Skip to contentSkip to conversation

mm: don't use ZONE_DMA unless CONFIG_ZONE_DMA is set in setup.c

PATCH In review Andres Salomon 4 participants 6 messages Mar 6, 2007 – Mar 7, 2007 Memory management

lore
Share & tools
lore

Andres Salomon

author
#0 lore raw

PATCH 1/1 mm: don't use ZONE_DMA unless CONFIG_ZONE_DMA is set in setup.c

If CONFIG_ZONE_DMA is ever undefined, ZONE_DMA will also not be defined, and setup.c won't compile. This wraps it with an #ifdef.

Signed-off-by Andres Salomon <[email protected]>
arch/i386/kernel/setup.c +2 -0
index 0b476e1..b69626e 100644
@@ -371,8 +371,10 @@ void __init zone_sizes_init(void)
 {
 	unsigned long max_zone_pfns[MAX_NR_ZONES];
 	memset(max_zone_pfns, 0, sizeof(max_zone_pfns));
+#ifdef CONFIG_ZONE_DMA
 	max_zone_pfns[ZONE_DMA] =
 		virt_to_phys((char *)MAX_DMA_ADDRESS) >> PAGE_SHIFT;
+#endif
 	max_zone_pfns[ZONE_NORMAL] = max_low_pfn;
 #ifdef CONFIG_HIGHMEM
 	max_zone_pfns[ZONE_HIGHMEM] = highend_pfn;
maintainer
#1 lore raw
Quoted 3 lines · On Tue, 06 Mar 2007 18:52:59 -0500 Andres Salomon <[email protected]> wrote:

If CONFIG_ZONE_DMA is ever undefined, ZONE_DMA will also not be defined, and setup.c won't compile. This wraps it with an #ifdef.

I guess if anyone tries to disable ZONE_DMA on i386 they'll pretty quickly discover that. But I don't think we need to "fix" it yet?

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

#2 lore raw
Quoted 9 lines · Andrew Morton
Quoted 3 lines · On Tue, 06 Mar 2007 18:52:59 -0500 Andres Salomon <[email protected]> wrote:

If CONFIG_ZONE_DMA is ever undefined, ZONE_DMA will also not be defined, and setup.c won't compile. This wraps it with an #ifdef.

I guess if anyone tries to disable ZONE_DMA on i386 they'll pretty quickly discover that. But I don't think we need to "fix" it yet?

CONFIG_ZONE_DMA isn't even optional on i386, so I'm curious how you could hit this compile failure.

Dave

Signature · 7 lines
http://www.codemonkey.org.uk

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to [email protected].  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>
author
#3 lore raw
Quoted 10 lines · Dave Jones
Quoted 9 lines · Andrew Morton
Quoted 3 lines · On Tue, 06 Mar 2007 18:52:59 -0500 Andres Salomon <[email protected]> wrote:

If CONFIG_ZONE_DMA is ever undefined, ZONE_DMA will also not be defined, and setup.c won't compile. This wraps it with an #ifdef.

I guess if anyone tries to disable ZONE_DMA on i386 they'll pretty quickly discover that. But I don't think we need to "fix" it yet?

Oh, it's certainly not urgent. I sent it simply for correctness reasons.

It would've been nice to see the ZONE_DMA removal patches just #define ZONE_DMA regardless, and include less #ifdefs scattered about; but at this point, I'd just as soon prefer to see a proper way to allocate things based on address constraints (as discussed in
http://www.gelato.unsw.edu.au/archives/linux-ia64/0609/19036.html).

Quoted 4 lines

CONFIG_ZONE_DMA isn't even optional on i386, so I'm curious how you could hit this compile failure.

Why, with custom code of course ;)

#4 lore raw
Quoted 5 lines · Andres Salomon

It would've been nice to see the ZONE_DMA removal patches just #define ZONE_DMA regardless, and include less #ifdefs scattered about; but at this point, I'd just as soon prefer to see a proper way to allocate things based on address constraints (as discussed in
http://www.gelato.unsw.edu.au/archives/linux-ia64/0609/19036.html).

Would you be willing to work on that? I can sent you a bunch of unfinished patches if you have the time.

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to [email protected]. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"[email protected]"> [email protected] </a>

author
#5 lore raw
Quoted 10 lines · Christoph Lameter
Quoted 5 lines · Andres Salomon

It would've been nice to see the ZONE_DMA removal patches just #define ZONE_DMA regardless, and include less #ifdefs scattered about; but at this point, I'd just as soon prefer to see a proper way to allocate things based on address constraints (as discussed in
http://www.gelato.unsw.edu.au/archives/linux-ia64/0609/19036.html).

Would you be willing to work on that? I can sent you a bunch of unfinished patches if you have the time.

Sure, I'd be willing to work on it. Whether or not I have the time is questionable, but it wouldn't hurt to send them..