http://landley.net/notes.html#16-03-2013
Let's jump back to the 3.6 kernel release, where I was reverting all
the irq routing stuff back to what it looked like in 3.3 or so, and
both scsi and the ethernet controller worked. In that context, we got
the following boot messages:
PCI: enabling device 0000:00:0d.0 (0100 -> 0103)
sym0: <895a> rev 0x0 at pci 0000:00:0d.0 irq 27
sym0: No NVRAM, ID 7, Fast-40, LVD, parity checking
sym0: SCSI BUS has been reset.
...
8139cp: 8139cp: 10/100 PCI Ethernet driver v1.3 (Mar 22, 2004)
PCI: enabling device 0000:00:0c.0 (0100 -> 0103)
8139cp 0000:00:0c.0 eth0: RTL-8139C+ at 0xc8874400, 52:54:00:12:34:56, IRQ 27
So both the scsi and ethernet were sharing IRQ 27.
After they moved the IRQ controller start from 0 to 32 (for not
obvious reason), this means the corresponding IRQ is now 27+32=59. And
this is the one that works.
Here are the new boot messages after the fix I checked in yesterday:
PCI: enabling device 0000:00:0d.0 (0100 -> 0103)
sym0: <895a> rev 0x0 at pci 0000:00:0d.0 irq 59
sym0: No NVRAM, ID 7, Fast-40, LVD, parity checking
sym0: SCSI BUS has been reset.
...
8139cp: 8139cp: 10/100 PCI Ethernet driver v1.3 (Mar 22, 2004)
PCI: enabling device 0000:00:0c.0 (0100 -> 0103)
8139cp 0000:00:0c.0 eth0: RTL-8139C+ at 0xd0874400, 52:54:00:12:34:56, IRQ 62
Spot anything? Like the fact that 62 != 59? And thus, when I run the
nativel linux from scratch build and it tries to distribute
compilation over distcc, we get:
=== zlib(2 of 48)
Checking for gcc...
irq 59: nobody cared (try booting with the "irqpoll" option)
Backtrace:
[<c00113e8>] (dump_backtrace+0x0/0x110) from [<c001152c>] (dump_stack+0x18/0x1c)
r6:00000000 r5:c02eabdc r4:c02eabdc
[<c0011514>] (dump_stack+0x0/0x1c) from [<c004d384>]
(__report_bad_irq+0x28/0xb0)
... (mondo useless stack dump that just says the IRQ came from hardware)
Disabling IRQ #59
... (more stack dump)
8139cp 0000:00:0c.0 eth0: Transmit timeout, status d 2b 5 80ff
8139cp 0000:00:0c.0 eth0: Transmit timeout, status d 2b 5 80ff
8139cp 0000:00:0c.0 eth0: Transmit timeout, status d 2b 5 80ff
Oddly, QEMU's scsi device still seems to be usable so at a guess they
didn't actually disable IRQ 59? Dunno...
So let's annotate and see:
slot=12 pin=1 irq=62
slot=13 pin=1 irq=59
Hi Viresh,
(1) Recently I have done a lot of testing for the new RCU functionality
(new RCU FNOHZ implementation, updated RCU no-callback implementation)
which is scheduled for 3.11/3.12 mainline. In my tests it helped to keep
the A15's quiet and showed power savings for some workloads.
During last Linaro Connect people started to ask me if we can bring
these features in already earlier.
Would it be possible for you to merge in 'idlenocb.2013.03.18b' (based
on v3.9-rc2) (or newer) from
'git://git.kernel.org/pub/scm/linux/kernel/git/paulmck/linux-rcu.git'
into your big.LITTLE MP tree? I talked to Paul McKenney about this and
he would like to have this additional test coverage.
I have performed the following tests: System comes up and pm-qa
(cpufreq, cpuidle, cpuhotplug) on nano and FTS (mpbasic, mpcore,
mpextended, mploadbalance, mpregression, pmump) on Android FS look good.
