linux-kernel - Re: [patch 5/5] arm64: compat: vdso: Use legacy syscalls as fallback

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: <68c3f5ca-6c38-4a82-4586-6b33fd54390a@arm.com>
Date:   Sun, 28 Jul 2019 16:35:44 +0100
From:   Vincenzo Frascino <vincenzo.frascino@....com>
To:     Thomas Gleixner <tglx@...utronix.de>,
        LKML <linux-kernel@...r.kernel.org>
Cc:     x86@...nel.org, Andy Lutomirski <luto@...nel.org>,
        Sean Christopherson <sean.j.christopherson@...el.com>,
        Kees Cook <keescook@...omium.org>,
        Paul Bolle <pebolle@...cali.nl>, Will Deacon <will@...nel.org>
Subject: Re: [patch 5/5] arm64: compat: vdso: Use legacy syscalls as fallback

On 7/28/19 2:12 PM, Thomas Gleixner wrote:
> The generic VDSO implementation uses the Y2038 safe clock_gettime64() and
> clock_getres_time64() syscalls as fallback for 32bit VDSO. This breaks
> seccomp setups because these syscalls might be not (yet) allowed.
> 
> Implement the 32bit variants which use the legacy syscalls and select the
> variant in the core library.
> 
> The 64bit time variants are not removed because they are required for the
> time64 based vdso accessors.
>

Reviewed-by: Vincenzo Frascino <vincenzo.frascino@....com>
Tested-by: Vincenzo Frascino <vincenzo.frascino@....com>

> Reported-by: Sean Christopherson <sean.j.christopherson@...el.com>
> Reported-by: Paul Bolle <pebolle@...cali.nl>
> Suggested-by: Andy Lutomirski <luto@...nel.org>
> Fixes: 00b26474c2f1 ("lib/vdso: Provide generic VDSO implementation")
> Signed-off-by: Thomas Gleixner <tglx@...utronix.de>
> ---
> Note:
> 
> The NULL pointer check in the getres/getres32() fallbacks is just wrong.
> The proper return code for a NULL pointer is -EFAULT. How did this ever
> pass any posix test? Also it just should go away because any other invalid
> pointer will be caught in the syscall anyway. The clockid check is also
> part of the syscall so no point in having it here again. Handling calls
> with invalid arguments is not really a performance critical operation.
> 
> ---
>  arch/arm64/include/asm/vdso/compat_gettimeofday.h |   40 ++++++++++++++++++++++
>  1 file changed, 40 insertions(+)
> 
> --- a/arch/arm64/include/asm/vdso/compat_gettimeofday.h
> +++ b/arch/arm64/include/asm/vdso/compat_gettimeofday.h
> @@ -16,6 +16,8 @@
>  
>  #define VDSO_HAS_CLOCK_GETRES		1
>  
> +#define VDSO_HAS_32BIT_FALLBACK		1
> +
>  static __always_inline
>  int gettimeofday_fallback(struct __kernel_old_timeval *_tv,
>  			  struct timezone *_tz)
> @@ -52,6 +54,23 @@ long clock_gettime_fallback(clockid_t _c
>  }
>  
>  static __always_inline
> +long clock_gettime32_fallback(clockid_t _clkid, struct old_timespec32 *_ts)
> +{
> +	register struct old_timespec32 *ts asm("r1") = _ts;
> +	register clockid_t clkid asm("r0") = _clkid;
> +	register long ret asm ("r0");
> +	register long nr asm("r7") = __NR_compat_clock_gettime;
> +
> +	asm volatile(
> +	"	swi #0\n"
> +	: "=r" (ret)
> +	: "r" (clkid), "r" (ts), "r" (nr)
> +	: "memory");
> +
> +	return ret;
> +}
> +
> +static __always_inline
>  int clock_getres_fallback(clockid_t _clkid, struct __kernel_timespec *_ts)
>  {
>  	register struct __kernel_timespec *ts asm("r1") = _ts;
> @@ -61,6 +80,27 @@ int clock_getres_fallback(clockid_t _clk
>  
>  	/* The checks below are required for ABI consistency with arm */
>  	if ((_clkid >= MAX_CLOCKS) && (_ts == NULL))
> +		return -EINVAL;
> +
> +	asm volatile(
> +	"       swi #0\n"
> +	: "=r" (ret)
> +	: "r" (clkid), "r" (ts), "r" (nr)
> +	: "memory");
> +
> +	return ret;
> +}
> +
> +static __always_inline
> +int clock_getres32_fallback(clockid_t _clkid, struct old_timespec32 *_ts)
> +{
> +	register struct old_timespec32 *ts asm("r1") = _ts;
> +	register clockid_t clkid asm("r0") = _clkid;
> +	register long ret asm ("r0");
> +	register long nr asm("r7") = __NR_compat_clock_getres;
> +
> +	/* The checks below are required for ABI consistency with arm */
> +	if ((_clkid >= MAX_CLOCKS) && (_ts == NULL))
>  		return -EINVAL;
>  
>  	asm volatile(
> 
> 

-- 
Regards,
Vincenzo