Re: [PATCH] i386 timer changes for 2.5.41

From: Greg KH (greg@kroah.com)
Date: Thu Oct 10 2002 - 13:30:18 EST


# This is a BitKeeper generated patch for the following project:
# Project Name: Linux kernel tree
# This patch format is intended for GNU patch command version 2.5 or higher.
# This patch includes the following deltas:
# ChangeSet 1.750 -> 1.751
# include/asm-i386/timer.h 1.1 -> 1.2
# arch/i386/kernel/timers/timer.c 1.1 -> 1.2
# arch/i386/Makefile 1.19 -> 1.20
# arch/i386/kernel/time.c 1.17 -> 1.18
# arch/i386/kernel/timers/timer_pit.c 1.18 -> 1.19
# arch/i386/kernel/timers/timer_tsc.c 1.18 -> 1.19
# (new) -> 1.1 arch/i386/kernel/timers/Makefile
#
# The following is the BitKeeper ChangeSet Log
# --------------------------------------------
# 02/10/10 johnstul@us.ibm.com 1.751
# i386 timer core: intergrate the new timer code to use the two different timer files.
# --------------------------------------------
#
diff -Nru a/arch/i386/Makefile b/arch/i386/Makefile
--- a/arch/i386/Makefile Thu Oct 10 11:21:10 2002
+++ b/arch/i386/Makefile Thu Oct 10 11:21:10 2002
@@ -53,7 +53,7 @@
 
 libs-y += arch/i386/lib/
 core-y += arch/i386/kernel/ arch/i386/mm/ \
- arch/i386/$(MACHINE)/
+ arch/i386/$(MACHINE)/ arch/i386/kernel/timers/
 drivers-$(CONFIG_MATH_EMULATION) += arch/i386/math-emu/
 drivers-$(CONFIG_PCI) += arch/i386/pci/
 
diff -Nru a/arch/i386/kernel/time.c b/arch/i386/kernel/time.c
--- a/arch/i386/kernel/time.c Thu Oct 10 11:21:10 2002
+++ b/arch/i386/kernel/time.c Thu Oct 10 11:21:10 2002
@@ -53,6 +53,7 @@
 #include <asm/mpspec.h>
 #include <asm/uaccess.h>
 #include <asm/processor.h>
+#include <asm/timer.h>
 
 #include <linux/mc146818rtc.h>
 #include <linux/timex.h>
@@ -78,18 +79,10 @@
 
 spinlock_t rtc_lock = SPIN_LOCK_UNLOCKED;
 
-
-#define TICK_SIZE (tick_nsec / 1000)
-
 spinlock_t i8253_lock = SPIN_LOCK_UNLOCKED;
 EXPORT_SYMBOL(i8253_lock);
 
-#ifndef CONFIG_X86_TSC
-#else
-
-#define do_gettimeoffset() do_fast_gettimeoffset()
-
-#endif
+struct timer_opts* timer;
 
 /*
  * This version of gettimeofday has microsecond resolution
@@ -101,7 +94,7 @@
         unsigned long usec, sec;
 
         read_lock_irqsave(&xtime_lock, flags);
- usec = do_gettimeoffset();
+ usec = timer->get_offset();
         {
                 unsigned long lost = jiffies - wall_jiffies;
                 if (lost)
@@ -129,7 +122,7 @@
          * wall time. Discover what correction gettimeofday() would have
          * made, and then undo it!
          */
- tv->tv_usec -= do_gettimeoffset();
+ tv->tv_usec -= timer->get_offset();
         tv->tv_usec -= (jiffies - wall_jiffies) * (1000000 / HZ);
 
         while (tv->tv_usec < 0) {
@@ -275,8 +268,6 @@
 #endif
 }
 
-static int use_tsc;
-
 /*
  * This is the same as the above, except we _also_ save the current
  * Time Stamp Counter value at the time of the timer interrupt, so that
@@ -284,8 +275,6 @@
  */
 void timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
 {
- int count;
-
         /*
          * Here we are in the timer irq handler. We just have irqs locally
          * disabled but we don't know if the timer_bh is running on the other
@@ -295,7 +284,7 @@
          */
         write_lock(&xtime_lock);
 
-
+ timer->mark_offset();
  
         do_timer_interrupt(irq, NULL, regs);
 
@@ -345,6 +334,7 @@
         return mktime(year, mon, day, hour, min, sec);
 }
 
+/* XXX this driverfs stuff should probably go elsewhere later -john */
 static struct sys_device device_i8253 = {
         .name = "rtc",
         .id = 0,
@@ -368,5 +358,6 @@
         xtime.tv_nsec = 0;
 
