +184,28 @@ struct boot_params {
>	__u8  _pad9[276];				/* 0xeec */
>  } __attribute__((packed));
>
> +static inline unsigned int boot_loader_type(const struct boot_params *_bp)
> +{
> +	unsigned int _type = _bp->type_of_loader >> 4;
> +	if (_type == BOOTLOADER_EXTENDED_ID)
> +		_type = _bp->ext_loader_type + BOOTLOADER_EXTID_BASE
> +	return _type;
> +}
> +static inline unsigned int boot_loader_ver(const struct boot_params *_bp)
> +{
> +	return (_bp->type_of_loader & 0x0f) + (_bp->ext_loader_ver << 4);
> +}
> +static inline void set_boot_loader(struct boot_params *_bp,
> +				   unsigned int _type, unsigned int _ver)
> +{
> +	if (_type >= BOOTLOADER_EXTID_BASE) {
> +		_bp->ext_loader_type = _type - BOOTLOADER_EXTID_BASE;
> +		_type = BOOTLOADER_EXTENDED_ID;
> +	}
> +	_bp->ext_loader_ver = _ver >> 4;
> +	_bp->type_of_loader = _type + (_ver & 0x0f);
> +}

Ack, with a s/_bp/bp, s/_type/type and s/_ver/ver side note, as this
isn't a macro and thus there's no need to obfuscate the inline
function's parameter name to avoid unintended shadowing?

Thanks,

	Ingo