| 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
| ||
|
Date: Wed, 22 Sep 2021 09:45:42 +0100
From: Richard Palethorpe <rpalethorpe@...e.de>
To: Arnd Bergmann <arnd@...db.de>
Cc: Linux FS-devel Mailing List <linux-fsdevel@...r.kernel.org>,
Linux API <linux-api@...r.kernel.org>,
linux-aio <linux-aio@...ck.org>,
"y2038 Mailman List" <y2038@...ts.linaro.org>,
Andy Lutomirski <luto@...nel.org>,
Thomas Gleixner <tglx@...utronix.de>,
Ingo Molnar <mingo@...hat.com>, Borislav Petkov <bp@...en8.de>,
the arch/x86 maintainers <x86@...nel.org>,
"H. Peter Anvin" <hpa@...or.com>,
Deepa Dinamani <deepa.kernel@...il.com>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
LTP List <ltp@...ts.linux.it>
Subject: ia32 signed long treated as x64 unsigned int by __ia32_sys*
Richard Palethorpe <rpalethorpe@...e.de> writes:
> Richard Palethorpe <rpalethorpe@...e.de> writes:
>
>> Hello Arnd,
>>
>> Arnd Bergmann <arnd@...db.de> writes:
>>
>>> On Tue, Sep 21, 2021 at 3:01 PM Richard Palethorpe <rpalethorpe@...e.com> wrote:
>>>>
>>>> The LTP test io_pgetevents02 fails in 32bit compat mode because an
>>>> nr_max of -1 appears to be treated as a large positive integer. This
>>>> causes pgetevents_time64 to return an event. The test expects the call
>>>> to fail and errno to be set to EINVAL.
>>>>
>>>> Using the compat syscall fixes the issue.
>>>>
>>>> Fixes: 7a35397f8c06 ("io_pgetevents: use __kernel_timespec")
>>>> Signed-off-by: Richard Palethorpe <rpalethorpe@...e.com>
>>>
>>> Thanks a lot for finding this, indeed there is definitely a mistake that
>>> this function is defined and not used, but I don't yet see how it would
>>> get to the specific failure you report.
>>>
>>> Between the two implementations, I can see a difference in the
>>> handling of the signal mask, but that should only affect architectures
>>> with incompatible compat_sigset_t, i.e. big-endian or
>>> _COMPAT_NSIG_WORDS!=_NSIG_WORDS, and the latter is
>>> never true for currently supported architectures. On x86, there is
>>> no difference in the sigset at all.
>>>
>>> The negative 'nr' and 'min_nr' arguments that you list as causing
>>> the problem /should/ be converted by the magic
>>> SYSCALL_DEFINE6() definition. If this is currently broken, I would
>>> expect other syscalls to be affected as well.
>>
>> That is what I thought, but I couldn't think of another explanation for
>> it.
>>
>>>
>>> Have you tried reproducing this on non-x86 architectures? If I
>>> misremembered how the compat conversion in SYSCALL_DEFINE6()
>>> works, then all architectures that support CONFIG_COMPAT have
>>> to be fixed.
>>>
>>> Arnd
>>
>> No, but I suppose I can try it on ARM or PowerPC. I suppose printing the
>> arguments would be a good idea too.
>
> It appears it really is failing to sign extend the s32 to s64. I added
> the following printks
>
> modified fs/aio.c
> @@ -2054,6 +2054,7 @@ static long do_io_getevents(aio_context_t ctx_id,
> long ret = -EINVAL;
>
> if (likely(ioctx)) {
> + printk("comparing %ld <= %ld\n", min_nr, nr);
> if (likely(min_nr <= nr && min_nr >= 0))
> ret = read_events(ioctx, min_nr, nr, events, until);
> percpu_ref_put(&ioctx->users);
> @@ -2114,6 +2115,8 @@ SYSCALL_DEFINE6(io_pgetevents,
> bool interrupted;
> int ret;
>
> + printk("io_pgetevents(%lx, %ld, %ld, ...)\n", ctx_id, min_nr, nr);
> +
> if (timeout && unlikely(get_timespec64(&ts, timeout)))
> return -EFAULT;
>
> Then the output is:
>
> [ 11.252268] io_pgetevents(f7f19000, 4294967295, 1, ...)
> [ 11.252401] comparing 4294967295 <= 1
> io_pgetevents02.c:114: TPASS: invalid min_nr: io_pgetevents() failed as expected: EINVAL (22)
> [ 11.252610] io_pgetevents(f7f19000, 1, 4294967295, ...)
> [ 11.252748] comparing 1 <= 4294967295
> io_pgetevents02.c:103: TFAIL: invalid max_nr: io_pgetevents() passed unexpectedly
and below is the macro expansion for the automatically generated 32bit to
64bit io_pgetevents. I believe it is casting u32 to s64, which appears
to mean there is no sign extension. I don't know if this is the expected
behaviour?
For the manually written compat version we cast back to s32 which is
what fixes the issue.
long __ia32_sys_io_pgetevents(const struct pt_regs *regs) {
return __se_sys_io_pgetevents((unsigned int)regs->bx, (unsigned int)regs->cx,
(unsigned int)regs->dx, (unsigned int)regs->si,
(unsigned int)regs->di, (unsigned int)regs->bp);
}
static long __se_sys_io_pgetevents(
__typeof(__builtin_choose_expr(
(__builtin_types_compatible_p(typeof((aio_context_t)0), typeof(0LL)) ||
__builtin_types_compatible_p(typeof((aio_context_t)0), typeof(0ULL))),
0LL, 0L)) ctx_id,
__typeof(__builtin_choose_expr(
(__builtin_types_compatible_p(typeof((long)0), typeof(0LL)) ||
__builtin_types_compatible_p(typeof((long)0), typeof(0ULL))),
0LL, 0L)) min_nr,
__typeof(__builtin_choose_expr(
(__builtin_types_compatible_p(typeof((long)0), typeof(0LL)) ||
__builtin_types_compatible_p(typeof((long)0), typeof(0ULL))),
0LL, 0L)) nr,
__typeof(__builtin_choose_expr(
(__builtin_types_compatible_p(typeof((struct io_event *)0),
typeof(0LL)) ||
__builtin_types_compatible_p(typeof((struct io_event *)0),
typeof(0ULL))),
0LL, 0L)) events,
__typeof(__builtin_choose_expr(
(__builtin_types_compatible_p(typeof((struct __kernel_timespec *)0),
typeof(0LL)) ||
__builtin_types_compatible_p(typeof((struct __kernel_timespec *)0),
typeof(0ULL))),
0LL, 0L)) timeout,
__typeof(__builtin_choose_expr(
(__builtin_types_compatible_p(typeof((const struct __aio_sigset *)0),
typeof(0LL)) ||
__builtin_types_compatible_p(typeof((const struct __aio_sigset *)0),
typeof(0ULL))),
0LL, 0L)) usig)
{
long ret = __do_sys_io_pgetevents(
(aio_context_t)ctx_id, (long)min_nr, (long)nr, (struct io_event *)events,
(struct __kernel_timespec *)timeout, (const struct __aio_sigset
*)usig);
...
}
--
Thank you,
Richard.
Powered by blists - more mailing lists