s390/udelay: make udelay have busy loop semantics

When using systemtap it was observed that our udelay implementation is
rather suboptimal if being called from a kprobe handler installed by
systemtap.

The problem observed when a kprobe was installed on lock_acquired().
When the probe was hit the kprobe handler did call udelay, which set
up an (internal) timer and reenabled interrupts (only the clock comparator
interrupt) and waited for the interrupt.
This is an optimization to avoid that the cpu is busy looping while waiting
that enough time passes. The problem is that the interrupt handler still
does call irq_enter()/irq_exit() which then again can lead to a deadlock,
since some accounting functions may take locks as well.

If one of these locks is the same, which caused lock_acquired() to be
called, we have a nice deadlock.

This patch reworks the udelay code for the interrupts disabled case to
immediately leave the low level interrupt handler when the clock
comparator interrupt happens. That way no C code is being called and the
deadlock cannot happen anymore.

Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Reviewed-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

diff --git a/arch/s390/include/asm/idle.h b/arch/s390/include/asm/idle.h
index 113cd963dbbef445b99fc6d51fe74171bdcc4e4e..51ff96d9f2877ca6f78d08e12b6ffe2b9a6282b8 100644 (file)
--- a/arch/s390/include/asm/idle.h
+++ b/arch/s390/include/asm/idle.h
@@ -24,4 +24,6 @@ struct s390_idle_data {
 extern struct device_attribute dev_attr_idle_count;
 extern struct device_attribute dev_attr_idle_time_us;
 
+void psw_idle(struct s390_idle_data *, unsigned long);
+
 #endif /* _S390_IDLE_H */

diff --git a/arch/s390/include/asm/irq.h b/arch/s390/include/asm/irq.h
index ff95d15a2384760bab3be77bab26eda0ab848a09..33d26424c2f5af471a1dc2ccf52284c5b4c506f2 100644 (file)
--- a/arch/s390/include/asm/irq.h
+++ b/arch/s390/include/asm/irq.h
@@ -96,6 +96,19 @@ enum irq_subclass {
        IRQ_SUBCLASS_SERVICE_SIGNAL = 9,
 };
 
+#define CR0_IRQ_SUBCLASS_MASK                                    \
+       ((1UL << (63 - 30))  /* Warning Track */                | \
+        (1UL << (63 - 48))  /* Malfunction Alert */            | \
+        (1UL << (63 - 49))  /* Emergency Signal */             | \
+        (1UL << (63 - 50))  /* External Call */                | \
+        (1UL << (63 - 52))  /* Clock Comparator */             | \
+        (1UL << (63 - 53))  /* CPU Timer */                    | \
+        (1UL << (63 - 54))  /* Service Signal */               | \
+        (1UL << (63 - 57))  /* Interrupt Key */                | \
+        (1UL << (63 - 58))  /* Measurement Alert */            | \
+        (1UL << (63 - 59))  /* Timing Alert */                 | \
+        (1UL << (63 - 62))) /* IUCV */
+
 void irq_subclass_register(enum irq_subclass subclass);
 void irq_subclass_unregister(enum irq_subclass subclass);
 

diff --git a/arch/s390/include/asm/processor.h b/arch/s390/include/asm/processor.h
index 5900877bb19cbf8858892d226b68a7f10138af63..9e1cdcf8d12418cd0b396b59b0df6934edaac03a 100644 (file)
--- a/arch/s390/include/asm/processor.h
+++ b/arch/s390/include/asm/processor.h
@@ -15,11 +15,13 @@
 #define CIF_ASCE               1       /* user asce needs fixup / uaccess */
 #define CIF_NOHZ_DELAY         2       /* delay HZ disable for a tick */
 #define CIF_FPU                        3       /* restore vector registers */
+#define CIF_IGNORE_IRQ         4       /* ignore interrupt (for udelay) */
 
 #define _CIF_MCCK_PENDING      (1<<CIF_MCCK_PENDING)
 #define _CIF_ASCE              (1<<CIF_ASCE)
 #define _CIF_NOHZ_DELAY                (1<<CIF_NOHZ_DELAY)
 #define _CIF_FPU               (1<<CIF_FPU)
+#define _CIF_IGNORE_IRQ                (1<<CIF_IGNORE_IRQ)
 