 
+ timer = select_timer();
         time_init_hook();
 }
diff -Nru a/arch/i386/kernel/timers/Makefile b/arch/i386/kernel/timers/Makefile
--- /dev/null Wed Dec 31 16:00:00 1969
+++ b/arch/i386/kernel/timers/Makefile Thu Oct 10 11:21:10 2002
@@ -0,0 +1,10 @@
+#
+# Makefile for x86 timers
+#
+
+obj-y := timer.o
+
+obj-y += timer_tsc.o
+obj-y += timer_pit.o
+
+include $(TOPDIR)/Rules.make
diff -Nru a/arch/i386/kernel/timers/timer.c b/arch/i386/kernel/timers/timer.c
--- a/arch/i386/kernel/timers/timer.c Thu Oct 10 11:21:10 2002
+++ b/arch/i386/kernel/timers/timer.c Thu Oct 10 11:21:10 2002
@@ -2,11 +2,15 @@
 #include <asm/timer.h>
 
 /* list of externed timers */
-/* eg: extern struct timer_opts timer_XXX*/;
+extern struct timer_opts timer_pit;
+extern struct timer_opts timer_tsc;
 
 /* list of timers, ordered by preference, NULL terminated */
 static struct timer_opts* timers[] = {
- /* eg: &timer_XXX */
+ &timer_tsc,
+#ifndef CONFIG_X86_TSC
+ &timer_pit,
+#endif
         NULL,
 };
 
diff -Nru a/arch/i386/kernel/timers/timer_pit.c b/arch/i386/kernel/timers/timer_pit.c
--- a/arch/i386/kernel/timers/timer_pit.c Thu Oct 10 11:21:10 2002
+++ b/arch/i386/kernel/timers/timer_pit.c Thu Oct 10 11:21:10 2002
@@ -1,3 +1,28 @@
+/*
+ * This code largely moved from arch/i386/kernel/time.c.
+ * See comments there for proper credits.
+ */
+
+#include <linux/spinlock.h>
+#include <linux/module.h>
+#include <linux/device.h>
+#include <asm/timer.h>
+#include <asm/io.h>
+
+extern spinlock_t i8259A_lock;
+extern spinlock_t i8253_lock;
+#include "do_timer.h"
+
+static int init_pit(void)
+{
+ return 0;
+}
+
+static void mark_offset_pit(void)
+{
+ /* nothing needed */
+}
+
 
 /* This function must be called with interrupts disabled
  * It was inspired by Steve McCanne's microtime-i386 for BSD. -- jrs
@@ -31,7 +56,7 @@
  * comp.protocols.time.ntp!
  */
 
-static unsigned long do_slow_gettimeoffset(void)
+static unsigned long get_offset_pit(void)
 {
         int count;
 
@@ -94,4 +119,10 @@
         return count;
 }
 
-static unsigned long (*do_gettimeoffset)(void) = do_slow_gettimeoffset;
+
+/* tsc timer_opts struct */
+struct timer_opts timer_pit = {
+ .init = init_pit,
+ .mark_offset = mark_offset_pit,
+ .get_offset = get_offset_pit,
+};
diff -Nru a/arch/i386/kernel/timers/timer_tsc.c b/arch/i386/kernel/timers/timer_tsc.c
--- a/arch/i386/kernel/timers/timer_tsc.c Thu Oct 10 11:21:10 2002
+++ b/arch/i386/kernel/timers/timer_tsc.c Thu Oct 10 11:21:10 2002
@@ -1,3 +1,20 @@
+/*
+ * This code largely moved from arch/i386/kernel/time.c.
+ * See comments there for proper credits.
+ */
+
+#include <linux/spinlock.h>
+#include <linux/init.h>
+#include <linux/timex.h>
+#include <linux/errno.h>
+
+#include <asm/timer.h>
+#include <asm/io.h>
+
+extern int x86_udelay_tsc;
+extern spinlock_t i8253_lock;
+
+static int use_tsc;
 /* Number of usecs that the last interrupt was delayed */
 static int delay_at_last_interrupt;
 
@@ -10,7 +27,7 @@
  */
 unsigned long fast_gettimeoffset_quotient;
 
-static inline unsigned long do_fast_gettimeoffset(void)
+static unsigned long get_offset_tsc(void)
 {
         register unsigned long eax, edx;
 
@@ -39,36 +56,35 @@
         return delay_at_last_interrupt + edx;
 }
 
+static void mark_offset_tsc(void)
+{
+ int count;
+ /*
+ * It is important that these two operations happen almost at
+ * the same time. We do the RDTSC stuff first, since it's
+ * faster. To avoid any inconsistencies, we need interrupts
+ * disabled locally.
+ */
 
-
-if (use_tsc)
- {
- /*
- * It is important that these two operations happen almost at
- * the same time. We do the RDTSC stuff first, since it's
- * faster. To avoid any inconsistencies, we need interrupts
- * disabled locally.
- */
-
- /*
- * Interrupts are just disabled locally since the timer irq
- * has the SA_INTERRUPT flag set. -arca
- */
+ /*
+ * Interrupts are just disabled locally since the timer irq
+ * has the SA_INTERRUPT flag set. -arca
+ */
         
