From: Sebastian Hesselbarth <sebastian.hesselbarth@gmail.com>
To: "Sören Brinkmann" <soren.brinkmann@xilinx.com>
Cc: Mike Turquette <mturquette@linaro.org>,
Russell King <linux@arm.linux.org.uk>,
Arnd Bergmann <arnd@arndb.de>,
Michal Simek <michal.simek@xilinx.com>,
linux-arm-kernel@lists.infradead.org,
linux-kernel@vger.kernel.org,
Steffen Trumtrar <s.trumtrar@pengutronix.de>
Subject: Re: [RFC 17/17] clk: zynq: remove call to of_clk_init
Date: Fri, 23 Aug 2013 11:30:18 +0200 [thread overview]
Message-ID: <52172BAA.7020009@gmail.com> (raw)
In-Reply-To: <c4d10bdf-6506-4edc-92cc-2acc67b86602@VA3EHSMHS032.ehs.local>
[-- Attachment #1: Type: text/plain, Size: 2132 bytes --]
On 08/23/13 02:59, Sören Brinkmann wrote:
> On Thu, Aug 22, 2013 at 05:26:47PM -0700, Sören Brinkmann wrote:
>> On Tue, Aug 20, 2013 at 04:04:31AM +0200, Sebastian Hesselbarth wrote:
>>> With arch/arm calling of_clk_init(NULL) from time_init(), we can now
>>> remove it from corresponding drivers/clk code.
>>
>> I think that would break Zynq.
>> If I see this correctly you call of_clk_init() from common code,
>> _before_ the SOC specific time init function is called.
>> The problem is, that we have code setting up a global pointer which is
>> required by zynq_clk_setup() which is triggered when of_clk_init() is
>> called.
>>
>> Let me try to illustrate the current call graph:
>>
>> time_init()
>> zynq_timer_init() // this machines init_time()
>> zynq_slcr_init() // setup System Level Control Registers including a global pointer
>> zynq_clock_init()
>> of_clk_init()
>> zynq_clk_setup() // requires pointer setup in zynq_slcr_init()
>> ...
>>
>> IIUC, your series would change this to:
>> time_init()
>> of_clk_init()
>> zynq_clk_setup() // SLCR pointer is not setup/NULL
>> ...
>> zynq_timer_init()
>> zynq_slcr_init() // now the pointer becomes valid
>
> I guess we could move zynq_slcr_init() into init_irq(). I'll give that a
> shot tomorrow.
Sören,
thanks for looking into this. I also had a look at the files in
question. Based on Steffen's proposal, I prepared a diff that should do
the trick. It moves zynq_slcr_init() to early_init, instead of reusing
another hook that has magic cow powers (it calls irqchip_init that zynq
also wants sooner or later).
Also, it removes zynq_clock_init() and let zynq_clk_setup() map the
register itself by finding the node and use of_iomap(). I realized that
clock registers are quite separated within slcr, so you can consider
to have your own node for the clk-provider. As Steffen is proposing
this but mentioned incompatible DT changes, I chose that intermediate
step above.
It would be great, if you test the diff and prepare a patch out of
it, that I pick-up in the patch set. That way, we also have your
Signed-off on it.
