Skip to contentSkip to conversation

General CHRP/MPC5K2 platform support patch

PATCH NAK Nicolas DET 9 participants 40 messages Oct 25, 2006 – Oct 28, 2006 PowerPC

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 General CHRP/MPC5K2 platform support patch

This patch add support for MPC52xx/CHRP/OFW platform.

It contains the minimal changes required: platform/chrp/ and the interrupt controller code.

This one is slightly different than the one I submitted previously.
http://ozlabs.org/pipermail/linuxppc-dev/2006-October/027133.html

* /proc/ppc64/rtas/ I left it for now. It's not really the concern of this patch.

* mpc5k2_bestcomm.c (& helpers): I've been in contact with others people using bestcomm and please to see the development was still active. I'm already using latest code on my board but this is not include in this patch. Indeed, it needs some changes for ARCH=powerpc. Moreover, we have a different approach for the tasks. I'll be pleased to work with the bestcomm folks.

* mpc5k2.c: As far as I understood, using of_platform_device will allow 'smart' of tree parsing. I did not implement it yet.

* mpc5k2_pic.c : The thingy ;-). I moved it into arch/powerpc/sysdev as suggested. This code is 99% copy/paste from the ARCH=ppc. I added an irq_host with the appropriate xlate and co. However, I have not done more as this code has been written by others. I think we should work together to move to new Linux style irq. I stay ready to test or code upcoming revision/

* arch/powerpc/platform/*. Here, only small changes has been made. I added _CHRP_E5K2 (in asm/processor.h), an entry in chrp_names() and the correct detection in arch/powerpc/platform/setup.c / chrp_setup_arch(). in chrp_init_IRQ(), I set ppc_md.get_irq to mpc52xx_get_irq and call mpc52xx_init_irq().

* fec driver: I did not wrote it and just made small modification to make it works with our bestcomm API. However, we will soon you the same. I think this is out of topic for his patch.

* serial stuff / IBP Freq. As this frequency is MPC52xx specific. I think it would make sense to add a new func mpc52xx_getipbfreq(void). The way it would be implemented may depend of the architecture. Our firmware contains an 'ipb-freq' property in '/builtin/'.

arch/powerpc/platforms/chrp/setup.c +14 -1
diff -uprN a/arch/powerpc/platforms/chrp/setup.c b/arch/powerpc/platforms/chrp/setup.c
@@ -51,6 +51,7 @@
 #include <asm/mpic.h>
 #include <asm/rtas.h>
 #include <asm/xmon.h>
+#include <asm/mpc52xx.h>
 
 #include "chrp.h"
 
@@ -101,7 +102,8 @@ static const char *chrp_names[] = {
 	"Motorola",
 	"IBM or Longtrail",
 	"Genesi Pegasos",
-	"Total Impact Briq"
+	"Total Impact Briq",
+	"bPlan Efika"
 };
 
 void chrp_show_cpuinfo(struct seq_file *m)
@@ -260,6 +262,8 @@ void __init chrp_setup_arch(void)
 		machine = get_property(root, "model", NULL);
 	if (machine && strncmp(machine, "Pegasos", 7) == 0) {
 		_chrp_type = _CHRP_Pegasos;
+	} else if (machine && strncmp(machine, "EFIKA5K2", 8) == 0) {
+		_chrp_type =_CHRP_E5K2;
 	} else if (machine && strncmp(machine, "IBM", 3) == 0) {
 		_chrp_type = _CHRP_IBM;
 	} else if (machine && strncmp(machine, "MOT", 3) == 0) {
@@ -494,6 +498,12 @@ void __init chrp_init_IRQ(void)
 #if defined(CONFIG_VT) && defined(CONFIG_INPUT_ADBHID) && defined(XMON)
 	struct device_node *kbd;
 #endif
+	if (_chrp_type ==  _CHRP_E5K2) {
+		ppc_md.get_irq = mpc52xx_get_irq;
+		mpc52xx_init_irq();
+		return;
+	}
+
 	chrp_find_openpic();
 	chrp_find_8259();
 
@@ -530,6 +540,9 @@ chrp_init2(void)
 	chrp_nvram_init();
 #endif
 
+	if (_chrp_type == _CHRP_E5K2)
+		return;
+
 	request_region(0x20,0x20,"pic1");
 	request_region(0xa0,0x20,"pic2");
 	request_region(0x00,0x20,"dma1");
include/asm-powerpc/processor.h +1 -0
diff -uprN a/include/asm-powerpc/processor.h b/include/asm-powerpc/processor.h
@@ -33,6 +33,7 @@
 #define _CHRP_IBM	0x05	/* IBM chrp, the longtrail and longtrail 2 */
 #define _CHRP_Pegasos	0x06	/* Genesi/bplan's Pegasos and Pegasos2 */
 #define _CHRP_briq	0x07	/* TotalImpact's briQ */
+#define _CHRP_E5K2	0x08	/* bPlan's Efika 5k2*/
 
 #if defined(__KERNEL__) && defined(CONFIG_PPC32)
 
include/asm-ppc/mpc52xx.h +2 -2
diff -uprN a/include/asm-ppc/mpc52xx.h b/include/asm-ppc/mpc52xx.h
@@ -119,7 +119,7 @@ enum ppc_sys_devices {
 #define MPC52xx_SDMA_IRQ_NUM	17
 #define MPC52xx_PERP_IRQ_NUM	23
 
-#define MPC52xx_CRIT_IRQ_BASE	1
+#define MPC52xx_CRIT_IRQ_BASE	0
 #define MPC52xx_MAIN_IRQ_BASE	(MPC52xx_CRIT_IRQ_BASE + MPC52xx_CRIT_IRQ_NUM)
 #define MPC52xx_SDMA_IRQ_BASE	(MPC52xx_MAIN_IRQ_BASE + MPC52xx_MAIN_IRQ_NUM)
 #define MPC52xx_PERP_IRQ_BASE	(MPC52xx_SDMA_IRQ_BASE + MPC52xx_SDMA_IRQ_NUM)
@@ -415,7 +415,7 @@ struct mpc52xx_cdm {
 #ifndef __ASSEMBLY__
 
 extern void mpc52xx_init_irq(void);
-extern int mpc52xx_get_irq(void);
+extern unsigned int mpc52xx_get_irq(void);
 
 extern unsigned long mpc52xx_find_end_of_memory(void);
 extern void mpc52xx_set_bat(void);
arch/powerpc/sysdev/mpc52xx_pic.c +375 -0

375 lines not on this page · raw message

arch/powerpc/sysdev/Makefile +1 -0
@@ -13,6 +13,7 @@ obj-$(CONFIG_FSL_SOC)		+= fsl_soc.o
 obj-$(CONFIG_PPC_TODC)		+= todc.o
 obj-$(CONFIG_TSI108_BRIDGE)	+= tsi108_pci.o tsi108_dev.o
 obj-$(CONFIG_QUICC_ENGINE)	+= qe_lib/
+obj-$(CONFIG_PPC_CHRP)		+= mpc52xx_pic.o
 
 ifeq ($(CONFIG_PPC_MERGE),y)
 obj-$(CONFIG_PPC_I8259)		+= i8259.o

begin:vcard
fn:Nicolas DET ( bplan GmbH )
n:DET;Nicolas
org:bplan GmbH
adr:;;;;;;Germany
email;internet:[email protected]
title:Software Entwicklung
tel;work:+49 6171 9187 - 31
x-mozilla-html:FALSE
url:http://www.bplan-gmbh.de
version:2.1
end:vcard

#1 lore raw

arch/powerpc/platforms/chrp/setup.c void __init chrp_setup_arch(void) · line 266

Quoted 4 lines · Nicolas DET
arch/powerpc/platforms/chrp/setup.c +2 -0
 	if (machine && strncmp(machine, "Pegasos", 7) == 0) {
 		_chrp_type = _CHRP_Pegasos;
+	} else if (machine && strncmp(machine, "EFIKA5K2", 8) == 0) {
+		_chrp_type =_CHRP_E5K2;

This whole _chrp_type thing, and having to do different things based on the root-node model property, is really only a workaround for older machines with inadequate device trees. Decisions about things like which interrupt controller driver(s) to instantiate should be taken based on properties in the appropriate device-tree nodes, for instance the model and compatible properties in the node(s) for the interrupt controller(s). In fact I'd like to get rid of _chrp_type completely.

include/asm-ppc/mpc52xx.h enum ppc_sys_devices { · line 122

Quoted 9 lines

diff -uprN a/include/asm-ppc/mpc52xx.h b/include/asm-ppc/mpc52xx.h

include/asm-ppc/mpc52xx.h +1 -1
@@ -119,7 +119,7 @@ enum ppc_sys_devices {
 #define MPC52xx_SDMA_IRQ_NUM	17
 #define MPC52xx_PERP_IRQ_NUM	23
 
-#define MPC52xx_CRIT_IRQ_BASE	1
+#define MPC52xx_CRIT_IRQ_BASE	0

What is this going to do to other 52xx users?

Paul.

#2 lore raw
Quoted 7 lines · Paul Mackerras

arch/powerpc/platforms/chrp/setup.c void __init chrp_setup_arch(void) · line 266

Quoted 4 lines · Nicolas DET
      if (machine && strncmp(machine, "Pegasos", 7) == 0) {
              _chrp_type = _CHRP_Pegasos;
arch/powerpc/platforms/chrp/setup.c +2 -0
+     } else if (machine && strncmp(machine, "EFIKA5K2", 8) == 0) {
+             _chrp_type =_CHRP_E5K2;
Quoted 8 lines

This whole _chrp_type thing, and having to do different things based on the root-node model property, is really only a workaround for older machines with inadequate device trees. Decisions about things like which interrupt controller driver(s) to instantiate should be taken based on properties in the appropriate device-tree nodes, for instance the model and compatible properties in the node(s) for the interrupt controller(s). In fact I'd like to get rid of _chrp_type completely.

What are the implications anyway for the bplan board being CHRP vs the way the rest of the embedded ppcs are set up? Will having this setup as a CHRP system conflict w/ non-CHRP embedded boards? (I'm assuming that CHRP requires more capable firmware than u-boot passing in a fdt)

Quoted 8 lines
Quoted 8 lines

diff -uprN a/include/asm-ppc/mpc52xx.h b/include/asm-ppc/mpc52xx.h

include/asm-ppc/mpc52xx.h 2006-10-25 19:11:55.000000000 +0200 renamed from include/asm-ppc/mpc52xx.h 2006-10-25 19:07:48.000000000 +0200 +0 -0
@@ -119,7 +119,7 @@ enum ppc_sys_devices {
 #define MPC52xx_SDMA_IRQ_NUM 17
 #define MPC52xx_PERP_IRQ_NUM 23
 
Quoted 4 lines

include/asm-ppc/mpc52xx.h 2006-10-25 19:11:55.000000000 +0200 enum ppc_sys_devices { · line 122

Quoted 2 lines
include/asm-ppc/mpc52xx.h 2006-10-25 19:11:55.000000000 +0200 renamed from include/asm-ppc/mpc52xx.h 2006-10-25 19:07:48.000000000 +0200 +1 -1
-#define MPC52xx_CRIT_IRQ_BASE        1
+#define MPC52xx_CRIT_IRQ_BASE        0

What is this going to do to other 52xx users?

Shouldn't be a big deal if all drivers use the macros; but as you pointed out on IRC, the interrupt stuff needs to be reworked for the interrupt mapping interface anyway

g.

Signature · 4 lines
Grant Likely, B.Sc. P.Eng.
Secret Lab Technologies Ltd.
[email protected]
(403) 399-0195
#3 lore raw
Quoted 5 lines

What are the implications anyway for the bplan board being CHRP vs the way the rest of the embedded ppcs are set up? Will having this setup as a CHRP system conflict w/ non-CHRP embedded boards? (I'm assuming that CHRP requires more capable firmware than u-boot passing in a fdt)

Not really :) The main thing they "buy" by going CHRP is that they use RTAS for PCI config space access and thus don't have to write the host bridge code. Since their firmware initializes everything properly, there is little need for non-generic platform code and that sort of generic platform code is what CHRP provides.

With the device-tree thingy done properly nowadays, there should be little difference between platforms unless you end up dealing with really special things.

There should be no conflict with non-CHRP boards using the MPC5200 chip provided they do things right which is what I've been advocating so far. That includes making sure everybody agrees on the format of the MPC5200 specific bits in the device-tree:

- Format of the interrupt cells (they have come up upon my suggestion
to a 3-cell based format (could probably be compressed a bit but heh) and that should be documented publically)

- Binding (set of required properties and their definitions) for on
chip devices, and more specifically, for representing the bestcomm and the links between devices and associated bestcomm tasks.

If the above is properly agreed upon, then the code for dealing with MPC5200 specific devices will purely rely on those bits, and thus should be useable from whatever platform it's included in, wether it's CHRP, or something else.

Unfortunately, at this point, Nicolas seems to be unwilling to open up his device-tree publically.

Ben.

#4 lore raw
Quoted 4 lines · Benjamin Herrenschmidt

- Format of the interrupt cells (they have come up upon my suggestion
to a 3-cell based format (could probably be compressed a bit but heh) and that should be documented publically)

It's some kind of 'documented' in mpc52xx_pic.c/xlate()

Quoted 3 lines

- Binding (set of required properties and their definitions) for on
chip devices, and more specifically, for representing the bestcomm and the links between devices and associated bestcomm tasks.

Bestcomm is something a bit 'weird'. I think we are the only to use this approach (task code loaded by the OFW and not by the OS). As said earlier, I've been in contact with an bestcomm coders and I think we would work together on this subject.

begin:vcard
fn:Nicolas DET ( bplan GmbH )
n:DET;Nicolas
org:bplan GmbH
adr:;;;;;;Germany
email;internet:[email protected]
title:Software Entwicklung
tel;work:+49 6171 9187 - 31
x-mozilla-html:FALSE
url:http://www.bplan-gmbh.de
version:2.1
end:vcard

#5 lore raw
In reply to Grant Likely
Quoted 16 lines · Paul Mackerras

arch/powerpc/platforms/chrp/setup.c void __init chrp_setup_arch(void) · line 266

Quoted 4 lines · Nicolas DET
arch/powerpc/platforms/chrp/setup.c +2 -0
 	if (machine && strncmp(machine, "Pegasos", 7) == 0) {
 		_chrp_type = _CHRP_Pegasos;
+	} else if (machine && strncmp(machine, "EFIKA5K2", 8) == 0) {
+		_chrp_type =_CHRP_E5K2;

This whole _chrp_type thing, and having to do different things based on the root-node model property, is really only a workaround for older machines with inadequate device trees. Decisions about things like which interrupt controller driver(s) to instantiate should be taken based on properties in the appropriate device-tree nodes, for instance the model and compatible properties in the node(s) for the interrupt controller(s). In fact I'd like to get rid of _chrp_type completely.

I remove this part of the patch which is only cosmetic (good CHRP name).

Quoted 7 lines
Quoted 7 lines

diff -uprN a/include/asm-ppc/mpc52xx.h b/include/asm-ppc/mpc52xx.h

include/asm-ppc/mpc52xx.h +0 -0
@@ -119,7 +119,7 @@ enum ppc_sys_devices {
 #define MPC52xx_SDMA_IRQ_NUM	17
 #define MPC52xx_PERP_IRQ_NUM	23
 
Quoted 5 lines

include/asm-ppc/mpc52xx.h enum ppc_sys_devices { · line 122

Quoted 2 lines
include/asm-ppc/mpc52xx.h +1 -1
-#define MPC52xx_CRIT_IRQ_BASE	1
+#define MPC52xx_CRIT_IRQ_BASE	0

What is this going to do to other 52xx users?

IRQ 0 may cause issue in earlier kernel as far as I understood. New one, using the new IRQ stuff, just works (at least works great here).

This is a revert change from:
http://www.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=e21b9f2e9a580ce7375ec58953c1bb19aabe0db4
Regards,

begin:vcard
fn:Nicolas DET ( bplan GmbH )
n:DET;Nicolas
org:bplan GmbH
adr:;;;;;;Germany
email;internet:[email protected]
title:Software Entwicklung
tel;work:+49 6171 9187 - 31
x-mozilla-html:FALSE
url:http://www.bplan-gmbh.de
version:2.1
end:vcard

#6 lore raw
In reply to Nicolas DET

On Wed, 2006-10-25 at 21:05 +0200, Nicolas DET wrote:

In addition to various whitespace damage in the patch...

arch/powerpc/platforms/chrp/setup.c void __init chrp_init_IRQ(void) · line 505

Quoted 5 lines
arch/powerpc/platforms/chrp/setup.c +5 -0
+	if (_chrp_type ==  _CHRP_E5K2) {
+		ppc_md.get_irq = mpc52xx_get_irq;
+		mpc52xx_init_irq();
+		return;
+	}

As I wrote, the above should be unnecessary.

arch/powerpc/platforms/chrp/setup.c void __init chrp_init_IRQ(void) · line 508

Quoted 2 lines
arch/powerpc/platforms/chrp/setup.c +0 -0
 	chrp_find_openpic();
 	chrp_find_8259();

Just add a

	chrp_find_mpc5200pic();

Which will set itself as the default controller and set ppc_md.get_irq if none have done it before.

arch/powerpc/platforms/chrp/setup.c chrp_init2(void) · line 544

Quoted 6 lines
arch/powerpc/platforms/chrp/setup.c +2 -0
@@ -530,6 +540,9 @@ chrp_init2(void)
 	chrp_nvram_init();
 #endif
 
+	if (_chrp_type == _CHRP_E5K2)
+		return;

Why that ? Not that we really need those request_region() anyway, but it's highly recommended that your firmware doesn't allocate anything in that "legacy" portion of the IO space anyway, so requesting those regions will do no harm.

arch/powerpc/sysdev/mpc52xx_pic.c line 51

Quoted 11 lines
arch/powerpc/sysdev/mpc52xx_pic.c +7 -0
 	request_region(0x20,0x20,"pic1");
 	request_region(0xa0,0x20,"pic2");
 	request_region(0x00,0x20,"dma1");
 
+static void
+mpc52xx_ic_disable(unsigned int virq)
+{
+        u32 val;
+	int irq;
+
+	irq = irq_map[virq].hwirq;

You should test if the result is valid just in case you were called with a bogus irq number

Quoted 16 lines
arch/powerpc/sysdev/mpc52xx_pic.c +16 -0
+	pr_debug("%s: irq=%d\n", __func__, irq);
+
+        if (irq == MPC52xx_IRQ0) {
+                val = in_be32(&intr->ctrl);
+                val &= ~(1 << 11);
+                out_be32(&intr->ctrl, val);
+        } else if (irq < MPC52xx_IRQ1) {
+                BUG();
+        } else if (irq <= MPC52xx_IRQ3) {
+                val = in_be32(&intr->ctrl);
+                val &= ~(1 << (10 - (irq - MPC52xx_IRQ1)));
+                out_be32(&intr->ctrl, val);
+        } else if (irq < MPC52xx_SDMA_IRQ_BASE) {
+                val = in_be32(&intr->main_mask);
+                val |= 1 << (16 - (irq - MPC52xx_MAIN_IRQ_BASE));
+                out_be32(&intr->main_mask, val);

arch/powerpc/sysdev/mpc52xx_pic.c line 78

Quoted 10 lines
arch/powerpc/sysdev/mpc52xx_pic.c +10 -0
+        } else if (irq < MPC52xx_PERP_IRQ_BASE) {
+                val = in_be32(&sdma->IntMask);
+                val |= 1 << (irq - MPC52xx_SDMA_IRQ_BASE);
+                out_be32(&sdma->IntMask, val);
+        } else {
+                val = in_be32(&intr->per_mask);
+                val |= 1 << (31 - (irq - MPC52xx_PERP_IRQ_BASE));
+                out_be32(&intr->per_mask, val);
+        }
+}

You may want to chose a different encoding for your HW irqs to make the above less horrible. For example, have an irq category in some top bits and the actual bit mask pre-mashed in the bottom bits.

And same comments for enable().

arch/powerpc/sysdev/mpc52xx_pic.c line 172

Quoted 6 lines
arch/powerpc/sysdev/mpc52xx_pic.c +6 -0
+static void
+mpc52xx_ic_disable_and_ack(unsigned int irq)
+{
+        mpc52xx_ic_disable(irq);
+        mpc52xx_ic_ack(irq);
+}

You don't need the above. It will be done for you by the core code once you properly adapt to genirq.

arch/powerpc/sysdev/mpc52xx_pic.c line 179

Quoted 6 lines
arch/powerpc/sysdev/mpc52xx_pic.c +6 -0
+static void
+mpc52xx_ic_end(unsigned int irq)
+{
+        if (!(irq_desc[irq].status & (IRQ_DISABLED | IRQ_INPROGRESS)))
+                mpc52xx_ic_enable(irq);
+}

The above is not necessary anymore, you need to properly adapt to genirq which you haven't done yet.

arch/powerpc/sysdev/mpc52xx_pic.c line 187

Quoted 7 lines
arch/powerpc/sysdev/mpc52xx_pic.c +7 -0
+static struct irq_chip mpc52xx_irqchip = {
+	.name	= " MPC52xx  ",
+	.enable		= mpc52xx_ic_enable,
+	.disable	= mpc52xx_ic_disable,
+	.ack		= mpc52xx_ic_disable_and_ack,
+	.end		= mpc52xx_ic_end,
+};

As I said, you haven't properly adapted to genirq. That is

enable -> unmask
disable -> mask
ack stays and doesn't do disable
end is gone

You also need to call set_irq_chip_and_handler(), possibly in your host map() function to set the right flow handler for your interrupts.

arch/powerpc/sysdev/mpc52xx_pic.c line 197

Quoted 9 lines
arch/powerpc/sysdev/mpc52xx_pic.c +9 -0
+static int mpc52xx_irqhost_match(struct irq_host *h, struct device_node *node)
+{
+	pr_debug("%s: node=%p\n", __func__, node);
+
+	if ( device_is_compatible(node, "mpc52xx-pic") )
+		return 1;
+
+	return device_is_compatible(node, "mpc5200-pic");
+}

You probably need only one of the above statements. It depends on how the device-tree binding is defined for the MPC52xx which, for the 10000th time, should be done PUBLICALLY

arch/powerpc/sysdev/mpc52xx_pic.c line 267

Quoted 6 lines
arch/powerpc/sysdev/mpc52xx_pic.c +6 -0
+int mpc52xx_irqhost_map(struct irq_host *h, unsigned int virq, irq_hw_number_t hw)
+{
+	pr_debug("%s: v=%d, hw=%d\n", __func__, virq, (int) hw);
+
+	return 0;
+}

As I said earlier, the above needs to at least set a flow handler. Also, if you intend to have different flow handlers for edge and level irqs, then you'll also probably need to implement set_irq_type(). You may want to look at the IPI driver for that (and beware of the lock problem with set_irq_type() vs. set_irq_handler(), see the IPIC patch that went on the list recently)

arch/powerpc/sysdev/mpc52xx_pic.c line 319

Quoted 12 lines
arch/powerpc/sysdev/mpc52xx_pic.c +12 -0
+        /* Zero a bunch of the priority settings.  */
+        out_be32(&intr->per_pri1, 0);
+        out_be32(&intr->per_pri2, 0);
+        out_be32(&intr->per_pri3, 0);
+        out_be32(&intr->main_pri1, 0);
+        out_be32(&intr->main_pri2, 0);
+        /* Initialize irq_desc[i].handler's with mpc52xx_ic. */
+        for (i = 0; i < NR_IRQS; i++) {
+                irq_desc[i].chip = &mpc52xx_irqchip;
+                irq_desc[i].status = IRQ_LEVEL;
+		
+        }

The above should go. With the port to genirq, you get descriptors as irqs get mapped, from your map callback, which is where you set the chip and flow handler.

