Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1233594

[PATCH 4.2 028/134] arm64: entry: always restore x0 from the stack on syscall return

From Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Newsgroups linux.kernel
Subject [PATCH 4.2 028/134] arm64: entry: always restore x0 from the stack on syscall return
Date 2015-09-27 00:30 +0200
Message-ID <qd68G-4RL-31@gated-at.bofh.it> (permalink)
References <qd4Jz-2Ha-3@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw


4.2-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Will Deacon <will.deacon@arm.com>

commit 412fcb6cebd758d080cacd5a41a0cbc656ea5fce upstream.

We have a micro-optimisation on the fast syscall return path where we
take care to keep x0 live with the return value from the syscall so that
we can avoid restoring it from the stack. The benefit of doing this is
fairly suspect, since we will be restoring x1 from the stack anyway
(which lives adjacent in the pt_regs structure) and the only additional
cost is saving x0 back to pt_regs after the syscall handler, which could
be seen as a poor man's prefetch.

More importantly, this causes issues with the context tracking code.

The ct_user_enter macro ends up branching into C code, which is free to
use x0 as a scratch register and consequently leads to us returning junk
back to userspace as the syscall return value. Rather than special case
the context-tracking code, this patch removes the questionable
optimisation entirely.

Cc: Larry Bassel <larry.bassel@linaro.org>
Cc: Kevin Hilman <khilman@linaro.org>
Reviewed-by: Catalin Marinas <catalin.marinas@arm.com>
Reported-by: Hanjun Guo <hanjun.guo@linaro.org>
Tested-by: Hanjun Guo <hanjun.guo@linaro.org>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
 arch/arm64/kernel/entry.S |   17 ++++++-----------
 1 file changed, 6 insertions(+), 11 deletions(-)

--- a/arch/arm64/kernel/entry.S
+++ b/arch/arm64/kernel/entry.S
@@ -116,7 +116,7 @@
 	*/
 	.endm
 
-	.macro	kernel_exit, el, ret = 0
+	.macro	kernel_exit, el
 	ldp	x21, x22, [sp, #S_PC]		// load ELR, SPSR
 	.if	\el == 0
 	ct_user_enter
@@ -146,11 +146,7 @@
 	.endif
 	msr	elr_el1, x21			// set up the return data
 	msr	spsr_el1, x22
-	.if	\ret
-	ldr	x1, [sp, #S_X1]			// preserve x0 (syscall return)
-	.else
 	ldp	x0, x1, [sp, #16 * 0]
-	.endif
 	ldp	x2, x3, [sp, #16 * 1]
 	ldp	x4, x5, [sp, #16 * 2]
 	ldp	x6, x7, [sp, #16 * 3]
@@ -613,22 +609,21 @@ ENDPROC(cpu_switch_to)
  */
 ret_fast_syscall:
 	disable_irq				// disable interrupts
+	str	x0, [sp, #S_X0]			// returned x0
 	ldr	x1, [tsk, #TI_FLAGS]		// re-check for syscall tracing
 	and	x2, x1, #_TIF_SYSCALL_WORK
 	cbnz	x2, ret_fast_syscall_trace
 	and	x2, x1, #_TIF_WORK_MASK
-	cbnz	x2, fast_work_pending
+	cbnz	x2, work_pending
 	enable_step_tsk x1, x2
-	kernel_exit 0, ret = 1
+	kernel_exit 0
 ret_fast_syscall_trace:
 	enable_irq				// enable interrupts
-	b	__sys_trace_return
+	b	__sys_trace_return_skipped	// we already saved x0
 
 /*
  * Ok, we need to do extra processing, enter the slow path.
  */
-fast_work_pending:
-	str	x0, [sp, #S_X0]			// returned x0
 work_pending:
 	tbnz	x1, #TIF_NEED_RESCHED, work_resched
 	/* TIF_SIGPENDING, TIF_NOTIFY_RESUME or TIF_FOREIGN_FPSTATE case */
@@ -652,7 +647,7 @@ ret_to_user:
 	cbnz	x2, work_pending
 	enable_step_tsk x1, x2
 no_work_pending:
-	kernel_exit 0, ret = 0
+	kernel_exit 0
 ENDPROC(ret_to_user)
 
 /*


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Back to linux.kernel | Previous | NextPrevious in thread | Next in thread | Find similar | Unroll thread