(2) On top of the merge, we must assure that the TC2 default config file
gets changed:
Change the config file:
* Set CONFIG_RCU_FAST_NO_HZ=y
* Set CONFIG_RCU_NOCB_CPU=y
* Set CONFIG_RCU_NOCB_CPU_ALL=y
diff -Naur config_orig .config
--- config_orig 2013-03-13 14:15:10.574248125 +0000
+++ .config 2013-03-13 14:15:34.334248346 +0000
@@ -93,9 +93,12 @@
CONFIG_RCU_FANOUT=32
CONFIG_RCU_FANOUT_LEAF=16
# CONFIG_RCU_FANOUT_EXACT is not set
-# CONFIG_RCU_FAST_NO_HZ is not set
+CONFIG_RCU_FAST_NO_HZ=y
# CONFIG_TREE_RCU_TRACE is not set
-# CONFIG_RCU_NOCB_CPU is not set
+CONFIG_RCU_NOCB_CPU=y
+# CONFIG_RCU_NOCB_CPU_NONE is not set
+# CONFIG_RCU_NOCB_CPU_ZERO is not set
+CONFIG_RCU_NOCB_CPU_ALL=y
CONFIG_IKCONFIG=y
CONFIG_IKCONFIG_PROC=y
CONFIG_LOG_BUF_SHIFT=16
Cheers,
-- Dietmar
-- IMPORTANT NOTICE: The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the information in any medium. Thank you.
Hi Guys,
[All this is not yet mainlined and is WIP]
Linaro and ARM are developing Linux solutions for big LITTLE Architecture. And
I maintain all the stable stuff on this here: [1] and this gets pulled
in by Andrey for
every linux-linaro release.
[1] by itself isn't much usable as it doesn't have many bits which are
pulled in by
Andrey, like: tc2 support from tixy and Android stuff from John.
As this is a stable release it contains what works best and it may
contain stuff that
is never going to be mainlined.
So, i have started to maintain another release for people who are only
interested in
working with latest upstream effort in this direction. It is created
with following steps:
- Checkout a new branch based of linux-linaro latest release tag
- git revert Merge with big-LITTLE-MP-master-v*
- git merge with [2]
And we finally get [3] which only contains all the big LITTLE MP
efforts that are
under review on LKML.
[3] has all the bits required to test it on TC2 as it is based of
linux-linaro release.
Interested people may check it out to give it a try.
--
viresh
[1]: https://git.linaro.org/gitweb?p=arm/big.LITTLE/mp.git;a=shortlog;h=refs/hea…
[2]: https://git.linaro.org/gitweb?p=arm/big.LITTLE/mp.git;a=shortlog;h=refs/hea…
[3]: https://git.linaro.org/gitweb?p=arm/big.LITTLE/mp.git;a=shortlog;h=refs/hea…
Hello,
This series implements reentrancy for the common clk implementation of
the clk.h api. Making reentrant calls into the clock framework is both
necessary and desirable for many use cases such as enabling off-chip
clocks via i2c. The first patch in the series implements this.
A neat side effect of reentrancy is that it is possible for platforms
using voltage regulators controlled via i2c to register rate-change
notifier handlers to scale voltage as a function of clock rate. This is
an effective way to implement dynamic voltage & frequency scaling.
Patch #2 implements a helper function for registering such a notifier
handler.
The third patch in the series demonstrates dvfs on OMAP platforms by
modifying the cpufreq-omap driver; it migrates the voltage scaling logic
out of the cpufreq driver's .target callback and registers callbacks via
the helper introduced in patch #2.
Patches four and five are purely test coverage. And what better way to
test than to muck with fragile PLL programming code? These patches test
out a lot of the aforementioned reentrancy in the OMAP3+ DPLL code.
They are not for merging, but as a demonstration of what is now
possible.
Finally, I know that Documentation/clk.txt needs an update for these
changes but I wanted this on the list before I flew out to LCA 2013.
I'll provide that update during or after the conference.