You should also never completley override status like that. Just or-in the bits you need. Look at what other PICs in arch/powerpc do.

arch/powerpc/sysdev/mpc52xx_pic.c line 327

Quoted 7 lines
arch/powerpc/sysdev/mpc52xx_pic.c +7 -0
+#define IRQn_MODE(intr_ctrl,irq) (((intr_ctrl) >> (22-(i<<1))) & 0x03)
+        for (i=0 ; i<4 ; i++) {
+                int mode;
+                mode = IRQn_MODE(intr_ctrl,i);
+                if ((mode == 0x1) || (mode == 0x2))
+                        irq_desc[i?MPC52xx_IRQ1+i-1:MPC52xx_IRQ0].status = 0;
+        }

What is the above about ?

arch/powerpc/sysdev/mpc52xx_pic.c line 336

Quoted 8 lines
arch/powerpc/sysdev/mpc52xx_pic.c +8 -0
+	/*
+	 * As last step, add an irq host to translate the real
+	 * hw irq information provided by the ofw to linux virq
+	*/
+
+	mpc52xx_irqhost = irq_alloc_host(IRQ_HOST_MAP_LINEAR, 0, &mpc52xx_irqhost_ops, -1);
+	pr_debug("%s: mpc52xx_irqhost =%p\n", __func__, mpc52xx_irqhost );
+}

Ususally we do that first but heh, it doesn't really matter. However:

passing a count of 0 when creating a linear revmap is totally bogus (in fact, I'm surprised your code works).

Also, as far as the invalid irq is concerned, you should probably define it using a symbolic constant (properly typed) so you can use it elsewhere in the code to test the result of the revmap functions among others.

Quoted 27 lines
arch/powerpc/sysdev/mpc52xx_pic.c +27 -0
+unsigned int
+mpc52xx_get_irq(void)
+{
+        u32 status;
+	int virq;
+        int irq = NO_IRQ_IGNORE;
+
+        status = in_be32(&intr->enc_status);
+        if (status & 0x00000400)
+        {		/* critical */
+                irq = (status >> 8) & 0x3;
+                if (irq == 2)			/* high priority peripheral */
+                        goto peripheral;
+                irq += MPC52xx_CRIT_IRQ_BASE;
+        } else if (status & 0x00200000)
+        {		/* main */
+                irq = (status >> 16) & 0x1f;
+                if (irq == 4)			/* low priority peripheral */
+                        goto peripheral;
+                irq += MPC52xx_MAIN_IRQ_BASE;
+        } else if (status & 0x20000000)
+        {		/* peripheral */
+peripheral:
+                irq = (status >> 24) & 0x1f;
+                if (irq == 0) {			/* bestcomm */
+                        status = in_be32(&sdma->IntPend);
+                        irq = ffs(status) + MPC52xx_SDMA_IRQ_BASE-1;

arch/powerpc/sysdev/mpc52xx_pic.c line 374

Quoted 10 lines
arch/powerpc/sysdev/mpc52xx_pic.c +10 -0
+                } else
+                        irq += MPC52xx_PERP_IRQ_BASE;
+
+        }
+
+	virq = irq_linear_revmap(mpc52xx_irqhost, irq);
+	pr_debug("%s: irq=%d -> %d\n", __func__, irq, virq);
+
+        return virq;
+}

I'm surprised the revmap is working at all... considering the issues you have above. You are lucky but things should be fixed anyway.

arch/powerpc/sysdev/Makefile obj-$(CONFIG_FSL_SOC) += fsl_soc.o · line 16

Quoted 7 lines
arch/powerpc/sysdev/Makefile +1 -0
@@ -13,6 +13,7 @@ obj-$(CONFIG_FSL_SOC)		+= fsl_soc.o
 obj-$(CONFIG_PPC_TODC)		+= todc.o
 obj-$(CONFIG_TSI108_BRIDGE)	+= tsi108_pci.o tsi108_dev.o
 obj-$(CONFIG_QUICC_ENGINE)	+= qe_lib/
+obj-$(CONFIG_PPC_CHRP)		+= mpc52xx_pic.o

That's not the proper way to do it. Instead, define an invisible CONFIG_MPC52xx_PIC and have the chrp platform select it.

arch/powerpc/sysdev/Makefile obj-$(CONFIG_FSL_SOC) += fsl_soc.o · line 19

Quoted 2 lines
arch/powerpc/sysdev/Makefile +0 -0
 ifeq ($(CONFIG_PPC_MERGE),y)
 obj-$(CONFIG_PPC_I8259)		+= i8259.o

Ben.

#7 lore raw
Quoted 22 lines · Benjamin Herrenschmidt

On Wed, 2006-10-25 at 21:05 +0200, Nicolas DET wrote:

In addition to various whitespace damage in the patch...

Review

Quoted 5 lines
(quoted diff) +5 -0
+	if (_chrp_type ==  _CHRP_E5K2) {
+		ppc_md.get_irq = mpc52xx_get_irq;
+		mpc52xx_init_irq();
+		return;
+	}

As I wrote, the above should be unnecessary.

Review

Quoted 2 lines
(quoted diff) +0 -0
 	chrp_find_openpic();
 	chrp_find_8259();

Just add a

	chrp_find_mpc5200pic();

Which will set itself as the default controller and set ppc_md.get_irq if none have done it before.

Quoted 7 lines

Review chrp_init2(void) · line 544

Quoted 6 lines
(quoted diff) +2 -0
@@ -530,6 +540,9 @@ chrp_init2(void)
 	chrp_nvram_init();
 #endif
 
+	if (_chrp_type == _CHRP_E5K2)
+		return;

Done.

Quoted 4 lines
Quoted 4 lines
(quoted diff) +4 -0
+static void
+mpc52xx_ic_disable(unsigned int virq)
+{
+        u32 val;
Quoted 9 lines

Review chrp_init2(void)

Quoted 3 lines
(quoted diff) +3 -0
+	int irq;
+
+	irq = irq_map[virq].hwirq;

You should test if the result is valid just in case you were called with [...]
I'm surprised the revmap is working at all... considering the issues you have above. You are lucky but things should be fixed anyway.

As said in the mail. I'm waiting for the opinion of the original authors and the mainteners. What's why I did not made much change in this file. In the current patch, I know do:
mpc52xx_irqhost = irq_alloc_host(IRQ_HOST_MAP_LINEAR, NR_IRQS, &mpc52xx_irqhost_ops, -1);

Which should be ok.

Quoted 8 lines

arch/powerpc/sysdev/Makefile obj-$(CONFIG_FSL_SOC) += fsl_soc.o · line 16

Quoted 7 lines
arch/powerpc/sysdev/Makefile +1 -0
@@ -13,6 +13,7 @@ obj-$(CONFIG_FSL_SOC)		+= fsl_soc.o
 obj-$(CONFIG_PPC_TODC)		+= todc.o
 obj-$(CONFIG_TSI108_BRIDGE)	+= tsi108_pci.o tsi108_dev.o
 obj-$(CONFIG_QUICC_ENGINE)	+= qe_lib/
+obj-$(CONFIG_PPC_CHRP)		+= mpc52xx_pic.o

Done.

Regards

arch/powerpc/sysdev/Makefile +1 -0
@@ -13,6 +13,7 @@ obj-$(CONFIG_FSL_SOC)		+= fsl_soc.o
 obj-$(CONFIG_PPC_TODC)		+= todc.o
 obj-$(CONFIG_TSI108_BRIDGE)	+= tsi108_pci.o tsi108_dev.o
 obj-$(CONFIG_QUICC_ENGINE)	+= qe_lib/
+obj-$(CONFIG_PPC_MPC52xx_PIC)	+= mpc52xx_pic.o
 
 ifeq ($(CONFIG_PPC_MERGE),y)
 obj-$(CONFIG_PPC_I8259)		+= i8259.o
arch/powerpc/sysdev/mpc52xx_pic.c +371 -0

371 lines not on this page · raw message

arch/powerpc/Kconfig +6 -0
@@ -384,6 +384,12 @@ config PPC_CHRP
 	select PPC_RTAS
 	select PPC_MPC106
 	select PPC_UDBG_16550
+	select PPC_MPC52xx_PIC
+	default y
+
+config PPC_MPC52xx_PIC
+        bool
+	depends on PPC_CHRP
 	default y
 
 config PPC_PMAC
arch/powerpc/platforms/chrp/setup.c +52 -2

72 lines not on this page · raw message

include/asm-ppc/mpc52xx.h +2 -2
diff -uprN a/include/asm-ppc/mpc52xx.h b/include/asm-ppc/mpc52xx.h
@@ -119,7 +119,7 @@ enum ppc_sys_devices {
 #define MPC52xx_SDMA_IRQ_NUM	17
 #define MPC52xx_PERP_IRQ_NUM	23
 
-#define MPC52xx_CRIT_IRQ_BASE	1
+#define MPC52xx_CRIT_IRQ_BASE	0
 #define MPC52xx_MAIN_IRQ_BASE	(MPC52xx_CRIT_IRQ_BASE + MPC52xx_CRIT_IRQ_NUM)
 #define MPC52xx_SDMA_IRQ_BASE	(MPC52xx_MAIN_IRQ_BASE + MPC52xx_MAIN_IRQ_NUM)
 #define MPC52xx_PERP_IRQ_BASE	(MPC52xx_SDMA_IRQ_BASE + MPC52xx_SDMA_IRQ_NUM)
@@ -415,7 +415,7 @@ struct mpc52xx_cdm {
 #ifndef __ASSEMBLY__
 
 extern void mpc52xx_init_irq(void);
-extern int mpc52xx_get_irq(void);
+extern unsigned int mpc52xx_get_irq(void);
 
 extern unsigned long mpc52xx_find_end_of_memory(void);
 extern void mpc52xx_set_bat(void);

begin:vcard
fn:Nicolas DET ( bplan GmbH )
n:DET;Nicolas
org:bplan GmbH
adr:;;;;;;Germany
email;internet:[email protected]
title:Software Entwicklung
tel;work:+49 6171 9187 - 31
x-mozilla-html:FALSE
url:http://www.bplan-gmbh.de
version:2.1
end:vcard

#8 lore raw
Quoted 5 lines

rent patch, I know do:
mpc52xx_irqhost = irq_alloc_host(IRQ_HOST_MAP_LINEAR, NR_IRQS, &mpc52xx_irqhost_ops, -1);

Which should be ok.

NR_IRQ, it's a bit violent :) It's also the max number of virtual IRQs and thus has nothing to do with your HW numbers. Use the max number of HW IRQs here.

Ben.

#9 lore raw
Quoted 13 lines · Benjamin Herrenschmidt
Quoted 4 lines

rent patch, I know do:
mpc52xx_irqhost = irq_alloc_host(IRQ_HOST_MAP_LINEAR, NR_IRQS, &mpc52xx_irqhost_ops, -1);

Which should be ok.

NR_IRQ, it's a bit violent :) It's also the max number of virtual IRQs and thus has nothing to do with your HW numbers. Use the max number of HW IRQs here.

Ben.

arch/powerpc/sysdev/Makefile +1 -0
@@ -13,6 +13,7 @@ obj-$(CONFIG_FSL_SOC)		+= fsl_soc.o
 obj-$(CONFIG_PPC_TODC)		+= todc.o
 obj-$(CONFIG_TSI108_BRIDGE)	+= tsi108_pci.o tsi108_dev.o
 obj-$(CONFIG_QUICC_ENGINE)	+= qe_lib/
+obj-$(CONFIG_PPC_MPC52xx_PIC)	+= mpc52xx_pic.o
 
 ifeq ($(CONFIG_PPC_MERGE),y)
 obj-$(CONFIG_PPC_I8259)		+= i8259.o
arch/powerpc/sysdev/mpc52xx_pic.c +371 -0

371 lines not on this page · raw message

arch/powerpc/Kconfig +6 -0
@@ -384,6 +384,12 @@ config PPC_CHRP
 	select PPC_RTAS
 	select PPC_MPC106
 	select PPC_UDBG_16550
+	select PPC_MPC52xx_PIC
+	default y
+
+config PPC_MPC52xx_PIC
+        bool
+	depends on PPC_CHRP
 	default y
 
 config PPC_PMAC
arch/powerpc/platforms/chrp/setup.c +52 -2

72 lines not on this page · raw message

include/asm-ppc/mpc52xx.h +2 -2
diff -uprN a/include/asm-ppc/mpc52xx.h b/include/asm-ppc/mpc52xx.h
@@ -119,7 +119,7 @@ enum ppc_sys_devices {
 #define MPC52xx_SDMA_IRQ_NUM	17
 #define MPC52xx_PERP_IRQ_NUM	23
 
-#define MPC52xx_CRIT_IRQ_BASE	1
+#define MPC52xx_CRIT_IRQ_BASE	0
 #define MPC52xx_MAIN_IRQ_BASE	(MPC52xx_CRIT_IRQ_BASE + MPC52xx_CRIT_IRQ_NUM)
 #define MPC52xx_SDMA_IRQ_BASE	(MPC52xx_MAIN_IRQ_BASE + MPC52xx_MAIN_IRQ_NUM)
 #define MPC52xx_PERP_IRQ_BASE	(MPC52xx_SDMA_IRQ_BASE + MPC52xx_SDMA_IRQ_NUM)
@@ -415,7 +415,7 @@ struct mpc52xx_cdm {
 #ifndef __ASSEMBLY__
 
 extern void mpc52xx_init_irq(void);
-extern int mpc52xx_get_irq(void);
+extern unsigned int mpc52xx_get_irq(void);
 
 extern unsigned long mpc52xx_find_end_of_memory(void);
 extern void mpc52xx_set_bat(void);

begin:vcard
fn:Nicolas DET ( bplan GmbH )
n:DET;Nicolas
org:bplan GmbH
adr:;;;;;;Germany
email;internet:[email protected]
title:Software Entwicklung
tel;work:+49 6171 9187 - 31
x-mozilla-html:FALSE
url:http://www.bplan-gmbh.de
version:2.1
end:vcard

#10 lore raw

(minor) Whitespace is still inconsistent; mostly spaces used where they should be tabs and a few lines longer than 80 chars. Running the source through scripts/Lindent will help.

arch/powerpc/sysdev/mpc52xx_pic.c 2006-10-26 10:55:44.000000000 +0200 line 6

Quoted 9 lines · Nicolas DET
arch/powerpc/sysdev/mpc52xx_pic.c 2006-10-26 10:55:44.000000000 +0200 renamed from arch/powerpc/sysdev/mpc52xx_pic.c 2006-10-25 19:07:24.000000000 +0200 +6 -0
@@ -0,0 +1,371 @@
+/*
+ * arch/powerpc/sysdev/mpc52xx_pic.c
+ *
+ * Programmable Interrupt Controller functions for the Freescale MPC52xx
+ * embedded CPU.
+ * Modified for CHRP Efika 5K2

Don't really need to add the "Modified for.." line; That's what git is for. Besides this code is modified for more than just the Efika now. :)

arch/powerpc/sysdev/mpc52xx_pic.c 2006-10-26 10:55:44.000000000 +0200 line 11

Quoted 5 lines
arch/powerpc/sysdev/mpc52xx_pic.c 2006-10-26 10:55:44.000000000 +0200 renamed from arch/powerpc/sysdev/mpc52xx_pic.c 2006-10-25 19:07:24.000000000 +0200 +5 -0
+ *
+ * Maintainer : Sylvain Munaut <[email protected]>
+ *
+ * Based on (well, mostly copied from) the code from the 2.4 kernel by
+ * Dale Farnsworth <[email protected]> and Kent Borg.

Ditto for this; but I know you didn't add it. :) Although this kind of comment is more justified because it captures information not maintained in git. (Namely where the original file came from.)

arch/powerpc/sysdev/mpc52xx_pic.c 2006-10-26 10:55:44.000000000 +0200 line 14

Quoted 3 lines
arch/powerpc/sysdev/mpc52xx_pic.c 2006-10-26 10:55:44.000000000 +0200 renamed from arch/powerpc/sysdev/mpc52xx_pic.c 2006-10-25 19:07:24.000000000 +0200 +3 -0
+ *
+ * Copyright (C) 2004 Sylvain Munaut <[email protected]>
+ * Copyright (C) 2003 Montavista Software, Inc

You should add a bplan copyright line too.

arch/powerpc/sysdev/mpc52xx_pic.c 2006-10-26 10:55:44.000000000 +0200 line 21

Quoted 7 lines
arch/powerpc/sysdev/mpc52xx_pic.c 2006-10-26 10:55:44.000000000 +0200 renamed from arch/powerpc/sysdev/mpc52xx_pic.c 2006-10-25 19:07:24.000000000 +0200 +7 -0
+ *
+ * This file is licensed under the terms of the GNU General Public License
+ * version 2. This program is licensed "as is" without any warranty of any
+ * kind, whether express or implied.
+ */
+
+//#define DEBUG

C++ style comment bad. :)

Quoted 22 lines
arch/powerpc/sysdev/mpc52xx_pic.c 2006-10-26 10:55:44.000000000 +0200 renamed from arch/powerpc/sysdev/mpc52xx_pic.c 2006-10-25 19:07:24.000000000 +0200 +22 -0
+
+#include <linux/stddef.h>
+#include <linux/init.h>
+#include <linux/sched.h>
+#include <linux/signal.h>
+#include <linux/stddef.h>
+#include <linux/delay.h>
+#include <linux/irq.h>
+
+#include <asm/io.h>
+#include <asm/processor.h>
+#include <asm/system.h>
+#include <asm/irq.h>
+#include <asm/prom.h>
+
+#include <asm/mpc52xx.h>
+
+static struct mpc52xx_intr __iomem *intr;
+static struct mpc52xx_sdma __iomem *sdma;
+
+static struct irq_host *mpc52xx_irqhost = NULL;
+

(snip)

arch/powerpc/Kconfig 2006-10-26 11:32:54.000000000 +0200 config PPC_CHRP · line 393

Quoted 13 lines
arch/powerpc/Kconfig 2006-10-26 11:32:54.000000000 +0200 renamed from arch/powerpc/Kconfig 2006-10-25 19:07:23.000000000 +0200 +6 -0
@@ -384,6 +384,12 @@ config PPC_CHRP
        select PPC_RTAS
        select PPC_MPC106
        select PPC_UDBG_16550
+       select PPC_MPC52xx_PIC
+       default y
+
+config PPC_MPC52xx_PIC
+        bool
+       depends on PPC_CHRP
        default y

This isn't good for embedded 52xx boards. Anything u-boot based cannot be CHRP. Please drop the "depends on" line

Quoted 20 lines
arch/powerpc/platforms/chrp/setup.c 2006-10-26 10:59:39.000000000 +0200 renamed from arch/powerpc/platforms/chrp/setup.c 2006-10-25 19:07:23.000000000 +0200 +7 -0
@@ -51,6 +51,7 @@
 #include <asm/mpic.h>
 #include <asm/rtas.h>
 #include <asm/xmon.h>
+#include <asm/mpc52xx.h>
 
 #include "chrp.h"
 
@@ -435,6 +436,51 @@ static struct irqaction xmon_irqaction =
 };
 #endif
 
+
+struct device_node *find_mpc52xx_pic(void)
+{
+       struct device_node *dev;
+       const char *piccompatible_list[] =
+       {

arch/powerpc/platforms/chrp/setup.c 2006-10-26 10:59:39.000000000 +0200 static struct irqaction xmon_irqaction = · line 454

Quoted 10 lines
arch/powerpc/platforms/chrp/setup.c 2006-10-26 10:59:39.000000000 +0200 renamed from arch/powerpc/platforms/chrp/setup.c 2006-10-25 19:07:23.000000000 +0200 +10 -0
+               "mpc5200-interrupt-controller",
+               "mpc52xx-interrupt-controller",
+               "mpc52xx-pic",
+               "mpc5200-pic",
+               "5200-interrupt-controller",
+               "52xx-interrupt-controller",
+               "52xx-pic",
+               "5200-pic",
+               NULL
+       };

Considering Efika is the *only* CHRP 52xx board out there at the moment, and only a handful of embedded folks trying to move it to arch/powerpc, surely we can come to an agreement now on what this thing is named. :)

Seriously, just recommend a name to use and everyone will use it. Providing a list of names means that every name in your list will be used.

