linux-kernel - Re: [PATCH v3 4/4] x86/syscall: use int everywhere for system call numbers

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

Message-ID: <f2e4c3c3-08c3-eae1-803a-aa85d7e75ca0@kernel.org>
Date:   Sat, 15 May 2021 08:37:12 -0700
From:   Andy Lutomirski <luto@...nel.org>
To:     "H. Peter Anvin" <hpa@...or.com>, Ingo Molnar <mingo@...hat.com>,
        Thomas Gleixner <tglx@...utronix.de>,
        Borislav Petkov <bp@...en8.de>
Cc:     Linux Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v3 4/4] x86/syscall: use int everywhere for system call
 numbers

On 5/14/21 6:10 PM, H. Peter Anvin wrote:
> From: "H. Peter Anvin (Intel)" <hpa@...or.com>
> 
> System call numbers are defined as int, so use int everywhere for
> system call numbers. This patch is strictly a cleanup; it should not
> change anything user visible; all ABI changes have been done in the
> preceeding patches.
> 
> Signed-off-by: H. Peter Anvin (Intel) <hpa@...or.com>
> ---
>  arch/x86/entry/common.c        | 93 ++++++++++++++++++++++++----------
>  arch/x86/include/asm/syscall.h |  2 +-
>  2 files changed, 66 insertions(+), 29 deletions(-)
> 
> diff --git a/arch/x86/entry/common.c b/arch/x86/entry/common.c
> index f51bc17262db..714804f0970c 100644
> --- a/arch/x86/entry/common.c
> +++ b/arch/x86/entry/common.c
> @@ -36,49 +36,87 @@
>  #include <asm/irq_stack.h>
>  
>  #ifdef CONFIG_X86_64
> -__visible noinstr void do_syscall_64(struct pt_regs *regs, unsigned long nr)
> +
> +static __always_inline bool do_syscall_x64(struct pt_regs *regs, int nr)
> +{
> +	/*
> +	 * Convert negative numbers to very high and thus out of range
> +	 * numbers for comparisons. Use unsigned long to slightly
> +	 * improve the array_index_nospec() generated code.
> +	 */
> +	unsigned long unr = nr;
> +
> +	if (likely(unr < NR_syscalls)) {
> +		unr = array_index_nospec(unr, NR_syscalls);
> +		regs->ax = sys_call_table[unr](regs);
> +		return true;
> +	}
> +	return false;
> +}

How much do you like micro-optimization?  You could be silly^Wclever and
add a new syscall handler:

long skip_syscall(struct pt_regs *regs)
{
	return regs->ax;
}

and prepend this to the syscall tables -- it would be a sort-of-real
syscall -1.  Then the call sequence becomes:

int adjusted_nr = nr + 1 (or nr - x32bit + 1);

if (likely(nr < NR_adjusted_syscalls)) {
   unr = array_index_nospec...;
   regs->ax = sys_call_table[unr](regs);  /* might be a no-op! */
} else {
    regs->ax = -ENOSYS;
}

which removes a branch from the fast path.