linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
To: Grant Likely <grant.likely@secretlab.ca>
Cc: linux-kernel@vger.kernel.org, sodaville@linutronix.de,
	x86@kernel.org, devicetree-discuss@lists.ozlabs.org
Subject: Re: [PATCH v2 05/15] x86/dtb: add early parsing of APIC and IO APIC
Date: Tue, 18 Jan 2011 15:56:47 +0100	[thread overview]
Message-ID: <4D35AA2F.3000503@linutronix.de> (raw)
In-Reply-To: <20110111221401.GB2131@angua.secretlab.ca>

Grant Likely wrote:
> Hi Sebastian,
Hi Grant,

>> diff --git a/arch/x86/include/asm/prom.h b/arch/x86/include/asm/prom.h
>> index 9076ae4..3bc8ed5 100644
>> --- a/arch/x86/include/asm/prom.h
>> +++ b/arch/x86/include/asm/prom.h
>> @@ -22,10 +22,17 @@
>>  #include <asm/irq_controller.h>
>>  
>>  #ifdef CONFIG_OF
>> +extern int of_ioapic;
>> +extern u64 initial_dtb;
>>  extern void add_dtb(u64 data);
>> +void x86_dtb_find_config(void);
>> +void x86_dtb_get_config(unsigned int unused);
>>  void add_interrupt_host(struct irq_domain *ih);
>>  #else
>>  static inline void add_dtb(u64 data) { }
>> +#define x86_dtb_find_config x86_init_noop
>> +#define x86_dtb_get_config x86_init_uint_noop
>> +#define of_ioapic 0
> 
> Nit: Personally, I prefer static inlines over preprocessor macros, but
> that isn't anything that will block this patch.

Right. x86 does similar for ioapic function callbacks for CONFIG_IOAPIC
case. So I keep this unless x86 folks want it different.

>> +#ifdef CONFIG_X86_IO_APIC
>> +static void __init dtb_add_ioapic(struct device_node *dn)
>> +{
>> +	unsigned int ioapic_id;
>> +	const __be32 *cell;
>> +	int len;
>> +	struct resource r;
>> +	int ret;
>> +
>> +	cell = of_get_property(dn, "id", &len);
>> +	if (!cell) {
>> +		printk(KERN_ERR "Node %s is missing id property.\n",
>> +				dn->full_name);
>> +		return;
>> +	}
>> +	ioapic_id = of_read_ulong(cell, len / 4);
> 
> This looks like poor usage practise for the device tree.  'id' or any
> kind of enumeration property in a device tree node is strongly
> discouraged.  As much as possible, let Linux dynamically enumerate
> devices rather than trying to explicitly specify the numbering.  Since
> you can explicitly describe the relationship between device nodes in
> the tree, you should find that explicitly specifying numbers is almost
> never required.

Good point. I removed the id property and let linux statically enumerate
it. I don't see any relationship and spec. says only to keep it unique.

Sebastian

  reply	other threads:[~2011-01-18 14:57 UTC|newest]