Quoted 34 lines
arch/powerpc/platforms/chrp/setup.c 2006-10-26 10:59:39.000000000 +0200 renamed from arch/powerpc/platforms/chrp/setup.c 2006-10-25 19:07:23.000000000 +0200 +29 -0
+
+       /* Look for an MPC52xx interrupt controller */
+       for_each_node_by_type(dev, "interrupt-controller")
+       {
+               const char **piccompatible_entry = piccompatible_list;
+
+               for(piccompatible_entry = piccompatible_list; *piccompatible_entry; piccompatible_entry++ )
+               {
+                       if (device_is_compatible(dev, *piccompatible_entry ))
+                               return dev;
+               }
+       }
+
+       return NULL;
+}
+
+static int  __init chrp_find_mpc52xx_pic(void)
+{
+       if (find_mpc52xx_pic())
+       {
+               printk(KERN_INFO "Found MPC52xx Interrupt Controller\n");
+               ppc_md.get_irq = mpc52xx_get_irq;
+               mpc52xx_init_irq();
+               return 0;
+       }
+
+       return -ENODEV;
+}
+
 static void __init chrp_find_8259(void)
 {
        struct device_node *np, *pic = NULL;
@@ -494,8 +540,12 @@ void __init chrp_init_IRQ(void)
 #if defined(CONFIG_VT) && defined(CONFIG_INPUT_ADBHID) && defined(XMON)

arch/powerpc/platforms/chrp/setup.c 2006-10-26 10:59:39.000000000 +0200 void __init chrp_init_IRQ(void) · line 548

Quoted 10 lines
arch/powerpc/platforms/chrp/setup.c 2006-10-26 10:59:39.000000000 +0200 renamed from arch/powerpc/platforms/chrp/setup.c 2006-10-25 19:07:23.000000000 +0200 +6 -2
        struct device_node *kbd;
 #endif
-       chrp_find_openpic();
-       chrp_find_8259();
+
+       if ( chrp_find_mpc52xx_pic() < 0)
+       {
+               chrp_find_openpic();
+               chrp_find_8259();
+       }

Why? Why not just add the call to chrp_find_mpc52xx_pic() before or after the other two chrp_find_ calls? Will calling them cause things to break?

Keep the code simple.

Cheers,
g.

Signature · 4 lines
Grant Likely, B.Sc. P.Eng.
Secret Lab Technologies Ltd.
[email protected]
(403) 399-0195
#11 lore raw
Quoted 10 lines · Grant Likely

arch/powerpc/platforms/chrp/setup.c void __init chrp_init_IRQ(void) · line 548

Quoted 8 lines · Nicolas DET
arch/powerpc/platforms/chrp/setup.c +6 -2
-       chrp_find_openpic();
-       chrp_find_8259();
+
+       if ( chrp_find_mpc52xx_pic() < 0)
+       {
+               chrp_find_openpic();
+               chrp_find_8259();
+       }
Quoted 5 lines

Why? Why not just add the call to chrp_find_mpc52xx_pic() before or after the other two chrp_find_ calls? Will calling them cause things to break?

Keep the code simple.

... as possible; but no simpler.

:)

g.

Signature · 4 lines
Grant Likely, B.Sc. P.Eng.
Secret Lab Technologies Ltd.
[email protected]
(403) 399-0195
#12 lore raw
Quoted 11 lines · Grant Likely
Quoted 10 lines · Grant Likely
Quoted 8 lines · Nicolas DET
(quoted diff) +6 -2
-       chrp_find_openpic();
-       chrp_find_8259();
+
+       if ( chrp_find_mpc52xx_pic() < 0)
+       {
+               chrp_find_openpic();
+               chrp_find_8259();
+       }
Quoted 4 lines
Quoted 4 lines

Why? Why not just add the call to chrp_find_mpc52xx_pic() before or after the other two chrp_find_ calls? Will calling them cause things to break?

As decided on the IRC channel. a patch which remove the check but fix chrp_find_8259().

Regards,

arch/powerpc/sysdev/Makefile +1 -0
@@ -13,6 +13,7 @@ obj-$(CONFIG_FSL_SOC)		+= fsl_soc.o
 obj-$(CONFIG_PPC_TODC)		+= todc.o
 obj-$(CONFIG_TSI108_BRIDGE)	+= tsi108_pci.o tsi108_dev.o
 obj-$(CONFIG_QUICC_ENGINE)	+= qe_lib/
+obj-$(CONFIG_PPC_MPC52xx_PIC)	+= mpc52xx_pic.o
 
 ifeq ($(CONFIG_PPC_MERGE),y)
 obj-$(CONFIG_PPC_I8259)		+= i8259.o
arch/powerpc/sysdev/mpc52xx_pic.c +368 -0

368 lines not on this page · raw message

arch/powerpc/Kconfig +5 -0
@@ -384,6 +384,11 @@ config PPC_CHRP
 	select PPC_RTAS
 	select PPC_MPC106
 	select PPC_UDBG_16550
+	select PPC_MPC52xx_PIC
+	default y
+
+config PPC_MPC52xx_PIC
+        bool
 	default y
 
 config PPC_PMAC
arch/powerpc/platforms/chrp/setup.c +53 -2

79 lines not on this page · raw message

include/asm-ppc/mpc52xx.h +2 -2
diff -uprN a/include/asm-ppc/mpc52xx.h b/include/asm-ppc/mpc52xx.h
@@ -119,7 +119,7 @@ enum ppc_sys_devices {
 #define MPC52xx_SDMA_IRQ_NUM	17
 #define MPC52xx_PERP_IRQ_NUM	23
 
-#define MPC52xx_CRIT_IRQ_BASE	1
+#define MPC52xx_CRIT_IRQ_BASE	0
 #define MPC52xx_MAIN_IRQ_BASE	(MPC52xx_CRIT_IRQ_BASE + MPC52xx_CRIT_IRQ_NUM)
 #define MPC52xx_SDMA_IRQ_BASE	(MPC52xx_MAIN_IRQ_BASE + MPC52xx_MAIN_IRQ_NUM)
 #define MPC52xx_PERP_IRQ_BASE	(MPC52xx_SDMA_IRQ_BASE + MPC52xx_SDMA_IRQ_NUM)
@@ -415,7 +415,7 @@ struct mpc52xx_cdm {
 #ifndef __ASSEMBLY__
 
 extern void mpc52xx_init_irq(void);
-extern int mpc52xx_get_irq(void);
+extern unsigned int mpc52xx_get_irq(void);
 
 extern unsigned long mpc52xx_find_end_of_memory(void);
 extern void mpc52xx_set_bat(void);

begin:vcard
fn:Nicolas DET ( bplan GmbH )
n:DET;Nicolas
org:bplan GmbH
adr:;;;;;;Germany
email;internet:[email protected]
title:Software Entwicklung
tel;work:+49 6171 9187 - 31
x-mozilla-html:FALSE
url:http://www.bplan-gmbh.de
version:2.1
end:vcard

#13 lore raw

Hi Nicolas,

I don't have much to add to what's already been posted, so here's a couple of details,
mainly compatibility concerns with the legacy arch/ppc tree.

include/asm-ppc/mpc52xx.h enum ppc_sys_devices { · line 122

Quoted 10 lines

diff -uprN a/include/asm-ppc/mpc52xx.h b/include/asm-ppc/mpc52xx.h

include/asm-ppc/mpc52xx.h +1 -1
@@ -119,7 +119,7 @@ enum ppc_sys_devices {
 #define MPC52xx_SDMA_IRQ_NUM	17
 #define MPC52xx_PERP_IRQ_NUM	23
 
-#define MPC52xx_CRIT_IRQ_BASE	1
+#define MPC52xx_CRIT_IRQ_BASE	0

As explained in IRC, I'm not sure about this.
I was previously talked into applying this to avoid using interrupt number 0. In the LITE5200 IRQ0 is used for PCI. It's interrupt number is equal to MPC52xx_CRIT_IRQ_BASE and if it's 0 some drivers just don't work.
That's when using the arch/ppc code. Apparently this is no longer an issue in arch/powerpc (someone to confirm and explain me why ? I would guess the irq_linear_revmap thingie ), but can a "standard" lite5200 be booted in a arch/powerpc kernel ?

So if no critical, I would postpone that change until it's confirmed a lite5200 can boot ...

include/asm-ppc/mpc52xx.h struct mpc52xx_cdm { · line 418

Quoted 2 lines
include/asm-ppc/mpc52xx.h +1 -1
-extern int mpc52xx_get_irq(void);
+extern unsigned int mpc52xx_get_irq(void);

Mmmh, the one in arch/ppc is not unsigned but I guess the warning is fine.

Finally, some triviality : The patch adds some space on empty lines. Also, the indentation is sometimes done with space and sometimes with space ...

Sylvain

#14 lore raw
Quoted 6 lines · Sylvain Munaut

Hi Nicolas,

I don't have much to add to what's already been posted, so here's a couple of details,
mainly compatibility concerns with the legacy arch/ppc tree.

I'd rather not clobber the code to be compatible with arch/ppc... once 52xx is ported over, I'm happy to completely drop support for arch/ppc 52xx. If we want to keep dual support for a while and we are changing APIs too much, we can just fork the drivers for a kernel version of two until we are confident enough to drop the arch/ppc version.

Quoted 7 lines
Quoted 7 lines

diff -uprN a/include/asm-ppc/mpc52xx.h b/include/asm-ppc/mpc52xx.h

include/asm-ppc/mpc52xx.h +0 -0
@@ -119,7 +119,7 @@ enum ppc_sys_devices {
 #define MPC52xx_SDMA_IRQ_NUM	17
 #define MPC52xx_PERP_IRQ_NUM	23
 
Quoted 8 lines

include/asm-ppc/mpc52xx.h enum ppc_sys_devices { · line 122

Quoted 2 lines
include/asm-ppc/mpc52xx.h +1 -1
-#define MPC52xx_CRIT_IRQ_BASE	1
+#define MPC52xx_CRIT_IRQ_BASE	0

As explained in IRC, I'm not sure about this.
I was previously talked into applying this to avoid using interrupt number 0. In the LITE5200 IRQ0 is used for PCI. It's interrupt number is equal to MPC52xx_CRIT_IRQ_BASE and if it's 0 some drivers just don't work.

arch/powerpc has completely virtualized interrupt numbering. The "hardware" numbers supported by a given interrupt controller are generally not visible to a driver and can have any value including 0. The system will assign virtual numbers when interrupts are "mapped" and those will never be 0.

Quoted 4 lines

That's when using the arch/ppc code. Apparently this is no longer an issue in arch/powerpc (someone to confirm and explain me why ? I would guess the irq_linear_revmap thingie ), but can a "standard" lite5200 be booted in a arch/powerpc kernel ?

Once the code has been ported over :-)

Quoted 2 lines

So if no critical, I would postpone that change until it's confirmed a lite5200 can boot ...

It's not been ported yet I presume. Anyway, as I said, we shouldn't bother too much about what is in arch/ppc. If we think we are changing things in a conflicting way, then we should fork the drivers and includes (you can have the same include in both asm-ppc and asm-powerpc, the one matching you building ARCH will be used and they can have different definitions).

Quoted 4 lines

include/asm-ppc/mpc52xx.h struct mpc52xx_cdm { · line 418

Quoted 2 lines
include/asm-ppc/mpc52xx.h +1 -1
-extern int mpc52xx_get_irq(void);
+extern unsigned int mpc52xx_get_irq(void);

Mmmh, the one in arch/ppc is not unsigned but I guess the warning is fine.

We should instead create a new mpc52xx.h in include/asm-powerpc and leave arch/ppc alone

Quoted 2 lines

Finally, some triviality : The patch adds some space on empty lines. Also, the indentation is sometimes done with space and sometimes with space ...

Yeah, I've noticed that the patch is fairly whitespace damaged. It's also incorrect in many ways concerning the hooking to genirq, I'm surprised it works at all in fact.

It might be useful if somebody got me one of those boards (either efika or lite5200b) so I can do some proper rework of that stuff.

Ben.

#15 lore raw
In reply to Nicolas DET

I've already made these comments on IRC; but I'll repeat them here for posterity. :)

General:
- Run this code through Lindent; fixes up spacing and line endings

Quoted 20 lines · Nicolas DET
arch/powerpc/platforms/chrp/setup.c 2006-10-26 18:56:34.000000000 +0200 renamed from arch/powerpc/platforms/chrp/setup.c 2006-10-25 19:07:23.000000000 +0200 +7 -0
@@ -51,6 +51,7 @@
 #include <asm/mpic.h>
 #include <asm/rtas.h>
 #include <asm/xmon.h>
+#include <asm/mpc52xx.h>
 
 #include "chrp.h"
 
@@ -435,6 +436,51 @@ static struct irqaction xmon_irqaction =
 };
 #endif
 
+
+struct device_node *find_mpc52xx_pic(void)
+{
+       struct device_node *dev;
+       const char *piccompatible_list[] =
+       {

arch/powerpc/platforms/chrp/setup.c 2006-10-26 18:56:34.000000000 +0200 static struct irqaction xmon_irqaction = · line 454

Quoted 10 lines
arch/powerpc/platforms/chrp/setup.c 2006-10-26 18:56:34.000000000 +0200 renamed from arch/powerpc/platforms/chrp/setup.c 2006-10-25 19:07:23.000000000 +0200 +10 -0
+               "mpc5200-interrupt-controller",
+               "mpc52xx-interrupt-controller",
+               "mpc52xx-pic",
+               "mpc5200-pic",
+               "5200-interrupt-controller",
+               "52xx-interrupt-controller",
+               "52xx-pic",
+               "5200-pic",
+               NULL
+       };

As you mentioned on IRC that benh suggested "mpc52xx-pic"; Just call it mpc52xx-pic and be done with it. Don't need to loop over this table. If incompatible variants appear in the future; this can change.

Quoted 36 lines
arch/powerpc/platforms/chrp/setup.c 2006-10-26 18:56:34.000000000 +0200 renamed from arch/powerpc/platforms/chrp/setup.c 2006-10-25 19:07:23.000000000 +0200 +29 -0
+
+       /* Look for an MPC52xx interrupt controller */
+       for_each_node_by_type(dev, "interrupt-controller")
+       {
+               const char **piccompatible_entry = piccompatible_list;
+
+               for(piccompatible_entry = piccompatible_list; *piccompatible_entry; piccompatible_entry++ )
+               {
+                       if (device_is_compatible(dev, *piccompatible_entry ))
+                               return dev;
+               }
+       }
+
+       return NULL;
+}
+
+static int  __init chrp_find_mpc52xx_pic(void)
+{
+       if (find_mpc52xx_pic())
+       {
+               printk(KERN_INFO "Found MPC52xx Interrupt Controller\n");
+               ppc_md.get_irq = mpc52xx_get_irq;
+               mpc52xx_init_irq();
+               return 0;
+       }
+
+       return -ENODEV;
+}
+
 static void __init chrp_find_8259(void)
 {
        struct device_node *np, *pic = NULL;
@@ -473,8 +519,11 @@ static void __init chrp_find_8259(void)
                break;
        }
        if (np == NULL)

arch/powerpc/platforms/chrp/setup.c 2006-10-26 18:56:34.000000000 +0200 static void __init chrp_find_8259(void) · line 529

Quoted 10 lines
arch/powerpc/platforms/chrp/setup.c 2006-10-26 18:56:34.000000000 +0200 renamed from arch/powerpc/platforms/chrp/setup.c 2006-10-25 19:07:23.000000000 +0200 +5 -2
-               printk(KERN_WARNING "Cannot find PCI interrupt acknowledge"
-                      " address, polling\n");
+       {
+               printk(KERN_WARNING "Cannot find PCI/i8259 interrupt acknowledge"
+                      " Fix your tree!\n");
+               return;
+       }
 
        i8259_init(pic, chrp_int_ack);
        if (ppc_md.get_irq == NULL)

Missing the fixup to prevent i8258 from clobbering the 52xx-pic. You do need to have that change in this patch; or have this patch dependent on another patch that fixes it. Otherwise, as you know, this patch is worthless! :)

And on the other side of this coin; I don't think that the message change in chrp_find_8259 has anything to do with adding mpc52xx. :) (Unfortunately, I have tendencies towards perfectionism)

And on the flip side, I would (*

Quoted 13 lines
arch/powerpc/platforms/chrp/setup.c 2006-10-26 18:56:34.000000000 +0200 renamed from arch/powerpc/platforms/chrp/setup.c 2006-10-25 19:07:23.000000000 +0200 +2 -0
@@ -494,6 +543,8 @@ void __init chrp_init_IRQ(void)
 #if defined(CONFIG_VT) && defined(CONFIG_INPUT_ADBHID) && defined(XMON)
        struct device_node *kbd;
 #endif
+
+       chrp_find_mpc52xx_pic();
        chrp_find_openpic();
        chrp_find_8259();
 

diff -uprN a/include/asm-ppc/mpc52xx.h b/include/asm-ppc/mpc52xx.h

include/asm-ppc/mpc52xx.h 2006-10-25 19:11:55.000000000 +0200 enum ppc_sys_devices { · line 122

Quoted 5 lines
include/asm-ppc/mpc52xx.h 2006-10-25 19:11:55.000000000 +0200 renamed from include/asm-ppc/mpc52xx.h 2006-10-25 19:07:48.000000000 +0200 +1 -1
@@ -119,7 +119,7 @@ enum ppc_sys_devices {
 #define MPC52xx_SDMA_IRQ_NUM   17
 #define MPC52xx_PERP_IRQ_NUM   23
 
-#define MPC52xx_CRIT_IRQ_BASE  1
+#define MPC52xx_CRIT_IRQ_BASE  0

Is this *strictly* necessary for adding mpc52xx-pic support to arch/powerpc? If not, then move it to a separate patch.

include/asm-ppc/mpc52xx.h 2006-10-25 19:11:55.000000000 +0200 struct mpc52xx_cdm { · line 418

Quoted 9 lines
include/asm-ppc/mpc52xx.h 2006-10-25 19:11:55.000000000 +0200 renamed from include/asm-ppc/mpc52xx.h 2006-10-25 19:07:48.000000000 +0200 +1 -1
 #define MPC52xx_MAIN_IRQ_BASE  (MPC52xx_CRIT_IRQ_BASE + MPC52xx_CRIT_IRQ_NUM)
 #define MPC52xx_SDMA_IRQ_BASE  (MPC52xx_MAIN_IRQ_BASE + MPC52xx_MAIN_IRQ_NUM)
 #define MPC52xx_PERP_IRQ_BASE  (MPC52xx_SDMA_IRQ_BASE + MPC52xx_SDMA_IRQ_NUM)
@@ -415,7 +415,7 @@ struct mpc52xx_cdm {
 #ifndef __ASSEMBLY__
 
 extern void mpc52xx_init_irq(void);
-extern int mpc52xx_get_irq(void);
+extern unsigned int mpc52xx_get_irq(void);

Yeah, you really shouldn't do this without fixing the arch/ppc side too.

g.

Signature · 4 lines
Grant Likely, B.Sc. P.Eng.
Secret Lab Technologies Ltd.
[email protected]
(403) 399-0195
#16 lore raw
Quoted 8 lines · Grant Likely
Quoted 8 lines

diff -uprN a/include/asm-ppc/mpc52xx.h b/include/asm-ppc/mpc52xx.h

include/asm-ppc/mpc52xx.h 2006-10-25 19:11:55.000000000 +0200 renamed from include/asm-ppc/mpc52xx.h 2006-10-25 19:07:48.000000000 +0200 +0 -0
@@ -119,7 +119,7 @@ enum ppc_sys_devices {
 #define MPC52xx_SDMA_IRQ_NUM   17
 #define MPC52xx_PERP_IRQ_NUM   23
 
Quoted 5 lines

include/asm-ppc/mpc52xx.h 2006-10-25 19:11:55.000000000 +0200 enum ppc_sys_devices { · line 122

Quoted 2 lines
include/asm-ppc/mpc52xx.h 2006-10-25 19:11:55.000000000 +0200 renamed from include/asm-ppc/mpc52xx.h 2006-10-25 19:07:48.000000000 +0200 +1 -1
-#define MPC52xx_CRIT_IRQ_BASE  1
+#define MPC52xx_CRIT_IRQ_BASE  0

Is this *strictly* necessary for adding mpc52xx-pic support to arch/powerpc? If not, then move it to a separate patch.

Removed as it does not hurt ARCH=powerpc.

Bye

arch/powerpc/sysdev/Makefile +1 -0
@@ -13,6 +13,7 @@ obj-$(CONFIG_FSL_SOC)		+= fsl_soc.o
 obj-$(CONFIG_PPC_TODC)		+= todc.o
 obj-$(CONFIG_TSI108_BRIDGE)	+= tsi108_pci.o tsi108_dev.o
 obj-$(CONFIG_QUICC_ENGINE)	+= qe_lib/
+obj-$(CONFIG_PPC_MPC52xx_PIC)	+= mpc52xx_pic.o
 
 ifeq ($(CONFIG_PPC_MERGE),y)
 obj-$(CONFIG_PPC_I8259)		+= i8259.o
arch/powerpc/sysdev/mpc52xx_pic.c +368 -0

368 lines not on this page · raw message

arch/powerpc/Kconfig +5 -0
@@ -384,6 +384,11 @@ config PPC_CHRP
 	select PPC_RTAS
 	select PPC_MPC106
 	select PPC_UDBG_16550
+	select PPC_MPC52xx_PIC
+	default y
+
+config PPC_MPC52xx_PIC
+        bool
 	default y
 
 config PPC_PMAC
arch/powerpc/platforms/chrp/setup.c +53 -2

79 lines not on this page · raw message

begin:vcard
fn:Nicolas DET ( bplan GmbH )
n:DET;Nicolas
org:bplan GmbH
adr:;;;;;;Germany
email;internet:[email protected]
title:Software Entwicklung
tel;work:+49 6171 9187 - 31
x-mozilla-html:FALSE
url:http://www.bplan-gmbh.de
version:2.1
end:vcard

#17 lore raw
Quoted 10 lines · Nicolas DET
Quoted 8 lines · Grant Likely
Quoted 8 lines

diff -uprN a/include/asm-ppc/mpc52xx.h b/include/asm-ppc/mpc52xx.h

include/asm-ppc/mpc52xx.h 2006-10-25 19:11:55.000000000 +0200 renamed from include/asm-ppc/mpc52xx.h 2006-10-25 19:07:48.000000000 +0200 +0 -0
@@ -119,7 +119,7 @@ enum ppc_sys_devices {
 #define MPC52xx_SDMA_IRQ_NUM   17
 #define MPC52xx_PERP_IRQ_NUM   23
 
Quoted 8 lines
Quoted 5 lines

include/asm-ppc/mpc52xx.h 2006-10-25 19:11:55.000000000 +0200 enum ppc_sys_devices { · line 122

Quoted 2 lines
include/asm-ppc/mpc52xx.h 2006-10-25 19:11:55.000000000 +0200 renamed from include/asm-ppc/mpc52xx.h 2006-10-25 19:07:48.000000000 +0200 +1 -1
-#define MPC52xx_CRIT_IRQ_BASE  1
+#define MPC52xx_CRIT_IRQ_BASE  0

Is this *strictly* necessary for adding mpc52xx-pic support to arch/powerpc? If not, then move it to a separate patch.

Removed as it does not hurt ARCH=powerpc.

