rpms/kernel/F-7 linux-2.6-ppc-pegasos-keyboard-8042-check.patch, NONE, 1.1 linux-2.6-sched-cfs-v2.6.22.5-v20.2.patch, NONE, 1.1 linux-2.6-timekeeping-fixes.patch, NONE, 1.1 patch-2.6.22.5.bz2.sign, NONE, 1.1 .cvsignore, 1.630, 1.631 kernel-2.6.spec, 1.3326, 1.3327 sources, 1.593, 1.594 upstream, 1.517, 1.518 linux-2.6-sched-cfs.patch, 1.11, NONE patch-2.6.22.4.bz2.sign, 1.1, NONE patch-2.6.22.5-rc1.bz2, 1.1, NONE
Chuck Ebbert (cebbert)
fedora-extras-commits at redhat.com
Thu Aug 23 18:21:28 UTC 2007
Author: cebbert
Update of /cvs/pkgs/rpms/kernel/F-7
In directory cvs-int.fedora.redhat.com:/tmp/cvs-serv5133
Modified Files:
.cvsignore kernel-2.6.spec sources upstream
Added Files:
linux-2.6-ppc-pegasos-keyboard-8042-check.patch
linux-2.6-sched-cfs-v2.6.22.5-v20.2.patch
linux-2.6-timekeeping-fixes.patch patch-2.6.22.5.bz2.sign
Removed Files:
linux-2.6-sched-cfs.patch patch-2.6.22.4.bz2.sign
patch-2.6.22.5-rc1.bz2
Log Message:
* Thu Aug 23 2007 Chuck Ebbert <cebbert at redhat.com>
- Linux 2.6.22.5
- CFS scheduler v20.2
- fix Pegasos PS/2 port detection
linux-2.6-ppc-pegasos-keyboard-8042-check.patch:
--- NEW FILE linux-2.6-ppc-pegasos-keyboard-8042-check.patch ---
commit f5d834fc34e61f1a40435981062000e5d2b2baa8
tree 3ed0b4a2ba4870b733e66483b01d69903142aa59
parent 16782a604c458e1edcefca52457a82395b788bed
author Alan Curry <pacman at theworld.com> 1185326912 +1000
committer Paul Mackerras <paulus at samba.org> 1185430665 +1000
[POWERPC] Fix Pegasos keyboard detection
As of 2.6.22 the kernel doesn't recognize the i8042 keyboard/mouse
controller on the PegasosPPC. This is because of a feature/bug in the
OF device tree: the "device_type" attribute is an empty string instead
of "8042" as the kernel expects. This adds a secondary detection
which looks for a device whose *name* is "8042" if there is no device
whose *type* is "8042".
Signed-off-by: Alan Curry <pacman at world.std.com>
Acked-by: Benjamin Herrenschmidt <benh at kernel.crashing.org>
Signed-off-by: Paul Mackerras <paulus at samba.org>
---
arch/powerpc/kernel/setup-common.c | 4 ++++
1 files changed, 4 insertions(+)
--- linux-2.6.22.orig/arch/powerpc/kernel/setup-common.c
+++ linux-2.6.22/arch/powerpc/kernel/setup-common.c
@@ -487,6 +487,10 @@ int check_legacy_ioport(unsigned long ba
switch(base_port) {
case I8042_DATA_REG:
np = of_find_node_by_type(NULL, "8042");
+ /* Pegasos has no device_type on its 8042 node, look for the
+ * name instead */
+ if (!np)
+ np = of_find_node_by_name(NULL, "8042");
break;
case FDC_BASE: /* FDC1 */
np = of_find_node_by_type(NULL, "fdc");
linux-2.6-sched-cfs-v2.6.22.5-v20.2.patch:
--- NEW FILE linux-2.6-sched-cfs-v2.6.22.5-v20.2.patch ---
Index: linux-cfs-2.6.22.5.q/Documentation/kernel-parameters.txt
===================================================================
--- linux-cfs-2.6.22.5.q.orig/Documentation/kernel-parameters.txt
+++ linux-cfs-2.6.22.5.q/Documentation/kernel-parameters.txt
@@ -1009,49 +1009,6 @@ and is between 256 and 4096 characters.
mga= [HW,DRM]
- migration_cost=
- [KNL,SMP] debug: override scheduler migration costs
- Format: <level-1-usecs>,<level-2-usecs>,...
- This debugging option can be used to override the
- default scheduler migration cost matrix. The numbers
- are indexed by 'CPU domain distance'.
- E.g. migration_cost=1000,2000,3000 on an SMT NUMA
- box will set up an intra-core migration cost of
- 1 msec, an inter-core migration cost of 2 msecs,
- and an inter-node migration cost of 3 msecs.
-
- WARNING: using the wrong values here can break
- scheduler performance, so it's only for scheduler
- development purposes, not production environments.
-
- migration_debug=
- [KNL,SMP] migration cost auto-detect verbosity
- Format=<0|1|2>
- If a system's migration matrix reported at bootup
- seems erroneous then this option can be used to
- increase verbosity of the detection process.
- We default to 0 (no extra messages), 1 will print
- some more information, and 2 will be really
- verbose (probably only useful if you also have a
- serial console attached to the system).
-
- migration_factor=
- [KNL,SMP] multiply/divide migration costs by a factor
- Format=<percent>
- This debug option can be used to proportionally
- increase or decrease the auto-detected migration
- costs for all entries of the migration matrix.
- E.g. migration_factor=150 will increase migration
- costs by 50%. (and thus the scheduler will be less
- eager migrating cache-hot tasks)
- migration_factor=80 will decrease migration costs
- by 20%. (thus the scheduler will be more eager to
- migrate tasks)
-
- WARNING: using the wrong values here can break
- scheduler performance, so it's only for scheduler
- development purposes, not production environments.
