| 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
| ||
|
Message-Id: <1274615328-27953-1-git-send-email-tom.leiming@gmail.com> Date: Sun, 23 May 2010 19:48:48 +0800 From: tom.leiming@...il.com To: linux@....linux.org.uk Cc: linux-arm-kernel@...ts.infradead.org, linux-embedded@...r.kernel.org, a.p.zijlstra@...llo.nl, linux-kernel@...r.kernel.org, Ming Lei <tom.leiming@...il.com> Subject: [RESEND PATCH] ARM: fix 'unannotated irqs-on' lockdep warning From: Ming Lei <tom.leiming@...il.com> This patch fixes the 'unannotated irqs-on' lockdep warning[1]: This patch introduces macro of trace_ret_hardirqs_on, which will call trace_hardirqs_on if I flag in the stored CPSR is zero. In the path of returning to user-space, the patch replaces disable_irq with disable_irq_notrace in ret_to_user to avoid the 'unannotated irqs-on' lockdep warning since hardirqs is always enabled before returning to user. In the path of returning to kernel-space, the patch still replaces disable_irq with disable_irq_notrace in __xxx_svc handler and adds trace_hardirqs_on before calling svc_exit to trace the possible hardirqs on event for avoiding the possible 'unannotated irqs-on' lockdep warning. The patch not only fixes the warning[1] to make lockdep usable on ARM, but also improves the efficiency if lockdep is enabled: - remove one calling for asm_trace_hardirqs_on before returning to user-space - convert asm_trace_hardirqs_on to trace_hardirqs_on in the path of returning to kernel space from exception and __only__ call it if harirqs will be enabled before returning. [1] CPU: Testing write buffer coherency: ok ------------[ cut here ]------------ WARNING: at kernel/lockdep.c:3145 check_flags+0xcc/0x1dc() Modules linked in: [<c0035120>] (unwind_backtrace+0x0/0xf8) from [<c0355374>] (dump_stack+0x20/0x24) [<c0355374>] (dump_stack+0x20/0x24) from [<c0060c04>] (warn_slowpath_common+0x58/0x70) [<c0060c04>] (warn_slowpath_common+0x58/0x70) from [<c0060c3c>] (warn_slowpath_null+0x20/0x24) [<c0060c3c>] (warn_slowpath_null+0x20/0x24) from [<c008f224>] (check_flags+0xcc/0x1dc) [<c008f224>] (check_flags+0xcc/0x1dc) from [<c00945dc>] (lock_acquire+0x50/0x140) [<c00945dc>] (lock_acquire+0x50/0x140) from [<c0358434>] (_raw_spin_lock+0x50/0x88) [<c0358434>] (_raw_spin_lock+0x50/0x88) from [<c00fd114>] (set_task_comm+0x2c/0x60) [<c00fd114>] (set_task_comm+0x2c/0x60) from [<c007e184>] (kthreadd+0x30/0x108) [<c007e184>] (kthreadd+0x30/0x108) from [<c0030104>] (kernel_thread_exit+0x0/0x8) ---[ end trace 1b75b31a2719ed1c ]--- possible reason: unannotated irqs-on. irq event stamp: 3 hardirqs last enabled at (2): [<c0059bb0>] finish_task_switch+0x48/0xb0 hardirqs last disabled at (3): [<c002f0b0>] ret_slow_syscall+0xc/0x1c softirqs last enabled at (0): [<c005f3e0>] copy_process+0x394/0xe5c softirqs last disabled at (0): [<(null)>] (null) Signed-off-by: Ming Lei <tom.leiming@...il.com> --- arch/arm/kernel/entry-armv.S | 31 +++++++++++++++++-------------- arch/arm/kernel/entry-common.S | 2 +- arch/arm/kernel/entry-header.S | 7 +++++++ 3 files changed, 25 insertions(+), 15 deletions(-) diff --git a/arch/arm/kernel/entry-armv.S b/arch/arm/kernel/entry-armv.S index 7ee48e7..42d33cf 100644 --- a/arch/arm/kernel/entry-armv.S +++ b/arch/arm/kernel/entry-armv.S @@ -204,13 +204,15 @@ __dabt_svc: @ @ IRQs off again before pulling preserved data off the stack @ - disable_irq + disable_irq_notrace @ @ restore SPSR and restart the instruction + @ trace possible hardirqs on event before returning @ - ldr r2, [sp, #S_PSR] - svc_exit r2 @ return from exception + ldr r4, [sp, #S_PSR] + trace_ret_hardirqs_on r4 + svc_exit r4 @ return from exception UNWIND(.fnend ) ENDPROC(__dabt_svc) @@ -235,10 +237,7 @@ __irq_svc: blne svc_preempt #endif ldr r4, [sp, #S_PSR] @ irqs are already disabled -#ifdef CONFIG_TRACE_IRQFLAGS - tst r4, #PSR_I_BIT - bleq trace_hardirqs_on -#endif + trace_ret_hardirqs_on r4 svc_exit r4 @ return from exception UNWIND(.fnend ) ENDPROC(__irq_svc) @@ -291,13 +290,15 @@ __und_svc: @ @ IRQs off again before pulling preserved data off the stack @ -1: disable_irq +1: disable_irq_notrace @ @ restore SPSR and restart the instruction + @ trace possible hardirqs on event before returning @ - ldr r2, [sp, #S_PSR] @ Get SVC cpsr - svc_exit r2 @ return from exception + ldr r4, [sp, #S_PSR] @ Get SVC cpsr + trace_ret_hardirqs_on r4 + svc_exit r4 @ return from exception UNWIND(.fnend ) ENDPROC(__und_svc) @@ -327,13 +328,15 @@ __pabt_svc: @ @ IRQs off again before pulling preserved data off the stack @ - disable_irq + disable_irq_notrace @ @ restore SPSR and restart the instruction - @ - ldr r2, [sp, #S_PSR] - svc_exit r2 @ return from exception + @ trace possible hardirqs on event before returning + @ + ldr r4, [sp, #S_PSR] + trace_ret_hardirqs_on r4 + svc_exit r4 @ return from exception UNWIND(.fnend ) ENDPROC(__pabt_svc) diff --git a/arch/arm/kernel/entry-common.S b/arch/arm/kernel/entry-common.S index 2c1db77..ed471a7 100644 --- a/arch/arm/kernel/entry-common.S +++ b/arch/arm/kernel/entry-common.S @@ -58,7 +58,7 @@ work_resched: */ ENTRY(ret_to_user) ret_slow_syscall: - disable_irq @ disable interrupts + disable_irq_notrace @ disable interrupts ldr r1, [tsk, #TI_FLAGS] tst r1, #_TIF_WORK_MASK bne work_pending diff --git a/arch/arm/kernel/entry-header.S b/arch/arm/kernel/entry-header.S index d93f976..ac51ce3 100644 --- a/arch/arm/kernel/entry-header.S +++ b/arch/arm/kernel/entry-header.S @@ -30,6 +30,13 @@ #error "Please fix" #endif + .macro trace_ret_hardirqs_on, rspsr +#ifdef CONFIG_TRACE_IRQFLAGS + tst \rspsr, #PSR_I_BIT + bleq trace_hardirqs_on +#endif + .endm + .macro zero_fp #ifdef CONFIG_FRAME_POINTER mov fp, #0 -- 1.6.2.5 -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@...r.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists