2005-03-03 09:57:36

by Jaka Močnik

Subject: [PATCH] trivial fix for 2.6.11, initializing a few spin locks

this patch for 2.6.11 simply initializes a few spin locks that are being
reported as accessed prior to initalization on an embedded ppc system.

--- cut here ---
--- linux-2.6.11/drivers/serial/cpm_uart/cpm_uart_core.c 2004-12-24 22:35:27.000000000 +0100
+++ linux-2.6.11-sgn/drivers/serial/cpm_uart/cpm_uart_core.c 2005-03-03 10:08:02.000000000 +0100
@@ -864,11 +864,12 @@
.irq = SMC1_IRQ,
.ops = &cpm_uart_pops,
.iotype = SERIAL_IO_MEM,
+ .lock = SPIN_LOCK_UNLOCKED,
},
.flags = FLAG_SMC,
.tx_nrfifos = TX_NUM_FIFO,
.tx_fifosize = TX_BUF_SIZE,
- .rx_nrfifos = RX_NUM_FIFO,
+ .rx_nrfifos = RX_NUM_FIFO,
.rx_fifosize = RX_BUF_SIZE,
.set_lineif = smc1_lineif,
},
@@ -877,11 +878,12 @@
.irq = SMC2_IRQ,
.ops = &cpm_uart_pops,
.iotype = SERIAL_IO_MEM,
+ .lock = SPIN_LOCK_UNLOCKED,
},
.flags = FLAG_SMC,
.tx_nrfifos = TX_NUM_FIFO,
.tx_fifosize = TX_BUF_SIZE,
- .rx_nrfifos = RX_NUM_FIFO,
+ .rx_nrfifos = RX_NUM_FIFO,
.rx_fifosize = RX_BUF_SIZE,
.set_lineif = smc2_lineif,
#ifdef CONFIG_SERIAL_CPM_ALT_SMC2
@@ -893,10 +895,11 @@
.irq = SCC1_IRQ,
.ops = &cpm_uart_pops,
.iotype = SERIAL_IO_MEM,
+ .lock = SPIN_LOCK_UNLOCKED,
},
.tx_nrfifos = TX_NUM_FIFO,
.tx_fifosize = TX_BUF_SIZE,
- .rx_nrfifos = RX_NUM_FIFO,
+ .rx_nrfifos = RX_NUM_FIFO,
.rx_fifosize = RX_BUF_SIZE,
.set_lineif = scc1_lineif,
},
@@ -905,10 +908,11 @@
.irq = SCC2_IRQ,
.ops = &cpm_uart_pops,
.iotype = SERIAL_IO_MEM,
+ .lock = SPIN_LOCK_UNLOCKED,
},
.tx_nrfifos = TX_NUM_FIFO,
.tx_fifosize = TX_BUF_SIZE,
- .rx_nrfifos = RX_NUM_FIFO,
+ .rx_nrfifos = RX_NUM_FIFO,
.rx_fifosize = RX_BUF_SIZE,
.set_lineif = scc2_lineif,
},
@@ -917,10 +921,11 @@
.irq = SCC3_IRQ,
.ops = &cpm_uart_pops,
.iotype = SERIAL_IO_MEM,
+ .lock = SPIN_LOCK_UNLOCKED,
},
.tx_nrfifos = TX_NUM_FIFO,
.tx_fifosize = TX_BUF_SIZE,
- .rx_nrfifos = RX_NUM_FIFO,
+ .rx_nrfifos = RX_NUM_FIFO,
.rx_fifosize = RX_BUF_SIZE,
.set_lineif = scc3_lineif,
},
@@ -929,10 +934,11 @@
.irq = SCC4_IRQ,
.ops = &cpm_uart_pops,
.iotype = SERIAL_IO_MEM,
+ .lock = SPIN_LOCK_UNLOCKED,
},
.tx_nrfifos = TX_NUM_FIFO,
.tx_fifosize = TX_BUF_SIZE,
- .rx_nrfifos = RX_NUM_FIFO,
+ .rx_nrfifos = RX_NUM_FIFO,
.rx_fifosize = RX_BUF_SIZE,
.set_lineif = scc4_lineif,
},
--- linux-2.6.11/drivers/net/gianfar.c 2005-03-03 10:36:51.000000000 +0100
+++ linux-2.6.11-sgn/drivers/net/gianfar.c 2005-03-03 10:36:38.822996013 +0100
@@ -377,6 +377,8 @@
ADVERTISED_1000baseT_Full);
mii_info->autoneg = 1;

+ spin_lock_init(&mii_info->mdio_lock);
+
mii_info->mii_id = priv->einfo->phyid;

mii_info->dev = dev;
--- cut here ---

Signed-off-by: Jaka Močnik <[email protected]>

regards,
jaKa

--

w3: http://fish.homeunix.org/people/jaka
email: [email protected]

2005-03-03 12:01:00

by Alexey Dobriyan

Subject: Re: [PATCH] trivial fix for 2.6.11, initializing a few spin locks

On Thursday 03 March 2005 11:57, Jaka Močnik wrote:
> this patch for 2.6.11 simply initializes a few spin locks that are being
> reported as accessed prior to initalization on an embedded ppc system.

Please, split'em. Put Signed-off-by right after changelog comment and
_before_ the patch.

> --- cut here ---

This may screw scripts. "--- " sequence is magic.

Oh, and don't put 2.6.11 in Subject and changelog.

Alexey