-
mousedev.tap_time=
[MOUSE] Maximum time between finger touching and
leaving touchpad surface for touch to be considered
Index: linux-cfs-2.6.22.5.q/Documentation/sched-design-CFS.txt
===================================================================
--- /dev/null
+++ linux-cfs-2.6.22.5.q/Documentation/sched-design-CFS.txt
@@ -0,0 +1,119 @@
+
+This is the CFS scheduler.
+
+80% of CFS's design can be summed up in a single sentence: CFS basically
+models an "ideal, precise multi-tasking CPU" on real hardware.
+
+"Ideal multi-tasking CPU" is a (non-existent :-)) CPU that has 100%
+physical power and which can run each task at precise equal speed, in
+parallel, each at 1/nr_running speed. For example: if there are 2 tasks
+running then it runs each at 50% physical power - totally in parallel.
+
+On real hardware, we can run only a single task at once, so while that
+one task runs, the other tasks that are waiting for the CPU are at a
+disadvantage - the current task gets an unfair amount of CPU time. In
+CFS this fairness imbalance is expressed and tracked via the per-task
+p->wait_runtime (nanosec-unit) value. "wait_runtime" is the amount of
+time the task should now run on the CPU for it to become completely fair
+and balanced.
+
+( small detail: on 'ideal' hardware, the p->wait_runtime value would
+ always be zero - no task would ever get 'out of balance' from the
+ 'ideal' share of CPU time. )
+
+CFS's task picking logic is based on this p->wait_runtime value and it
+is thus very simple: it always tries to run the task with the largest
+p->wait_runtime value. In other words, CFS tries to run the task with
+the 'gravest need' for more CPU time. So CFS always tries to split up
+CPU time between runnable tasks as close to 'ideal multitasking
+hardware' as possible.
+
+Most of the rest of CFS's design just falls out of this really simple
+concept, with a few add-on embellishments like nice levels,
+multiprocessing and various algorithm variants to recognize sleepers.
+
+In practice it works like this: the system runs a task a bit, and when
+the task schedules (or a scheduler tick happens) the task's CPU usage is
+'accounted for': the (small) time it just spent using the physical CPU
+is deducted from p->wait_runtime. [minus the 'fair share' it would have
+gotten anyway]. Once p->wait_runtime gets low enough so that another
+task becomes the 'leftmost task' of the time-ordered rbtree it maintains
+(plus a small amount of 'granularity' distance relative to the leftmost
+task so that we do not over-schedule tasks and trash the cache) then the
+new leftmost task is picked and the current task is preempted.
+
+The rq->fair_clock value tracks the 'CPU time a runnable task would have
+fairly gotten, had it been runnable during that time'. So by using
+rq->fair_clock values we can accurately timestamp and measure the
+'expected CPU time' a task should have gotten. All runnable tasks are
+sorted in the rbtree by the "rq->fair_clock - p->wait_runtime" key, and
+CFS picks the 'leftmost' task and sticks to it. As the system progresses
+forwards, newly woken tasks are put into the tree more and more to the
+right - slowly but surely giving a chance for every task to become the
+'leftmost task' and thus get on the CPU within a deterministic amount of
+time.
+
+Some implementation details:
+
+ - the introduction of Scheduling Classes: an extensible hierarchy of
+ scheduler modules. These modules encapsulate scheduling policy
+ details and are handled by the scheduler core without the core
+ code assuming about them too much.
+
+ - sched_fair.c implements the 'CFS desktop scheduler': it is a
+ replacement for the vanilla scheduler's SCHED_OTHER interactivity
+ code.
+
+ I'd like to give credit to Con Kolivas for the general approach here:
+ he has proven via RSDL/SD that 'fair scheduling' is possible and that
+ it results in better desktop scheduling. Kudos Con!
+
+ The CFS patch uses a completely different approach and implementation
+ from RSDL/SD. My goal was to make CFS's interactivity quality exceed
+ that of RSDL/SD, which is a high standard to meet :-) Testing
+ feedback is welcome to decide this one way or another. [ and, in any
+ case, all of SD's logic could be added via a kernel/sched_sd.c module
+ as well, if Con is interested in such an approach. ]
+
+ CFS's design is quite radical: it does not use runqueues, it uses a
+ time-ordered rbtree to build a 'timeline' of future task execution,
+ and thus has no 'array switch' artifacts (by which both the vanilla
+ scheduler and RSDL/SD are affected).
+
+ CFS uses nanosecond granularity accounting and does not rely on any
+ jiffies or other HZ detail. Thus the CFS scheduler has no notion of
+ 'timeslices' and has no heuristics whatsoever. There is only one
+ central tunable:
+
+ /proc/sys/kernel/sched_granularity_ns
+
+ which can be used to tune the scheduler from 'desktop' (low
+ latencies) to 'server' (good batching) workloads. It defaults to a
+ setting suitable for desktop workloads. SCHED_BATCH is handled by the
+ CFS scheduler module too.
+
+ Due to its design, the CFS scheduler is not prone to any of the
+ 'attacks' that exist today against the heuristics of the stock
+ scheduler: fiftyp.c, thud.c, chew.c, ring-test.c, massive_intr.c all
+ work fine and do not impact interactivity and produce the expected
+ behavior.
+
+ the CFS scheduler has a much stronger handling of nice levels and
+ SCHED_BATCH: both types of workloads should be isolated much more
+ agressively than under the vanilla scheduler.