Sebastian
[-- Attachment #2: arm-of_clk_init-zynq-proposal.diff --]
[-- Type: text/x-patch, Size: 3117 bytes --]
diff --git a/arch/arm/mach-zynq/common.c b/arch/arm/mach-zynq/common.c
index 5f25256..a5a3969 100644
--- a/arch/arm/mach-zynq/common.c
+++ b/arch/arm/mach-zynq/common.c
@@ -19,8 +19,6 @@
#include <linux/cpumask.h>
#include <linux/platform_device.h>
#include <linux/clk.h>
-#include <linux/clk/zynq.h>
-#include <linux/clocksource.h>
#include <linux/of_address.h>
#include <linux/of_irq.h>
#include <linux/of_platform.h>
@@ -58,10 +56,9 @@ static void __init zynq_init_machine(void)
of_platform_bus_probe(NULL, zynq_of_bus_ids, NULL);
}
-static void __init zynq_timer_init(void)
+static void __init zynq_init_early(void)
{
zynq_slcr_init();
- clocksource_of_init();
}
static struct map_desc zynq_cortex_a9_scu_map __initdata = {
@@ -104,8 +101,8 @@ static const char * const zynq_dt_match[] = {
DT_MACHINE_START(XILINX_EP107, "Xilinx Zynq Platform")
.smp = smp_ops(zynq_smp_ops),
.map_io = zynq_map_io,
+ .init_early = zynq_init_early,
.init_machine = zynq_init_machine,
- .init_time = zynq_timer_init,
.dt_compat = zynq_dt_match,
.restart = zynq_system_reset,
MACHINE_END
diff --git a/arch/arm/mach-zynq/slcr.c b/arch/arm/mach-zynq/slcr.c
index 1836d5a..59ad09f 100644
--- a/arch/arm/mach-zynq/slcr.c
+++ b/arch/arm/mach-zynq/slcr.c
@@ -106,8 +106,6 @@ int __init zynq_slcr_init(void)
pr_info("%s mapped to %p\n", np->name, zynq_slcr_base);
- zynq_clock_init(zynq_slcr_base);
-
of_node_put(np);
return 0;
diff --git a/drivers/clk/zynq/clkc.c b/drivers/clk/zynq/clkc.c
index 089d3e3..8fa0de7 100644
--- a/drivers/clk/zynq/clkc.c
+++ b/drivers/clk/zynq/clkc.c
@@ -21,6 +21,7 @@
#include <linux/clk/zynq.h>
#include <linux/clk-provider.h>
#include <linux/of.h>
+#include <linux/of_address.h>
#include <linux/slab.h>
#include <linux/string.h>
#include <linux/io.h>
@@ -203,9 +204,19 @@ static void __init zynq_clk_setup(struct device_node *np)
const char *periph_parents[4];
const char *swdt_ext_clk_mux_parents[2];
const char *can_mio_mux_parents[NUM_MIO_PINS];
+ struct device_node *slcrnp;
pr_info("Zynq clock init\n");
+ slcrnp = of_find_compatible_node(NULL, NULL, "xlnx,zynq-slcr");
+ if (WARN_ON(!slcrnp))
+ return;
+
+ zynq_slcr_base_priv = of_iomap(slcrnp, 0);
+ of_node_put(slcrnp);
+ if (WARN_ON(!zynq_slcr_base_priv))
+ return;
+
/* get clock output names from DT */
for (i = 0; i < clk_max; i++) {
if (of_property_read_string_index(np, "clock-output-names",
@@ -528,9 +539,3 @@ static void __init zynq_clk_setup(struct device_node *np)
}
CLK_OF_DECLARE(zynq_clkc, "xlnx,ps7-clkc", zynq_clk_setup);
-
-void __init zynq_clock_init(void __iomem *slcr_base)
-{
- zynq_slcr_base_priv = slcr_base;
- of_clk_init(NULL);
-}
diff --git a/include/linux/clk/zynq.h b/include/linux/clk/zynq.h
index e062d31..58b5516 100644
--- a/include/linux/clk/zynq.h
+++ b/include/linux/clk/zynq.h
@@ -22,8 +22,6 @@
#include <linux/spinlock.h>
-void zynq_clock_init(void __iomem *slcr);
-
struct clk *clk_register_zynq_pll(const char *name, const char *parent,
void __iomem *pll_ctrl, void __iomem *pll_status, u8 lock_index,
spinlock_t *lock);
next prev parent reply other threads:[~2013-08-23 9:30 UTC|newest]
