linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Qiuxu Zhuo <qiuxu.zhuo@intel.com>
To: tglx@linutronix.de, dave.hansen@linux.intel.com,
	tony.luck@intel.com, bp@alien8.de,
	kirill.shutemov@linux.intel.com, peterz@infradead.org,
	mingo@redhat.com
Cc: x86@kernel.org, hpa@zytor.com, linux-kernel@vger.kernel.org,
	qiuxu.zhuo@intel.com
Subject: [PATCH 1/1] x86/lib: Revert to _ASM_EXTABLE_UA() for {get,put}_user() fixups
Date: Mon, 29 Jan 2024 14:38:42 +0800	[thread overview]
Message-ID: <20240129063842.61584-1-qiuxu.zhuo@intel.com> (raw)

During the memory error injection test on kernels >= v6.4, kernel
panics occurred like below. However, this issue couldn't be
reproduced on kernels <= v6.3.

  mce: [Hardware Error]: CPU 296: Machine Check Exception: f Bank 1: bd80000000100134
  mce: [Hardware Error]: RIP 10:<ffffffff821b9776> {__get_user_nocheck_4+0x6/0x20}
  mce: [Hardware Error]: TSC 411a93533ed ADDR 346a8730040 MISC 86
  mce: [Hardware Error]: PROCESSOR 0:a06d0 TIME 1706000767 SOCKET 1 APIC 211 microcode 80001490
  mce: [Hardware Error]: Run the above through 'mcelog --ascii'
  mce: [Hardware Error]: Machine check: Data load in unrecoverable area of kernel
  Kernel panic - not syncing: Fatal local machine check

MCE can recover from an in-kernel #MC if the fixup type is EX_TYPE_UACCESS,
explicitly indicating that the kernel is attempting to access userspace
memory. However, MCE can only panic the kernel for an in-kernel #MC if the
fixup type is EX_TYPE_DEFAULT as it includes various fixup types that MCE
may not recover from.

ex_handler_uaccess() would throw a #GP WARN_ONCE() if users gave
non-canonical addresses (with bit 63 clear) to {get, put}_user(), which was
unexpected. Therefore, the previous commit 'b19b74bc99b1' replaced
_ASM_EXTABLE_UA() with _ASM_EXTABLE() for {get, put}_user() fixups where
the corresponding exception handler ex_handler_default() does not have
a #GP WARN_ONCE(). However, the new fixup type EX_TYPE_DEFAULT resulted
in the mentioned kernel panic.

The subsequent commit '6014bc27561f2' (merged in v6.4) added the check
gp_fault_address_ok() right before #GP WARN_ONCE() in ex_handler_uaccess().
This check can avoid the #GP WARN_ONCE() for a non-canonical address (with
bit63 clear) in ex_handler_uaccess().

So with the subsequent commit '6014bc27561f2', now it's OK to revert to
_ASM_EXTABLE_UA() for {get,put}_user() exception fixups as it can both
avoid the #GP WARN_ONCE() in ex_handler_uaccess() and fix the panic
mentioned above.

Fixes: b19b74bc99b1 ("x86/mm: Rework address range check in get_user() and put_user()")
Signed-off-by: Qiuxu Zhuo <qiuxu.zhuo@intel.com>
Reviewed-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
---
 arch/x86/lib/getuser.S | 24 ++++++++++++------------
 arch/x86/lib/putuser.S | 20 ++++++++++----------
 2 files changed, 22 insertions(+), 22 deletions(-)

diff --git a/arch/x86/lib/getuser.S b/arch/x86/lib/getuser.S
index 20ef350a60fb..10d5ed8b5990 100644
--- a/arch/x86/lib/getuser.S
+++ b/arch/x86/lib/getuser.S
@@ -163,23 +163,23 @@ SYM_CODE_END(__get_user_8_handle_exception)
 #endif
 
 /* get_user */
-	_ASM_EXTABLE(1b, __get_user_handle_exception)
-	_ASM_EXTABLE(2b, __get_user_handle_exception)
-	_ASM_EXTABLE(3b, __get_user_handle_exception)
+	_ASM_EXTABLE_UA(1b, __get_user_handle_exception)
+	_ASM_EXTABLE_UA(2b, __get_user_handle_exception)
+	_ASM_EXTABLE_UA(3b, __get_user_handle_exception)
 #ifdef CONFIG_X86_64
-	_ASM_EXTABLE(4b, __get_user_handle_exception)
+	_ASM_EXTABLE_UA(4b, __get_user_handle_exception)
 #else
-	_ASM_EXTABLE(4b, __get_user_8_handle_exception)
-	_ASM_EXTABLE(5b, __get_user_8_handle_exception)
+	_ASM_EXTABLE_UA(4b, __get_user_8_handle_exception)
+	_ASM_EXTABLE_UA(5b, __get_user_8_handle_exception)
 #endif
 