Lindent...

arch/powerpc/sysdev/Makefile +1 -0
@@ -13,6 +13,7 @@ obj-$(CONFIG_FSL_SOC)		+= fsl_soc.o
 obj-$(CONFIG_PPC_TODC)		+= todc.o
 obj-$(CONFIG_TSI108_BRIDGE)	+= tsi108_pci.o tsi108_dev.o
 obj-$(CONFIG_QUICC_ENGINE)	+= qe_lib/
+obj-$(CONFIG_PPC_MPC52xx_PIC)	+= mpc52xx_pic.o
 
 ifeq ($(CONFIG_PPC_MERGE),y)
 obj-$(CONFIG_PPC_I8259)		+= i8259.o
arch/powerpc/sysdev/mpc52xx_pic.c +363 -0

363 lines not on this page · raw message

arch/powerpc/Kconfig +5 -0
@@ -384,6 +384,11 @@ config PPC_CHRP
 	select PPC_RTAS
 	select PPC_MPC106
 	select PPC_UDBG_16550
+	select PPC_MPC52xx_PIC
+	default y
+
+config PPC_MPC52xx_PIC
+        bool
 	default y
 
 config PPC_PMAC
arch/powerpc/platforms/chrp/setup.c +53 -2

79 lines not on this page · raw message

begin:vcard
fn:Nicolas DET ( bplan GmbH )
n:DET;Nicolas
org:bplan GmbH
adr:;;;;;;Germany
email;internet:[email protected]
title:Software Entwicklung
tel;work:+49 6171 9187 - 31
x-mozilla-html:FALSE
url:http://www.bplan-gmbh.de
version:2.1
end:vcard

#18 lore raw
In reply to Nicolas DET

Any reason why this is not static nor __init ? :

Quoted 33 lines
arch/powerpc/platforms/chrp/setup.c +33 -0
+struct device_node *find_mpc52xx_pic(void)
+{
+	struct device_node *dev;
+	const char *piccompatible_list[] =
+	{
+		"mpc5200-interrupt-controller",
+		"mpc52xx-interrupt-controller",
+		"mpc52xx-pic",
+		"mpc5200-pic",
+		"5200-interrupt-controller",
+		"52xx-interrupt-controller",
+		"52xx-pic",
+		"5200-pic",
+		NULL
+	};
+
+	/* Look for an MPC52xx interrupt controller */
+	for_each_node_by_type(dev, "interrupt-controller")
+	{
+		const char **piccompatible_entry = piccompatible_list;
+
+		for(piccompatible_entry = piccompatible_list; *piccompatible_entry; piccompatible_entry++ )
+		{
+			if (device_is_compatible(dev, *piccompatible_entry ))
+				return dev;
+		}
+	}
+
+	return NULL;
+}
+
+static int  __init chrp_find_mpc52xx_pic(void)
+{

arch/powerpc/platforms/chrp/setup.c static struct irqaction xmon_irqaction = · line 482

Quoted 10 lines
arch/powerpc/platforms/chrp/setup.c +10 -0
+	if (find_mpc52xx_pic())
+	{
+		printk(KERN_INFO "Found MPC52xx Interrupt Controller\n");
+		ppc_md.get_irq = mpc52xx_get_irq;
+		mpc52xx_init_irq();
+		return 0;
+	}
+
+	return -ENODEV;
+}

I'm not sure I see any point in splitting in two functions like that

Quoted 4 lines

static void __init chrp_find_8259(void)
{

arch/powerpc/platforms/chrp/setup.c +0 -0
 	struct device_node *np, *pic = NULL;

arch/powerpc/platforms/chrp/setup.c static void __init chrp_find_8259(void) · line 526

Quoted 10 lines
arch/powerpc/platforms/chrp/setup.c +5 -2
@@ -473,8 +519,11 @@ static void __init chrp_find_8259(void)
 		break;
 	}
 	if (np == NULL)
-		printk(KERN_WARNING "Cannot find PCI interrupt acknowledge"
-		       " address, polling\n");
+	{
+		printk(KERN_WARNING "Cannot find PCI/i8259 interrupt acknowledge"
+		       " Fix your tree!\n");
+		return;
+	}

Unrelated changes/fix, should probably be in a separate patch.

Cheers,
Ben.

#19 lore raw
In reply to Grant Likely
Quoted 7 lines · Grant Likely

arch/powerpc/Kconfig config PPC_CHRP · line 393

Quoted 4 lines
arch/powerpc/Kconfig +3 -0
+config PPC_MPC52xx_PIC
+        bool
+       depends on PPC_CHRP
       default y

This isn't good for embedded 52xx boards. Anything u-boot based cannot be CHRP. Please drop the "depends on" line

This was a direct suggestion from benh, ppc_chrp should not be dropped, but ored with whatever you are using, or somethiung like it.

Friendly,

Sven Luther

#20 lore raw
In reply to Grant Likely
Quoted 13 lines · Grant Likely
Quoted 13 lines
(quoted diff) +13 -0
+
+struct device_node *find_mpc52xx_pic(void)
+{
+       struct device_node *dev;
+       const char *piccompatible_list[] =
+       {
+               "mpc5200-interrupt-controller",
+               "mpc52xx-interrupt-controller",
+               "mpc52xx-pic",
+               "mpc5200-pic",
+               "5200-interrupt-controller",
+               "52xx-interrupt-controller",
+               "52xx-pic",
Quoted 10 lines

Review

Quoted 3 lines
(quoted diff) +3 -0
+               "5200-pic",
+               NULL
+       };

Considering Efika is the *only* CHRP 52xx board out there at the moment, and only a handful of embedded folks trying to move it to arch/powerpc, surely we can come to an agreement now on what this thing is named. :)

Done

arch/powerpc/sysdev/Makefile +1 -0
@@ -13,6 +13,7 @@ obj-$(CONFIG_FSL_SOC)		+= fsl_soc.o
 obj-$(CONFIG_PPC_TODC)		+= todc.o
 obj-$(CONFIG_TSI108_BRIDGE)	+= tsi108_pci.o tsi108_dev.o
 obj-$(CONFIG_QUICC_ENGINE)	+= qe_lib/
+obj-$(CONFIG_PPC_MPC52xx_PIC)	+= mpc52xx_pic.o
 
 ifeq ($(CONFIG_PPC_MERGE),y)
 obj-$(CONFIG_PPC_I8259)		+= i8259.o
arch/powerpc/sysdev/mpc52xx_pic.c +363 -0

363 lines not on this page · raw message

arch/powerpc/Kconfig +5 -0
@@ -384,6 +384,11 @@ config PPC_CHRP
 	select PPC_RTAS
 	select PPC_MPC106
 	select PPC_UDBG_16550
+	select PPC_MPC52xx_PIC
+	default y
+
+config PPC_MPC52xx_PIC
+        bool
 	default y
 
 config PPC_PMAC
arch/powerpc/platforms/chrp/setup.c +46 -2

72 lines not on this page · raw message

begin:vcard
fn:Nicolas DET ( bplan GmbH )
n:DET;Nicolas
org:bplan GmbH
adr:;;;;;;Germany
email;internet:[email protected]
title:Software Entwicklung
tel;work:+49 6171 9187 - 31
x-mozilla-html:FALSE
url:http://www.bplan-gmbh.de
version:2.1
end:vcard

Acked-by
#21 lore raw

My comments are satisfied

Quoted 503 lines · Nicolas DET
Quoted 13 lines · Grant Likely
Quoted 13 lines
arch/powerpc/platforms/chrp/setup.c +13 -0
+
+struct device_node *find_mpc52xx_pic(void)
+{
+       struct device_node *dev;
+       const char *piccompatible_list[] =
+       {
+               "mpc5200-interrupt-controller",
+               "mpc52xx-interrupt-controller",
+               "mpc52xx-pic",
+               "mpc5200-pic",
+               "5200-interrupt-controller",
+               "52xx-interrupt-controller",
+               "52xx-pic",
Quoted 10 lines

arch/powerpc/platforms/chrp/setup.c static struct irqaction xmon_irqaction = · line 447

Quoted 3 lines
arch/powerpc/platforms/chrp/setup.c +3 -0
+               "5200-pic",
+               NULL
+       };

Considering Efika is the *only* CHRP 52xx board out there at the moment, and only a handful of embedded folks trying to move it to arch/powerpc, surely we can come to an agreement now on what this thing is named. :)

Done

arch/powerpc/sysdev/Makefile 2006-10-26 11:38:02.000000000 +0200 renamed from arch/powerpc/sysdev/Makefile 2006-10-25 19:07:24.000000000 +0200 +1 -0
@@ -13,6 +13,7 @@ obj-$(CONFIG_FSL_SOC)         += fsl_soc.o
 obj-$(CONFIG_PPC_TODC)         += todc.o
 obj-$(CONFIG_TSI108_BRIDGE)    += tsi108_pci.o tsi108_dev.o
 obj-$(CONFIG_QUICC_ENGINE)     += qe_lib/
+obj-$(CONFIG_PPC_MPC52xx_PIC)  += mpc52xx_pic.o
 
 ifeq ($(CONFIG_PPC_MERGE),y)
 obj-$(CONFIG_PPC_I8259)                += i8259.o
arch/powerpc/sysdev/mpc52xx_pic.c 2006-10-26 21:32:05.000000000 +0200 renamed from arch/powerpc/sysdev/mpc52xx_pic.c 1970-01-01 01:00:00.000000000 +0100 +363 -0
@@ -0,0 +1,363 @@
+/*
+ * arch/powerpc/sysdev/mpc52xx_pic.c
+ *
+ * Programmable Interrupt Controller functions for the Freescale MPC52xx
+ * embedded CPU.
+ * Modified for CHRP Efika 5K2
+ *
+ * Maintainer : Sylvain Munaut <[email protected]>
+ *
+ * Based on (well, mostly copied from) the code from the 2.4 kernel by
+ * Dale Farnsworth <[email protected]> and Kent Borg.
+ *
+ * Copyright (C) 2004 Sylvain Munaut <[email protected]>
+ * Copyright (C) 2003 Montavista Software, Inc
+ *
+ * This file is licensed under the terms of the GNU General Public License
+ * version 2. This program is licensed "as is" without any warranty of any
+ * kind, whether express or implied.
+ */
+
+#include <linux/stddef.h>
+#include <linux/init.h>
+#include <linux/sched.h>
+#include <linux/signal.h>
+#include <linux/stddef.h>
+#include <linux/delay.h>
+#include <linux/irq.h>
+
+#include <asm/io.h>
+#include <asm/processor.h>
+#include <asm/system.h>
+#include <asm/irq.h>
+#include <asm/prom.h>
+
+#include <asm/mpc52xx.h>
+
+static struct mpc52xx_intr __iomem *intr;
+static struct mpc52xx_sdma __iomem *sdma;
+
+static struct irq_host *mpc52xx_irqhost = NULL;
+
+static void mpc52xx_ic_disable(unsigned int virq)
+{
+       u32 val;
+       int irq;
+
+       irq = irq_map[virq].hwirq;
+
+       pr_debug("%s: irq=%d\n", __func__, irq);
+
+       if (irq == MPC52xx_IRQ0) {
+               val = in_be32(&intr->ctrl);
+               val &= ~(1 << 11);
+               out_be32(&intr->ctrl, val);
+       } else if (irq < MPC52xx_IRQ1) {
+               BUG();
+       } else if (irq <= MPC52xx_IRQ3) {
+               val = in_be32(&intr->ctrl);
+               val &= ~(1 << (10 - (irq - MPC52xx_IRQ1)));
+               out_be32(&intr->ctrl, val);
+       } else if (irq < MPC52xx_SDMA_IRQ_BASE) {
+               val = in_be32(&intr->main_mask);
+               val |= 1 << (16 - (irq - MPC52xx_MAIN_IRQ_BASE));
+               out_be32(&intr->main_mask, val);
+       } else if (irq < MPC52xx_PERP_IRQ_BASE) {
+               val = in_be32(&sdma->IntMask);
+               val |= 1 << (irq - MPC52xx_SDMA_IRQ_BASE);
+               out_be32(&sdma->IntMask, val);
+       } else {
+               val = in_be32(&intr->per_mask);
+               val |= 1 << (31 - (irq - MPC52xx_PERP_IRQ_BASE));
+               out_be32(&intr->per_mask, val);
+       }
+}
+
+static void mpc52xx_ic_enable(unsigned int virq)
+{
+       u32 val;
+       int irq;
+
+       irq = irq_map[virq].hwirq;
+
+       pr_debug("%s: irq=%d\n", __func__, irq);
+
+       if (irq == MPC52xx_IRQ0) {
+               val = in_be32(&intr->ctrl);
+               val |= 1 << 11;
+               out_be32(&intr->ctrl, val);
+       } else if (irq < MPC52xx_IRQ1) {
+               BUG();
+       } else if (irq <= MPC52xx_IRQ3) {
+               val = in_be32(&intr->ctrl);
+               val |= 1 << (10 - (irq - MPC52xx_IRQ1));
+               out_be32(&intr->ctrl, val);
+       } else if (irq < MPC52xx_SDMA_IRQ_BASE) {
+               val = in_be32(&intr->main_mask);
+               val &= ~(1 << (16 - (irq - MPC52xx_MAIN_IRQ_BASE)));
+               out_be32(&intr->main_mask, val);
+       } else if (irq < MPC52xx_PERP_IRQ_BASE) {
+               val = in_be32(&sdma->IntMask);
+               val &= ~(1 << (irq - MPC52xx_SDMA_IRQ_BASE));
+               out_be32(&sdma->IntMask, val);
+       } else {
+               val = in_be32(&intr->per_mask);
+               val &= ~(1 << (31 - (irq - MPC52xx_PERP_IRQ_BASE)));
+               out_be32(&intr->per_mask, val);
+       }
+}
+
+static void mpc52xx_ic_ack(unsigned int virq)
+{
+       u32 val;
+       int irq;
+
+       irq = irq_map[virq].hwirq;
+
+       pr_debug("%s: irq=%d\n", __func__, irq);
+
+       /*
+        * Only some irqs are reset here, others in interrupting hardware.
+        */
+
+       switch (irq) {
+       case MPC52xx_IRQ0:
+               val = in_be32(&intr->ctrl);
+               val |= 0x08000000;
+               out_be32(&intr->ctrl, val);
+               break;
+       case MPC52xx_CCS_IRQ:
+               val = in_be32(&intr->enc_status);
+               val |= 0x00000400;
+               out_be32(&intr->enc_status, val);
+               break;
+       case MPC52xx_IRQ1:
+               val = in_be32(&intr->ctrl);
+               val |= 0x04000000;
+               out_be32(&intr->ctrl, val);
+               break;
+       case MPC52xx_IRQ2:
+               val = in_be32(&intr->ctrl);
+               val |= 0x02000000;
+               out_be32(&intr->ctrl, val);
+               break;
+       case MPC52xx_IRQ3:
+               val = in_be32(&intr->ctrl);
+               val |= 0x01000000;
+               out_be32(&intr->ctrl, val);
+               break;
+       default:
+               if (irq >= MPC52xx_SDMA_IRQ_BASE
+                   && irq < (MPC52xx_SDMA_IRQ_BASE + MPC52xx_SDMA_IRQ_NUM)) {
+                       out_be32(&sdma->IntPend,
+                                1 << (irq - MPC52xx_SDMA_IRQ_BASE));
+               }
+
+               break;
+       }
+
+}
+
+static void mpc52xx_ic_disable_and_ack(unsigned int irq)
+{
+       mpc52xx_ic_disable(irq);
+       mpc52xx_ic_ack(irq);
+}
+
+static void mpc52xx_ic_end(unsigned int irq)
+{
+       if (!(irq_desc[irq].status & (IRQ_DISABLED | IRQ_INPROGRESS)))
+               mpc52xx_ic_enable(irq);
+}
+
+static struct irq_chip mpc52xx_irqchip = {
+       .name = " MPC52xx  ",
+       .enable = mpc52xx_ic_enable,
+       .disable = mpc52xx_ic_disable,
+       .ack = mpc52xx_ic_disable_and_ack,
+       .end = mpc52xx_ic_end,
+};
+
+extern struct device_node *find_mpc52xx_pic(void);
+static int mpc52xx_irqhost_match(struct irq_host *h, struct device_node *node)
+{
+       pr_debug("%s: %p vs %p\n", __func__, find_mpc52xx_pic(), node);
+       return find_mpc52xx_pic() == node;
+}
+
+static int mpc52xx_irqhost_xlate(struct irq_host *h, struct device_node *ct,
+                                u32 * intspec, unsigned int intsize,
+                                irq_hw_number_t * out_hwirq,
+                                unsigned int *out_flags)
+{
+       static unsigned char map_senses[4] = {
+               IRQ_TYPE_LEVEL_HIGH,
+               IRQ_TYPE_EDGE_FALLING,
+               IRQ_TYPE_EDGE_RISING,
+               IRQ_TYPE_LEVEL_LOW,
+       };
+
+       int intrvect_l1;
+       int intrvect_l2;
+       int intrvect_type;
+       int intrvect_linux;
+
+       pr_debug("%s:\n", __func__);
+
+       if (intsize != 3)
+               return -1;
+
+       intrvect_l1 = (int)intspec[0];
+       intrvect_l2 = (int)intspec[1];
+       intrvect_type = (int)intspec[2];
+
+       pr_debug("l1=%d, l2=%d, type=%d\n", intrvect_l1, intrvect_l2,
+                intrvect_type);
+
+       switch (intrvect_l1) {
+       case 0:         /* Critical */
+               intrvect_linux = MPC52xx_CRIT_IRQ_BASE;
+               break;
+
+       case 1:         /* Main */
+               intrvect_linux = MPC52xx_MAIN_IRQ_BASE;
+               break;
+
+       case 2:         /* Periph */
+               intrvect_linux = MPC52xx_PERP_IRQ_BASE;
+               break;
+
+       case 3:         /* Bestcomm */
+               intrvect_linux = MPC52xx_SDMA_IRQ_BASE;
+               break;
+
+       default:
+               if (printk_ratelimit())
+                       printk(KERN_ERR "Wrong L1 interrupt vector (%d)\n",
+                              intrvect_l1);
+
+               return -1;
+       }
+
+       intrvect_linux += intrvect_l2;
+
+       pr_debug("return %d\n", intrvect_linux);
+
+       *out_hwirq = intrvect_linux;
+       *out_flags = map_senses[intrvect_type];
+
+       return 0;
+
+}
+
+int mpc52xx_irqhost_map(struct irq_host *h, unsigned int virq,
+                       irq_hw_number_t hw)
+{
+       pr_debug("%s: v=%d, hw=%d\n", __func__, virq, (int)hw);
+
+       return 0;
+}
+
+void mpc52xx_irqhost_unmap(struct irq_host *h, unsigned int virq)
+{
+       pr_debug("%s: v=%d\n", __func__, virq);
+}
+
+static struct irq_host_ops mpc52xx_irqhost_ops = {
+       .match = mpc52xx_irqhost_match,
+       .xlate = mpc52xx_irqhost_xlate,
+       .map = mpc52xx_irqhost_map,
+       .unmap = mpc52xx_irqhost_unmap,
+};
+
+void __init mpc52xx_init_irq(void)
+{
+       int i;
+       u32 intr_ctrl;
+
+       /* Remap the necessary zones */
+       intr = ioremap(MPC52xx_PA(MPC52xx_INTR_OFFSET), MPC52xx_INTR_SIZE);
+       sdma = ioremap(MPC52xx_PA(MPC52xx_SDMA_OFFSET), MPC52xx_SDMA_SIZE);
+
+       if ((intr == NULL) || (sdma == NULL))
+               panic("Can't ioremap PIC/SDMA register or init_irq !");
+
+       /* Disable all interrupt sources. */
+       out_be32(&sdma->IntPend, 0xffffffff);   /* 1 means clear pending */
+       out_be32(&sdma->IntMask, 0xffffffff);   /* 1 means disabled */
+       out_be32(&intr->per_mask, 0x7ffffc00);  /* 1 means disabled */
+       out_be32(&intr->main_mask, 0x00010fff); /* 1 means disabled */
+       intr_ctrl = in_be32(&intr->ctrl);
+       intr_ctrl &= 0x00ff0000;        /* Keeps IRQ[0-3] config */
+       intr_ctrl |= 0x0f000000 |       /* clear IRQ 0-3 */
+           0x00001000 |        /* MEE master external enable */
+           0x00000000 |        /* 0 means disable IRQ 0-3 */
+           0x00000001;         /* CEb route critical normally */
+       out_be32(&intr->ctrl, intr_ctrl);
+
+       /* Zero a bunch of the priority settings.  */
+       out_be32(&intr->per_pri1, 0);
+       out_be32(&intr->per_pri2, 0);
+       out_be32(&intr->per_pri3, 0);
+       out_be32(&intr->main_pri1, 0);
+       out_be32(&intr->main_pri2, 0);
+       /* Initialize irq_desc[i].handler's with mpc52xx_ic. */
+       for (i = 0; i < NR_IRQS; i++) {
+               irq_desc[i].chip = &mpc52xx_irqchip;
+               irq_desc[i].status = IRQ_LEVEL;
+
+       }
+
+#define IRQn_MODE(intr_ctrl,irq) (((intr_ctrl) >> (22-(i<<1))) & 0x03)
+       for (i = 0; i < 4; i++) {
+               int mode;
+               mode = IRQn_MODE(intr_ctrl, i);
+               if ((mode == 0x1) || (mode == 0x2))
+                       irq_desc[i ? MPC52xx_IRQ1 + i -
+                                1 : MPC52xx_IRQ0].status = 0;
+       }
+
+       /*
+        * As last step, add an irq host to translate the real
+        * hw irq information provided by the ofw to linux virq
+        */
+
+       mpc52xx_irqhost =
+           irq_alloc_host(IRQ_HOST_MAP_LINEAR, NR_IRQS, &mpc52xx_irqhost_ops,
+                          -1);
+       pr_debug("%s: mpc52xx_irqhost =%p\n", __func__, mpc52xx_irqhost);
+}
+
+unsigned int mpc52xx_get_irq(void)
+{
+       u32 status;
+       int virq;
+       int irq = NO_IRQ_IGNORE;
+
+       status = in_be32(&intr->enc_status);
+       if (status & 0x00000400) {      /* critical */
+               irq = (status >> 8) & 0x3;
+               if (irq == 2)   /* high priority peripheral */
+                       goto peripheral;
+               irq += MPC52xx_CRIT_IRQ_BASE;
+       } else if (status & 0x00200000) {       /* main */
+               irq = (status >> 16) & 0x1f;
+               if (irq == 4)   /* low priority peripheral */
+                       goto peripheral;
+               irq += MPC52xx_MAIN_IRQ_BASE;
+       } else if (status & 0x20000000) {       /* peripheral */
+             peripheral:
+               irq = (status >> 24) & 0x1f;
+               if (irq == 0) { /* bestcomm */
+                       status = in_be32(&sdma->IntPend);
+                       irq = ffs(status) + MPC52xx_SDMA_IRQ_BASE - 1;
+               } else
+                       irq += MPC52xx_PERP_IRQ_BASE;
+
+       }
+
+       virq = irq_linear_revmap(mpc52xx_irqhost, irq);
+       pr_debug("%s: irq=%d -> %d\n", __func__, irq, virq);
+
+       return virq;
+}
arch/powerpc/Kconfig 2006-10-26 18:55:58.000000000 +0200 renamed from arch/powerpc/Kconfig 2006-10-25 19:07:23.000000000 +0200 +5 -0
@@ -384,6 +384,11 @@ config PPC_CHRP
        select PPC_RTAS
        select PPC_MPC106
        select PPC_UDBG_16550
+       select PPC_MPC52xx_PIC
+       default y
+
+config PPC_MPC52xx_PIC
+        bool
        default y
 
 config PPC_PMAC
arch/powerpc/platforms/chrp/setup.c 2006-10-26 21:47:29.000000000 +0200 renamed from arch/powerpc/platforms/chrp/setup.c 2006-10-25 19:07:23.000000000 +0200 +46 -2
@@ -51,6 +51,7 @@
 #include <asm/mpic.h>
 #include <asm/rtas.h>
 #include <asm/xmon.h>
+#include <asm/mpc52xx.h>
 