Thread overview: 118+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-08-20 2:04 [RFC 00/17] ARM: provide common arch init for DT clocks Sebastian Hesselbarth
2013-08-20 2:04 ` [RFC 01/17] clk: ensure __clk_of_table is only initialized once Sebastian Hesselbarth
2013-08-20 2:04 ` [RFC 02/17] ARM: call clk_of_init from time_init Sebastian Hesselbarth
2013-08-20 15:46 ` Stephen Warren
2013-08-20 19:47 ` Sebastian Hesselbarth
2013-08-20 19:52 ` Stephen Warren
2013-08-20 20:19 ` Sebastian Hesselbarth
2013-08-21 18:54 ` Arnd Bergmann
2013-08-22 7:25 ` Mike Turquette
2013-08-22 17:28 ` Arnd Bergmann
2013-08-20 20:44 ` Arnd Bergmann
2013-08-20 20:57 ` Stephen Warren
2013-08-21 18:51 ` Arnd Bergmann
2013-08-20 2:04 ` [RFC 03/17] ARM: dove: remove custom .init_time hook Sebastian Hesselbarth
2013-08-20 2:04 ` [RFC 04/17] ARM: exynos: " Sebastian Hesselbarth
2013-08-20 2:04 ` [RFC 05/17] ARM: highbank: " Sebastian Hesselbarth
2013-08-20 19:12 ` Rob Herring
2013-08-20 20:01 ` Sebastian Hesselbarth
2013-08-20 2:04 ` [RFC 06/17] ARM: imx: " Sebastian Hesselbarth
2013-08-20 7:26 ` Sascha Hauer
2013-08-20 8:48 ` Sebastian Hesselbarth
2013-08-20 9:10 ` Sascha Hauer
2013-08-20 9:20 ` Sebastian Hesselbarth
2013-08-20 10:20 ` Russell King - ARM Linux
2013-08-20 10:37 ` Sebastian Hesselbarth
2013-08-20 10:42 ` Russell King - ARM Linux
2013-08-20 10:46 ` Sebastian Hesselbarth
2013-08-20 9:23 ` [RFC v2] " Sebastian Hesselbarth
2013-08-20 2:04 ` [RFC 07/17] ARM: kirkwood: " Sebastian Hesselbarth
2013-08-20 2:04 ` [RFC 08/17] ARM: mvebu: " Sebastian Hesselbarth
2013-08-20 12:50 ` Ezequiel Garcia
2013-08-20 13:12 ` Sebastian Hesselbarth
2013-08-20 14:01 ` Ezequiel Garcia
2013-08-23 10:06 ` Gregory CLEMENT
2013-08-23 10:32 ` Sebastian Hesselbarth
2013-08-23 11:39 ` Gregory CLEMENT
2013-08-23 12:13 ` Sebastian Hesselbarth
2013-08-23 14:23 ` Gregory CLEMENT
2013-08-23 14:34 ` Gregory CLEMENT
2013-08-20 2:04 ` [RFC 09/17] ARM: mxs: " Sebastian Hesselbarth
2013-08-20 2:04 ` [RFC 10/17] ARM: nspire: " Sebastian Hesselbarth
2013-08-20 2:04 ` [RFC 11/17] ARM: rockchip: " Sebastian Hesselbarth
2013-08-20 2:04 ` [RFC 12/17] ARM: socfpga: " Sebastian Hesselbarth
2013-08-20 2:04 ` [RFC 13/17] ARM: sti: " Sebastian Hesselbarth
2013-08-20 2:04 ` [RFC 14/17] ARM: tegra: remove .init_time hook to clocksource_of_init Sebastian Hesselbarth
2013-08-20 2:04 ` [RFC 15/17] ARM: vexpress: remove custom .init_time hook Sebastian Hesselbarth
2013-08-20 2:04 ` [RFC 16/17] clk: vt8500: remove call to of_clk_init Sebastian Hesselbarth
2013-08-20 2:04 ` [RFC 17/17] clk: zynq: " Sebastian Hesselbarth
2013-08-23 0:26 ` Sören Brinkmann
2013-08-23 0:59 ` Sören Brinkmann
2013-08-23 7:32 ` Steffen Trumtrar
2013-08-23 16:00 ` Sören Brinkmann
2013-08-23 23:38 ` Steffen Trumtrar
2013-08-26 11:15 ` Michal Simek
2013-08-26 12:07 ` Steffen Trumtrar
2013-08-26 12:53 ` Sebastian Hesselbarth
2013-08-26 15:14 ` Michal Simek
2013-08-26 15:31 ` Steffen Trumtrar
2013-08-26 15:16 ` Sören Brinkmann
2013-08-26 15:33 ` Steffen Trumtrar
2013-08-23 9:30 ` Sebastian Hesselbarth [this message]
2013-08-23 17:19 ` Sören Brinkmann
2013-08-23 17:44 ` Sebastian Hesselbarth
2013-08-23 23:22 ` Steffen Trumtrar
2013-08-26 15:20 ` Sören Brinkmann
