Skip to contentSkip to conversation

pata_platform: fallout from set_mode() change

PATCH In review Al Viro 4 participants 5 messages Jan 30, 2007 libata

lore
Share & tools
lore

Al Viro

author
#0 lore raw

PATCH 1/1 pata_platform: fallout from set_mode() change

---
 drivers/ata/pata_platform.c |    3 ++-
 1 files changed, 2 insertions(+), 1 deletions(-)
drivers/ata/pata_platform.c +2 -1
index 443b1d8..40ae11c 100644
@@ -30,7 +30,7 @@ static int pio_mask = 1;
  * Provide our own set_mode() as we don't want to change anything that has
  * already been configured..
  */
-static void pata_platform_set_mode(struct ata_port *ap)
+static int pata_platform_set_mode(struct ata_port *ap, struct ata_device **unused)
 {
 	int i;
 
@@ -44,6 +44,7 @@ static void pata_platform_set_mode(struct ata_port *ap)
 			dev->flags |= ATA_DFLAG_PIO;
 		}
 	}
+	return 0;
 }
 
 static void pata_platform_host_stop(struct ata_host *host)
Signature · 1 line
1.5.0-rc2.GIT
#1 lore raw
Quoted 1 line · Al Viro

This is already in the queue via Andrew, who caught the same problem.

Sometimes it seems like you and Andrew are the only people to build on non-x86 these days...

Jeff

#2 lore raw
Quoted 8 lines · Jeff Garzik
Quoted 1 line · Al Viro

This is already in the queue via Andrew, who caught the same problem.

Sometimes it seems like you and Andrew are the only people to build on non-x86 these days...

This is the first I've heard of this or the changes that caused this, while I do build current git on a pretty much daily basis, -mm doesn't get nearly as much testing.

There are many people who build on non-x86 every day, there are rather less people who have the bandwidth for -mm as well as current git or for tracking down changes that haven't made it upstream to make sure everything plays nice.

If you want immediate action, I suggest you CC the authors of the drivers when there's changes they should be made aware of, rather than deferring to hand-waving generalizations about how no one does any testing after the fact.

#3 lore raw
Quoted 16 lines · Paul Mundt
Quoted 6 lines · Jeff Garzik
Quoted 1 line · Al Viro

This is already in the queue via Andrew, who caught the same problem.

Sometimes it seems like you and Andrew are the only people to build on non-x86 these days...

This is the first I've heard of this or the changes that caused this, while I do build current git on a pretty much daily basis, -mm doesn't get nearly as much testing.

There are many people who build on non-x86 every day, there are rather less people who have the bandwidth for -mm as well as current git or for tracking down changes that haven't made it upstream to make sure everything plays nice.

Quoted 4 lines

If you want immediate action, I suggest you CC the authors of the drivers when there's changes they should be made aware of, rather than deferring to hand-waving generalizations about how no one does any testing after the fact.

The change in question is in the current -git. But my comment was largely self-critical, since /I/ applied the patch in question that broke on non-x86.

Jeff