Two previous (and considerably more insane) attempts at this,
v1: http://article.gmane.org/gmane.linux.kernel/1327866
v2: http://marc.info/?l=linux-kernel&m=134507429302463&w=2
Mike Turquette (5):
clk: allow reentrant calls into the clk framework
clk: notifier handler for dynamic voltage scaling
cpufreq: omap: scale regulator from clk notifier
HACK: set_parent callback for OMAP4 non-core DPLLs
HACK: omap: opp: add fake 400MHz OPP to bypass MPU
arch/arm/mach-omap2/cclock44xx_data.c | 1 +
arch/arm/mach-omap2/clock.h | 1 +
arch/arm/mach-omap2/dpll3xxx.c | 107 ++++++++++----
arch/arm/mach-omap2/opp4xxx_data.c | 18 +++
drivers/clk/Makefile | 1 +
drivers/clk/clk.c | 254 ++++++++++++++++++++++++---------
drivers/clk/dvfs.c | 125 ++++++++++++++++
drivers/cpufreq/omap-cpufreq.c | 82 +++--------
include/linux/clk.h | 27 +++-
9 files changed, 459 insertions(+), 157 deletions(-)
create mode 100644 drivers/clk/dvfs.c
--
1.7.10.4
Hi,
There are many scripts in use that parse the web interface for
snapshots and releases.linaro.org. This is far from ideal because
changes to the web interface can result in the scripts breaking. To
alleviate this problem an API has been introduced. The API is very
simple, just two new URLs and a new header when downloading files.
<server>/api/ls/<path to file> returns a JSON document containing all the
data shown by a file listing on the web interface. It can target
either a directory of a file.
<server>/api/license/<path to file> returns a JSON document containing
the licence information for the file pointed to. Both the license text
and the digest used to accept the license are returned. Including the
digest means that the choice of digest used internally can change without
re-writing clients - to the client this is just a magic string and they
don't need to care how it is generated.
When downloading a file you simply include the header
LICENSE_ACCEPTED, its value being a space separated list of license
digests representing the licenses that you have accepted.
e.g.
(license text and digest removed for brevity)
http://snapshots.linaro.org/api/ls/quantal/hwpacks/snowball/latest
{
"files": [
{
"mtime": "15-Mar-2013 08:29",
"name":
"hwpack_linaro-snowball_20130315-269_armhf_supported.manifest.txt",
"size": "762",
"type": "text",
"url":
"/quantal/hwpacks/snowball/latest/hwpack_linaro-snowball_20130315-269_armhf_supported.manifest.txt"
},
{
"mtime": "15-Mar-2013 08:31",
"name":
"hwpack_linaro-snowball_20130315-269_armhf_supported.tar.gz",
"size": "18.1M",
"type": "other",
"url":
"/quantal/hwpacks/snowball/latest/hwpack_linaro-snowball_20130315-269_armhf_supported.tar.gz"
}
]
}
http://snapshots.linaro.org/api/license/quantal/hwpacks/snowball/latest/hwp…
{
"licenses": [
{
"text": "license text...",
"digest": "license digest..."
}
]
}
wget --header="LICENSE_ACCEPTED: license digest..."
http://snapshots.linaro.org/quantal/hwpacks/snowball/latest/hwpack_linaro-s…
There is a example client script that illustrates how to use the API here:
http://bazaar.launchpad.net/~linaro-automation/linaro-license-protection/tr…
Usage: download.py <URL>
If URL points to a directory all files are downloaded.
If URL points to a file, just that file is downloaded.
It prints the license(s) for each file and waits for the user to
accept each one. It also stores which licenses the user has accepted
so they don't need to be re-accepted if the text hasn't changed.
--
James Tunnicliffe
Add notifier calls in clk_prepare and clk_unprepare so drivers which are
interested in knowing that clk_prepare/unprepare call can act accordingly.
The existing "clk_set_rate" notifier is not enough for normal DVFS
inplementation since clock might be enabled/disabled at runtime. Adding
these notifiers is useful on DVFS core which take clk_prepare as a hint
on that the notified clock might be enabled later so it can raise voltage
to a safe level before enabling the clock, and take clk_unprepare as a
hint that the clock has been disabled and is safe to lower the voltage.