2013-08-27 21:27 ` [PATCH RFC v2 00/16] ARM: provide common arch init for DT clocks Sebastian Hesselbarth
2013-08-27 21:27 ` [PATCH RFC v2 01/16] ARM: call clk_of_init from time_init Sebastian Hesselbarth
2013-08-27 22:19 ` Sören Brinkmann
2013-08-27 22:58 ` Sebastian Hesselbarth
2013-08-27 23:20 ` Sören Brinkmann
2013-08-29 13:45 ` Arnd Bergmann
2013-09-04 19:32 ` Sebastian Hesselbarth
2013-09-04 20:41 ` Sören Brinkmann
2013-09-04 20:52 ` Sebastian Hesselbarth
2013-09-04 20:59 ` Sören Brinkmann
2013-09-05 9:34 ` Arnd Bergmann
2013-08-27 21:27 ` [PATCH RFC v2 02/16] ARM: dove: remove custom .init_time hook Sebastian Hesselbarth
2013-08-27 21:27 ` [PATCH RFC v2 03/16] ARM: exynos: " Sebastian Hesselbarth
2013-08-27 21:27 ` [PATCH RFC v2 04/16] ARM: highbank: " Sebastian Hesselbarth
2013-08-27 23:13 ` Mike Turquette
2013-08-27 21:27 ` [PATCH RFC v2 05/16] ARM: imx: " Sebastian Hesselbarth
2013-08-29 11:00 ` Shawn Guo
2013-08-27 21:28 ` [PATCH RFC v2 06/16] ARM: kirkwood: " Sebastian Hesselbarth
2013-08-27 21:28 ` [PATCH RFC v2 07/16] ARM: mvebu: " Sebastian Hesselbarth
2013-08-28 6:28 ` Gregory CLEMENT
2013-08-27 21:28 ` [PATCH RFC v2 08/16] ARM: mxs: " Sebastian Hesselbarth
2013-08-27 23:16 ` Mike Turquette
2013-08-29 11:08 ` Shawn Guo
2013-08-29 11:32 ` Sebastian Hesselbarth
2013-08-29 12:58 ` Shawn Guo
2013-08-27 21:28 ` [PATCH RFC v2 09/16] ARM: nspire: " Sebastian Hesselbarth
2013-08-27 21:28 ` [PATCH RFC v2 10/16] ARM: rockchip: " Sebastian Hesselbarth
2013-08-27 21:28 ` [PATCH RFC v2 11/16] ARM: socfpga: remove call to of_clk_init Sebastian Hesselbarth
2013-08-29 15:38 ` Dinh Nguyen
2013-08-29 17:23 ` Arnd Bergmann
2013-08-27 21:28 ` [PATCH RFC v2 12/16] ARM: sti: remove custom .init_time hook Sebastian Hesselbarth
2013-08-29 16:29 ` [STLinux Kernel] " Srinivas KANDAGATLA
2013-08-27 21:28 ` [PATCH RFC v2 13/16] ARM: tegra: split tegra_pmc_init() in two Sebastian Hesselbarth
2013-08-27 21:59 ` Stephen Warren
2013-08-27 22:09 ` Sebastian Hesselbarth
2013-09-11 19:56 ` Stephen Warren
2013-09-12 6:21 ` Sebastian Hesselbarth
2013-09-12 16:32 ` Stephen Warren
2013-08-27 21:28 ` [PATCH RFC v2 14/16] ARM: vexpress: remove custom .init_time hook Sebastian Hesselbarth
2013-08-29 13:35 ` Arnd Bergmann
2013-08-29 18:16 ` Sebastian Hesselbarth
2013-08-30 10:02 ` Jon Medhurst (Tixy)
2013-08-30 11:39 ` Russell King - ARM Linux
2013-08-30 13:12 ` Pawel Moll
2013-08-27 21:28 ` [PATCH RFC v2 15/16] clk: vt8500: remove call to of_clk_init Sebastian Hesselbarth
2013-08-27 23:16 ` Mike Turquette
2013-08-29 13:42 ` Arnd Bergmann
2013-08-27 21:28 ` [PATCH RFC v2 16/16] ARM: zynq: Don't call of_clk_init() Sebastian Hesselbarth
2013-08-29 13:37 ` Arnd Bergmann
2013-08-29 14:21 ` Michal Simek
2013-08-29 17:20 ` Arnd Bergmann
2013-08-29 18:14 ` Michal Simek
2013-08-29 18:20 ` Sebastian Hesselbarth
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=52172BAA.7020009@gmail.com \
--to=sebastian.hesselbarth@gmail.com \
--cc=arnd@arndb.de \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux@arm.linux.org.uk \
--cc=michal.simek@xilinx.com \
--cc=mturquette@linaro.org \
--cc=s.trumtrar@pengutronix.de \
--cc=soren.brinkmann@xilinx.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).