kernel/linux-2.6-silence-noise.patch
Kyle McMartin 472d463edc update to 3.0-rc4-git3
drop linux-3.0-fix-uts-release.patch and just perl the Makefile
instead

linux-2.6-silence-noise.patch: fix context
2011-06-23 15:27:07 -04:00

74 lines
2.5 KiB
Diff

From b4e96f34c17e5a79cd28774cc722bb33e7e02c6e Mon Sep 17 00:00:00 2001
From: Peter Jones <pjones@redhat.com>
Date: Thu, 25 Sep 2008 16:23:33 -0400
Subject: [PATCH] Don't print an error message just because there's no i8042 chip.
Some systems, such as EFI-based Apple systems, won't necessarily have an
i8042 to initialize. We shouldn't be printing an error message in this
case, since not detecting the chip is the correct behavior.
---
drivers/input/serio/i8042.c | 4 +---
1 files changed, 1 insertions(+), 3 deletions(-)
diff --git a/drivers/input/serio/i8042.c b/drivers/input/serio/i8042.c
index 170f71e..4f3e632 100644
--- a/drivers/input/serio/i8042.c
+++ b/drivers/input/serio/i8042.c
@@ -701,10 +701,8 @@ static int __devinit i8042_check_aux(void)
static int i8042_controller_check(void)
{
- if (i8042_flush() == I8042_BUFFER_SIZE) {
- pr_err("No controller found\n");
+ if (i8042_flush() == I8042_BUFFER_SIZE)
return -ENODEV;
- }
return 0;
}
--
1.6.0.1
Socket fuzzers like sfuzz will trigger this printk a lot, even though it's
ratelimited. It isn't particularly useful, so just remove it.
Signed-off-by: Dave Jones <davej@redhat.com>
--- a/net/can/af_can.c
+++ b/net/can/af_can.c
@@ -157,13 +157,9 @@ static int can_create(struct net *net, struct socket *sock, int protocol,
err = request_module("can-proto-%d", protocol);
/*
- * In case of error we only print a message but don't
- * return the error code immediately. Below we will
- * return -EPROTONOSUPPORT
+ * In case of error we but don't return the error code immediately.
+ * Below we will return -EPROTONOSUPPORT
*/
- if (err && printk_ratelimit())
- printk(KERN_ERR "can: request_module "
- "(can-proto-%d) failed.\n", protocol);
cp = can_get_proto(protocol);
}
This was removed in revision 1.6 of linux-2.6-silence-noise.patch
in ye olde CVS tree. I have no idea why. Originally the pr_debug in
device_pm_remove was nuked as well, but that seems to have gotten lost in
the r1.634 of kernel.spec (2.6.26-rc2-git5.)
diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c
index 2a52270..bacbdd2 100644
--- a/drivers/base/power/main.c
+++ b/drivers/base/power/main.c
@@ -87,8 +87,6 @@ void device_pm_unlock(void)
*/
void device_pm_add(struct device *dev)
{
- pr_debug("PM: Adding info for %s:%s\n",
- dev->bus ? dev->bus->name : "No Bus", dev_name(dev));
mutex_lock(&dpm_list_mtx);
if (dev->parent && dev->parent->power.is_prepared)
dev_warn(dev, "parent %s should not be sleeping\n",