Messages in this thread | | | Subject | Re: [PATCH v2] x86/fault: Fix wrong signal when vsyscall fails with pkey | From | Jiashuo Liang <> | Date | Sun, 19 Sep 2021 12:39:27 +0800 |
| |
On Mon, 2021-08-23 at 07:55 -0700, Dave Hansen wrote: > On 7/29/21 8:01 PM, Jiashuo Liang wrote: >> The function __bad_area_nosemaphore() calls kernelmode_fixup_or_oops() with >> parameter "signal" being "pkey", which will send a signal numbered "pkey". >> >> This bug appears when the kernel fail to access user-given memory pages >> that are protected by pkey, so it can go through the do_user_addr_fault() >> path and pass the !user_mode() check in __bad_area_nosemaphore(). Most >> cases will simply run the kernel fixup code to make an -EFAULT. But when >> another condition current->thread.sig_on_uaccess_err is met, which is >> only used to emulate vsyscall, we will generate the wrong signal. >> >> A new parameter "pkey" is added to kernelmode_fixup_or_oops() to fix it. >> >> Fixes: 5042d40a264c ("x86/fault: Bypass no_context() for implicit kernel faults from usermode") >> Signed-off-by: Jiashuo Liang <liangjs@pku.edu.cn> > > This is pretty obscure, only affecting vsyscall emulation, but: > > Acked-by: Dave Hansen <dave.hansen@linux.intel.com>
Ping... Is there anyone reviewing/committing this patch?
Thanks!
| |