+
+ ( another detail: due to nanosec accounting and timeline sorting,
+ sched_yield() support is very simple under CFS, and in fact under
+ CFS sched_yield() behaves much better than under any other
+ scheduler i have tested so far. )
+
+ - sched_rt.c implements SCHED_FIFO and SCHED_RR semantics, in a simpler
+ way than the vanilla scheduler does. It uses 100 runqueues (for all
+ 100 RT priority levels, instead of 140 in the vanilla scheduler)
+ and it needs no expired array.
+
+ - reworked/sanitized SMP load-balancing: the runqueue-walking
+ assumptions are gone from the load-balancing code now, and
+ iterators of the scheduling modules are used. The balancing code got
+ quite a bit simpler as a result.
+
Index: linux-cfs-2.6.22.5.q/arch/i386/kernel/smpboot.c
===================================================================
--- linux-cfs-2.6.22.5.q.orig/arch/i386/kernel/smpboot.c
+++ linux-cfs-2.6.22.5.q/arch/i386/kernel/smpboot.c
@@ -941,17 +941,6 @@ exit:
}
#endif
-static void smp_tune_scheduling(void)
-{
- if (cpu_khz) {
- /* cache size in kB */
- long cachesize = boot_cpu_data.x86_cache_size;
-
- if (cachesize > 0)
- max_cache_size = cachesize * 1024;
- }
-}
-
/*
* Cycle through the processors sending APIC IPIs to boot each.
[...8403 lines suppressed...]
+{
+ unsigned long long now = sched_clock(), delta = 0;
+
+ if (t->sched_info.last_queued)
+ delta = now - t->sched_info.last_queued;
+ sched_info_dequeued(t);
+ t->sched_info.run_delay += delta;
+ t->sched_info.last_arrival = now;
+ t->sched_info.pcnt++;
+
+ rq_sched_info_arrive(task_rq(t), delta);
+}
+
+/*
+ * Called when a process is queued into either the active or expired
+ * array. The time is noted and later used to determine how long we
+ * had to wait for us to reach the cpu. Since the expired queue will
+ * become the active queue after active queue is empty, without dequeuing
+ * and requeuing any tasks, we are interested in queuing to either. It
+ * is unusual but not impossible for tasks to be dequeued and immediately
+ * requeued in the same or another array: this can happen in sched_yield(),
+ * set_user_nice(), and even load_balance() as it moves tasks from runqueue
+ * to runqueue.
+ *
+ * This function is only called from enqueue_task(), but also only updates
+ * the timestamp if it is already not set. It's assumed that
+ * sched_info_dequeued() will clear that stamp when appropriate.
+ */
+static inline void sched_info_queued(struct task_struct *t)
+{
+ if (unlikely(sched_info_on()))
+ if (!t->sched_info.last_queued)
+ t->sched_info.last_queued = sched_clock();
+}
+
+/*
+ * Called when a process ceases being the active-running process, either
+ * voluntarily or involuntarily. Now we can calculate how long we ran.
+ */
+static inline void sched_info_depart(struct task_struct *t)
+{
+ unsigned long long delta = sched_clock() - t->sched_info.last_arrival;
+
+ t->sched_info.cpu_time += delta;
+ rq_sched_info_depart(task_rq(t), delta);
+}
+
+/*
+ * Called when tasks are switched involuntarily due, typically, to expiring
+ * their time slice. (This may also be called when switching to or from
+ * the idle task.) We are only called when prev != next.
+ */
+static inline void
+__sched_info_switch(struct task_struct *prev, struct task_struct *next)
+{
+ struct rq *rq = task_rq(prev);
+
+ /*
+ * prev now departs the cpu. It's not interesting to record
+ * stats about how efficient we were at scheduling the idle
+ * process, however.
+ */
+ if (prev != rq->idle)
+ sched_info_depart(prev);
+
+ if (next != rq->idle)
+ sched_info_arrive(next);
+}
+static inline void
+sched_info_switch(struct task_struct *prev, struct task_struct *next)
+{
+ if (unlikely(sched_info_on()))
+ __sched_info_switch(prev, next);
+}
+#else
+#define sched_info_queued(t) do { } while (0)
+#define sched_info_switch(t, next) do { } while (0)
+#endif /* CONFIG_SCHEDSTATS || CONFIG_TASK_DELAY_ACCT */
+
Index: linux-cfs-2.6.22.5.q/kernel/softirq.c
===================================================================
--- linux-cfs-2.6.22.5.q.orig/kernel/softirq.c
+++ linux-cfs-2.6.22.5.q/kernel/softirq.c
@@ -488,7 +488,6 @@ void __init softirq_init(void)
static int ksoftirqd(void * __bind_cpu)
{
- set_user_nice(current, 19);
current->flags |= PF_NOFREEZE;
set_current_state(TASK_INTERRUPTIBLE);
Index: linux-cfs-2.6.22.5.q/kernel/sysctl.c
===================================================================
