From: Lai Jiangshan <jiangshanlai@gmail.com>
To: linux-kernel@vger.kernel.org
Cc: x86@kernel.org, Lai Jiangshan <laijs@linux.alibaba.com>,
Andy Lutomirski <luto@kernel.org>,
Thomas Gleixner <tglx@linutronix.de>,
Ingo Molnar <mingo@redhat.com>, Borislav Petkov <bp@alien8.de>,
"H. Peter Anvin" <hpa@zytor.com>
Subject: [PATCH V3 02/49] x86/entry: Use the correct fence macro after swapgs in kernel CR3
Date: Thu, 14 Oct 2021 11:13:25 +0800 [thread overview]
Message-ID: <20211014031413.14471-3-jiangshanlai@gmail.com> (raw)
In-Reply-To: <20211014031413.14471-1-jiangshanlai@gmail.com>
From: Lai Jiangshan <laijs@linux.alibaba.com>
The commit c75890700455 ("x86/entry/64: Remove unneeded kernel CR3
switching") removes a CR3 write in the faulting path of load_gs_index.
But the path's FENCE_SWAPGS_USER_ENTRY has no fence operation if PTI
is enabled. Rahter, it depends on the CR3 write of SWITCH_TO_KERNEL_CR3.
So the path should use FENCE_SWAPGS_KERNEL_ENTRY if SWITCH_TO_KERNEL_CR3
is removed.
Fixes: c75890700455 ("x86/entry/64: Remove unneeded kernel CR3 switching")
Signed-off-by: Lai Jiangshan <laijs@linux.alibaba.com>
---
arch/x86/entry/entry_64.S | 16 +++++++++-------
1 file changed, 9 insertions(+), 7 deletions(-)
diff --git a/arch/x86/entry/entry_64.S b/arch/x86/entry/entry_64.S
index 1f98188e83ef..9d468c8877e2 100644
--- a/arch/x86/entry/entry_64.S
+++ b/arch/x86/entry/entry_64.S
@@ -994,11 +994,6 @@ SYM_CODE_START_LOCAL(error_entry)
pushq %r12
ret
-.Lerror_entry_done_lfence:
- FENCE_SWAPGS_KERNEL_ENTRY
-.Lerror_entry_done:
- ret
-
/*
* There are two places in the kernel that can potentially fault with
* usergs. Handle them here. B stepping K8s sometimes report a
@@ -1021,8 +1016,15 @@ SYM_CODE_START_LOCAL(error_entry)
* .Lgs_change's error handler with kernel gsbase.
*/
SWAPGS
- FENCE_SWAPGS_USER_ENTRY
- jmp .Lerror_entry_done
+
+ /*
+ * The above code has no serializing instruction. So do an lfence
+ * to prevent GS speculation, regardless of whether it is kernel
+ * gsbase or user gsbase.
