All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Randy.Dunlap" <rddunlap@osdl.org>
To: kernel-janitors@vger.kernel.org
Subject: Re: [Kernel-janitors] [PATCH] IO-APIC debug message reducti
Date: Wed, 07 Jul 2004 21:03:13 +0000	[thread overview]
Message-ID: <20040707140313.390abfa1.rddunlap@osdl.org> (raw)
In-Reply-To: <1089145628.2957.9.camel@tigger>

[-- Attachment #1: Type: text/plain, Size: 4319 bytes --]

On Wed, 07 Jul 2004 21:29:36 +0100 Mark Broadbent wrote:

| Hi,
| 
| > > +static int ioapic_verbosity = 0;
| > > +
| >
| > no need to init static variables to zero, not doing so has the 
| > advantage of moving this variable to the .bss kernel section.
| 
| Does this mean it will get set to zero automagically or must it
| be set explicitly in the code?

The BSS section of data is cleared to zero by kernel init...
(somewhere :)

| As for the other comments. I have removed unneeded parens around the if
| statements and brackets around the defined constants.  I have used a
| ioapic_printk define to replace the original printk except for the
| functions labelled print_* that break out at function start, this is
| done to reduce the size of the resulting object code (saves ~1.5kB).
| 
| Thanks for the comments, a revised patch is given below:
| 
| 
| Index: linux-2.6.7/arch/i386/kernel/io_apic.c
| ===================================================================
| --- linux-2.6.7/arch/i386/kernel/io_apic.c	(revision 1)
| +++ linux-2.6.7/arch/i386/kernel/io_apic.c	(working copy)
| @@ -84,7 +84,23 @@
|  #define vector_to_irq(vector)	(vector)
|  #endif
|  
| +#define IOAPIC_QUIET   0
| +#define IOAPIC_VERBOSE 1
| +#define IOAPIC_DEBUG   2
| +
|  /*
| + * Define the default level of output to be very little
| + * This can be turned up by using apic=verbose for more
| + * information and apic=debug for _lots_ of information.
| + */
| +static int ioapic_verbosity;
| +
| +#define ioapic_printk(v, s, a...) do {       \
| +		if ((v) >= ioapic_verbosity) \
| +			printk(s, ##a);      \
| +	} while (0)
| +
| +/*
|   * The common case is 1:1 IRQ<->pin mappings. Sometimes there are
|   * shared ISA-space IRQs, so we have to support them. We are super
|   * fast in the common case, and fast for shared ISA-space IRQs.
| @@ -745,13 +761,15 @@
|  		pirq_entries[i] = -1;
|  
|  	pirqs_enabled = 1;
| -	printk(KERN_INFO "PIRQ redirection, working around broken
| MP-BIOS.\n");
| +	ioapic_printk(IOAPIC_VERBOSE, KERN_INFO 
| +			"PIRQ redirection, working around broken MP-BIOS.\n");
|  	max = MAX_PIRQS;
|  	if (ints[0] < MAX_PIRQS)
|  		max = ints[0];
|  
|  	for (i = 0; i < max; i++) {
| -		printk(KERN_DEBUG "... PIRQ%d -> IRQ %d\n", i, ints[i+1]);
| +		ioapic_printk(IOAPIC_VERBOSE, KERN_DEBUG 
| +				"... PIRQ%d -> IRQ %d\n", i, ints[i+1]);

I'm a little confused about having both a verbosity level and a
printk message level here.  Did you make sure that your uses of
them are consistent?  Can the KERN_ level be removed or automatically
inserted in ioapic_printk() ?
Or maybe it's OK as is...

|  		/*
|  		 * PIRQs are mapped upside down, usually.
|  		 */
| @@ -1123,10 +1157,12 @@
|  	if ((pin >= 16) && (pin <= 23)) {
|  		if (pirq_entries[pin-16] != -1) {
|  			if (!pirq_entries[pin-16]) {
| -				printk(KERN_DEBUG "disabling PIRQ%d\n", pin-16);
| +				ioapic_printk(IOAPIC_VERBOSE, KERN_DEBUG +						"disabling

Patch is mangled here....

| PIRQ%d\n", pin-16);
|  			} else {
|  				irq = pirq_entries[pin-16];
| -				printk(KERN_DEBUG "using PIRQ%d -> IRQ %d\n",
| +				ioapic_printk(IOAPIC_VERBOSE, KERN_DEBUG +						"using PIRQ%d ->
| IRQ %d\n",

and here...

|  						pin-16, irq);
|  			}
|  		}
| @@ -2246,13 +2303,14 @@
|  	enable_8259A_irq(0);
|  
|  	if (timer_irq_works()) {
| -		printk(" works.\n");
| +		ioapic_printk(IOAPIC_VERBOSE, " works.\n");
|  		return;
|  	}
|  	apic_write_around(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_FIXED |
| vector);
| -	printk(" failed.\n");
| +	ioapic_printk(IOAPIC_VERBOSE, " failed.\n");

Print always ?

| -	printk(KERN_INFO "...trying to set up timer as ExtINT IRQ...");
| +	ioapic_printk(IOAPIC_VERBOSE, KERN_INFO 
| +			"...trying to set up timer as ExtINT IRQ...");
|  
|  	timer_ack = 0;
|  	init_8259A(0);
| @@ -2262,11 +2320,12 @@
|  	unlock_ExtINT_logic();
|  
|  	if (timer_irq_works()) {
| -		printk(" works.\n");
| +		ioapic_printk(IOAPIC_VERBOSE, " works.\n");
|  		return;
|  	}
| -	printk(" failed :(.\n");
| -	panic("IO-APIC + timer doesn't work! pester mingo@redhat.com");
| +	ioapic_printk(IOAPIC_VERBOSE, " failed :(.\n");

I'd prefer to see this print even if booted with ioapic=quiet.

| +	panic("IO-APIC + timer doesn't work! ensure you boot with "
| +		"ioapic=debug and pester mingo@redhat.com");
|  }
|  
|  /*


--
~Randy

[-- Attachment #2: Type: text/plain, Size: 167 bytes --]

_______________________________________________
Kernel-janitors mailing list
Kernel-janitors@lists.osdl.org
http://lists.osdl.org/mailman/listinfo/kernel-janitors

  parent reply	other threads:[~2004-07-07 21:03 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-07-06 20:27 [Kernel-janitors] [PATCH] IO-APIC debug message reduction at Mark Broadbent
2004-07-07  0:25 ` Arnaldo Carvalho de Melo
2004-07-07 20:29 ` Re: [Kernel-janitors] [PATCH] IO-APIC debug message reducti Mark Broadbent
2004-07-07 21:03 ` Randy.Dunlap [this message]
2004-07-07 21:11 ` Luiz Fernando N. Capitulino
2004-07-08 18:45 ` Mark Broadbent
2004-07-08 23:09 ` Jeff Garzik
2004-07-10  0:56 ` Mark Broadbent
2004-07-11 10:38 ` maximilian attems
2004-07-13 19:20 ` Mark Broadbent
2004-07-13 19:27 ` Jeff Garzik
2004-07-14  5:24 ` Randy.Dunlap
2004-07-14  8:04 ` Mark Broadbent

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=20040707140313.390abfa1.rddunlap@osdl.org \
    --to=rddunlap@osdl.org \
    --cc=kernel-janitors@vger.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.