--- linux-cfs-2.6.22.5.q.orig/kernel/sysctl.c
+++ linux-cfs-2.6.22.5.q/kernel/sysctl.c
@@ -206,8 +206,84 @@ static ctl_table root_table[] = {
{ .ctl_name = 0 }
};
+static unsigned long min_sched_granularity_ns = 100000; /* 100 usecs */
+static unsigned long max_sched_granularity_ns = 1000000000; /* 1 second */
+static unsigned long min_wakeup_granularity_ns; /* 0 usecs */
+static unsigned long max_wakeup_granularity_ns = 1000000000; /* 1 second */
+
static ctl_table kern_table[] = {
{
+ .ctl_name = CTL_UNNUMBERED,
+ .procname = "sched_granularity_ns",
+ .data = &sysctl_sched_granularity,
+ .maxlen = sizeof(unsigned int),
+ .mode = 0644,
+ .proc_handler = &proc_dointvec_minmax,
+ .strategy = &sysctl_intvec,
+ .extra1 = &min_sched_granularity_ns,
+ .extra2 = &max_sched_granularity_ns,
+ },
+ {
+ .ctl_name = CTL_UNNUMBERED,
+ .procname = "sched_wakeup_granularity_ns",
+ .data = &sysctl_sched_wakeup_granularity,
+ .maxlen = sizeof(unsigned int),
+ .mode = 0644,
+ .proc_handler = &proc_dointvec_minmax,
+ .strategy = &sysctl_intvec,
+ .extra1 = &min_wakeup_granularity_ns,
+ .extra2 = &max_wakeup_granularity_ns,
+ },
+ {
+ .ctl_name = CTL_UNNUMBERED,
+ .procname = "sched_batch_wakeup_granularity_ns",
+ .data = &sysctl_sched_batch_wakeup_granularity,
+ .maxlen = sizeof(unsigned int),
+ .mode = 0644,
+ .proc_handler = &proc_dointvec_minmax,
+ .strategy = &sysctl_intvec,
+ .extra1 = &min_wakeup_granularity_ns,
+ .extra2 = &max_wakeup_granularity_ns,
+ },
+ {
+ .ctl_name = CTL_UNNUMBERED,
+ .procname = "sched_stat_granularity_ns",
+ .data = &sysctl_sched_stat_granularity,
+ .maxlen = sizeof(unsigned int),
+ .mode = 0644,
+ .proc_handler = &proc_dointvec_minmax,
+ .strategy = &sysctl_intvec,
+ .extra1 = &min_wakeup_granularity_ns,
+ .extra2 = &max_wakeup_granularity_ns,
+ },
+ {
+ .ctl_name = CTL_UNNUMBERED,
+ .procname = "sched_runtime_limit_ns",
+ .data = &sysctl_sched_runtime_limit,
+ .maxlen = sizeof(unsigned int),
+ .mode = 0644,
+ .proc_handler = &proc_dointvec_minmax,
+ .strategy = &sysctl_intvec,
+ .extra1 = &min_sched_granularity_ns,
+ .extra2 = &max_sched_granularity_ns,
+ },
+ {
+ .ctl_name = CTL_UNNUMBERED,
+ .procname = "sched_child_runs_first",
+ .data = &sysctl_sched_child_runs_first,
+ .maxlen = sizeof(unsigned int),
+ .mode = 0644,
+ .proc_handler = &proc_dointvec,
+ },
+ {
+ .ctl_name = CTL_UNNUMBERED,
+ .procname = "sched_features",
+ .data = &sysctl_sched_features,
+ .maxlen = sizeof(unsigned int),
+ .mode = 0644,
+ .proc_handler = &proc_dointvec,
+ },
+ {
.ctl_name = KERN_PANIC,
.procname = "panic",
.data = &panic_timeout,
Index: linux-cfs-2.6.22.5.q/lib/Kconfig.debug
===================================================================
--- linux-cfs-2.6.22.5.q.orig/lib/Kconfig.debug
+++ linux-cfs-2.6.22.5.q/lib/Kconfig.debug
@@ -105,6 +105,15 @@ config DETECT_SOFTLOCKUP
can be detected via the NMI-watchdog, on platforms that
support it.)
+config SCHED_DEBUG
+ bool "Collect scheduler debugging info"
+ depends on DEBUG_KERNEL && PROC_FS
+ default y
+ help
+ If you say Y here, the /proc/sched_debug file will be provided
+ that can help debug the scheduler. The runtime overhead of this
+ option is minimal.
+
config SCHEDSTATS
bool "Collect scheduler statistics"
depends on DEBUG_KERNEL && PROC_FS
linux-2.6-timekeeping-fixes.patch:
--- NEW FILE linux-2.6-timekeeping-fixes.patch ---
Gitweb: http://git.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=2c6b47de17c75d553de3e2fb426d8298d2074585
Commit: 2c6b47de17c75d553de3e2fb426d8298d2074585
Parent: 0de085bb474f64e4fdb2f1ff3268590792648c7b
Author: john stultz <johnstul at us.ibm.com>
AuthorDate: Tue Jul 24 17:47:43 2007 -0700
Committer: Linus Torvalds <torvalds at woody.linux-foundation.org>
CommitDate: Wed Jul 25 10:09:20 2007 -0700
Cleanup non-arch xtime uses, use get_seconds() or current_kernel_time().
This avoids use of the kernel-internal "xtime" variable directly outside
of the actual time-related functions. Instead, use the helper functions
that we already have available to us.
This doesn't actually change any behaviour, but this will allow us to
fix the fact that "xtime" isn't updated very often with CONFIG_NO_HZ
(because much of the realtime information is maintained as separate
offsets to 'xtime'), which has caused interfaces that use xtime directly
to get a time that is out of sync with the real-time clock by up to a
third of a second or so.