 /* __get_user */
-	_ASM_EXTABLE(6b, __get_user_handle_exception)
-	_ASM_EXTABLE(7b, __get_user_handle_exception)
-	_ASM_EXTABLE(8b, __get_user_handle_exception)
+	_ASM_EXTABLE_UA(6b, __get_user_handle_exception)
+	_ASM_EXTABLE_UA(7b, __get_user_handle_exception)
+	_ASM_EXTABLE_UA(8b, __get_user_handle_exception)
 #ifdef CONFIG_X86_64
-	_ASM_EXTABLE(9b, __get_user_handle_exception)
+	_ASM_EXTABLE_UA(9b, __get_user_handle_exception)
 #else
-	_ASM_EXTABLE(9b, __get_user_8_handle_exception)
-	_ASM_EXTABLE(10b, __get_user_8_handle_exception)
+	_ASM_EXTABLE_UA(9b, __get_user_8_handle_exception)
+	_ASM_EXTABLE_UA(10b, __get_user_8_handle_exception)
 #endif
diff --git a/arch/x86/lib/putuser.S b/arch/x86/lib/putuser.S
index 2877f5934177..975c9c18263d 100644
--- a/arch/x86/lib/putuser.S
+++ b/arch/x86/lib/putuser.S
@@ -133,15 +133,15 @@ SYM_CODE_START_LOCAL(__put_user_handle_exception)
 	RET
 SYM_CODE_END(__put_user_handle_exception)
 
-	_ASM_EXTABLE(1b, __put_user_handle_exception)
-	_ASM_EXTABLE(2b, __put_user_handle_exception)
-	_ASM_EXTABLE(3b, __put_user_handle_exception)
-	_ASM_EXTABLE(4b, __put_user_handle_exception)
-	_ASM_EXTABLE(5b, __put_user_handle_exception)
-	_ASM_EXTABLE(6b, __put_user_handle_exception)
-	_ASM_EXTABLE(7b, __put_user_handle_exception)
-	_ASM_EXTABLE(9b, __put_user_handle_exception)
+	_ASM_EXTABLE_UA(1b, __put_user_handle_exception)
+	_ASM_EXTABLE_UA(2b, __put_user_handle_exception)
+	_ASM_EXTABLE_UA(3b, __put_user_handle_exception)
+	_ASM_EXTABLE_UA(4b, __put_user_handle_exception)
+	_ASM_EXTABLE_UA(5b, __put_user_handle_exception)
+	_ASM_EXTABLE_UA(6b, __put_user_handle_exception)
+	_ASM_EXTABLE_UA(7b, __put_user_handle_exception)
+	_ASM_EXTABLE_UA(9b, __put_user_handle_exception)
 #ifdef CONFIG_X86_32
-	_ASM_EXTABLE(8b, __put_user_handle_exception)
-	_ASM_EXTABLE(10b, __put_user_handle_exception)
+	_ASM_EXTABLE_UA(8b, __put_user_handle_exception)
+	_ASM_EXTABLE_UA(10b, __put_user_handle_exception)
 #endif
-- 
2.17.1


             reply	other threads:[~2024-01-29  6:39 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-01-29  6:38 Qiuxu Zhuo [this message]
2024-01-29 10:50 ` [tip: x86/urgent] x86/lib: Revert to _ASM_EXTABLE_UA() for {get,put}_user() fixups tip-bot2 for Qiuxu Zhuo

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20240129063842.61584-1-qiuxu.zhuo@intel.com \
    --to=qiuxu.zhuo@intel.com \
    --cc=bp@alien8.de \
    --cc=dave.hansen@linux.intel.com \
    --cc=hpa@zytor.com \
    --cc=kirill.shutemov@linux.intel.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@redhat.com \
    --cc=peterz@infradead.org \
    --cc=tglx@linutronix.de \
    --cc=tony.luck@intel.com \
    --cc=x86@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).