The added notifier events are:
PRE_CLK_PREPARE
POST_CLK_PREPARE
ABORT_CLK_PREPARE
PRE_CLK_UNPREPARE
POST_CLK_UNPREPARE
Signed-off-by: Bill Huang <bilhuang(a)nvidia.com>
---
drivers/clk/clk.c | 88 ++++++++++++++++++++++++++++++---------------------
include/linux/clk.h | 5 +++
2 files changed, 57 insertions(+), 36 deletions(-)
diff --git a/drivers/clk/clk.c b/drivers/clk/clk.c
index ed87b24..ac07c6e 100644
--- a/drivers/clk/clk.c
+++ b/drivers/clk/clk.c
@@ -516,6 +516,42 @@ struct clk *__clk_lookup(const char *name)
/*** clk api ***/
+/**
+ * __clk_notify - call clk notifier chain
+ * @clk: struct clk * that is changing rate
+ * @msg: clk notifier type (see include/linux/clk.h)
+ * @old_rate: old clk rate
+ * @new_rate: new clk rate
+ *
+ * Triggers a notifier call chain on the clk rate-change notification
+ * for 'clk'. Passes a pointer to the struct clk and the previous
+ * and current rates to the notifier callback. Intended to be called by
+ * internal clock code only. Returns NOTIFY_DONE from the last driver
+ * called if all went well, or NOTIFY_STOP or NOTIFY_BAD immediately if
+ * a driver returns that.
+ */
+static int __clk_notify(struct clk *clk, unsigned long msg,
+ unsigned long old_rate, unsigned long new_rate)
+{
+ struct clk_notifier *cn;
+ struct clk_notifier_data cnd;
+ int ret = NOTIFY_DONE;
+
+ cnd.clk = clk;
+ cnd.old_rate = old_rate;
+ cnd.new_rate = new_rate;
+
+ list_for_each_entry(cn, &clk_notifier_list, node) {
+ if (cn->clk == clk) {
+ ret = srcu_notifier_call_chain(&cn->notifier_head, msg,
+ &cnd);
+ break;
+ }
+ }
+
+ return ret;
+}
+
void __clk_unprepare(struct clk *clk)
{
if (!clk)
@@ -549,7 +585,14 @@ void __clk_unprepare(struct clk *clk)
void clk_unprepare(struct clk *clk)
{
mutex_lock(&prepare_lock);
+
+ if (clk->notifier_count)
+ __clk_notify(clk, PRE_CLK_UNPREPARE, clk->rate, clk->rate);
+
__clk_unprepare(clk);
+ if (clk->notifier_count)
+ __clk_notify(clk, POST_CLK_UNPREPARE, clk->rate, clk->rate);
+
mutex_unlock(&prepare_lock);
}
EXPORT_SYMBOL_GPL(clk_unprepare);
@@ -597,7 +640,16 @@ int clk_prepare(struct clk *clk)
int ret;
mutex_lock(&prepare_lock);
+
+ if (clk->notifier_count)
+ __clk_notify(clk, PRE_CLK_PREPARE, clk->rate, clk->rate);
+
ret = __clk_prepare(clk);
+ if (!ret && clk->notifier_count)
+ __clk_notify(clk, POST_CLK_PREPARE, clk->rate, clk->rate);
+ else if (clk->notifier_count)
+ __clk_notify(clk, ABORT_CLK_PREPARE, clk->rate, clk->rate);
+
mutex_unlock(&prepare_lock);
return ret;
@@ -749,42 +801,6 @@ long clk_round_rate(struct clk *clk, unsigned long rate)
EXPORT_SYMBOL_GPL(clk_round_rate);
/**
- * __clk_notify - call clk notifier chain
- * @clk: struct clk * that is changing rate
- * @msg: clk notifier type (see include/linux/clk.h)
- * @old_rate: old clk rate
- * @new_rate: new clk rate
- *
- * Triggers a notifier call chain on the clk rate-change notification
- * for 'clk'. Passes a pointer to the struct clk and the previous
- * and current rates to the notifier callback. Intended to be called by
- * internal clock code only. Returns NOTIFY_DONE from the last driver
- * called if all went well, or NOTIFY_STOP or NOTIFY_BAD immediately if
- * a driver returns that.