Signed-off-by: John Stultz <johnstul at us.ibm.com>
Cc: Ingo Molnar <mingo at elte.hu>
Cc: Thomas Gleixner <tglx at linutronix.de>
Signed-off-by: Linus Torvalds <torvalds at linux-foundation.org>
---
drivers/rtc/class.c | 5 +++--
drivers/s390/net/ctcmain.c | 6 +++---
drivers/s390/net/netiucv.c | 4 ++--
include/linux/time.h | 2 +-
kernel/acct.c | 2 +-
kernel/hrtimer.c | 2 +-
kernel/time.c | 16 ----------------
kernel/time/timekeeping.c | 16 ++++++++++++++++
kernel/tsacct.c | 2 +-
net/rxrpc/af_rxrpc.c | 2 +-
net/rxrpc/ar-connection.c | 4 ++--
net/rxrpc/ar-transport.c | 4 ++--
net/rxrpc/rxkad.c | 2 +-
13 files changed, 34 insertions(+), 33 deletions(-)
diff --git a/drivers/rtc/class.c b/drivers/rtc/class.c
index 8b3cd31..10ab3b7 100644
--- a/drivers/rtc/class.c
+++ b/drivers/rtc/class.c
@@ -46,6 +46,7 @@ static int rtc_suspend(struct device *dev, pm_message_t mesg)
{
struct rtc_device *rtc = to_rtc_device(dev);
struct rtc_time tm;
+ struct timespec ts = current_kernel_time();
if (strncmp(rtc->dev.bus_id,
CONFIG_RTC_HCTOSYS_DEVICE,
@@ -57,8 +58,8 @@ static int rtc_suspend(struct device *dev, pm_message_t mesg)
/* RTC precision is 1 second; adjust delta for avg 1/2 sec err */
set_normalized_timespec(&delta,
- xtime.tv_sec - oldtime,
- xtime.tv_nsec - (NSEC_PER_SEC >> 1));
+ ts.tv_sec - oldtime,
+ ts.tv_nsec - (NSEC_PER_SEC >> 1));
return 0;
}
diff --git a/drivers/s390/net/ctcmain.c b/drivers/s390/net/ctcmain.c
index b20fd06..92e8a37 100644
--- a/drivers/s390/net/ctcmain.c
+++ b/drivers/s390/net/ctcmain.c
@@ -674,7 +674,7 @@ ch_action_txdone(fsm_instance * fi, int event, void *arg)
int first = 1;
int i;
unsigned long duration;
- struct timespec done_stamp = xtime;
+ struct timespec done_stamp = current_kernel_time();
DBF_TEXT(trace, 4, __FUNCTION__);
@@ -730,7 +730,7 @@ ch_action_txdone(fsm_instance * fi, int event, void *arg)
spin_unlock(&ch->collect_lock);
ch->ccw[1].count = ch->trans_skb->len;
fsm_addtimer(&ch->timer, CTC_TIMEOUT_5SEC, CH_EVENT_TIMER, ch);
- ch->prof.send_stamp = xtime;
+ ch->prof.send_stamp = current_kernel_time();
rc = ccw_device_start(ch->cdev, &ch->ccw[0],
(unsigned long) ch, 0xff, 0);
ch->prof.doios_multi++;
@@ -2281,7 +2281,7 @@ transmit_skb(struct channel *ch, struct sk_buff *skb)
fsm_newstate(ch->fsm, CH_STATE_TX);
fsm_addtimer(&ch->timer, CTC_TIMEOUT_5SEC, CH_EVENT_TIMER, ch);
spin_lock_irqsave(get_ccwdev_lock(ch->cdev), saveflags);
- ch->prof.send_stamp = xtime;
+ ch->prof.send_stamp = current_kernel_time();
rc = ccw_device_start(ch->cdev, &ch->ccw[ccw_idx],
(unsigned long) ch, 0xff, 0);
spin_unlock_irqrestore(get_ccwdev_lock(ch->cdev), saveflags);
diff --git a/drivers/s390/net/netiucv.c b/drivers/s390/net/netiucv.c
index 3d28e1a..2688894 100644
--- a/drivers/s390/net/netiucv.c
+++ b/drivers/s390/net/netiucv.c
@@ -753,7 +753,7 @@ static void conn_action_txdone(fsm_instance *fi, int event, void *arg)
header.next = 0;
memcpy(skb_put(conn->tx_buff, NETIUCV_HDRLEN), &header, NETIUCV_HDRLEN);
- conn->prof.send_stamp = xtime;
+ conn->prof.send_stamp = current_kernel_time();
txmsg.class = 0;
txmsg.tag = 0;
rc = iucv_message_send(conn->path, &txmsg, 0, 0,
@@ -1185,7 +1185,7 @@ static int netiucv_transmit_skb(struct iucv_connection *conn,
memcpy(skb_put(nskb, NETIUCV_HDRLEN), &header, NETIUCV_HDRLEN);
fsm_newstate(conn->fsm, CONN_STATE_TX);
- conn->prof.send_stamp = xtime;
+ conn->prof.send_stamp = current_kernel_time();
msg.tag = 1;
msg.class = 0;
diff --git a/include/linux/time.h b/include/linux/time.h
index e6aea51..71181df 100644
--- a/include/linux/time.h
+++ b/include/linux/time.h
@@ -107,7 +107,7 @@ static inline unsigned long get_seconds(void)
struct timespec current_kernel_time(void);
#define CURRENT_TIME (current_kernel_time())
-#define CURRENT_TIME_SEC ((struct timespec) { xtime.tv_sec, 0 })
+#define CURRENT_TIME_SEC ((struct timespec) { get_seconds(), 0 })
extern void do_gettimeofday(struct timeval *tv);
extern int do_settimeofday(struct timespec *tv);
diff --git a/kernel/acct.c b/kernel/acct.c
index 70d0d88..24f0f8b 100644
--- a/kernel/acct.c
+++ b/kernel/acct.c
@@ -468,7 +468,7 @@ static void do_acct_process(struct file *file)
}
#endif
do_div(elapsed, AHZ);
- ac.ac_btime = xtime.tv_sec - elapsed;
+ ac.ac_btime = get_seconds() - elapsed;
/* we really need to bite the bullet and change layout */
ac.ac_uid = current->uid;
ac.ac_gid = current->gid;
diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c
index eb1ddeb..a7bb05e 100644
--- a/kernel/hrtimer.c
+++ b/kernel/hrtimer.c
@@ -144,7 +144,7 @@ static void hrtimer_get_softirq_time(struct hrtimer_cpu_base *base)
#ifdef CONFIG_NO_HZ
getnstimeofday(&xts);
#else
- xts = xtime;
+ xts = current_kernel_time();
#endif
tom = wall_to_monotonic;
} while (read_seqretry(&xtime_lock, seq));
diff --git a/kernel/time.c b/kernel/time.c
index 5b81da0..2289a8d 100644
--- a/kernel/time.c
+++ b/kernel/time.c
@@ -215,22 +215,6 @@ asmlinkage long sys_adjtimex(struct timex __user *txc_p)
return copy_to_user(txc_p, &txc, sizeof(struct timex)) ? -EFAULT : ret;
}
-inline struct timespec current_kernel_time(void)
-{
- struct timespec now;
- unsigned long seq;
-
- do {
- seq = read_seqbegin(&xtime_lock);
-
- now = xtime;
- } while (read_seqretry(&xtime_lock, seq));
-
- return now;
-}
-
-EXPORT_SYMBOL(current_kernel_time);
-
/**
* current_fs_time - Return FS time
* @sb: Superblock.
diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c
index 88c8102..07a3f14 100644
--- a/kernel/time/timekeeping.c
+++ b/kernel/time/timekeeping.c
@@ -509,3 +509,19 @@ void monotonic_to_bootbased(struct timespec *ts)
change_clocksource();
update_vsyscall(&xtime, clock);
}
+
+struct timespec current_kernel_time(void)
+{
+ struct timespec now;
+ unsigned long seq;
+
+ do {
+ seq = read_seqbegin(&xtime_lock);
+
+ now = xtime;
+ } while (read_seqretry(&xtime_lock, seq));
+
+ return now;
+}
+
+EXPORT_SYMBOL(current_kernel_time);
diff --git a/kernel/tsacct.c b/kernel/tsacct.c
index 658f638..c122131 100644
--- a/kernel/tsacct.c
+++ b/kernel/tsacct.c
@@ -39,7 +39,7 @@ void bacct_add_tsk(struct taskstats *stats, struct task_struct *tsk)
ac_etime = timespec_to_ns(&ts);
do_div(ac_etime, NSEC_PER_USEC);
stats->ac_etime = ac_etime;
- stats->ac_btime = xtime.tv_sec - ts.tv_sec;
+ stats->ac_btime = get_seconds() - ts.tv_sec;
if (thread_group_leader(tsk)) {
stats->ac_exitcode = tsk->exit_code;
if (tsk->flags & PF_FORKNOEXEC)
diff --git a/net/rxrpc/af_rxrpc.c b/net/rxrpc/af_rxrpc.c
index 16a68df..c58fa0d 100644
--- a/net/rxrpc/af_rxrpc.c
+++ b/net/rxrpc/af_rxrpc.c
@@ -787,7 +787,7 @@ static int __init af_rxrpc_init(void)
BUILD_BUG_ON(sizeof(struct rxrpc_skb_priv) > sizeof(dummy_skb->cb));
- rxrpc_epoch = htonl(xtime.tv_sec);
+ rxrpc_epoch = htonl(get_seconds());
ret = -ENOMEM;
rxrpc_call_jar = kmem_cache_create(
diff --git a/net/rxrpc/ar-connection.c b/net/rxrpc/ar-connection.c
index 482750e..372b244 100644
--- a/net/rxrpc/ar-connection.c
+++ b/net/rxrpc/ar-connection.c
@@ -791,7 +791,7 @@ void rxrpc_put_connection(struct rxrpc_connection *conn)
ASSERTCMP(atomic_read(&conn->usage), >, 0);
- conn->put_time = xtime.tv_sec;
+ conn->put_time = get_seconds();
if (atomic_dec_and_test(&conn->usage)) {
_debug("zombie");
rxrpc_queue_delayed_work(&rxrpc_connection_reap, 0);
@@ -835,7 +835,7 @@ void rxrpc_connection_reaper(struct work_struct *work)
_enter("");
- now = xtime.tv_sec;
+ now = get_seconds();
earliest = ULONG_MAX;
write_lock_bh(&rxrpc_connection_lock);
diff --git a/net/rxrpc/ar-transport.c b/net/rxrpc/ar-transport.c
index d43d78f..bb282a6 100644
--- a/net/rxrpc/ar-transport.c
+++ b/net/rxrpc/ar-transport.c
@@ -183,7 +183,7 @@ void rxrpc_put_transport(struct rxrpc_transport *trans)
ASSERTCMP(atomic_read(&trans->usage), >, 0);
- trans->put_time = xtime.tv_sec;
+ trans->put_time = get_seconds();
if (unlikely(atomic_dec_and_test(&trans->usage)))
_debug("zombie");
/* let the reaper determine the timeout to avoid a race with
@@ -219,7 +219,7 @@ static void rxrpc_transport_reaper(struct work_struct *work)
_enter("");
- now = xtime.tv_sec;
+ now = get_seconds();
earliest = ULONG_MAX;
/* extract all the transports that have been dead too long */
diff --git a/net/rxrpc/rxkad.c b/net/rxrpc/rxkad.c
index 5ec7051..ac3cabd 100644
--- a/net/rxrpc/rxkad.c
+++ b/net/rxrpc/rxkad.c
@@ -916,7 +916,7 @@ static int rxkad_decrypt_ticket(struct rxrpc_connection *conn,
issue = be32_to_cpu(stamp);
}
p += 4;
- now = xtime.tv_sec;
+ now = get_seconds();
_debug("KIV ISSUE: %lx [%lx]", issue, now);
/* check the ticket is in date */
Gitweb: http://git.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=17c38b7490b3f0300c7812aefdae2ddda7ab4112
Commit: 17c38b7490b3f0300c7812aefdae2ddda7ab4112
Parent: 2c6b47de17c75d553de3e2fb426d8298d2074585
Author: john stultz <johnstul at us.ibm.com>
AuthorDate: Tue Jul 24 18:38:34 2007 -0700
Committer: Linus Torvalds <torvalds at woody.linux-foundation.org>
CommitDate: Wed Jul 25 10:17:44 2007 -0700
Cache xtime every call to update_wall_time
This avoids xtime lag seen with dynticks, because while 'xtime' itself
is still not updated often, we keep a 'xtime_cache' variable around that
contains the approximate real-time that _is_ updated each time we do a
'update_wall_time()', and is thus never off by more than one tick.