 #include "chrp.h"
 
@@ -435,6 +436,44 @@ static struct irqaction xmon_irqaction =
 };
 #endif
 
+
+struct device_node *find_mpc52xx_pic(void)
+{
+       struct device_node *dev;
+       const char *piccompatible_list[] =
+       {
+               "mpc5200-pic",
+               NULL
+       };
+
+       /* Look for an MPC52xx interrupt controller */
+       for_each_node_by_type(dev, "interrupt-controller")
+       {
+               const char **piccompatible_entry = piccompatible_list;
+
+               for(piccompatible_entry = piccompatible_list; *piccompatible_entry; piccompatible_entry++ )
+               {
+                       if (device_is_compatible(dev, *piccompatible_entry ))
+                               return dev;
+               }
+       }
+
+       return NULL;
+}
+
+static int  __init chrp_find_mpc52xx_pic(void)
+{
+       if (find_mpc52xx_pic())
+       {
+               printk(KERN_INFO "Found MPC52xx Interrupt Controller\n");
+               ppc_md.get_irq = mpc52xx_get_irq;
+               mpc52xx_init_irq();
+               return 0;
+       }
+
+       return -ENODEV;
+}
+
 static void __init chrp_find_8259(void)
 {
        struct device_node *np, *pic = NULL;
@@ -473,8 +512,11 @@ static void __init chrp_find_8259(void)
                break;
        }
        if (np == NULL)
-               printk(KERN_WARNING "Cannot find PCI interrupt acknowledge"
-                      " address, polling\n");
+       {
+               printk(KERN_WARNING "Cannot find PCI/i8259 interrupt acknowledge"
+                      " Fix your tree!\n");
+               return;
+       }
 
        i8259_init(pic, chrp_int_ack);
        if (ppc_md.get_irq == NULL)
@@ -494,6 +536,8 @@ void __init chrp_init_IRQ(void)
 #if defined(CONFIG_VT) && defined(CONFIG_INPUT_ADBHID) && defined(XMON)
        struct device_node *kbd;
 #endif
+
+       chrp_find_mpc52xx_pic();
        chrp_find_openpic();
        chrp_find_8259();
 
Signature · 4 lines
Grant Likely, B.Sc. P.Eng.
Secret Lab Technologies Ltd.
[email protected]
(403) 399-0195
#22 lore raw

FWIW, Looks good to me too.

It will need some changes when we'll get rid of the need to stay compatible with arch/ppc. This will hopefully be soon and getting this in should help get things moving imho.

Sylvain

Quoted 534 lines · Grant Likely

My comments are satisfied

Quoted 527 lines · Nicolas DET
Quoted 13 lines · Grant Likely
Quoted 13 lines
arch/powerpc/platforms/chrp/setup.c +13 -0
+
+struct device_node *find_mpc52xx_pic(void)
+{
+       struct device_node *dev;
+       const char *piccompatible_list[] =
+       {
+               "mpc5200-interrupt-controller",
+               "mpc52xx-interrupt-controller",
+               "mpc52xx-pic",
+               "mpc5200-pic",
+               "5200-interrupt-controller",
+               "52xx-interrupt-controller",
+               "52xx-pic",
Quoted 10 lines

arch/powerpc/platforms/chrp/setup.c static struct irqaction xmon_irqaction = · line 447

Quoted 3 lines
arch/powerpc/platforms/chrp/setup.c +3 -0
+               "5200-pic",
+               NULL
+       };

Considering Efika is the *only* CHRP 52xx board out there at the moment, and only a handful of embedded folks trying to move it to arch/powerpc, surely we can come to an agreement now on what this thing is named. :)

Done

--- a/arch/powerpc/sysdev/Makefile 2006-10-25 19:07:24.000000000
+0200
+++ b/arch/powerpc/sysdev/Makefile 2006-10-26 11:38:02.000000000
+0200

arch/powerpc/sysdev/Makefile +8 -1
@@ -13,6 +13,7 @@ obj-$(CONFIG_FSL_SOC)         += fsl_soc.o
 obj-$(CONFIG_PPC_TODC)         += todc.o
 obj-$(CONFIG_TSI108_BRIDGE)    += tsi108_pci.o tsi108_dev.o
 obj-$(CONFIG_QUICC_ENGINE)     += qe_lib/
+obj-$(CONFIG_PPC_MPC52xx_PIC)  += mpc52xx_pic.o
 
 ifeq ($(CONFIG_PPC_MERGE),y)
 obj-$(CONFIG_PPC_I8259)                += i8259.o
--- a/arch/powerpc/sysdev/mpc52xx_pic.c 1970-01-01 01:00:00.000000000
+0100
+++ b/arch/powerpc/sysdev/mpc52xx_pic.c 2006-10-26 21:32:05.000000000
+0200
@@ -0,0 +1,363 @@
+/*
+ * arch/powerpc/sysdev/mpc52xx_pic.c
+ *
+ * Programmable Interrupt Controller functions for the Freescale

MPC52xx
+ * embedded CPU.
+ * Modified for CHRP Efika 5K2
+ *
+ * Maintainer : Sylvain Munaut <[email protected]>
+ *
+ * Based on (well, mostly copied from) the code from the 2.4 kernel by
+ * Dale Farnsworth <[email protected]> and Kent Borg.
+ *
+ * Copyright (C) 2004 Sylvain Munaut <[email protected]>
+ * Copyright (C) 2003 Montavista Software, Inc
+ *
+ * This file is licensed under the terms of the GNU General Public License
+ * version 2. This program is licensed "as is" without any warranty of any
+ * kind, whether express or implied.
+ */

arch/powerpc/sysdev/mpc52xx_pic.c +101 -0
+
+#include <linux/stddef.h>
+#include <linux/init.h>
+#include <linux/sched.h>
+#include <linux/signal.h>
+#include <linux/stddef.h>
+#include <linux/delay.h>
+#include <linux/irq.h>
+
+#include <asm/io.h>
+#include <asm/processor.h>
+#include <asm/system.h>
+#include <asm/irq.h>
+#include <asm/prom.h>
+
+#include <asm/mpc52xx.h>
+
+static struct mpc52xx_intr __iomem *intr;
+static struct mpc52xx_sdma __iomem *sdma;
+
+static struct irq_host *mpc52xx_irqhost = NULL;
+
+static void mpc52xx_ic_disable(unsigned int virq)
+{
+       u32 val;
+       int irq;
+
+       irq = irq_map[virq].hwirq;
+
+       pr_debug("%s: irq=%d\n", __func__, irq);
+
+       if (irq == MPC52xx_IRQ0) {
+               val = in_be32(&intr->ctrl);
+               val &= ~(1 << 11);
+               out_be32(&intr->ctrl, val);
+       } else if (irq < MPC52xx_IRQ1) {
+               BUG();
+       } else if (irq <= MPC52xx_IRQ3) {
+               val = in_be32(&intr->ctrl);
+               val &= ~(1 << (10 - (irq - MPC52xx_IRQ1)));
+               out_be32(&intr->ctrl, val);
+       } else if (irq < MPC52xx_SDMA_IRQ_BASE) {
+               val = in_be32(&intr->main_mask);
+               val |= 1 << (16 - (irq - MPC52xx_MAIN_IRQ_BASE));
+               out_be32(&intr->main_mask, val);
+       } else if (irq < MPC52xx_PERP_IRQ_BASE) {
+               val = in_be32(&sdma->IntMask);
+               val |= 1 << (irq - MPC52xx_SDMA_IRQ_BASE);
+               out_be32(&sdma->IntMask, val);
+       } else {
+               val = in_be32(&intr->per_mask);
+               val |= 1 << (31 - (irq - MPC52xx_PERP_IRQ_BASE));
+               out_be32(&intr->per_mask, val);
+       }
+}
+
+static void mpc52xx_ic_enable(unsigned int virq)
+{
+       u32 val;
+       int irq;
+
+       irq = irq_map[virq].hwirq;
+
+       pr_debug("%s: irq=%d\n", __func__, irq);
+
+       if (irq == MPC52xx_IRQ0) {
+               val = in_be32(&intr->ctrl);
+               val |= 1 << 11;
+               out_be32(&intr->ctrl, val);
+       } else if (irq < MPC52xx_IRQ1) {
+               BUG();
+       } else if (irq <= MPC52xx_IRQ3) {
+               val = in_be32(&intr->ctrl);
+               val |= 1 << (10 - (irq - MPC52xx_IRQ1));
+               out_be32(&intr->ctrl, val);
+       } else if (irq < MPC52xx_SDMA_IRQ_BASE) {
+               val = in_be32(&intr->main_mask);
+               val &= ~(1 << (16 - (irq - MPC52xx_MAIN_IRQ_BASE)));
+               out_be32(&intr->main_mask, val);
+       } else if (irq < MPC52xx_PERP_IRQ_BASE) {
+               val = in_be32(&sdma->IntMask);
+               val &= ~(1 << (irq - MPC52xx_SDMA_IRQ_BASE));
+               out_be32(&sdma->IntMask, val);
+       } else {
+               val = in_be32(&intr->per_mask);
+               val &= ~(1 << (31 - (irq - MPC52xx_PERP_IRQ_BASE)));
+               out_be32(&intr->per_mask, val);
+       }
+}
+
+static void mpc52xx_ic_ack(unsigned int virq)
+{
+       u32 val;
+       int irq;
+
+       irq = irq_map[virq].hwirq;
+
+       pr_debug("%s: irq=%d\n", __func__, irq);
+
+       /*
+        * Only some irqs are reset here, others in interrupting

hardware.

arch/powerpc/sysdev/mpc52xx_pic.c +31 -0
+        */
+
+       switch (irq) {
+       case MPC52xx_IRQ0:
+               val = in_be32(&intr->ctrl);
+               val |= 0x08000000;
+               out_be32(&intr->ctrl, val);
+               break;
+       case MPC52xx_CCS_IRQ:
+               val = in_be32(&intr->enc_status);
+               val |= 0x00000400;
+               out_be32(&intr->enc_status, val);
+               break;
+       case MPC52xx_IRQ1:
+               val = in_be32(&intr->ctrl);
+               val |= 0x04000000;
+               out_be32(&intr->ctrl, val);
+               break;
+       case MPC52xx_IRQ2:
+               val = in_be32(&intr->ctrl);
+               val |= 0x02000000;
+               out_be32(&intr->ctrl, val);
+               break;
+       case MPC52xx_IRQ3:
+               val = in_be32(&intr->ctrl);
+               val |= 0x01000000;
+               out_be32(&intr->ctrl, val);
+               break;
+       default:
+               if (irq >= MPC52xx_SDMA_IRQ_BASE
+                   && irq < (MPC52xx_SDMA_IRQ_BASE +

MPC52xx_SDMA_IRQ_NUM)) {

arch/powerpc/sysdev/mpc52xx_pic.c +31 -0
+                       out_be32(&sdma->IntPend,
+                                1 << (irq - MPC52xx_SDMA_IRQ_BASE));
+               }
+
+               break;
+       }
+
+}
+
+static void mpc52xx_ic_disable_and_ack(unsigned int irq)
+{
+       mpc52xx_ic_disable(irq);
+       mpc52xx_ic_ack(irq);
+}
+
+static void mpc52xx_ic_end(unsigned int irq)
+{
+       if (!(irq_desc[irq].status & (IRQ_DISABLED | IRQ_INPROGRESS)))
+               mpc52xx_ic_enable(irq);
+}
+
+static struct irq_chip mpc52xx_irqchip = {
+       .name = " MPC52xx  ",
+       .enable = mpc52xx_ic_enable,
+       .disable = mpc52xx_ic_disable,
+       .ack = mpc52xx_ic_disable_and_ack,
+       .end = mpc52xx_ic_end,
+};
+
+extern struct device_node *find_mpc52xx_pic(void);
+static int mpc52xx_irqhost_match(struct irq_host *h, struct

device_node *node)

arch/powerpc/sysdev/mpc52xx_pic.c +6 -0
+{
+       pr_debug("%s: %p vs %p\n", __func__, find_mpc52xx_pic(), node);
+       return find_mpc52xx_pic() == node;
+}
+
+static int mpc52xx_irqhost_xlate(struct irq_host *h, struct

device_node *ct,

arch/powerpc/sysdev/mpc52xx_pic.c +48 -0
+                                u32 * intspec, unsigned int intsize,
+                                irq_hw_number_t * out_hwirq,
+                                unsigned int *out_flags)
+{
+       static unsigned char map_senses[4] = {
+               IRQ_TYPE_LEVEL_HIGH,
+               IRQ_TYPE_EDGE_FALLING,
+               IRQ_TYPE_EDGE_RISING,
+               IRQ_TYPE_LEVEL_LOW,
+       };
+
+       int intrvect_l1;
+       int intrvect_l2;
+       int intrvect_type;
+       int intrvect_linux;
+
+       pr_debug("%s:\n", __func__);
+
+       if (intsize != 3)
+               return -1;
+
+       intrvect_l1 = (int)intspec[0];
+       intrvect_l2 = (int)intspec[1];
+       intrvect_type = (int)intspec[2];
+
+       pr_debug("l1=%d, l2=%d, type=%d\n", intrvect_l1, intrvect_l2,
+                intrvect_type);
+
+       switch (intrvect_l1) {
+       case 0:         /* Critical */
+               intrvect_linux = MPC52xx_CRIT_IRQ_BASE;
+               break;
+
+       case 1:         /* Main */
+               intrvect_linux = MPC52xx_MAIN_IRQ_BASE;
+               break;
+
+       case 2:         /* Periph */
+               intrvect_linux = MPC52xx_PERP_IRQ_BASE;
+               break;
+
+       case 3:         /* Bestcomm */
+               intrvect_linux = MPC52xx_SDMA_IRQ_BASE;
+               break;
+
+       default:
+               if (printk_ratelimit())
+                       printk(KERN_ERR "Wrong L1 interrupt vector

(%d)\n",

arch/powerpc/sysdev/mpc52xx_pic.c +43 -0
+                              intrvect_l1);
+
+               return -1;
+       }
+
+       intrvect_linux += intrvect_l2;
+
+       pr_debug("return %d\n", intrvect_linux);
+
+       *out_hwirq = intrvect_linux;
+       *out_flags = map_senses[intrvect_type];
+
+       return 0;
+
+}
+
+int mpc52xx_irqhost_map(struct irq_host *h, unsigned int virq,
+                       irq_hw_number_t hw)
+{
+       pr_debug("%s: v=%d, hw=%d\n", __func__, virq, (int)hw);
+
+       return 0;
+}
+
+void mpc52xx_irqhost_unmap(struct irq_host *h, unsigned int virq)
+{
+       pr_debug("%s: v=%d\n", __func__, virq);
+}
+
+static struct irq_host_ops mpc52xx_irqhost_ops = {
+       .match = mpc52xx_irqhost_match,
+       .xlate = mpc52xx_irqhost_xlate,
+       .map = mpc52xx_irqhost_map,
+       .unmap = mpc52xx_irqhost_unmap,
+};
+
+void __init mpc52xx_init_irq(void)
+{
+       int i;
+       u32 intr_ctrl;
+
+       /* Remap the necessary zones */
+       intr = ioremap(MPC52xx_PA(MPC52xx_INTR_OFFSET),

MPC52xx_INTR_SIZE);
+ sdma = ioremap(MPC52xx_PA(MPC52xx_SDMA_OFFSET),
MPC52xx_SDMA_SIZE);

arch/powerpc/sysdev/mpc52xx_pic.c +6 -0
+
+       if ((intr == NULL) || (sdma == NULL))
+               panic("Can't ioremap PIC/SDMA register or init_irq !");
+
+       /* Disable all interrupt sources. */
+       out_be32(&sdma->IntPend, 0xffffffff);   /* 1 means clear

pending */

arch/powerpc/sysdev/mpc52xx_pic.c +40 -0
+       out_be32(&sdma->IntMask, 0xffffffff);   /* 1 means disabled */
+       out_be32(&intr->per_mask, 0x7ffffc00);  /* 1 means disabled */
+       out_be32(&intr->main_mask, 0x00010fff); /* 1 means disabled */
+       intr_ctrl = in_be32(&intr->ctrl);
+       intr_ctrl &= 0x00ff0000;        /* Keeps IRQ[0-3] config */
+       intr_ctrl |= 0x0f000000 |       /* clear IRQ 0-3 */
+           0x00001000 |        /* MEE master external enable */
+           0x00000000 |        /* 0 means disable IRQ 0-3 */
+           0x00000001;         /* CEb route critical normally */
+       out_be32(&intr->ctrl, intr_ctrl);
+
+       /* Zero a bunch of the priority settings.  */
+       out_be32(&intr->per_pri1, 0);
+       out_be32(&intr->per_pri2, 0);
+       out_be32(&intr->per_pri3, 0);
+       out_be32(&intr->main_pri1, 0);
+       out_be32(&intr->main_pri2, 0);
+       /* Initialize irq_desc[i].handler's with mpc52xx_ic. */
+       for (i = 0; i < NR_IRQS; i++) {
+               irq_desc[i].chip = &mpc52xx_irqchip;
+               irq_desc[i].status = IRQ_LEVEL;
+
+       }
+
+#define IRQn_MODE(intr_ctrl,irq) (((intr_ctrl) >> (22-(i<<1))) & 0x03)
+       for (i = 0; i < 4; i++) {
+               int mode;
+               mode = IRQn_MODE(intr_ctrl, i);
+               if ((mode == 0x1) || (mode == 0x2))
+                       irq_desc[i ? MPC52xx_IRQ1 + i -
+                                1 : MPC52xx_IRQ0].status = 0;
+       }
+
+       /*
+        * As last step, add an irq host to translate the real
+        * hw irq information provided by the ofw to linux virq
+        */
+
+       mpc52xx_irqhost =
+           irq_alloc_host(IRQ_HOST_MAP_LINEAR, NR_IRQS,

&mpc52xx_irqhost_ops,

(quoted diff) +2 -0
+                          -1);
+       pr_debug("%s: mpc52xx_irqhost =%p\n", __func__,

mpc52xx_irqhost);

arch/powerpc/sysdev/mpc52xx_pic.c +35 -0
+}
+
+unsigned int mpc52xx_get_irq(void)
+{
+       u32 status;
+       int virq;
+       int irq = NO_IRQ_IGNORE;
+
+       status = in_be32(&intr->enc_status);
+       if (status & 0x00000400) {      /* critical */
+               irq = (status >> 8) & 0x3;
+               if (irq == 2)   /* high priority peripheral */
+                       goto peripheral;
+               irq += MPC52xx_CRIT_IRQ_BASE;
+       } else if (status & 0x00200000) {       /* main */
+               irq = (status >> 16) & 0x1f;
+               if (irq == 4)   /* low priority peripheral */
+                       goto peripheral;
+               irq += MPC52xx_MAIN_IRQ_BASE;
+       } else if (status & 0x20000000) {       /* peripheral */
+             peripheral:
+               irq = (status >> 24) & 0x1f;
+               if (irq == 0) { /* bestcomm */
+                       status = in_be32(&sdma->IntPend);
+                       irq = ffs(status) + MPC52xx_SDMA_IRQ_BASE - 1;
+               } else
+                       irq += MPC52xx_PERP_IRQ_BASE;
+
+       }
+
+       virq = irq_linear_revmap(mpc52xx_irqhost, irq);
+       pr_debug("%s: irq=%d -> %d\n", __func__, irq, virq);
+
+       return virq;
+}
arch/powerpc/Kconfig 2006-10-26 18:55:58.000000000 +0200 renamed from arch/powerpc/Kconfig 2006-10-25 19:07:23.000000000 +0200 +5 -1
@@ -384,6 +384,11 @@ config PPC_CHRP
        select PPC_RTAS
        select PPC_MPC106
        select PPC_UDBG_16550
+       select PPC_MPC52xx_PIC
+       default y
+
+config PPC_MPC52xx_PIC
+        bool
        default y
 
 config PPC_PMAC
--- a/arch/powerpc/platforms/chrp/setup.c       2006-10-25

19:07:23.000000000 +0200
+++ b/arch/powerpc/platforms/chrp/setup.c 2006-10-26
21:47:29.000000000 +0200

arch/powerpc/Kconfig 2006-10-26 18:55:58.000000000 +0200 renamed from arch/powerpc/Kconfig 2006-10-25 19:07:23.000000000 +0200 +39 -0
@@ -51,6 +51,7 @@
 #include <asm/mpic.h>
 #include <asm/rtas.h>
 #include <asm/xmon.h>
+#include <asm/mpc52xx.h>
 
 #include "chrp.h"
 
@@ -435,6 +436,44 @@ static struct irqaction xmon_irqaction =
 };
 #endif
 
+
+struct device_node *find_mpc52xx_pic(void)
+{
+       struct device_node *dev;
+       const char *piccompatible_list[] =
+       {
+               "mpc5200-pic",
+               NULL
+       };
+
+       /* Look for an MPC52xx interrupt controller */
+       for_each_node_by_type(dev, "interrupt-controller")
+       {
+               const char **piccompatible_entry = piccompatible_list;
+
+               for(piccompatible_entry = piccompatible_list;
 *piccompatible_entry; piccompatible_entry++ )
+               {
+                       if (device_is_compatible(dev,
 *piccompatible_entry ))
+                               return dev;
+               }
+       }
+
+       return NULL;
+}
+
+static int  __init chrp_find_mpc52xx_pic(void)
+{
+       if (find_mpc52xx_pic())
+       {
+               printk(KERN_INFO "Found MPC52xx Interrupt
 Controller\n");
+               ppc_md.get_irq = mpc52xx_get_irq;
+               mpc52xx_init_irq();
+               return 0;
+       }
+
+       return -ENODEV;
+}
+

static void __init chrp_find_8259(void)

 {
        struct device_node *np, *pic = NULL;
arch/powerpc/Kconfig 2006-10-26 18:55:58.000000000 +0200 renamed from arch/powerpc/Kconfig 2006-10-25 19:07:23.000000000 +0200 +0 -1
@@ -473,8 +512,11 @@ static void __init chrp_find_8259(void)
                break;
        }
        if (np == NULL)
-               printk(KERN_WARNING "Cannot find PCI interrupt

acknowledge"

arch/powerpc/Kconfig 2006-10-26 18:55:58.000000000 +0200 renamed from arch/powerpc/Kconfig 2006-10-25 19:07:23.000000000 +0200 +2 -1
-                      " address, polling\n");
+       {
+               printk(KERN_WARNING "Cannot find PCI/i8259 interrupt

acknowledge"

arch/powerpc/Kconfig 2006-10-26 18:55:58.000000000 +0200 renamed from arch/powerpc/Kconfig 2006-10-25 19:07:23.000000000 +0200 +5 -0
+                      " Fix your tree!\n");
+               return;
+       }
 
        i8259_init(pic, chrp_int_ack);
        if (ppc_md.get_irq == NULL)
@@ -494,6 +536,8 @@ void __init chrp_init_IRQ(void)
 #if defined(CONFIG_VT) && defined(CONFIG_INPUT_ADBHID) && defined(XMON)
        struct device_node *kbd;
 #endif
+
+       chrp_find_mpc52xx_pic();
        chrp_find_openpic();
        chrp_find_8259();
 
#23 lore raw
In reply to Grant Likely
Quoted 3 lines · Grant Likely

My comments are satisfied

Nack.