- /* read Pentium cycle counter */
+ /* read Pentium cycle counter */
 
- rdtscl(last_tsc_low);
+ rdtscl(last_tsc_low);
 
- spin_lock(&i8253_lock);
- outb_p(0x00, 0x43); /* latch the count ASAP */
+ spin_lock(&i8253_lock);
+ outb_p(0x00, 0x43); /* latch the count ASAP */
 
- count = inb_p(0x40); /* read the latched count */
- count |= inb(0x40) << 8;
- spin_unlock(&i8253_lock);
+ count = inb_p(0x40); /* read the latched count */
+ count |= inb(0x40) << 8;
+ spin_unlock(&i8253_lock);
 
- count = ((LATCH-1) - count) * TICK_SIZE;
- delay_at_last_interrupt = (count + LATCH/2) / LATCH;
- }
+ count = ((LATCH-1) - count) * TICK_SIZE;
+ delay_at_last_interrupt = (count + LATCH/2) / LATCH;
+}
 
 
 /* ------ Calibrate the TSC -------
@@ -83,7 +99,6 @@
 #define CALIBRATE_LATCH (5 * LATCH)
 #define CALIBRATE_TIME (5 * 1000020/HZ)
 
-#ifdef CONFIG_X86_TSC
 static unsigned long __init calibrate_tsc(void)
 {
        /* Set the Gate high, disable speaker */
@@ -148,7 +163,6 @@
 bad_ctc:
         return 0;
 }
-#endif /* CONFIG_X86_TSC */
 
 
 #ifdef CONFIG_CPU_FREQ
@@ -196,26 +210,22 @@
 #endif
 
 
-
-#ifdef CONFIG_X86_TSC
- extern int x86_udelay_tsc;
-#endif
-
-/*
- * If we have APM enabled or the CPU clock speed is variable
- * (CPU stops clock on HLT or slows clock to save power)
- * then the TSC timestamps may diverge by up to 1 jiffy from
- * 'real time' but nothing will break.
- * The most frequent case is that the CPU is "woken" from a halt
- * state by the timer interrupt itself, so we get 0 error. In the
- * rare cases where a driver would "wake" the CPU and request a
- * timestamp, the maximum error is < 1 jiffy. But timestamps are
- * still perfectly ordered.
- * Note that the TSC counter will be reset if APM suspends
- * to disk; this won't break the kernel, though, 'cuz we're
- * smart. See arch/i386/kernel/apm.c.
- */
-#ifdef CONFIG_X86_TSC
+static int init_tsc(void)
+{
+ /*
+ * If we have APM enabled or the CPU clock speed is variable
+ * (CPU stops clock on HLT or slows clock to save power)
+ * then the TSC timestamps may diverge by up to 1 jiffy from
+ * 'real time' but nothing will break.
+ * The most frequent case is that the CPU is "woken" from a halt
+ * state by the timer interrupt itself, so we get 0 error. In the
+ * rare cases where a driver would "wake" the CPU and request a
+ * timestamp, the maximum error is < 1 jiffy. But timestamps are
+ * still perfectly ordered.
+ * Note that the TSC counter will be reset if APM suspends
+ * to disk; this won't break the kernel, though, 'cuz we're
+ * smart. See arch/i386/kernel/apm.c.
+ */
          /*
           * Firstly we have to do a CPU check for chips with
           * a potentially buggy TSC. At this point we haven't run
@@ -239,9 +249,6 @@
                          * and just enable this for the next intel chips ?
                          */
                         x86_udelay_tsc = 1;
-#ifndef do_gettimeoffset
- do_gettimeoffset = do_fast_gettimeoffset;
-#endif
 
                         /* report CPU clock rate in Hz.
                          * The formula is (10^6 * 2^32) / (2^32 * 1 / (clocks/us)) =
@@ -257,6 +264,17 @@
 #ifdef CONFIG_CPU_FREQ
                         cpufreq_register_notifier(&time_cpufreq_notifier_block, CPUFREQ_TRANSITION_NOTIFIER);
 #endif
+ return 0;
                 }
         }
-#endif /* CONFIG_X86_TSC */
+ return -ENODEV;
+}
+
+/************************************************************/
+
+/* tsc timer_opts struct */
+struct timer_opts timer_tsc = {
+ .init = init_tsc,
+ .mark_offset = mark_offset_tsc,
+ .get_offset = get_offset_tsc,
+};
diff -Nru a/include/asm-i386/timer.h b/include/asm-i386/timer.h
--- a/include/asm-i386/timer.h Thu Oct 10 11:21:10 2002
+++ b/include/asm-i386/timer.h Thu Oct 10 11:21:10 2002
@@ -16,5 +16,7 @@
         unsigned long (*get_offset)(void);
 };
 
+#define TICK_SIZE (tick_nsec / 1000)
+
 extern struct timer_opts* select_timer(void);
 #endif
-
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/



This archive was generated by hypermail 2b29 : Tue Oct 15 2002 - 22:00:38 EST