IOW, this restores the original semantics for 'xtime' users, as long as
you use the proper abstraction functions (ie 'current_kernel_time()' or
'get_seconds()' depending on whether you want a timespec or just the
seconds field).
[ Updated Patch. As penance for my sins I've also yanked another #ifdef
that was added to avoid the xtime lag w/ hrtimers. ]
Signed-off-by: John Stultz <johnstul at us.ibm.com>
Cc: Ingo Molnar <mingo at elte.hu>
Cc: Thomas Gleixner <tglx at linutronix.de>
Signed-off-by: Linus Torvalds <torvalds at linux-foundation.org>
---
include/linux/time.h | 6 +-----
kernel/hrtimer.c | 4 ----
kernel/time/timekeeping.c | 26 +++++++++++++++++++++++---
3 files changed, 24 insertions(+), 12 deletions(-)
diff --git a/include/linux/time.h b/include/linux/time.h
index 71181df..6a5f503 100644
--- a/include/linux/time.h
+++ b/include/linux/time.h
@@ -99,11 +99,7 @@ extern int update_persistent_clock(struct timespec now);
extern int no_sync_cmos_clock __read_mostly;
void timekeeping_init(void);
-static inline unsigned long get_seconds(void)
-{
- return xtime.tv_sec;
-}
-
+unsigned long get_seconds(void);
struct timespec current_kernel_time(void);
#define CURRENT_TIME (current_kernel_time())
diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c
index a7bb05e..c21ca6b 100644
--- a/kernel/hrtimer.c
+++ b/kernel/hrtimer.c
@@ -141,11 +141,7 @@ static void hrtimer_get_softirq_time(struct hrtimer_cpu_base *base)
do {
seq = read_seqbegin(&xtime_lock);
-#ifdef CONFIG_NO_HZ
- getnstimeofday(&xts);
-#else
xts = current_kernel_time();
-#endif
tom = wall_to_monotonic;
} while (read_seqretry(&xtime_lock, seq));
diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c
index 07a3f14..acc417b 100644
--- a/kernel/time/timekeeping.c
+++ b/kernel/time/timekeeping.c
@@ -47,6 +47,19 @@ EXPORT_SYMBOL(xtime_lock);
EXPORT_SYMBOL(xtime);
+#ifdef CONFIG_NO_HZ
+static struct timespec xtime_cache __attribute__ ((aligned (16)));
+static inline void update_xtime_cache(u64 nsec)
+{
+ xtime_cache = xtime;
+ timespec_add_ns(&xtime_cache, nsec);
+}
+#else
+#define xtime_cache xtime
+/* We do *not* want to evaluate the argument for this case */
+#define update_xtime_cache(n) do { } while (0)
+#endif
+
static struct clocksource *clock; /* pointer to current clocksource */
@@ -478,6 +490,8 @@ void update_wall_time(void)
xtime.tv_nsec = (s64)clock->xtime_nsec >> clock->shift;
clock->xtime_nsec -= (s64)xtime.tv_nsec << clock->shift;
+ update_xtime_cache(cyc2ns(clock, offset));
+
/* check to see if there is a new clocksource to use */
change_clocksource();
update_vsyscall(&xtime, clock);
@@ -510,6 +524,13 @@ void monotonic_to_bootbased(struct timespec *ts)
ts->tv_sec += total_sleep_time;
}
+unsigned long get_seconds(void)
+{
+ return xtime_cache.tv_sec;
+}
+EXPORT_SYMBOL(get_seconds);
+
+
struct timespec current_kernel_time(void)
{
struct timespec now;
@@ -518,10 +539,9 @@ struct timespec current_kernel_time(void)
do {
seq = read_seqbegin(&xtime_lock);
- now = xtime;
+ now = xtime_cache;
} while (read_seqretry(&xtime_lock, seq));
return now;
}
-
EXPORT_SYMBOL(current_kernel_time);
--- NEW FILE patch-2.6.22.5.bz2.sign ---
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (GNU/Linux)
Comment: See http://www.kernel.org/signature.html for info
iD8DBQBGzM+KyGugalF9Dw4RAhc0AKCEaNlBFN/fm638yAojZ3s8jTF0mgCfTeNU
zn3f3FQg2WHeUk9mPbuMJHI=
=lZSp
-----END PGP SIGNATURE-----
Index: .cvsignore
===================================================================
RCS file: /cvs/pkgs/rpms/kernel/F-7/.cvsignore,v
retrieving revision 1.630
retrieving revision 1.631
diff -u -r1.630 -r1.631
--- .cvsignore 21 Aug 2007 18:40:05 -0000 1.630
+++ .cvsignore 23 Aug 2007 18:20:55 -0000 1.631
@@ -3,5 +3,4 @@
temp-*
kernel-2.6.22
linux-2.6.22.tar.bz2
-patch-2.6.22.3.bz2
-patch-2.6.22.4.bz2
+patch-2.6.22.5.bz2
Index: kernel-2.6.spec
===================================================================
RCS file: /cvs/pkgs/rpms/kernel/F-7/kernel-2.6.spec,v
retrieving revision 1.3326
retrieving revision 1.3327
diff -u -r1.3326 -r1.3327
--- kernel-2.6.spec 22 Aug 2007 22:31:20 -0000 1.3326
+++ kernel-2.6.spec 23 Aug 2007 18:20:55 -0000 1.3327
@@ -29,8 +29,8 @@
## If this is a released kernel ##
%if 0%{?released_kernel}
-# Do we have a 2.6.21.y update to apply?