The irq code doesn't properly use the genirq infrastructure. It's not setting flow handlers, not implementing set_irq_type, not using the new mask/unmask instead of the toplevel enable/disable...

Ben.

Quoted 506 lines · Nicolas DET
Quoted 503 lines
Quoted 13 lines · Grant Likely
Quoted 13 lines
arch/powerpc/platforms/chrp/setup.c +13 -0
+
+struct device_node *find_mpc52xx_pic(void)
+{
+       struct device_node *dev;
+       const char *piccompatible_list[] =
+       {
+               "mpc5200-interrupt-controller",
+               "mpc52xx-interrupt-controller",
+               "mpc52xx-pic",
+               "mpc5200-pic",
+               "5200-interrupt-controller",
+               "52xx-interrupt-controller",
+               "52xx-pic",
Quoted 10 lines

arch/powerpc/platforms/chrp/setup.c static struct irqaction xmon_irqaction = · line 447

Quoted 3 lines
arch/powerpc/platforms/chrp/setup.c +3 -0
+               "5200-pic",
+               NULL
+       };

Considering Efika is the *only* CHRP 52xx board out there at the moment, and only a handful of embedded folks trying to move it to arch/powerpc, surely we can come to an agreement now on what this thing is named. :)

Done

arch/powerpc/sysdev/Makefile 2006-10-26 11:38:02.000000000 +0200 renamed from arch/powerpc/sysdev/Makefile 2006-10-25 19:07:24.000000000 +0200 +1 -0
@@ -13,6 +13,7 @@ obj-$(CONFIG_FSL_SOC)         += fsl_soc.o
 obj-$(CONFIG_PPC_TODC)         += todc.o
 obj-$(CONFIG_TSI108_BRIDGE)    += tsi108_pci.o tsi108_dev.o
 obj-$(CONFIG_QUICC_ENGINE)     += qe_lib/
+obj-$(CONFIG_PPC_MPC52xx_PIC)  += mpc52xx_pic.o
 
 ifeq ($(CONFIG_PPC_MERGE),y)
 obj-$(CONFIG_PPC_I8259)                += i8259.o
arch/powerpc/sysdev/mpc52xx_pic.c 2006-10-26 21:32:05.000000000 +0200 renamed from arch/powerpc/sysdev/mpc52xx_pic.c 1970-01-01 01:00:00.000000000 +0100 +363 -0
@@ -0,0 +1,363 @@
+/*
+ * arch/powerpc/sysdev/mpc52xx_pic.c
+ *
+ * Programmable Interrupt Controller functions for the Freescale MPC52xx
+ * embedded CPU.
+ * Modified for CHRP Efika 5K2
+ *
+ * Maintainer : Sylvain Munaut <[email protected]>
+ *
+ * Based on (well, mostly copied from) the code from the 2.4 kernel by
+ * Dale Farnsworth <[email protected]> and Kent Borg.
+ *
+ * Copyright (C) 2004 Sylvain Munaut <[email protected]>
+ * Copyright (C) 2003 Montavista Software, Inc
+ *
+ * This file is licensed under the terms of the GNU General Public License
+ * version 2. This program is licensed "as is" without any warranty of any
+ * kind, whether express or implied.
+ */
+
+#include <linux/stddef.h>
+#include <linux/init.h>
+#include <linux/sched.h>
+#include <linux/signal.h>
+#include <linux/stddef.h>
+#include <linux/delay.h>
+#include <linux/irq.h>
+
+#include <asm/io.h>
+#include <asm/processor.h>
+#include <asm/system.h>
+#include <asm/irq.h>
+#include <asm/prom.h>
+
+#include <asm/mpc52xx.h>
+
+static struct mpc52xx_intr __iomem *intr;
+static struct mpc52xx_sdma __iomem *sdma;
+
+static struct irq_host *mpc52xx_irqhost = NULL;
+
+static void mpc52xx_ic_disable(unsigned int virq)
+{
+       u32 val;
+       int irq;
+
+       irq = irq_map[virq].hwirq;
+
+       pr_debug("%s: irq=%d\n", __func__, irq);
+
+       if (irq == MPC52xx_IRQ0) {
+               val = in_be32(&intr->ctrl);
+               val &= ~(1 << 11);
+               out_be32(&intr->ctrl, val);
+       } else if (irq < MPC52xx_IRQ1) {
+               BUG();
+       } else if (irq <= MPC52xx_IRQ3) {
+               val = in_be32(&intr->ctrl);
+               val &= ~(1 << (10 - (irq - MPC52xx_IRQ1)));
+               out_be32(&intr->ctrl, val);
+       } else if (irq < MPC52xx_SDMA_IRQ_BASE) {
+               val = in_be32(&intr->main_mask);
+               val |= 1 << (16 - (irq - MPC52xx_MAIN_IRQ_BASE));
+               out_be32(&intr->main_mask, val);
+       } else if (irq < MPC52xx_PERP_IRQ_BASE) {
+               val = in_be32(&sdma->IntMask);
+               val |= 1 << (irq - MPC52xx_SDMA_IRQ_BASE);
+               out_be32(&sdma->IntMask, val);
+       } else {
+               val = in_be32(&intr->per_mask);
+               val |= 1 << (31 - (irq - MPC52xx_PERP_IRQ_BASE));
+               out_be32(&intr->per_mask, val);
+       }
+}
+
+static void mpc52xx_ic_enable(unsigned int virq)
+{
+       u32 val;
+       int irq;
+
+       irq = irq_map[virq].hwirq;
+
+       pr_debug("%s: irq=%d\n", __func__, irq);
+
+       if (irq == MPC52xx_IRQ0) {
+               val = in_be32(&intr->ctrl);
+               val |= 1 << 11;
+               out_be32(&intr->ctrl, val);
+       } else if (irq < MPC52xx_IRQ1) {
+               BUG();
+       } else if (irq <= MPC52xx_IRQ3) {
+               val = in_be32(&intr->ctrl);
+               val |= 1 << (10 - (irq - MPC52xx_IRQ1));
+               out_be32(&intr->ctrl, val);
+       } else if (irq < MPC52xx_SDMA_IRQ_BASE) {
+               val = in_be32(&intr->main_mask);
+               val &= ~(1 << (16 - (irq - MPC52xx_MAIN_IRQ_BASE)));
+               out_be32(&intr->main_mask, val);
+       } else if (irq < MPC52xx_PERP_IRQ_BASE) {
+               val = in_be32(&sdma->IntMask);
+               val &= ~(1 << (irq - MPC52xx_SDMA_IRQ_BASE));
+               out_be32(&sdma->IntMask, val);
+       } else {
+               val = in_be32(&intr->per_mask);
+               val &= ~(1 << (31 - (irq - MPC52xx_PERP_IRQ_BASE)));
+               out_be32(&intr->per_mask, val);
+       }
+}
+
+static void mpc52xx_ic_ack(unsigned int virq)
+{
+       u32 val;
+       int irq;
+
+       irq = irq_map[virq].hwirq;
+
+       pr_debug("%s: irq=%d\n", __func__, irq);
+
+       /*
+        * Only some irqs are reset here, others in interrupting hardware.
+        */
+
+       switch (irq) {
+       case MPC52xx_IRQ0:
+               val = in_be32(&intr->ctrl);
+               val |= 0x08000000;
+               out_be32(&intr->ctrl, val);
+               break;
+       case MPC52xx_CCS_IRQ:
+               val = in_be32(&intr->enc_status);
+               val |= 0x00000400;
+               out_be32(&intr->enc_status, val);
+               break;
+       case MPC52xx_IRQ1:
+               val = in_be32(&intr->ctrl);
+               val |= 0x04000000;
+               out_be32(&intr->ctrl, val);
+               break;
+       case MPC52xx_IRQ2:
+               val = in_be32(&intr->ctrl);
+               val |= 0x02000000;
+               out_be32(&intr->ctrl, val);
+               break;
+       case MPC52xx_IRQ3:
+               val = in_be32(&intr->ctrl);
+               val |= 0x01000000;
+               out_be32(&intr->ctrl, val);
+               break;
+       default:
+               if (irq >= MPC52xx_SDMA_IRQ_BASE
+                   && irq < (MPC52xx_SDMA_IRQ_BASE + MPC52xx_SDMA_IRQ_NUM)) {
+                       out_be32(&sdma->IntPend,
+                                1 << (irq - MPC52xx_SDMA_IRQ_BASE));
+               }
+
+               break;
+       }
+
+}
+
+static void mpc52xx_ic_disable_and_ack(unsigned int irq)
+{
+       mpc52xx_ic_disable(irq);
+       mpc52xx_ic_ack(irq);
+}
+
+static void mpc52xx_ic_end(unsigned int irq)
+{
+       if (!(irq_desc[irq].status & (IRQ_DISABLED | IRQ_INPROGRESS)))
+               mpc52xx_ic_enable(irq);
+}
+
+static struct irq_chip mpc52xx_irqchip = {
+       .name = " MPC52xx  ",
+       .enable = mpc52xx_ic_enable,
+       .disable = mpc52xx_ic_disable,
+       .ack = mpc52xx_ic_disable_and_ack,
+       .end = mpc52xx_ic_end,
+};
+
+extern struct device_node *find_mpc52xx_pic(void);
+static int mpc52xx_irqhost_match(struct irq_host *h, struct device_node *node)
+{
+       pr_debug("%s: %p vs %p\n", __func__, find_mpc52xx_pic(), node);
+       return find_mpc52xx_pic() == node;
+}
+
+static int mpc52xx_irqhost_xlate(struct irq_host *h, struct device_node *ct,
+                                u32 * intspec, unsigned int intsize,
+                                irq_hw_number_t * out_hwirq,
+                                unsigned int *out_flags)
+{
+       static unsigned char map_senses[4] = {
+               IRQ_TYPE_LEVEL_HIGH,
+               IRQ_TYPE_EDGE_FALLING,
+               IRQ_TYPE_EDGE_RISING,
+               IRQ_TYPE_LEVEL_LOW,
+       };
+
+       int intrvect_l1;
+       int intrvect_l2;
+       int intrvect_type;
+       int intrvect_linux;
+
+       pr_debug("%s:\n", __func__);
+
+       if (intsize != 3)
+               return -1;
+
+       intrvect_l1 = (int)intspec[0];
+       intrvect_l2 = (int)intspec[1];
+       intrvect_type = (int)intspec[2];
+
+       pr_debug("l1=%d, l2=%d, type=%d\n", intrvect_l1, intrvect_l2,
+                intrvect_type);
+
+       switch (intrvect_l1) {
+       case 0:         /* Critical */
+               intrvect_linux = MPC52xx_CRIT_IRQ_BASE;
+               break;
+
+       case 1:         /* Main */
+               intrvect_linux = MPC52xx_MAIN_IRQ_BASE;
+               break;
+
+       case 2:         /* Periph */
+               intrvect_linux = MPC52xx_PERP_IRQ_BASE;
+               break;
+
+       case 3:         /* Bestcomm */
+               intrvect_linux = MPC52xx_SDMA_IRQ_BASE;
+               break;
+
+       default:
+               if (printk_ratelimit())
+                       printk(KERN_ERR "Wrong L1 interrupt vector (%d)\n",
+                              intrvect_l1);
+
+               return -1;
+       }
+
+       intrvect_linux += intrvect_l2;
+
+       pr_debug("return %d\n", intrvect_linux);
+
+       *out_hwirq = intrvect_linux;
+       *out_flags = map_senses[intrvect_type];
+
+       return 0;
+
+}
+
+int mpc52xx_irqhost_map(struct irq_host *h, unsigned int virq,
+                       irq_hw_number_t hw)
+{
+       pr_debug("%s: v=%d, hw=%d\n", __func__, virq, (int)hw);
+
+       return 0;
+}
+
+void mpc52xx_irqhost_unmap(struct irq_host *h, unsigned int virq)
+{
+       pr_debug("%s: v=%d\n", __func__, virq);
+}
+
+static struct irq_host_ops mpc52xx_irqhost_ops = {
+       .match = mpc52xx_irqhost_match,
+       .xlate = mpc52xx_irqhost_xlate,
+       .map = mpc52xx_irqhost_map,
+       .unmap = mpc52xx_irqhost_unmap,
+};
+
+void __init mpc52xx_init_irq(void)
+{
+       int i;
+       u32 intr_ctrl;
+
+       /* Remap the necessary zones */
+       intr = ioremap(MPC52xx_PA(MPC52xx_INTR_OFFSET), MPC52xx_INTR_SIZE);
+       sdma = ioremap(MPC52xx_PA(MPC52xx_SDMA_OFFSET), MPC52xx_SDMA_SIZE);
+
+       if ((intr == NULL) || (sdma == NULL))
+               panic("Can't ioremap PIC/SDMA register or init_irq !");
+
+       /* Disable all interrupt sources. */
+       out_be32(&sdma->IntPend, 0xffffffff);   /* 1 means clear pending */
+       out_be32(&sdma->IntMask, 0xffffffff);   /* 1 means disabled */
+       out_be32(&intr->per_mask, 0x7ffffc00);  /* 1 means disabled */
+       out_be32(&intr->main_mask, 0x00010fff); /* 1 means disabled */
+       intr_ctrl = in_be32(&intr->ctrl);
+       intr_ctrl &= 0x00ff0000;        /* Keeps IRQ[0-3] config */
+       intr_ctrl |= 0x0f000000 |       /* clear IRQ 0-3 */
+           0x00001000 |        /* MEE master external enable */
+           0x00000000 |        /* 0 means disable IRQ 0-3 */
+           0x00000001;         /* CEb route critical normally */
+       out_be32(&intr->ctrl, intr_ctrl);
+
+       /* Zero a bunch of the priority settings.  */
+       out_be32(&intr->per_pri1, 0);
+       out_be32(&intr->per_pri2, 0);
+       out_be32(&intr->per_pri3, 0);
+       out_be32(&intr->main_pri1, 0);
+       out_be32(&intr->main_pri2, 0);
+       /* Initialize irq_desc[i].handler's with mpc52xx_ic. */
+       for (i = 0; i < NR_IRQS; i++) {
+               irq_desc[i].chip = &mpc52xx_irqchip;
+               irq_desc[i].status = IRQ_LEVEL;
+
+       }
+
+#define IRQn_MODE(intr_ctrl,irq) (((intr_ctrl) >> (22-(i<<1))) & 0x03)
+       for (i = 0; i < 4; i++) {
+               int mode;
+               mode = IRQn_MODE(intr_ctrl, i);
+               if ((mode == 0x1) || (mode == 0x2))
+                       irq_desc[i ? MPC52xx_IRQ1 + i -
+                                1 : MPC52xx_IRQ0].status = 0;
+       }
+
+       /*
+        * As last step, add an irq host to translate the real
+        * hw irq information provided by the ofw to linux virq
+        */
+
+       mpc52xx_irqhost =
+           irq_alloc_host(IRQ_HOST_MAP_LINEAR, NR_IRQS, &mpc52xx_irqhost_ops,
+                          -1);
+       pr_debug("%s: mpc52xx_irqhost =%p\n", __func__, mpc52xx_irqhost);
+}
+
+unsigned int mpc52xx_get_irq(void)
+{
+       u32 status;
+       int virq;
+       int irq = NO_IRQ_IGNORE;
+
+       status = in_be32(&intr->enc_status);
+       if (status & 0x00000400) {      /* critical */
+               irq = (status >> 8) & 0x3;
+               if (irq == 2)   /* high priority peripheral */
+                       goto peripheral;
+               irq += MPC52xx_CRIT_IRQ_BASE;
+       } else if (status & 0x00200000) {       /* main */
+               irq = (status >> 16) & 0x1f;
+               if (irq == 4)   /* low priority peripheral */
+                       goto peripheral;
+               irq += MPC52xx_MAIN_IRQ_BASE;
+       } else if (status & 0x20000000) {       /* peripheral */
+             peripheral:
+               irq = (status >> 24) & 0x1f;
+               if (irq == 0) { /* bestcomm */
+                       status = in_be32(&sdma->IntPend);
+                       irq = ffs(status) + MPC52xx_SDMA_IRQ_BASE - 1;
+               } else
+                       irq += MPC52xx_PERP_IRQ_BASE;
+
+       }
+
+       virq = irq_linear_revmap(mpc52xx_irqhost, irq);
+       pr_debug("%s: irq=%d -> %d\n", __func__, irq, virq);
+
+       return virq;
+}
arch/powerpc/Kconfig 2006-10-26 18:55:58.000000000 +0200 renamed from arch/powerpc/Kconfig 2006-10-25 19:07:23.000000000 +0200 +5 -0
@@ -384,6 +384,11 @@ config PPC_CHRP
        select PPC_RTAS
        select PPC_MPC106
        select PPC_UDBG_16550
+       select PPC_MPC52xx_PIC
+       default y
+
+config PPC_MPC52xx_PIC
+        bool
        default y
 
 config PPC_PMAC
arch/powerpc/platforms/chrp/setup.c 2006-10-26 21:47:29.000000000 +0200 renamed from arch/powerpc/platforms/chrp/setup.c 2006-10-25 19:07:23.000000000 +0200 +46 -2
@@ -51,6 +51,7 @@
 #include <asm/mpic.h>
 #include <asm/rtas.h>
 #include <asm/xmon.h>
+#include <asm/mpc52xx.h>
 
 #include "chrp.h"
 
@@ -435,6 +436,44 @@ static struct irqaction xmon_irqaction =
 };
 #endif
 
+
+struct device_node *find_mpc52xx_pic(void)
+{
+       struct device_node *dev;
+       const char *piccompatible_list[] =
+       {
+               "mpc5200-pic",
+               NULL
+       };
+
+       /* Look for an MPC52xx interrupt controller */
+       for_each_node_by_type(dev, "interrupt-controller")
+       {
+               const char **piccompatible_entry = piccompatible_list;
+
+               for(piccompatible_entry = piccompatible_list; *piccompatible_entry; piccompatible_entry++ )
+               {
+                       if (device_is_compatible(dev, *piccompatible_entry ))
+                               return dev;
+               }
+       }
+
+       return NULL;
+}
+
+static int  __init chrp_find_mpc52xx_pic(void)
+{
+       if (find_mpc52xx_pic())
+       {
+               printk(KERN_INFO "Found MPC52xx Interrupt Controller\n");
+               ppc_md.get_irq = mpc52xx_get_irq;
+               mpc52xx_init_irq();
+               return 0;
+       }
+
+       return -ENODEV;
+}
+
 static void __init chrp_find_8259(void)
 {
        struct device_node *np, *pic = NULL;
@@ -473,8 +512,11 @@ static void __init chrp_find_8259(void)
                break;
        }
        if (np == NULL)
-               printk(KERN_WARNING "Cannot find PCI interrupt acknowledge"
-                      " address, polling\n");
+       {
+               printk(KERN_WARNING "Cannot find PCI/i8259 interrupt acknowledge"
+                      " Fix your tree!\n");
+               return;
+       }
 
        i8259_init(pic, chrp_int_ack);
        if (ppc_md.get_irq == NULL)
@@ -494,6 +536,8 @@ void __init chrp_init_IRQ(void)
 #if defined(CONFIG_VT) && defined(CONFIG_INPUT_ADBHID) && defined(XMON)
        struct device_node *kbd;
 #endif
+
+       chrp_find_mpc52xx_pic();
        chrp_find_openpic();
        chrp_find_8259();
 
#24 lore raw
Quoted 12 lines · Benjamin Herrenschmidt
Quoted 3 lines · Grant Likely

My comments are satisfied

Nack.

The irq code doesn't properly use the genirq infrastructure. It's not setting flow handlers, not implementing set_irq_type, not using the new mask/unmask instead of the toplevel enable/disable...

Done.

We agree on IRC last night to postpone this implementation later on. However, here it is.

This patch compiles fine when applied to the kernel 2.6.19-rc3. I also tested gcc 3.3.5 and 4.1.2.

I made some tests and did not found any issue. However, more testing/stressing/torturing and also with on others platforms is required to fully validate the driver.