- */
-static int __clk_notify(struct clk *clk, unsigned long msg,
- unsigned long old_rate, unsigned long new_rate)
-{
- struct clk_notifier *cn;
- struct clk_notifier_data cnd;
- int ret = NOTIFY_DONE;
-
- cnd.clk = clk;
- cnd.old_rate = old_rate;
- cnd.new_rate = new_rate;
-
- list_for_each_entry(cn, &clk_notifier_list, node) {
- if (cn->clk == clk) {
- ret = srcu_notifier_call_chain(&cn->notifier_head, msg,
- &cnd);
- break;
- }
- }
-
- return ret;
-}
-
-/**
* __clk_recalc_rates
* @clk: first clk in the subtree
* @msg: notification type (see include/linux/clk.h)
diff --git a/include/linux/clk.h b/include/linux/clk.h
index b3ac22d..41d567d 100644
--- a/include/linux/clk.h
+++ b/include/linux/clk.h
@@ -43,6 +43,11 @@ struct clk;
#define PRE_RATE_CHANGE BIT(0)
#define POST_RATE_CHANGE BIT(1)
#define ABORT_RATE_CHANGE BIT(2)
+#define PRE_CLK_PREPARE BIT(3)
+#define POST_CLK_PREPARE BIT(4)
+#define ABORT_CLK_PREPARE BIT(5)
+#define PRE_CLK_UNPREPARE BIT(6)
+#define POST_CLK_UNPREPARE BIT(7)
/**
* struct clk_notifier - associate a clk with a notifier
--
1.7.9.5
Linaro 13.03 Release (Calendar Week 13): Here is test result summary for
Linux Linaro ubuntu Quantal image on following boards:
https://docs.google.com/a/linaro.org/spreadsheet/ccc?key=0AgB-fT5LL31CdEtLT…
1) ARM Versatile Express A9;
2) Samsung Arndale;
3) TI Panda 4430;
4) TI Panda 4460;
5) ST Ericsson Snowball.
Synopsis: Kernel version on all boards has been upgraded to "3.9.0-1"; TI
Panda 4460 booting failed.
1. ARM Versatile Express A9 + Linux Linaro Quantal (Column O):
https://docs.google.com/a/linaro.org/spreadsheet/ccc?key=0AroPySpr4FnEdFNmV…
It keeps exactly same status since calendar week 50 last year: only "Halt"
& "Device Tree" test failed, all other features work well.
2. Samsung Arndale + Linux Linaro Quantal (Column I):
https://docs.google.com/a/linaro.org/spreadsheet/ccc?key=0AgB-fT5LL31CdGZJS…
It keeps same status as last week: system hangs during the Power Management
test, HDMI display is unavailable. All other features work well.
3. TI Panda 4430 + Linux Linaro Quantal (Column O):
https://docs.google.com/a/linaro.org/spreadsheet/ccc?key=0AroPySpr4FnEdEwwZ…
WiFi, HDMI and DVI-D are unavailable. All other features work well.
4. TI Panda 4460 + Linux Linaro Quantal (Column O):
https://docs.google.com/a/linaro.org/spreadsheet/ccc?key=0AroPySpr4FnEdEwwZ…
System booting failed since 3 weeks ago, all test cases are blocked. No
release for this board in this month.
5. ST Ericsson Snowball + Linux Linaro Quantal (Column P):
https://docs.google.com/a/linaro.org/spreadsheet/ccc?key=0AroPySpr4FnEdFJ4X…
SD/MMC test works well now, I/O error has been eliminated. Kernel panic has
gone in power management test but another test case "thermal_03.1" failed
there. "Reboot" & HDMI display are still unavailable. All other features
work well.
For the previous week test summary (Calendar Week 12), please refer to
attachment.
Thank you.
Best Regards
Botao Sun
A second test rebuild of raring ringtail was started yesterday for the amd64,
i386 and armhf architectures. Currently running, and with mostly idle buildds
over easter and the current freeze it will hopefully finish within ten days.
Results can be seen at
http://people.ubuntuwire.org/~wgrant/rebuild-ftbfs-test/test-rebuild-201303…
The archive for the test rebuild is
https://launchpad.net/ubuntu/+archive/test-rebuild-20130329/
Some common build failures are:
- underlinking: symbols used in linked object files, which formerly
were resolved by linked libraries. The fix almost always is to add
the library (mentioned in the error message) to the link command.
