[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20220113155014.GA26100@altlinux.org>
Date: Thu, 13 Jan 2022 18:50:14 +0300
From: "Dmitry V. Levin" <ldv@...linux.org>
To: Christophe Leroy <christophe.leroy@...roup.eu>
Cc: Benjamin Herrenschmidt <benh@...nel.crashing.org>,
Paul Mackerras <paulus@...ba.org>,
Michael Ellerman <mpe@...erman.id.au>,
linux-kernel@...r.kernel.org, linuxppc-dev@...ts.ozlabs.org,
strace-devel@...ts.strace.io
Subject: Re: [PATCH] powerpc/audit: Simplify syscall_get_arch()
On Fri, Aug 20, 2021 at 09:39:14AM +0000, Christophe Leroy wrote:
> Make use of is_32bit_task() and CONFIG_CPU_LITTLE_ENDIAN
> to simplify syscall_get_arch().
>
> Signed-off-by: Christophe Leroy <christophe.leroy@...roup.eu>
> ---
> arch/powerpc/include/asm/syscall.h | 15 +++++----------
> 1 file changed, 5 insertions(+), 10 deletions(-)
Unfortunately, this commit breaks PTRACE_GET_SYSCALL_INFO and therefore
breaks strace support of AUDIT_ARCH_PPC personality on ppc64.
I've got the following report from the test farm:
=====================================================
strace 5.16: tests-m32/test-suite.log
=====================================================
# TOTAL: 1184
# PASS: 138
# SKIP: 189
# XFAIL: 0
# FAIL: 857
# XPASS: 0
# ERROR: 0
$ hostname -f
gcc203.fsffrance.org
$ uname -a
Linux gcc203 5.15.0-2-powerpc64 #1 SMP Debian 5.15.5-2 (2021-12-18) ppc64 GNU/Linux
It boils down to
$ echo 'int main(){return 0;}' |gcc -m32 -xc -
$ strace --trace=none ./a.out
syscall_0xc0(0, 0x18357, 0x1, 0x2, 0x3, 0) = 0xf7930000
syscall_0xc0(0x390000, 0x2089c0, 0x5, 0x802, 0x3, 0) = 0x390000
syscall_0xc0(0x580000, 0x20000, 0x3, 0x812, 0x3, 0x1e0) = 0x580000
+++ exited with 0 +++
$ strace -qq --signal=none --trace=ptrace strace -qq --trace=none -o/dev/null ./a.out
...
ptrace(PTRACE_GET_SYSCALL_INFO, 1234567, 88, {op=PTRACE_SYSCALL_INFO_ENTRY, arch=AUDIT_ARCH_PPC64, instruction_pointer=0xf7b34594, stack_pointer=0xffecfbf0, entry={nr=192, args=[0, 0x18357, 0x1, 0x2, 0x3, 0]}}) = 80
ptrace(PTRACE_SYSCALL, 1234567, NULL, 0) = 0
ptrace(PTRACE_GET_SYSCALL_INFO, 1234567, 88, {op=PTRACE_SYSCALL_INFO_EXIT, arch=AUDIT_ARCH_PPC64, instruction_pointer=0xf7b34594, stack_pointer=0xffecfbf0, exit={rval=4155441152, is_error=0}}) = 33
) = 0xf7af0000
As you can see, arch=AUDIT_ARCH_PPC64 is wrong here and causes all the damage.
> diff --git a/arch/powerpc/include/asm/syscall.h b/arch/powerpc/include/asm/syscall.h
> index ba0f88f3a30d..ac766037e8a1 100644
> --- a/arch/powerpc/include/asm/syscall.h
> +++ b/arch/powerpc/include/asm/syscall.h
> @@ -116,16 +116,11 @@ static inline void syscall_set_arguments(struct task_struct *task,
>
> static inline int syscall_get_arch(struct task_struct *task)
> {
> - int arch;
> -
> - if (IS_ENABLED(CONFIG_PPC64) && !test_tsk_thread_flag(task, TIF_32BIT))
> - arch = AUDIT_ARCH_PPC64;
> + if (is_32bit_task())
> + return AUDIT_ARCH_PPC;
> + else if (IS_ENABLED(CONFIG_CPU_LITTLE_ENDIAN))
> + return AUDIT_ARCH_PPC64LE;
> else
> - arch = AUDIT_ARCH_PPC;
> -
> -#ifdef __LITTLE_ENDIAN__
> - arch |= __AUDIT_ARCH_LE;
> -#endif
> - return arch;
> + return AUDIT_ARCH_PPC64;
> }
> #endif /* _ASM_SYSCALL_H */
Please revert or fix.
--
ldv
Powered by blists - more mailing lists