+ */
+.Lerror_entry_done_lfence:
+ FENCE_SWAPGS_KERNEL_ENTRY
+ ret
.Lbstep_iret:
/* Fix truncated RIP */
--
2.19.1.6.gb485710b
next prev parent reply other threads:[~2021-10-14 3:14 UTC|newest]
Thread overview: 57+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-10-14 3:13 [PATCH V3 00/49] x86/entry/64: Convert a bunch of ASM entry code into C code Lai Jiangshan
2021-10-14 3:13 ` [PATCH V3 01/49] x86/entry: Add fence for kernel entry swapgs in paranoid_entry() Lai Jiangshan
2021-10-14 3:13 ` Lai Jiangshan [this message]
2021-10-14 3:13 ` [PATCH V3 03/49] x86/traps: Remove stack-protector from traps.c Lai Jiangshan
2021-10-14 3:13 ` [PATCH V3 04/49] x86/xen: Add xenpv_restore_regs_and_return_to_usermode() Lai Jiangshan
2021-10-16 18:08 ` Boris Ostrovsky
2021-10-14 3:13 ` [PATCH V3 05/49] x86/entry: Use swapgs and native_iret directly in swapgs_restore_regs_and_return_to_usermode Lai Jiangshan
2021-10-14 3:33 ` [PATCH V3 06/49] compiler_types.h: Add __noinstr_section() for noinstr Lai Jiangshan
2021-10-14 18:29 ` Nick Desaulniers
2021-10-14 18:35 ` Miguel Ojeda
2021-10-18 9:16 ` Lai Jiangshan
2021-10-18 11:29 ` Miguel Ojeda
2021-10-14 3:33 ` [PATCH V3 07/49] x86/entry: Introduce __entry_text for entry code written in C Lai Jiangshan
2021-10-14 3:33 ` [PATCH V3 08/49] x86/entry: Move PTI_USER_* to arch/x86/include/asm/processor-flags.h Lai Jiangshan
2021-10-14 3:33 ` [PATCH V3 09/49] x86: Remove unused kernel_to_user_p4dp() and user_to_kernel_p4dp() Lai Jiangshan
2021-10-14 3:33 ` [PATCH V3 10/49] x86: Replace PTI_PGTABLE_SWITCH_BIT with PTI_USER_PGTABLE_BIT Lai Jiangshan
2021-10-14 3:33 ` [PATCH V3 11/49] x86: Mark __native_read_cr3() & native_write_cr3() as __always_inline Lai Jiangshan
2021-10-14 3:33 ` [PATCH V3 12/49] x86/traps: Move the declaration of native_irq_return_iret into proto.h Lai Jiangshan
2021-10-14 3:33 ` [PATCH V3 13/49] x86/entry: Add arch/x86/entry/entry64.c for C entry code Lai Jiangshan
2021-10-14 3:33 ` [PATCH V3 14/49] x86/entry: Expose the address of .Lgs_change to entry64.c Lai Jiangshan
2021-10-14 3:33 ` [PATCH V3 15/49] x86/entry: Add C verion of SWITCH_TO_KERNEL_CR3 as switch_to_kernel_cr3() Lai Jiangshan
2021-10-14 3:33 ` [PATCH V3 16/49] x86/traps: Add fence_swapgs_{user,kernel}_entry() Lai Jiangshan
2021-10-14 3:33 ` [PATCH V3 17/49] x86/entry: Add C {user,kernel}_entry_swapgs_and_fence() Lai Jiangshan
2021-10-14 3:33 ` [PATCH V3 18/49] x86/traps: Move pt_regs only in fixup_bad_iret() Lai Jiangshan
2021-10-14 3:33 ` [PATCH V3 19/49] x86/entry: Switch the stack after error_entry() returns Lai Jiangshan
2021-10-14 3:33 ` [PATCH V3 20/49] x86/entry: move PUSH_AND_CLEAR_REGS out of error_entry Lai Jiangshan
2021-10-14 3:33 ` [PATCH V3 21/49] x86/entry: Move cld to the start of idtentry Lai Jiangshan
2021-10-14 3:33 ` [PATCH V3 22/49] x86/entry: Don't call error_entry for XENPV Lai Jiangshan
2021-10-14 3:33 ` [PATCH V3 23/49] x86/entry: Convert SWAPGS to swapgs in error_entry() Lai Jiangshan
2021-10-14 3:33 ` [PATCH V3 24/49] x86/entry: Implement the whole error_entry() as C code Lai Jiangshan