Thread overview: 71+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-12-17 15:33 Add device tree support for x86, v2 Sebastian Andrzej Siewior
2010-12-17 15:33 ` [PATCH 01/15] x86/e820: remove conditional early mapping in parse_e820_ext Sebastian Andrzej Siewior
2010-12-30  8:37   ` Grant Likely
2011-01-04 13:08     ` [PATCH v2 " Sebastian Andrzej Siewior
2011-01-14  8:14       ` Grant Likely
2011-01-14 10:57         ` Sebastian Andrzej Siewior
2011-01-14 21:43           ` Andres Salomon
2011-01-27 19:54             ` Daniel Drake
2011-02-01 13:21               ` Sebastian Andrzej Siewior
2011-02-01 18:36                 ` Andres Salomon
2011-02-02  3:10                   ` Grant Likely
2011-02-02 16:22                     ` Sebastian Andrzej Siewior
2011-02-03 18:40                     ` Andres Salomon
2011-02-03 19:44                       ` [sodaville] " H. Peter Anvin
2011-02-03 20:09                         ` Andres Salomon
2011-02-03 20:16                           ` H. Peter Anvin
2011-02-03 20:39                             ` Andres Salomon
2011-02-03 21:04                               ` H. Peter Anvin
2011-02-02 18:59               ` H. Peter Anvin
2010-12-17 15:33 ` [PATCH 02/15] x86: Add device tree support Sebastian Andrzej Siewior
2010-12-30  8:43   ` Grant Likely
2010-12-30 21:01     ` Grant Likely
2011-01-02  0:40       ` H. Peter Anvin
2011-01-03 12:20     ` Sebastian Andrzej Siewior
2011-01-03 18:05       ` [sodaville] " H. Peter Anvin
2010-12-17 15:33 ` [PATCH 03/15] x86/dtb: Add a device tree for CE4100 Sebastian Andrzej Siewior
2010-12-30  8:51   ` Grant Likely
2011-01-03 11:28     ` Sebastian Andrzej Siewior
2011-01-03 17:45       ` Grant Likely
2011-01-05  9:48         ` [PATCH v2 " Sebastian Andrzej Siewior
2011-01-05 10:01     ` [PATCH " Sebastian Andrzej Siewior
2011-01-05 23:20     ` David Gibson
2010-12-17 15:33 ` [PATCH 04/15] x86/dtb: add irq domain abstraction Sebastian Andrzej Siewior
2011-01-11 22:03   ` Grant Likely
2011-01-23 13:06     ` Sebastian Andrzej Siewior
2010-12-17 15:33 ` [PATCH 05/15] x86/dtb: add early parsing of APIC and IO APIC Sebastian Andrzej Siewior
2010-12-30  8:54   ` Grant Likely
2011-01-04 13:23     ` [PATCH v2 " Sebastian Andrzej Siewior
2011-01-11 22:14       ` Grant Likely
2011-01-18 14:56         ` Sebastian Andrzej Siewior [this message]
2010-12-17 15:33 ` [PATCH 06/15] x86/dtb: add support hpet Sebastian Andrzej Siewior
2011-01-11 22:26   ` Grant Likely
2010-12-17 15:33 ` [PATCH 07/15] of: move of_irq_map_pci() into generic code Sebastian Andrzej Siewior
2010-12-17 21:16   ` Benjamin Herrenschmidt
2011-01-04 14:27     ` [PATCH v2 " Sebastian Andrzej Siewior
2011-01-11 23:27       ` Grant Likely
2011-01-12 18:21         ` Sebastian Andrzej Siewior
2010-12-17 15:33 ` [PATCH 08/15] x86/dtb: add support for PCI devices backed by dtb nodes Sebastian Andrzej Siewior
2011-01-11 23:47   ` Grant Likely
2010-12-17 15:33 ` [PATCH 09/15] x86/dtb: Add generic bus probe Sebastian Andrzej Siewior
2011-01-11 23:48   ` Grant Likely
2010-12-17 15:33 ` [PATCH 10/15] x86/ioapic: Add OF bindings for IO-APIC Sebastian Andrzej Siewior
2011-01-11 23:53   ` Grant Likely
2011-01-12 17:07     ` Sebastian Andrzej Siewior
2011-01-12 17:19       ` [sodaville] " H. Peter Anvin
2011-01-13 10:38         ` Sebastian Andrzej Siewior
2010-12-17 15:33 ` [PATCH 11/15] x86/ce4100: use OF for ioapic Sebastian Andrzej Siewior
2011-01-11 23:54   ` Grant Likely
2010-12-17 15:33 ` [PATCH 12/15] of/address: use propper endianess in get_flags Sebastian Andrzej Siewior
2010-12-30  9:05   ` Grant Likely
2010-12-17 15:33 ` [PATCH 13/15] x86/rtc: don't register rtc if we have an OF node for it Sebastian Andrzej Siewior
2010-12-30  8:59   ` Grant Likely
2011-01-04 13:28     ` [PATCH v2 13/15] x86/rtc: don't register rtc if we the DT blob Sebastian Andrzej Siewior
2011-01-12  0:02       ` Grant Likely
2011-01-12 18:29         ` Sebastian Andrzej Siewior
2011-01-14 22:16           ` Grant Likely
2010-12-17 15:33 ` [PATCH 14/15] rtc/cmos: add OF bindings Sebastian Andrzej Siewior
2011-01-12  0:04   ` Grant Likely
2011-01-13 10:50     ` [PATCH v2] " Sebastian Andrzej Siewior
2010-12-17 15:33 ` [PATCH 15/15] x86/pci: remove warning Sebastian Andrzej Siewior
2011-02-18 16:06   ` [tip:x86/platform] x86/pci: Remove unused variable tip-bot for Sebastian Andrzej Siewior

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=4D35AA2F.3000503@linutronix.de \
    --to=bigeasy@linutronix.de \
    --cc=devicetree-discuss@lists.ozlabs.org \
    --cc=grant.likely@secretlab.ca \
    --cc=linux-kernel@vger.kernel.org \
    --cc=sodaville@linutronix.de \
    --cc=x86@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).