linux-kernel - Re: [PATCH v2 7/9] x86/vdso: abstract out vdso system call internals

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: <c0f19f86-043b-470a-9196-06323974ac54@zytor.com>
Date: Fri, 14 Nov 2025 13:08:58 -0800
From: "H. Peter Anvin" <hpa@...or.com>
To: Uros Bizjak <ubizjak@...il.com>,
        Alexey Makhalov <alexey.makhalov@...adcom.com>
Cc: "Jason A. Donenfeld" <Jason@...c4.com>,
        "Peter Zijlstra (Intel)" <peterz@...radead.org>,
        "Theodore Ts'o" <tytso@....edu>,
        Thomas Weißschuh <thomas.weissschuh@...utronix.de>,
        Xin Li <xin@...or.com>, Andrew Cooper <andrew.cooper3@...rix.com>,
        Andy Lutomirski <luto@...nel.org>, Ard Biesheuvel <ardb@...nel.org>,
        Borislav Petkov <bp@...en8.de>, Brian Gerst <brgerst@...il.com>,
        Dave Hansen <dave.hansen@...ux.intel.com>,
        Ingo Molnar <mingo@...hat.com>, James Morse <james.morse@....com>,
        Jarkko Sakkinen <jarkko@...nel.org>,
        Josh Poimboeuf <jpoimboe@...nel.org>, Kees Cook <kees@...nel.org>,
        Nam Cao <namcao@...utronix.de>, Oleg Nesterov <oleg@...hat.com>,
        Perry Yuan <perry.yuan@....com>, Thomas Gleixner <tglx@...utronix.de>,
        Thomas Huth <thuth@...hat.com>, linux-kernel@...r.kernel.org,
        linux-mm@...ck.org, linux-sgx@...r.kernel.org, x86@...nel.org
Subject: Re: [PATCH v2 7/9] x86/vdso: abstract out vdso system call internals

On 2025-11-13 20:40, H. Peter Anvin wrote:
> 
> For 64 bits, if you need frame pointer support *and* frob %rsp, using:
> 
> xchg %[arg],%%rbp
> <stuff>
> xchg %[arg],%%rbp
> 
> ... is probably easiest, with %[arg] in a register.
> 

Incidentally, ORC is one thing, but if you need DWARF information to be
correct for user space, it is definitely "fun".  With the xchg hack for 64
bits it is just a matter of using .cfi_def_cfa_register to change the frame
pointer around, but in the 32-bit case the best I could come up with was:

	.cfi_remember_state
	push %ebp
	.cfi_escape 0x0f,3,0x74,0,0x06	// CFA address in *%esp
	mov %eax, %ebp
	mov $syscall_no, %eax
	int $0x80
	pop %ebp
	.cfi_restore_state

Note that in both cases you need to *NOT* put in any of these CFI directives
when compiling without frame pointers (in which case you need to let gcc take
care of it if you want any hope of getting the CFI correct.)

	-hpa