From: Dave Jiang <djiang@mvista.com>
To: linux-kernel@vger.kernel.org
Cc: mgreer@mvista.com, akpm@osdl.org, djiang@mvista.com
Subject: [PATCH] MPSC serial driver tx locking
Date: Tue, 6 Mar 2007 16:24:48 -0700 [thread overview]
Message-ID: <20070306232448.GA2736@blade.az.mvista.com> (raw)
The MPSC serial driver assumes that interrupt is always on to pick up
the DMA transmit ops that aren't submitted while the DMA engine is active.
However when irqs are off for a period of time such as operations under kernel
crash dump console messages do not show up due to additional DMA ops are being
dropped. This makes console writes to process through all the tx DMAs
queued up before submitting a new request.
Also, the current locking mechanism does not protect the hardware
registers and ring buffer when a printk is done during the serial write
operations. The additional per port transmit lock provides a finer granular
locking and protects registers being clobbered while printks are nested within
UART writes.
Signed-off-by: Dave Jiang <djiang@mvista.com>
Signed-off-by: Mark A. Greer <mgreer@mvista.com>
---
drivers/serial/mpsc.c | 25 ++++++++++++++++++++++++-
1 files changed, 24 insertions(+), 1 deletions(-)
---
diff --git a/drivers/serial/mpsc.c b/drivers/serial/mpsc.c
index 3d2fcc5..d09f209 100644
--- a/drivers/serial/mpsc.c
+++ b/drivers/serial/mpsc.c
@@ -183,6 +183,7 @@ struct mpsc_port_info {
u8 *txb_p; /* Phys addr of txb */
int txr_head; /* Where new data goes */
int txr_tail; /* Where sent data comes off */
+ spinlock_t tx_lock; /* transmit lock */
/* Mirrored values of regs we can't read (if 'mirror_regs' set) */
u32 MPSC_MPCR_m;
@@ -1212,6 +1213,9 @@ mpsc_tx_intr(struct mpsc_port_info *pi)
{
struct mpsc_tx_desc *txre;
int rc = 0;
+ unsigned long iflags;
+
+ spin_lock_irqsave(&pi->tx_lock, iflags);
if (!mpsc_sdma_tx_active(pi)) {
txre = (struct mpsc_tx_desc *)(pi->txr +
@@ -1248,6 +1252,7 @@ mpsc_tx_intr(struct mpsc_port_info *pi)
mpsc_sdma_start_tx(pi); /* start next desc if ready */
}
+ spin_unlock_irqrestore(&pi->tx_lock, iflags);
return rc;
}
@@ -1338,11 +1343,16 @@ static void
mpsc_start_tx(struct uart_port *port)
{
struct mpsc_port_info *pi = (struct mpsc_port_info *)port;
+ unsigned long iflags;
+
+ spin_lock_irqsave(&pi->tx_lock, iflags);
mpsc_unfreeze(pi);
mpsc_copy_tx_data(pi);
mpsc_sdma_start_tx(pi);
+ spin_unlock_irqrestore(&pi->tx_lock, iflags);
+
pr_debug("mpsc_start_tx[%d]\n", port->line);
return;
}
@@ -1625,6 +1635,16 @@ mpsc_console_write(struct console *co, const char *s, uint count)
struct mpsc_port_info *pi = &mpsc_ports[co->index];
u8 *bp, *dp, add_cr = 0;
int i;
+ unsigned long iflags;
+
+ spin_lock_irqsave(&pi->tx_lock, iflags);
+
+ while (pi->txr_head != pi->txr_tail) {
+ while (mpsc_sdma_tx_active(pi))
+ udelay(100);
+ mpsc_sdma_intr_ack(pi);
+ mpsc_tx_intr(pi);
+ }
while (mpsc_sdma_tx_active(pi))
udelay(100);
@@ -1668,6 +1688,7 @@ mpsc_console_write(struct console *co, const char *s, uint count)
pi->txr_tail = (pi->txr_tail + 1) & (MPSC_TXR_ENTRIES - 1);
}
+ spin_unlock_irqrestore(&pi->tx_lock, iflags);
return;
}
@@ -2005,7 +2026,8 @@ mpsc_drv_probe(struct platform_device *dev)
if (!(rc = mpsc_drv_map_regs(pi, dev))) {
mpsc_drv_get_platform_data(pi, dev, dev->id);
- if (!(rc = mpsc_make_ready(pi)))
+ if (!(rc = mpsc_make_ready(pi))) {
+ spin_lock_init(&pi->tx_lock);
if (!(rc = uart_add_one_port(&mpsc_reg,
&pi->port)))
rc = 0;
@@ -2014,6 +2036,7 @@ mpsc_drv_probe(struct platform_device *dev)
(struct uart_port *)pi);
mpsc_drv_unmap_regs(pi);
}
+ }
else
mpsc_drv_unmap_regs(pi);
}
next reply other threads:[~2007-03-06 23:24 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-03-06 23:24 Dave Jiang [this message]
-- strict thread matches above, loose matches on Subject: below --
2006-05-10 0:35 Dave Jiang
2006-05-10 16:27 ` Russell King
2006-05-10 18:08 ` Dave Jiang
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=20070306232448.GA2736@blade.az.mvista.com \
--to=djiang@mvista.com \
--cc=akpm@osdl.org \
--cc=linux-kernel@vger.kernel.org \
--cc=mgreer@mvista.com \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®