 #ifndef __ASSEMBLY__
 

diff --git a/arch/s390/kernel/entry.S b/arch/s390/kernel/entry.S
index d653a87d66b58470bad04659d7a94e104fc593ca..857b6526d29833507c3abec126b90a24e491956f 100644 (file)
--- a/arch/s390/kernel/entry.S
+++ b/arch/s390/kernel/entry.S
@@ -563,6 +563,8 @@ ENTRY(io_int_handler)
        stmg    %r8,%r9,__PT_PSW(%r11)
        mvc     __PT_INT_CODE(12,%r11),__LC_SUBCHANNEL_ID
        xc      __PT_FLAGS(8,%r11),__PT_FLAGS(%r11)
+       TSTMSK  __LC_CPU_FLAGS,_CIF_IGNORE_IRQ
+       jo      .Lio_restore
        TRACE_IRQS_OFF
        xc      __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15)
 .Lio_loop:
@@ -738,6 +740,8 @@ ENTRY(ext_int_handler)
        mvc     __PT_INT_PARM(4,%r11),__LC_EXT_PARAMS
        mvc     __PT_INT_PARM_LONG(8,%r11),0(%r1)
        xc      __PT_FLAGS(8,%r11),__PT_FLAGS(%r11)
+       TSTMSK  __LC_CPU_FLAGS,_CIF_IGNORE_IRQ
+       jo      .Lio_restore
        TRACE_IRQS_OFF
        xc      __SF_BACKCHAIN(8,%r15),__SF_BACKCHAIN(%r15)
        lgr     %r2,%r11                # pass pointer to pt_regs

diff --git a/arch/s390/kernel/entry.h b/arch/s390/kernel/entry.h
index 3eecd0f7adb2ccc8a948dfe7bffc0909de1f94bb..b7019ab74070f8a8e692941a22711d71775d2a30 100644 (file)
--- a/arch/s390/kernel/entry.h
+++ b/arch/s390/kernel/entry.h
@@ -16,7 +16,6 @@ void io_int_handler(void);
 void mcck_int_handler(void);
 void restart_int_handler(void);
 void restart_call_handler(void);
-void psw_idle(struct s390_idle_data *, unsigned long);
 
 asmlinkage long do_syscall_trace_enter(struct pt_regs *regs);
 asmlinkage void do_syscall_trace_exit(struct pt_regs *regs);

diff --git a/arch/s390/lib/delay.c b/arch/s390/lib/delay.c
index 246a7eb4b680208dd680207aafc09d8f0deefdda..501dcd4ca4a068a5db21eed7284c390346d70546 100644 (file)
--- a/arch/s390/lib/delay.c
+++ b/arch/s390/lib/delay.c
@@ -12,8 +12,10 @@
 #include <linux/module.h>
 #include <linux/irqflags.h>
 #include <linux/interrupt.h>
+#include <linux/irq.h>
 #include <asm/vtimer.h>
 #include <asm/div64.h>
+#include <asm/idle.h>
 
 void __delay(unsigned long loops)
 {
@@ -30,26 +32,22 @@ EXPORT_SYMBOL(__delay);
 
 static void __udelay_disabled(unsigned long long usecs)
 {
-       unsigned long cr0, cr6, new;
-       u64 clock_saved, end;
+       unsigned long cr0, cr0_new, psw_mask;
+       struct s390_idle_data idle;
+       u64 end;
 
        end = get_tod_clock() + (usecs << 12);
-       clock_saved = local_tick_disable();
        __ctl_store(cr0, 0, 0);
-       __ctl_store(cr6, 6, 6);
-       new = (cr0 &  0xffff00e0) | 0x00000800;
-       __ctl_load(new , 0, 0);
-       new = 0;
-       __ctl_load(new, 6, 6);
-       lockdep_off();
-       do {
-               set_clock_comparator(end);
-               enabled_wait();
-       } while (get_tod_clock_fast() < end);
-       lockdep_on();
+       cr0_new = cr0 & ~CR0_IRQ_SUBCLASS_MASK;
+       cr0_new |= (1UL << (63 - 52)); /* enable clock comparator irq */
+       __ctl_load(cr0_new, 0, 0);
+       psw_mask = __extract_psw() | PSW_MASK_EXT | PSW_MASK_WAIT;
+       set_clock_comparator(end);
+       set_cpu_flag(CIF_IGNORE_IRQ);
+       psw_idle(&idle, psw_mask);
+       clear_cpu_flag(CIF_IGNORE_IRQ);
+       set_clock_comparator(S390_lowcore.clock_comparator);
        __ctl_load(cr0, 0, 0);
-       __ctl_load(cr6, 6, 6);
-       local_tick_enable(clock_saved);
 }
 
 static void __udelay_enabled(unsigned long long usecs)