Regards

arch/powerpc/sysdev/Makefile +1 -0
@@ -13,6 +13,7 @@ obj-$(CONFIG_FSL_SOC)		+= fsl_soc.o
 obj-$(CONFIG_PPC_TODC)		+= todc.o
 obj-$(CONFIG_TSI108_BRIDGE)	+= tsi108_pci.o tsi108_dev.o
 obj-$(CONFIG_QUICC_ENGINE)	+= qe_lib/
+obj-$(CONFIG_PPC_MPC52xx_PIC)	+= mpc52xx_pic.o
 
 ifeq ($(CONFIG_PPC_MERGE),y)
 obj-$(CONFIG_PPC_I8259)		+= i8259.o
arch/powerpc/sysdev/mpc52xx_pic.c +413 -0

413 lines not on this page · raw message

arch/powerpc/Kconfig +5 -0
@@ -384,6 +384,11 @@ config PPC_CHRP
 	select PPC_RTAS
 	select PPC_MPC106
 	select PPC_UDBG_16550
+	select PPC_MPC52xx_PIC
+	default y
+
+config PPC_MPC52xx_PIC
+        bool
 	default y
 
 config PPC_PMAC
arch/powerpc/platforms/chrp/setup.c +14 -0
@@ -51,6 +51,7 @@
 #include <asm/mpic.h>
 #include <asm/rtas.h>
 #include <asm/xmon.h>
+#include <asm/mpc52xx.h>
 
 #include "chrp.h"
 
@@ -435,6 +436,18 @@ static struct irqaction xmon_irqaction =
 };
 #endif
 
+static int __init chrp_find_mpc52xx_pic(void)
+{
+	if (find_mpc52xx_picnode()) {
+		printk(KERN_INFO "Found MPC52xx Interrupt Controller\n");
+		ppc_md.get_irq = mpc52xx_get_irq;
+		mpc52xx_init_irq();
+		return 0;
+	}
+
+	return -ENODEV;
+}
+
 static void __init chrp_find_8259(void)
 {
 	struct device_node *np, *pic = NULL;
@@ -496,6 +509,7 @@ void __init chrp_init_IRQ(void)
 #endif
 	chrp_find_openpic();
 	chrp_find_8259();
+	chrp_find_mpc52xx_pic();
 
 #ifdef CONFIG_SMP
 	/* Pegasos has no MPIC, those ops would make it crash. It might be an
include/asm-powerpc/mpc52xx.h +414 -0

414 lines not on this page · raw message

begin:vcard
fn:Nicolas DET ( bplan GmbH )
n:DET;Nicolas
org:bplan GmbH
adr:;;;;;;Germany
email;internet:[email protected]
title:Software Entwicklung
tel;work:+49 6171 9187 - 31
x-mozilla-html:FALSE
url:http://www.bplan-gmbh.de
version:2.1
end:vcard

#25 lore raw
Quoted 28 lines · Nicolas DET
Quoted 12 lines · Benjamin Herrenschmidt
Quoted 3 lines · Grant Likely

My comments are satisfied

Nack.

The irq code doesn't properly use the genirq infrastructure. It's not setting flow handlers, not implementing set_irq_type, not using the new mask/unmask instead of the toplevel enable/disable...

Done.

We agree on IRC last night to postpone this implementation later on. However, here it is.

This patch compiles fine when applied to the kernel 2.6.19-rc3. I also tested gcc 3.3.5 and 4.1.2.

I made some tests and did not found any issue. However, more testing/stressing/torturing and also with on others platforms is required to fully validate the driver.

/* Pegasos has no MPIC, those ops would make it crash. It might be an
diff -uprN a/include/asm-powerpc/mpc52xx.h b/include/asm-powerpc/mpc52xx.h

include/asm-powerpc/mpc52xx.h line 3

Quoted 3 lines
include/asm-powerpc/mpc52xx.h +3 -0
@@ -0,0 +1,414 @@
+/*
+ * include/asm-ppc/mpc52xx.h
+ * 

Fixed to powerpc

Quoted 28 lines
include/asm-powerpc/mpc52xx.h +28 -0
+#endif /* __ASSEMBLY__ */
+
+
+#ifdef CONFIG_PCI
+#define _IO_BASE	isa_io_base
+#define _ISA_MEM_BASE	isa_mem_base
+#define PCI_DRAM_OFFSET	pci_dram_offset
+#else
+#define _IO_BASE	0
+#define _ISA_MEM_BASE	0
+#define PCI_DRAM_OFFSET	0
+#endif
+
+
+/* ======================================================================== */
+/* PPC Sys devices definition                                               */
+/* ======================================================================== */
+
+enum ppc_sys_devices {
+	MPC52xx_MSCAN1,
+	MPC52xx_MSCAN2,
+	MPC52xx_SPI,
+	MPC52xx_USB,
+	MPC52xx_BDLC,
+	MPC52xx_PSC1,
+	MPC52xx_PSC2,
+	MPC52xx_PSC3,
+	MPC52xx_PSC4,

include/asm-powerpc/mpc52xx.h line 41

Quoted 10 lines
include/asm-powerpc/mpc52xx.h +10 -0
+	MPC52xx_PSC5,
+	MPC52xx_PSC6,
+	MPC52xx_FEC,
+	MPC52xx_ATA,
+	MPC52xx_I2C1,
+	MPC52xx_I2C2,
+	NUM_PPC_SYS_DEVS,
+};
+
+

Removed

arch/powerpc/sysdev/Makefile +1 -0
@@ -13,6 +13,7 @@ obj-$(CONFIG_FSL_SOC)		+= fsl_soc.o
 obj-$(CONFIG_PPC_TODC)		+= todc.o
 obj-$(CONFIG_TSI108_BRIDGE)	+= tsi108_pci.o tsi108_dev.o
 obj-$(CONFIG_QUICC_ENGINE)	+= qe_lib/
+obj-$(CONFIG_PPC_MPC52xx_PIC)	+= mpc52xx_pic.o
 
 ifeq ($(CONFIG_PPC_MERGE),y)
 obj-$(CONFIG_PPC_I8259)		+= i8259.o
arch/powerpc/sysdev/mpc52xx_pic.c +413 -0

413 lines not on this page · raw message

arch/powerpc/Kconfig +5 -0
@@ -384,6 +384,11 @@ config PPC_CHRP
 	select PPC_RTAS
 	select PPC_MPC106
 	select PPC_UDBG_16550
+	select PPC_MPC52xx_PIC
+	default y
+
+config PPC_MPC52xx_PIC
+        bool
 	default y
 
 config PPC_PMAC
arch/powerpc/platforms/chrp/setup.c +14 -0
@@ -51,6 +51,7 @@
 #include <asm/mpic.h>
 #include <asm/rtas.h>
 #include <asm/xmon.h>
+#include <asm/mpc52xx.h>
 
 #include "chrp.h"
 
@@ -435,6 +436,18 @@ static struct irqaction xmon_irqaction =
 };
 #endif
 
+static int __init chrp_find_mpc52xx_pic(void)
+{
+	if (find_mpc52xx_picnode()) {
+		printk(KERN_INFO "Found MPC52xx Interrupt Controller\n");
+		ppc_md.get_irq = mpc52xx_get_irq;
+		mpc52xx_init_irq();
+		return 0;
+	}
+
+	return -ENODEV;
+}
+
 static void __init chrp_find_8259(void)
 {
 	struct device_node *np, *pic = NULL;
@@ -496,6 +509,7 @@ void __init chrp_init_IRQ(void)
 #endif
 	chrp_find_openpic();
 	chrp_find_8259();
+	chrp_find_mpc52xx_pic();
 
 #ifdef CONFIG_SMP
 	/* Pegasos has no MPIC, those ops would make it crash. It might be an
include/asm-powerpc/mpc52xx.h +389 -0

389 lines not on this page · raw message

begin:vcard
fn:Nicolas DET ( bplan GmbH )
n:DET;Nicolas
org:bplan GmbH
adr:;;;;;;Germany
email;internet:[email protected]
title:Software Entwicklung
tel;work:+49 6171 9187 - 31
x-mozilla-html:FALSE
url:http://www.bplan-gmbh.de
version:2.1
end:vcard

#26 lore raw
Quoted 10 lines · Nicolas DET

include/asm-powerpc/mpc52xx.h line 3

Quoted 7 lines

diff -uprN a/include/asm-powerpc/mpc52xx.h b/include/asm-powerpc/mpc52xx.h

include/asm-powerpc/mpc52xx.h +3 -0
@@ -0,0 +1,414 @@
+/*
+ * include/asm-ppc/mpc52xx.h
+ * 

Fixed to powerpc

I thought we were removing the names of files from within the file itself, right?

Thanks,
jdl

#27 lore raw
Quoted 13 lines · On Fri, 27 Oct 2006 12:08:46 -0500 Jon Loeliger <[email protected]> wrote:
Quoted 9 lines · Nicolas DET

include/asm-powerpc/mpc52xx.h line 3

Quoted 7 lines

diff -uprN a/include/asm-powerpc/mpc52xx.h b/include/asm-powerpc/mpc52xx.h

include/asm-powerpc/mpc52xx.h +3 -0
@@ -0,0 +1,414 @@
+/*
+ * include/asm-ppc/mpc52xx.h
+ *

Fixed to powerpc

Quoted 2 lines

I thought we were removing the names of files from within the file itself, right?

Correct. They add nothing and become stale.

--
Cheers,
Stephen Rothwell [email protected]
http://www.canb.auug.org.au/~sfr/

#28 lore raw
In reply to Nicolas DET
Quoted 24 lines · Nicolas DET
arch/powerpc/sysdev/mpc52xx_pic.c +24 -0
+static void mpc52xx_ic_mask(unsigned int virq)
+{
+	u32 val;
+	int irq;
+	int l1irq;
+	int l2irq;
+
+	irq = irq_map[virq].hwirq;
+	l1irq = (irq & MPC52xx_IRQ_L1_MASK) >> MPC52xx_IRQ_L1_OFFSET;
+	l2irq = (irq & MPC52xx_IRQ_L2_MASK) >> MPC52xx_IRQ_L2_OFFSET;
+
+	pr_debug("%s: irq=%x. l1=%d, l2=%d\n", __func__, irq, l1irq, l2irq);
+
+	switch (l1irq) {
+	case MPC52xx_IRQ_L1_CRIT:
+		if (l2irq != 0)
+			BUG();
+
+		val = in_be32(&intr->ctrl);
+		val &= ~(1 << 11);
+		out_be32(&intr->ctrl, val);
+		break;
+
+	case MPC52xx_IRQ_L1_MAIN:

arch/powerpc/sysdev/mpc52xx_pic.c line 75

Quoted 10 lines
arch/powerpc/sysdev/mpc52xx_pic.c +10 -0
+		if ( (l2irq >= 1) && (l2irq <= 3) ) {
+			val = in_be32(&intr->ctrl);
+			val &= ~(1 << (11 - l2irq));
+			out_be32(&intr->ctrl, val);
+		} else {
+			val = in_be32(&intr->main_mask);
+			val |= 1 << (16 - l2irq);
+			out_be32(&intr->main_mask, val);
+		}
+		break;

Any reason why you do the above instead o defining two diferent levels instead. Also, L1_CRIT would fit in the L1_MAIN case too...

I don't see the point of having also XXX-l2, just put a bit number in L2 and be done with it.

The idea is to streamling the code, such that you can index an array with l1irq to get the register, and build a mask. No test, no switch case, etc...

Actually... the most performant way of doing all this is to have a different irq_chip (with a different set of ask,mask,unmask) for each "L1" so that they get right to the point. But I would settle for an array indexed by L1.

arch/powerpc/sysdev/mpc52xx_pic.c line 196

Quoted 5 lines
arch/powerpc/sysdev/mpc52xx_pic.c +5 -0
+static void mpc52xx_ic_mask_and_ack(unsigned int irq)
+{
+	mpc52xx_ic_mask(irq);
+	mpc52xx_ic_ack(irq);
+}

The above is unnuecessary. The core will call mask and ack.

arch/powerpc/sysdev/mpc52xx_pic.c line 203

Quoted 6 lines
arch/powerpc/sysdev/mpc52xx_pic.c +6 -0
+static struct irq_chip mpc52xx_irqchip = {
+	.typename = " MPC52xx  ",
+	.mask = mpc52xx_ic_mask,
+	.unmask = mpc52xx_ic_unmask,
+	.mask_ack = mpc52xx_ic_mask_and_ack,
+};

In the above, you need to provide ack.

arch/powerpc/sysdev/mpc52xx_pic.c line 209

Quoted 5 lines
arch/powerpc/sysdev/mpc52xx_pic.c +5 -0
+static int mpc52xx_irqhost_match(struct irq_host *h, struct device_node *node)
+{
+	pr_debug("%s: %p vs %p\n", __func__, find_mpc52xx_picnode(), node);
+	return find_mpc52xx_picnode() == node;
+}

Don't redo the whole find all the time. Put the node in your irq_host->host_data at init time and compare it there. That way, also, find_mpc53xx_picnode() thingy can just stay static to chrp/setup.c. The way you did it would be a problem if we had more than one platform using 52xx built in the same kernel.

Quoted 6 lines
arch/powerpc/sysdev/mpc52xx_pic.c +6 -0
+static int mpc52xx_islevel(int num)
+{
+	u32 ictl_req;
+	int ictl_type;
+
+	ictl_req = in_be32(&intr->ctrl);

arch/powerpc/sysdev/mpc52xx_pic.c line 267

Quoted 10 lines
arch/powerpc/sysdev/mpc52xx_pic.c +10 -0
+	ictl_type = (ictl_req >> 16) & 0x3;
+
+	switch (ictl_type) {
+	case 0:
+	case 3:
+		return 1;
+	}
+
+	return 0;
+}

You only have level/edge settings, not polarity ? Also, you aren't giving the user or device-tree the option to set the type manually....

Ideally, you should use the bits in IRQ_TYPE_SENSE_MASK to fully define the irq type/polarity and provide a set_irq_type() callback.

Also, beware that you cannot call set_irq_chip_and_handler() from within set_irq_type() due to a spinlock recursion issue. There is a patch floating on the list for IPIC fixing an issue of that sort, you may want to have a look.

In general, look at what others are doing, notably mpic and ipic.

You can also keep IRQ_LEVEL in "sync" with the other polarity bits, it's really only useful to display the polarity in /proc/interrupts.

Note that your map code would always OR the bit, never clear it. It should have probably cleared it before the switch/case.

arch/powerpc/sysdev/mpc52xx_pic.c line 316

Quoted 8 lines
arch/powerpc/sysdev/mpc52xx_pic.c +8 -0
+void mpc52xx_irqhost_unmap(struct irq_host *h, unsigned int virq)
+{
+	pr_debug("%s: v=%x\n", __func__, virq);
+
+	mpc52xx_ic_mask(virq);
+	set_irq_chip_and_handler(virq, NULL, NULL);
+	synchronize_irq(virq);
+}

The common code will do all of the above. Your unmap can be empty. (Or just don't do an unmap).

Quoted 34 lines
arch/powerpc/sysdev/mpc52xx_pic.c +34 -0
+static struct irq_host_ops mpc52xx_irqhost_ops = {
+	.match = mpc52xx_irqhost_match,
+	.xlate = mpc52xx_irqhost_xlate,
+	.map = mpc52xx_irqhost_map,
+	.unmap = mpc52xx_irqhost_unmap,
+};
+
+void __init mpc52xx_init_irq(void)
+{
+	int i;
+	u32 intr_ctrl;
+
+	/* Remap the necessary zones */
+	intr = ioremap(MPC52xx_PA(MPC52xx_INTR_OFFSET), MPC52xx_INTR_SIZE);
+	sdma = ioremap(MPC52xx_PA(MPC52xx_SDMA_OFFSET), MPC52xx_SDMA_SIZE);
+
+	if ((intr == NULL) || (sdma == NULL))
+		panic("Can't ioremap PIC/SDMA register or init_irq !");
+
+	/* Disable all interrupt sources. */
+	out_be32(&sdma->IntPend, 0xffffffff);	/* 1 means clear pending */
+	out_be32(&sdma->IntMask, 0xffffffff);	/* 1 means disabled */
+	out_be32(&intr->per_mask, 0x7ffffc00);	/* 1 means disabled */
+	out_be32(&intr->main_mask, 0x00010fff);	/* 1 means disabled */
+	intr_ctrl = in_be32(&intr->ctrl);
+	intr_ctrl &= 0x00ff0000;	/* Keeps IRQ[0-3] config */
+	intr_ctrl |= 0x0f000000 |	/* clear IRQ 0-3 */
+	    0x00001000 |	/* MEE master external enable */
+	    0x00000000 |	/* 0 means disable IRQ 0-3 */
+	    0x00000001;		/* CEb route critical normally */
+	out_be32(&intr->ctrl, intr_ctrl);
+
+	/* Zero a bunch of the priority settings.  */
+	out_be32(&intr->per_pri1, 0);

arch/powerpc/sysdev/mpc52xx_pic.c line 361

Quoted 10 lines
arch/powerpc/sysdev/mpc52xx_pic.c +10 -0
+	out_be32(&intr->per_pri2, 0);
+	out_be32(&intr->per_pri3, 0);
+	out_be32(&intr->main_pri1, 0);
+	out_be32(&intr->main_pri2, 0);
+	/* Initialize irq_desc[i].handler's with mpc52xx_ic. */
+	for (i = 0; i < NR_IRQS; i++) {
+		irq_desc[i].chip = &mpc52xx_irqchip;
+		irq_desc[i].status = IRQ_LEVEL;
+
+	}

The above is completely bogus and should be just removed. You should not touch the irq_desc array. You should only ever modify irq_desc's that have been assigned to you by your host->map callback.

arch/powerpc/sysdev/mpc52xx_pic.c line 371

Quoted 9 lines
arch/powerpc/sysdev/mpc52xx_pic.c +9 -0
+	/*
+	 * As last step, add an irq host to translate the real
+	 * hw irq information provided by the ofw to linux virq
+	 */
+
+	mpc52xx_irqhost =
+	    irq_alloc_host(IRQ_HOST_MAP_LINEAR, NR_IRQS, &mpc52xx_irqhost_ops,
+			   -1);
+}

As I said before, NR_IRQ is a poor choice for the size of the revmap. You should have a constant somewhere defining what is your max HW irq number. and use that +1, or a hw irq count.

Quoted 15 lines
arch/powerpc/platforms/chrp/setup.c +2 -0
@@ -51,6 +51,7 @@
 #include <asm/mpic.h>
 #include <asm/rtas.h>
 #include <asm/xmon.h>
+#include <asm/mpc52xx.h>
 
 #include "chrp.h"
 
@@ -435,6 +436,18 @@ static struct irqaction xmon_irqaction =
 };
 #endif
 
+static int __init chrp_find_mpc52xx_pic(void)

arch/powerpc/platforms/chrp/setup.c static struct irqaction xmon_irqaction = · line 449

Quoted 10 lines
arch/powerpc/platforms/chrp/setup.c +10 -0
+{
+	if (find_mpc52xx_picnode()) {
+		printk(KERN_INFO "Found MPC52xx Interrupt Controller\n");
+		ppc_md.get_irq = mpc52xx_get_irq;
+		mpc52xx_init_irq();
+		return 0;
+	}
+
+	return -ENODEV;
+}

Just get rif of find_mpic53xx_picnode. Do it locally and pass the device node to mpc52xx_init_irq().

Quoted 44 lines
arch/powerpc/platforms/chrp/setup.c +1 -0
 static void __init chrp_find_8259(void)
 {
 	struct device_node *np, *pic = NULL;
@@ -496,6 +509,7 @@ void __init chrp_init_IRQ(void)
 #endif
 	chrp_find_openpic();
 	chrp_find_8259();
+	chrp_find_mpc52xx_pic();
 
 #ifdef CONFIG_SMP
 	/* Pegasos has no MPIC, those ops would make it crash. It might be an

diff -uprN a/include/asm-powerpc/mpc52xx.h b/include/asm-powerpc/mpc52xx.h

include/asm-powerpc/mpc52xx.h +29 -0
@@ -0,0 +1,414 @@
+/*
+ * include/asm-powerpc/mpc52xx.h
+ * 
+ * Prototypes, etc. for the Freescale MPC52xx embedded cpu chips
+ * May need to be cleaned as the port goes on ...
+ *
+ *
+ * Maintainer : Sylvain Munaut <[email protected]>
+ *
+ * Originally written by Dale Farnsworth <[email protected]> 
+ * for the 2.4 kernel.
+ *
+ * Copyright (C) 2004-2005 Sylvain Munaut <[email protected]>
+ * Copyright (C) 2003 MontaVista, Software, Inc.
+ *
+ * This file is licensed under the terms of the GNU General Public License
+ * version 2. This program is licensed "as is" without any warranty of any
+ * kind, whether express or implied.
+ */
+
+#ifndef __ASM_POWERPC_MPC52xx_H__
+#define __ASM_POWERPC_MPC52xx_H__
+
+#ifndef __ASSEMBLY__
+#include <asm/types.h>
+#include <asm/prom.h>
+
+#endif /* __ASSEMBLY__ */
+

include/asm-powerpc/mpc52xx.h line 39

Quoted 10 lines
include/asm-powerpc/mpc52xx.h +10 -0
+
+#ifdef CONFIG_PCI
+#define _IO_BASE	isa_io_base
+#define _ISA_MEM_BASE	isa_mem_base
+#define PCI_DRAM_OFFSET	pci_dram_offset
+#else
+#define _IO_BASE	0
+#define _ISA_MEM_BASE	0
+#define PCI_DRAM_OFFSET	0
+#endif

Remove all of the above, it's bogus. You don't want to touch those things when CONFIG_MULTIPLATFORM. (That is, they are variables set programmatically, you can't just go #define them and common code already takes care of defining them anyway).

Ben.

#29 lore raw
In reply to Nicolas DET

Hi Nicolas,

Here is a few comments. I'm not very familiar with the new irq stuff so others
might have more insights.

Sylvain

arch/powerpc/sysdev/mpc52xx_pic.c line 2

Quoted 6 lines
arch/powerpc/sysdev/mpc52xx_pic.c +2 -0
@@ -0,0 +1,414 @@
+/*
+ * arch/powerpc/sysdev/mpc52xx_pic.c

Looks like jdl is right, we apparently don't do that any more ... So let's not ;)

arch/powerpc/sysdev/mpc52xx_pic.c line 10

Quoted 3 lines
arch/powerpc/sysdev/mpc52xx_pic.c +2 -0
+ * Based on (well, mostly copied from) the code from the 2.4 kernel by
+ * Dale Farnsworth <[email protected]> and Kent Borg.