Thread

[PATCH 4.2 000/134] 4.2.2-stable review Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:20 +0200
  [PATCH 4.2 049/134] powerpc/eeh: Probe after unbalanced kref check Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:20 +0200
  [PATCH 4.2 018/134] MIPS: CPS: use 32b accesses to GCRs Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:20 +0200
  [PATCH 4.2 032/134] arm64: compat: fix vfp save/restore across signal handlers in big-endian Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:20 +0200
  [PATCH 4.2 047/134] pinctrl: at91: fix null pointer dereference Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:20 +0200
  [PATCH 4.2 006/134] nfc: st-nci: Remove data from ack_pending_q when receiving a SYNC_ACK Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:20 +0200
  [PATCH 4.2 015/134] rtlwifi: rtl8821ae: Fix an expression that is always false Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:20 +0200
  [PATCH 4.2 030/134] of/fdt: make memblock maximum physical address arch configurable Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:20 +0200
  [PATCH 4.2 035/134] arm/arm64: KVM: vgic: Check for !irqchip_in_kernel() when mapping resources Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:20 +0200
  [PATCH 4.2 050/134] powerpc/eeh: Fix fenced PHB caused by eeh_slot_error_detail() Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:20 +0200
  [PATCH 4.2 007/134] nfc: st-nci: Free data with irrelevant NDLC PCB_SYNC value Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:20 +0200
  [PATCH 4.2 017/134] tg3: Fix temperature reporting Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:20 +0200
  [PATCH 4.2 053/134] powerpc: Uncomment and make enable_kernel_vsx() routine available Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:20 +0200
  [PATCH 4.2 003/134] nfc: st-nci: Fix non accurate comment for st_nci_i2c_read Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:20 +0200
  [PATCH 4.2 036/134] KVM: arm64: add workaround for Cortex-A57 erratum #852523 Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:20 +0200
  [PATCH 4.2 004/134] NFC: st21nfca: fix use of uninitialized variables in error path Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:20 +0200
  [PATCH 4.2 037/134] arm64: KVM: Disable virtual timer even if the guest is not using it Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:20 +0200
  [PATCH 4.2 008/134] nfc: netlink: Add check on NFC_ATTR_VENDOR_DATA Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:20 +0200
  [PATCH 4.2 048/134] powerpc/pseries: Fix corrupted pdn list Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:20 +0200
  [PATCH 4.2 059/134] powerpc/mm: Recompute hash value after a failed update Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:20 +0200
  [PATCH 4.2 021/134] mac80211: enable assoc check for mesh interfaces Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:30 +0200
  [PATCH 4.2 029/134] arm64: flush FP/SIMD state correctly after execve() Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:30 +0200
  [PATCH 4.2 019/134] MIPS: math-emu: Allow m{f,t}hc emulation on MIPS R6 Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:30 +0200
  [PATCH 4.2 024/134] cxl: Fix unbalanced pci_dev_get in cxl_probe Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:30 +0200
  [PATCH 4.2 002/134] nfc: st-nci: Fix typo when changing from st21nfcb to st-nci Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:30 +0200
  [PATCH 4.2 022/134] cxl: Allow release of contexts which have been OPENED but not STARTED Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:30 +0200
  [PATCH 4.2 027/134] arm64: kconfig: Move LIST_POISON to a safe value Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:30 +0200
  [PATCH 4.2 020/134] MIPS: math-emu: Emulate missing BC1{EQ,NE}Z instructions Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:30 +0200
  [PATCH 4.2 025/134] ext4: dont manipulate recovery flag when freezing no-journal fs Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:30 +0200
  [PATCH 4.2 028/134] arm64: entry: always restore x0 from the stack on syscall return Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:30 +0200
  [PATCH 4.2 010/134] nfc: nci: hci: Add check on skb nci_hci_send_cmd parameter Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:30 +0200
  [PATCH 4.2 026/134] Revert "ext4: remove block_device_ejected" Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:30 +0200
  [PATCH 4.2 023/134] cxl: Remove racy attempt to force EEH invocation in reset Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 00:30 +0200
  Re: [PATCH 4.2 000/134] 4.2.2-stable review Guenter Roeck <linux@roeck-us.net> - 2015-09-27 20:20 +0200
    Re: [PATCH 4.2 000/134] 4.2.2-stable review Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-27 20:40 +0200
  Re: [PATCH 4.2 000/134] 4.2.2-stable review Sudip Mukherjee <sudipm.mukherjee@gmail.com> - 2015-09-28 08:00 +0200
    Re: [PATCH 4.2 000/134] 4.2.2-stable review Greg Kroah-Hartman <gregkh@linuxfoundation.org> - 2015-09-28 16:30 +0200
      Re: [PATCH 4.2 000/134] 4.2.2-stable review Sudip Mukherjee <sudipm.mukherjee@gmail.com> - 2015-09-28 17:10 +0200
      Re: [PATCH 4.2 000/134] 4.2.2-stable review Guenter Roeck <linux@roeck-us.net> - 2015-09-28 17:10 +0200
  Re: [PATCH 4.2 000/134] 4.2.2-stable review Shuah Khan <shuahkh@osg.samsung.com> - 2015-09-28 19:40 +0200

csiph-web