- Tcl/Tk migration to multiarch locations. For autoconf based packages
it should be enough to adjust the configure arguments for the Tcl/Tk
libraries. The location for the header files didn't change.
- eglibc-2.17 changes: -Werror=unused-result used together with -Werror.
- Some glib/gtk/dbus related build and test failures, however there is
no common pattern yet.
Please help fixing the build failures for the final release.
Matthias
[ Please Cc me on replies ]
The README for flashbench shows numbers that only descend. Can someone help me decode the following alignment results? Perhaps the README could be updated, too?
root@spot-372b21:/var/volatile/media/sda1# ./flashbench -a /dev/mmcblk0 --blocksize=1024 --count=100
align 2147483648 pre 1.82ms on 2.74ms post 2.1ms diff 784µs
align 1073741824 pre 1.79ms on 2.68ms post 2.05ms diff 757µs
align 536870912 pre 1.78ms on 2.67ms post 2.05ms diff 755µs
align 268435456 pre 1.79ms on 2.66ms post 2.04ms diff 749µs
align 134217728 pre 1.79ms on 2.67ms post 2.04ms diff 753µs
align 67108864 pre 1.78ms on 2.66ms post 2.05ms diff 750µs
align 33554432 pre 1.79ms on 3.38ms post 2.19ms diff 1.39ms
align 16777216 pre 1.84ms on 3.56ms post 2.19ms diff 1.54ms
align 8388608 pre 1.85ms on 2.73ms post 2.04ms diff 786µs
align 4194304 pre 1.84ms on 3.56ms post 2.2ms diff 1.54ms
align 2097152 pre 1.83ms on 2.69ms post 2.26ms diff 648µs
align 1048576 pre 1.81ms on 2.85ms post 2.24ms diff 826µs
align 524288 pre 1.83ms on 2.95ms post 2.24ms diff 920µs
align 262144 pre 1.87ms on 2.18ms post 2.18ms diff 162µs
align 131072 pre 1.89ms on 2.2ms post 2.2ms diff 161µs
align 65536 pre 1.9ms on 2.21ms post 2.21ms diff 160µs
align 32768 pre 1.91ms on 2.21ms post 2.18ms diff 160µs
align 16384 pre 1.92ms on 2.22ms post 2.19ms diff 168µs
align 8192 pre 1.92ms on 2.22ms post 2.19ms diff 168µs
align 4096 pre 2.08ms on 2.08ms post 2.04ms diff 19.5µs
align 2048 pre 2.15ms on 2.15ms post 2.11ms diff 18.1µs
root@spot-372b21:/var/volatile/media/sda1# find /sys/block/mmcblk0/device/ -maxdepth 1 -type f | xargs head
==> /sys/block/mmcblk0/device/uevent <==
DRIVER=mmcblk
MMC_TYPE=SD
MMC_NAME=SD8GB
MODALIAS=mmc:block
==> /sys/block/mmcblk0/device/cid <==
4134325344384742300030018f00b367
==> /sys/block/mmcblk0/device/csd <==
400e00325b5900003b777f800a40009f
==> /sys/block/mmcblk0/device/scr <==
0235800000000000
==> /sys/block/mmcblk0/device/date <==
03/2011
==> /sys/block/mmcblk0/device/fwrev <==
0x0
==> /sys/block/mmcblk0/device/hwrev <==
0x3
==> /sys/block/mmcblk0/device/manfid <==
0x000041
==> /sys/block/mmcblk0/device/name <==
SD8GB
==> /sys/block/mmcblk0/device/oemid <==
0x3432
==> /sys/block/mmcblk0/device/serial <==
0x0030018f
==> /sys/block/mmcblk0/device/type <==
SD
root@spot-372b21:/var/volatile/media/sda1#
Hi folks,
I installed the panda-quantal_developer_20130128-58.img.gz on an old
Pandaboard A1 and apt-get update got a new kernel and have overwritten the
preEnv.txt boot=UUID=...hash... into boot=UUID=rootfs, only that rootfs is
not a device.
It fell back to busybox saying couldn't find root, so I re-wrote the file
to have the correct UUID and it booted correctly.
Not sure anyone has seen this, but seems like a simple bug to fix...
cheers,
--renato