That can be removed ... We can't blame Dale anymore if it doesn't work ;)

Quoted 9 lines
arch/powerpc/sysdev/mpc52xx_pic.c +6 -0
+
+static void mpc52xx_ic_mask_and_ack(unsigned int irq)
+{
+	mpc52xx_ic_mask(irq);
+	mpc52xx_ic_ack(irq);
+}

From kernel/irq/chip.c that's done automatically if mask_and_ack is NULL.

arch/powerpc/sysdev/mpc52xx_pic.c line 203

Quoted 7 lines
arch/powerpc/sysdev/mpc52xx_pic.c +7 -0
+
+static struct irq_chip mpc52xx_irqchip = {
+	.typename = " MPC52xx  ",
+	.mask = mpc52xx_ic_mask,
+	.unmask = mpc52xx_ic_unmask,
+	.mask_ack = mpc52xx_ic_mask_and_ack,
+};

Is it useful to implement set_type for IRQ[0-3] ? (Just asking ...)

arch/powerpc/sysdev/mpc52xx_pic.c line 361

Quoted 6 lines
arch/powerpc/sysdev/mpc52xx_pic.c +5 -0
+	for (i = 0; i < NR_IRQS; i++) {
+		irq_desc[i].chip = &mpc52xx_irqchip;
+		irq_desc[i].status = IRQ_LEVEL;
+
+	}

All LEVEL ?

arch/powerpc/sysdev/mpc52xx_pic.c line 371

Quoted 11 lines
arch/powerpc/sysdev/mpc52xx_pic.c +10 -0
+
+	/*
+	 * As last step, add an irq host to translate the real
+	 * hw irq information provided by the ofw to linux virq
+	 */
+
+	mpc52xx_irqhost =
+	    irq_alloc_host(IRQ_HOST_MAP_LINEAR, NR_IRQS, &mpc52xx_irqhost_ops,
+			   -1);
+}

NR_IRQS ? Might be time to do something better.

Quoted 6 lines

diff -uprN a/include/asm-powerpc/mpc52xx.h b/include/asm-powerpc/mpc52xx.h

include/asm-powerpc/mpc52xx.h +1 -0
@@ -0,0 +1,414 @@
+/*

include/asm-powerpc/mpc52xx.h line 11

Quoted 10 lines
include/asm-powerpc/mpc52xx.h +10 -0
+ * include/asm-ppc/mpc52xx.h
+ * 
+ * Prototypes, etc. for the Freescale MPC52xx embedded cpu chips
+ * May need to be cleaned as the port goes on ...
+ *
+ *
+ * Maintainer : Sylvain Munaut <[email protected]>
+ *
+ * Originally written by Dale Farnsworth <[email protected]> 
+ * for the 2.4 kernel.

Again, remove all that (just leave the first line of the description)

Quoted 6 lines
include/asm-powerpc/mpc52xx.h +5 -0
+
+/* ======================================================================== */
+/* IRQ mapping                                                              */
+/* ======================================================================== */
+

include/asm-powerpc/mpc52xx.h line 122

Quoted 10 lines
include/asm-powerpc/mpc52xx.h +10 -0
+#define MPC52xx_IRQ_L1_CRIT	0
+#define MPC52xx_IRQ_L1_MAIN	1
+#define MPC52xx_IRQ_L1_PERP	2
+#define MPC52xx_IRQ_L1_SDMA	3
+
+#define MPC52xx_IRQ_L1_OFFSET  (6)
+#define MPC52xx_IRQ_L1_MASK    (0xc0)
+
+#define MPC52xx_IRQ_L2_OFFSET  (0)
+#define MPC52xx_IRQ_L2_MASK    (0x3f)

As benh suggested on IRC, a L1 offset of 8 might be better for readability of the
hw irq numbers.

#30 lore raw
In reply to Nicolas DET

Thanks Nicolas, I'll dig through the PIC code tonight. I'm really not sure about the implications of this being a CHRP board vs. the Lite5200 using u-boot. ie. Should the Lite5200 also be a CHRP platform? Does u-boot provide enough capability (this is where my ignorance about CHRP shines through)

Cheers,
g.

Quoted 9 lines · Nicolas DET

* mpc5k2.c: As far as I understood, using of_platform_device will allow 'smart' of tree parsing. I did not implement it yet.

* mpc5k2_pic.c : The thingy ;-). I moved it into arch/powerpc/sysdev as suggested. This code is 99% copy/paste from the ARCH=ppc. I added an irq_host with the appropriate xlate and co. However, I have not done more as this code has been written by others. I think we should work together to move to new Linux style irq. I stay ready to test or code upcoming revision/

Why change the naming scheme from mpc52xx to mpc5k3? All other freescale supports follow the mpcXXxxx scheme.

Quoted 16 lines

* arch/powerpc/platform/*. Here, only small changes has been made. I added _CHRP_E5K2 (in asm/processor.h), an entry in chrp_names() and the correct detection in arch/powerpc/platform/setup.c / chrp_setup_arch(). in chrp_init_IRQ(), I set ppc_md.get_irq to mpc52xx_get_irq and call mpc52xx_init_irq().

* serial stuff / IBP Freq. As this frequency is MPC52xx specific. I think it would make sense to add a new func mpc52xx_getipbfreq(void). The way it would be implemented may depend of the architecture. Our firmware contains an 'ipb-freq' property in '/builtin/'.

diff -uprN a/arch/powerpc/platforms/chrp/setup.c b/arch/powerpc/platforms/chrp/setup.c

arch/powerpc/platforms/chrp/setup.c 2006-10-25 19:10:18.000000000 +0200 static const char *chrp_names[] = { · line 107

Quoted 7 lines
arch/powerpc/platforms/chrp/setup.c 2006-10-25 19:10:18.000000000 +0200 renamed from arch/powerpc/platforms/chrp/setup.c 2006-10-25 19:07:23.000000000 +0200 +2 -1
@@ -101,7 +102,8 @@ static const char *chrp_names[] = {
        "Motorola",
        "IBM or Longtrail",
        "Genesi Pegasos",
-       "Total Impact Briq"
+       "Total Impact Briq",
+       "bPlan Efika"
 };

I agree w/ Paul... the chrp_type stuff is stinky. :)

arch/powerpc/platforms/chrp/setup.c 2006-10-25 19:10:18.000000000 +0200 void __init chrp_init_IRQ(void) · line 508

Quoted 12 lines
arch/powerpc/platforms/chrp/setup.c 2006-10-25 19:10:18.000000000 +0200 renamed from arch/powerpc/platforms/chrp/setup.c 2006-10-25 19:07:23.000000000 +0200 +6 -0
@@ -494,6 +498,12 @@ void __init chrp_init_IRQ(void)
 #if defined(CONFIG_VT) && defined(CONFIG_INPUT_ADBHID) && defined(XMON)
        struct device_node *kbd;
 #endif
+       if (_chrp_type ==  _CHRP_E5K2) {
+               ppc_md.get_irq = mpc52xx_get_irq;
+               mpc52xx_init_irq();
+               return;
+       }
+
        chrp_find_openpic();
        chrp_find_8259();

Also, this only covers one 52xx board and does not match the convention in this function. Shouldn't this instead call a new function "chrp_find_mpc52xx_pic()?

arch/powerpc/platforms/chrp/setup.c 2006-10-25 19:10:18.000000000 +0200 chrp_init2(void) · line 548

Quoted 11 lines
arch/powerpc/platforms/chrp/setup.c 2006-10-25 19:10:18.000000000 +0200 renamed from arch/powerpc/platforms/chrp/setup.c 2006-10-25 19:07:23.000000000 +0200 +3 -0
@@ -530,6 +540,9 @@ chrp_init2(void)
        chrp_nvram_init();
 #endif
 
+       if (_chrp_type == _CHRP_E5K2)
+               return;
+
        request_region(0x20,0x20,"pic1");
        request_region(0xa0,0x20,"pic2");
        request_region(0x00,0x20,"dma1");

There's got to be a better way to go about this. If all the request_regions are not required by all CHRP platforms, then they should be enclosed by an if{} block. Just bailing from the function does not seem to be a good idea to me. For example, at the end of chrp_init2() there is a progress method called (if enabled). Bailing out prevents that progress message. (not a big issue, but I'm looking a convention consistency here)

include/asm-powerpc/processor.h 2006-10-25 19:11:54.000000000 +0200 line 36

Quoted 8 lines

diff -uprN a/include/asm-powerpc/processor.h b/include/asm-powerpc/processor.h

include/asm-powerpc/processor.h 2006-10-25 19:11:54.000000000 +0200 renamed from include/asm-powerpc/processor.h 2006-10-25 19:07:48.000000000 +0200 +1 -0
@@ -33,6 +33,7 @@
 #define _CHRP_IBM      0x05    /* IBM chrp, the longtrail and longtrail 2 */
 #define _CHRP_Pegasos  0x06    /* Genesi/bplan's Pegasos and Pegasos2 */
 #define _CHRP_briq     0x07    /* TotalImpact's briQ */
+#define _CHRP_E5K2     0x08    /* bPlan's Efika 5k2*/

Ick. :)

include/asm-ppc/mpc52xx.h 2006-10-25 19:11:55.000000000 +0200 enum ppc_sys_devices { · line 125

Quoted 12 lines

diff -uprN a/include/asm-ppc/mpc52xx.h b/include/asm-ppc/mpc52xx.h

include/asm-ppc/mpc52xx.h 2006-10-25 19:11:55.000000000 +0200 renamed from include/asm-ppc/mpc52xx.h 2006-10-25 19:07:48.000000000 +0200 +1 -1
@@ -119,7 +119,7 @@ enum ppc_sys_devices {
 #define MPC52xx_SDMA_IRQ_NUM   17
 #define MPC52xx_PERP_IRQ_NUM   23
 
-#define MPC52xx_CRIT_IRQ_BASE  1
+#define MPC52xx_CRIT_IRQ_BASE  0
 #define MPC52xx_MAIN_IRQ_BASE  (MPC52xx_CRIT_IRQ_BASE + MPC52xx_CRIT_IRQ_NUM)
 #define MPC52xx_SDMA_IRQ_BASE  (MPC52xx_MAIN_IRQ_BASE + MPC52xx_MAIN_IRQ_NUM)
 #define MPC52xx_PERP_IRQ_BASE  (MPC52xx_SDMA_IRQ_BASE + MPC52xx_SDMA_IRQ_NUM)

Just curious... why?

arch/powerpc/sysdev/mpc52xx_pic.c 2006-10-25 19:17:48.000000000 +0200

Quoted 2 lines
arch/powerpc/sysdev/mpc52xx_pic.c 2006-10-25 19:17:48.000000000 +0200 renamed from arch/powerpc/sysdev/mpc52xx_pic.c 1970-01-01 01:00:00.000000000 +0100 +0 -0

I'll skip commenting on PIC stuff as I'm still working through how it all works myself. It still needs to be worked out if we go to a two level interrupt # table in the device tree (interrupt_cells=3) and where they will be mapped back into Linux IRQ numbers.

Quoted 14 lines
arch/powerpc/sysdev/Makefile 2006-10-25 20:33:32.000000000 +0200 renamed from arch/powerpc/sysdev/Makefile 2006-10-25 19:07:24.000000000 +0200 +1 -0
@@ -13,6 +13,7 @@ obj-$(CONFIG_FSL_SOC)         += fsl_soc.o
 obj-$(CONFIG_PPC_TODC)         += todc.o
 obj-$(CONFIG_TSI108_BRIDGE)    += tsi108_pci.o tsi108_dev.o
 obj-$(CONFIG_QUICC_ENGINE)     += qe_lib/
+obj-$(CONFIG_PPC_CHRP)         += mpc52xx_pic.o
 
 ifeq ($(CONFIG_PPC_MERGE),y)
 obj-$(CONFIG_PPC_I8259)                += i8259.o
Signature · 4 lines
Grant Likely, B.Sc. P.Eng.
Secret Lab Technologies Ltd.
[email protected]
(403) 399-0195
#31 lore raw
Quoted 5 lines · Grant Likely

Thanks Nicolas, I'll dig through the PIC code tonight. I'm really not sure about the implications of this being a CHRP board vs. the Lite5200 using u-boot. ie. Should the Lite5200 also be a CHRP platform? Does u-boot provide enough capability (this is where my ignorance about CHRP shines through)

Bah, if you have no RTAS, don't bother being CHRP.

Quoted 4 lines
Quoted 4 lines

* serial stuff / IBP Freq. As this frequency is MPC52xx specific. I think it would make sense to add a new func mpc52xx_getipbfreq(void). The way it would be implemented may depend of the architecture. Our firmware contains an 'ipb-freq' property in '/builtin/'.

BTW. I forgot to comment on the above in my previous mail... As I said, there should be no such thing. Instead, the IPB freq should be in the device-tree.

#32 lore raw
Quoted 8 lines · Benjamin Herrenschmidt
Quoted 4 lines
Quoted 4 lines

* serial stuff / IBP Freq. As this frequency is MPC52xx specific. I think it would make sense to add a new func mpc52xx_getipbfreq(void). The way it would be implemented may depend of the architecture. Our firmware contains an 'ipb-freq' property in '/builtin/'.

BTW. I forgot to comment on the above in my previous mail... As I said, there should be no such thing. Instead, the IPB freq should be in the device-tree.

Euh, how is this different from having our "firmware contains an 'ipb-freq' property in '/builtin/'". There is no an ipb-freq property in the device tree, so you need a function to be able to get it from the the device tree, which i believe what is proposed here.

Friendly,

Sven Luther

#33 lore raw
In reply to Benjamin Herrenschmidt
Quoted 11 lines · Benjamin Herrenschmidt
Quoted 4 lines
Quoted 4 lines

* serial stuff / IBP Freq. As this frequency is MPC52xx specific. I think it would make sense to add a new func mpc52xx_getipbfreq(void). The way it would be implemented may depend of the architecture. Our firmware contains an 'ipb-freq' property in '/builtin/'.

BTW. I forgot to comment on the above in my previous mail... As I said, there should be no such thing. Instead, the IPB freq should be in the device-tree.

What do you mean exactly by 'the IPB freq should be in the device-tree'? It is as property.

Regards

begin:vcard
fn:Nicolas DET ( bplan GmbH )
n:DET;Nicolas
org:bplan GmbH
adr:;;;;;;Germany
email;internet:[email protected]
title:Software Entwicklung
tel;work:+49 6171 9187 - 31
x-mozilla-html:FALSE
url:http://www.bplan-gmbh.de
version:2.1
end:vcard

#34 lore raw
Quoted 2 lines · Nicolas DET

What do you mean exactly by 'the IPB freq should be in the device-tree'? It is as property.

Yeah, just that drivers should pick it up from there

Ben.

#35 lore raw
In reply to Nicolas DET
Quoted 8 lines

* mpc5k2_bestcomm.c (& helpers): I've been in contact with others people using bestcomm and please to see the development was still active. I'm already using latest code on my board but this is not include in this patch. Indeed, it needs some changes for ARCH=powerpc. Moreover, we have a different approach for the tasks. I'll be pleased to work with the bestcomm folks.

So you preload the bestcomm task code in your boot loader? Do those of us using other bootloaders still have the option of loading the task code later in linux?

What api are you using? Sylvain's proposed one based on Dale's with some additions from me? Or have you come up with your own?

#36 lore raw
Quoted 10 lines · John Rigby
Quoted 8 lines

* mpc5k2_bestcomm.c (& helpers): I've been in contact with others people using bestcomm and please to see the development was still active. I'm already using latest code on my board but this is not include in this patch. Indeed, it needs some changes for ARCH=powerpc. Moreover, we have a different approach for the tasks. I'll be pleased to work with the bestcomm folks.

So you preload the bestcomm task code in your boot loader?

Yes.

Quoted 3 lines

Do those
of us using other bootloaders still have the option of loading the task code later in linux?

Well, just reading the phy address from 0xf008000 to 0xf00c0000 (if I remenber correctly) should be enough to get the tasks code.

Quoted 3 lines

What api are you using? Sylvain's proposed one based on Dale's with some additions from me? Or have you come up with your own?

We implemented our own yet. However, we could moved to any.

Regards,

begin:vcard
fn:Nicolas DET ( bplan GmbH )
n:DET;Nicolas
org:bplan GmbH
adr:;;;;;;Germany
email;internet:[email protected]
title:Software Entwicklung
tel;work:+49 6171 9187 - 31
x-mozilla-html:FALSE
url:http://www.bplan-gmbh.de
version:2.1
end:vcard

#37 lore raw
In reply to John Rigby
Quoted 9 lines · John Rigby
Quoted 8 lines

* mpc5k2_bestcomm.c (& helpers): I've been in contact with others people using bestcomm and please to see the development was still active. I'm already using latest code on my board but this is not include in this patch. Indeed, it needs some changes for ARCH=powerpc. Moreover, we have a different approach for the tasks. I'll be pleased to work with the bestcomm folks.

Quoted 7 lines

So you preload the bestcomm task code in your boot loader? Do those of us using other bootloaders still have the option of loading the task code later in linux?

What api are you using? Sylvain's proposed one based on Dale's with some additions from me? Or have you come up with your own?

Hi John,

I've sent him the latest API you describe.

As for task preloading, I'm not too sure what Nicolas' plan is. I'm waiting to see
what he's proposing.

Since the bestcomm init is in a module, a different module could be implemented,
(like bestcomm-chrp5k2.ko) that instead of clearing SRAM and doing everything
from scrtch, would rely on the bootloader more. This would allow "custom" tasks
to be marked as "don't touch". So if the bootloader want to load something that
will not be altered by the linux driver (for whatever reason ...)

However, for the tasks like FEC and ATA and everything supported in the kernel
natively, I would still recommend reload the task code (anyway the driver is gonna
reset the task and everything ...) Trying to reuse the microcode loaded by the
bootloader is not a great idea IMHO. (Imagine, the fec driver is changed to use
the latest microcode, but the bootloader is not ... both have to be in perfect sync ...)

We would still have to compare the internal FDT and the preloaded one though since
they have to match ...

Sylvain

#38 lore raw
Quoted 9 lines

Since the bestcomm init is in a module, a different module could be implemented,
(like bestcomm-chrp5k2.ko) that instead of clearing SRAM and doing everything
from scrtch, would rely on the bootloader more. This would allow "custom" tasks
to be marked as "don't touch". So if the bootloader want to load something that
will not be altered by the linux driver (for whatever reason ...)

I don't think we should have two different modules.

We are talking here about arch/powerpc where the presence of a device-tree is mandatory. Thus, we should rely on that and define something based on that.

In any case, I strongly think a single module with a single API should be able to deal with all the cases we need to define for loading the tasks. As I defined in a separate email, that includes wether the tasks are preloaded, provided as a blob in the device-tree (possibly via the zImage wrapper, and thus still in the kernel tree), or hard coded in the kernel driver (I don't like that solution though).

It's mostly a matter on how the thing initializes anyway. Once it's up and running, the code path should be identical.

Quoted 10 lines

However, for the tasks like FEC and ATA and everything supported in the kernel
natively, I would still recommend reload the task code (anyway the driver is gonna
reset the task and everything ...) Trying to reuse the microcode loaded by the
bootloader is not a great idea IMHO. (Imagine, the fec driver is changed to use
the latest microcode, but the bootloader is not ... both have to be in perfect sync ...)

Is there some versionning available with the microcode ? I'm afraid there might not be (since that whole bestcomm business seems to be immune to good ideas from whoever design the whole chip), but that would be useful. I agree that having the task code coming from the firmware might not be the best idea if there is no versionning, but it nicely solves some of the problems related to distributing "binary blobs" ... I'm sure debian would have issues with bestcomm code shipped with the kernel....

If no versionning is available we might want to create one ourselves, via a wiki possibly and one person (you ?) responsible for distributing the "releases" and assigning them versions...

At which point, the device-tree can contain version information along with the task code or pointers to the code in sram.

Quoted 3 lines

We would still have to compare the internal FDT and the preloaded one though since
they have to match ...

Ben.

#39 lore raw
In reply to John Rigby
Quoted 9 lines · John Rigby
Quoted 8 lines

* mpc5k2_bestcomm.c (& helpers): I've been in contact with others people using bestcomm and please to see the development was still active. I'm already using latest code on my board but this is not include in this patch. Indeed, it needs some changes for ARCH=powerpc. Moreover, we have a different approach for the tasks. I'll be pleased to work with the bestcomm folks.

Quoted 6 lines

So you preload the bestcomm task code in your boot loader? Do those of us using other bootloaders still have the option of loading the task code later in linux?

What api are you using? Sylvain's proposed one based on Dale's with some additions from me? Or have you come up with your own?

I think we need to look into 3 options for the bestcomm task code:

- Preloaded by the firmware
- In the device-tree but not preloaded
- In the kernel

Though we might want to completely dismiss the 3rd option and decide that it has to be in the device-tree, wether it's provided by the zImage wrapper, or by the firmware.

Do you see any other case ?

So we need to define the content of the various device nodes for those 3 cases. The 3rd case is easy: probably nothing to put there, purely an in-kernel thing, unless the bestcomm task IDs related to a given cell are always the same.

The two first ones are similar, with the exception that the "blob" is in one case preloaded (thus we provide a property telling where it is in the sram) and in the other case, the blob is in the DT itself, thus we have a bit of "bootstrap" code that loads it. In both cases, tasks IDs are known and thus the device-tree might contain the necessary binding informations.

Now, in any case, the "drivers" need a way to link to their respective "tasks". That is, the function drivers (FEC, ...) need to use functions provided by the "bestcomm" driver to setup and use tasks. In order to do that the function drivers need to provide enough information of course to allow the bestcomm to properly find out the tasks.

That's why my idea is that we associate to "known" bestcomm tasks a name (a string), for example, "fec-tx" and "fec-rx" and the function drivers pass their device-node to the bestcomm drivers. The Bestcomm "driver" can then look into the device-tree node for a property "bestcomm-task-% s" where %s is the task "name", containing the task informations (ID, position in a table, whatever, I don't remember the details, it's up to you guys to define something).

That property can then be in the device own node, in the bestcomm node, or absent in the case where the whole bestcom ucode is embedded in the kernel itself, in which case the driver has an internal table to resolve that, but I wonder if we shouldn't make madantory to at least have it in the device-tree and thus ignore that case 3...

Ben.