2021-10-14 3:33 ` [PATCH V3 25/49] x86/entry: Use idtentry macro for entry_INT80_compat Lai Jiangshan
2021-10-14 3:33 ` [PATCH V3 26/49] x86/entry: Convert SWAPGS to swapgs in entry_SYSENTER_compat() Lai Jiangshan
2021-10-14 3:33 ` [PATCH V3 27/49] x86: Remove the definition of SWAPGS Lai Jiangshan
2021-10-14 3:33 ` [PATCH V3 28/49] x86/entry: Make paranoid_exit() callable Lai Jiangshan
2021-10-14 3:33 ` [PATCH V3 29/49] x86/entry: Call paranoid_exit() in asm_exc_nmi() Lai Jiangshan
2021-10-14 3:33 ` [PATCH V3 30/49] x86/entry: move PUSH_AND_CLEAR_REGS out of paranoid_entry Lai Jiangshan
2021-10-14 3:33 ` [PATCH V3 31/49] x86/entry: Add the C version ist_switch_to_kernel_cr3() Lai Jiangshan
2021-10-14 3:41 ` [PATCH V3 32/49] x86/entry: Add the C version ist_restore_cr3() Lai Jiangshan
2021-10-25 0:34 ` Lai Jiangshan
2021-10-14 3:41 ` [PATCH V3 33/49] x86/entry: Add the C version get_percpu_base() Lai Jiangshan
2021-10-14 3:41 ` [PATCH V3 34/49] x86/entry: Add the C version ist_switch_to_kernel_gsbase() Lai Jiangshan
2021-10-14 3:41 ` [PATCH V3 35/49] x86/entry: Implement the C version ist_paranoid_entry() Lai Jiangshan
2021-10-14 3:50 ` [PATCH V3 36/49] x86/entry: Implement the C version ist_paranoid_exit() Lai Jiangshan
2021-10-14 3:50 ` [PATCH V3 37/49] x86/entry: Add a C macro to define the function body for IST in .entry.text Lai Jiangshan
2021-10-14 3:50 ` [PATCH V3 38/49] x86/debug, mce: Use C entry code Lai Jiangshan
2021-10-14 3:50 ` [PATCH V3 39/49] x86/idtentry.h: Move the definitions *IDTENTRY_{MCE|DEBUG}* up Lai Jiangshan
2021-10-14 3:50 ` [PATCH V3 40/49] x86/nmi: Use DEFINE_IDTENTRY_NMI for nmi Lai Jiangshan
2021-10-14 3:50 ` [PATCH V3 41/49] x86/nmi: Use C entry code Lai Jiangshan
2021-10-14 3:50 ` [PATCH V3 42/49] x86/entry: Add a C macro to define the function body for IST in .entry.text with an error code Lai Jiangshan
2021-10-14 3:58 ` [PATCH V3 43/49] x86/doublefault: Use C entry code Lai Jiangshan
2021-10-14 3:58 ` [PATCH V3 44/49] x86/sev: Add and use ist_vc_switch_off_ist() Lai Jiangshan
2021-10-14 3:58 ` [PATCH V3 45/49] x86/sev: Use C entry code Lai Jiangshan
2021-10-14 3:58 ` [PATCH V3 46/49] x86/entry: Remove ASM function paranoid_entry() and paranoid_exit() Lai Jiangshan
2021-10-14 3:58 ` [PATCH V3 47/49] x86/entry: Remove the unused ASM macros Lai Jiangshan
2021-10-14 3:58 ` [PATCH V3 48/49] x86/entry: Remove save_ret from PUSH_AND_CLEAR_REGS Lai Jiangshan
2021-10-14 3:58 ` [PATCH V3 49/49] x86/syscall/64: Move the checking for sysret to C code Lai Jiangshan
2021-10-23 0:01 ` [PATCH V3 00/49] x86/entry/64: Convert a bunch of ASM entry code into " Lai Jiangshan
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=20211014031413.14471-3-jiangshanlai@gmail.com \
--to=jiangshanlai@gmail.com \
--cc=bp@alien8.de \
--cc=hpa@zytor.com \
--cc=laijs@linux.alibaba.com \
--cc=linux-kernel@vger.kernel.org \
--cc=luto@kernel.org \
--cc=mingo@redhat.com \
--cc=tglx@linutronix.de \
--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).