-%define stable_update 4
+# Do we have a 2.6.x.y update to apply?
+%define stable_update 5
# Set rpm version accordingly
%if 0%{?stable_update}
%define stablerev .%{stable_update}
@@ -496,8 +496,6 @@
%endif
-Patch02: patch-2.6.22.5-rc1.bz2
-
%if !%{nopatches}
# Revert -stable pieces we get from elsewhere
@@ -538,6 +536,8 @@
Patch134: linux-2.6-powerpc-generic-suspend-4-kill-pmu-sleep-notifier.patch
Patch135: linux-2.6-powerpc-generic-suspend-5-pmu-pm_ops.patch
Patch140: linux-2.6-pmac-zilog.patch
+Patch141: linux-2.6-ppc-pegasos-keyboard-8042-check.patch
+
Patch150: linux-2.6-build-nonintconfig.patch
Patch160: linux-2.6-execshield.patch
Patch170: linux-2.6-modsign-mpilib.patch
@@ -615,7 +615,8 @@
#Patch780: linux-2.6-clockevents-fix-resume-logic.patch
Patch800: linux-2.6-wakeups-hdaps.patch
Patch801: linux-2.6-wakeups.patch
-Patch900: linux-2.6-sched-cfs.patch
+Patch900: linux-2.6-sched-cfs-v2.6.22.5-v20.2.patch
+Patch901: linux-2.6-timekeeping-fixes.patch
Patch1000: linux-2.6-dmi-based-module-autoloading.patch
Patch1020: linux-2.6-usb-autosuspend-default-disable.patch
Patch1030: linux-2.6-nfs-nosharecache.patch
@@ -1074,15 +1075,15 @@
# builds (as used in the buildsystem).
ApplyPatch linux-2.6-build-nonintconfig.patch
-ApplyPatch patch-2.6.22.5-rc1.bz2
-
%if !%{nopatches}
# Revert -stable pieces we get from elsewhere
ApplyPatch linux-2.6-softmac-fix-essid-problem_R.patch -R
# Ingo's new scheduler.
-ApplyPatch linux-2.6-sched-cfs.patch
+ApplyPatch linux-2.6-sched-cfs-v2.6.22.5-v20.2.patch
+# apply timekeeping updates that were in the Fedora CFS patch
+ApplyPatch linux-2.6-timekeeping-fixes.patch
# Roland's utrace ptrace replacement.
ApplyPatch linux-2.6-utrace-tracehook.patch -F2
@@ -1147,6 +1148,8 @@
ApplyPatch linux-2.6-ibmvscsi-schizo.patch
# Move pmac_zilog to its newly-registered device number
ApplyPatch linux-2.6-pmac-zilog.patch
+# fix detection of i8042 on ppc pegasos platform
+ApplyPatch linux-2.6-ppc-pegasos-keyboard-8042-check.patch
# Sleep through /sys/power/state
# Patches from http://johannes.sipsolutions.net/patches/kernel/all/2007-07-25-09:02/
ApplyPatch linux-2.6-powerpc-generic-suspend-1-remove-dead-code.patch
@@ -2296,6 +2299,11 @@
%endif
%changelog
+* Thu Aug 23 2007 Chuck Ebbert <cebbert at redhat.com>
+- Linux 2.6.22.5
+- CFS scheduler v20.2
+- fix Pegasos PS/2 port detection
+
* Wed Aug 22 2007 Chuck Ebbert <cebbert at redhat.com>
- 2.6.22.5-rc1
- un-revert genirq changes
Index: sources
===================================================================
RCS file: /cvs/pkgs/rpms/kernel/F-7/sources,v
retrieving revision 1.593
retrieving revision 1.594
diff -u -r1.593 -r1.594
--- sources 21 Aug 2007 18:40:05 -0000 1.593
+++ sources 23 Aug 2007 18:20:55 -0000 1.594
@@ -1,2 +1,2 @@
2e230d005c002fb3d38a3ca07c0200d0 linux-2.6.22.tar.bz2
-9041cc4a8570b5762a68b23235b6acd4 patch-2.6.22.4.bz2
+27544a58763bbd4ce497a77658af744a patch-2.6.22.5.bz2
Index: upstream
===================================================================
RCS file: /cvs/pkgs/rpms/kernel/F-7/upstream,v
retrieving revision 1.517
retrieving revision 1.518
diff -u -r1.517 -r1.518
--- upstream 21 Aug 2007 18:40:05 -0000 1.517
+++ upstream 23 Aug 2007 18:20:55 -0000 1.518
@@ -1,2 +1,2 @@
linux-2.6.22.tar.bz2
-patch-2.6.22.4.bz2
+patch-2.6.22.5.bz2
--- linux-2.6-sched-cfs.patch DELETED ---
--- patch-2.6.22.4.bz2.sign DELETED ---
--- patch-2.6.22.5-rc1.bz2 DELETED ---
More information about the fedora-extras-commits
mailing list