[Top] [All Lists]

Re: [PATCH 5/5] Serial: UART driver changes for Cavium OCTEON.

To: Shinya Kuribayashi <>
Subject: Re: [PATCH 5/5] Serial: UART driver changes for Cavium OCTEON.
From: David Daney <>
Date: Mon, 01 Dec 2008 10:43:11 -0800
Cc:,,,, Tomaso Paoletti <>
In-reply-to: <>
Original-recipient: rfc822;
References: <> <> <>
User-agent: Thunderbird (X11/20080723)
Shinya Kuribayashi wrote:
Hi David,

David Daney wrote:
Cavium UART implementation won't work with the standard 8250 driver
as-is.  Define a new uart_config (PORT_OCTEON) and use it to enable
special handling required by the OCTEON's serial port.  Two new bug
types are defined.

You just added one new bug type, not two. ;-)

Unfortunately I overlooked the patch comment when I moved the other bug workaround to platform code.

However after your suggestion below...
I see your point.  Then let's look at this commit:

|commit 40b36daad0ac704e6d5c1b75789f371ef5b053c1
|Author: Alex Williamson <>
|Date:   Wed Feb 14 00:33:04 2007 -0800
|    [PATCH] 8250 UART backup timer
|    The patch below works around a minor bug found in the UART of the remote
|    management card used in many HP ia64 and parisc servers (aka the Diva
|    UARTs).  The problem is that the UART does not reassert the THRE interrupt
|    if it has been previously cleared and the IIR THRI bit is re-enabled.  This
|    can produce a very annoying failure mode when used as a serial console,
|    allowing a boot/reboot to hang indefinitely until an RX interrupt kicks it
|    into working again (ie.  an unattended reboot could stall).
|    To solve this problem, a backup timer is introduced that runs alongside the
|    standard interrupt driven mechanism.  This timer wakes up periodically,
|    checks for a hang condition and gets characters moving again.  This backup
|    mechanism is only enabled if the UART is detected as having this problem,
|    so systems without these UARTs will have no additional overhead.
|    This version of the patch incorporates previous comments from Pavel and
|    removes races in the bug detection code.  The test is now done before the
|    irq linking to prevent races with interrupt handler clearing the THRE
|    interrupt.  Short delays and syncs are also added to ensure the device is
|    able to update register state before the result is tested.
|    Aristeu says:
|      this was tested on the following HP machines and solved the problem:
|      rx2600, rx2620, rx1600 and rx1620s.
|    hpa says:
|      I have seen this same bug in soft UART IP from "a major vendor."
|    Signed-off-by: Alex Williamson <>
|    Cc: "H. Peter Anvin" <>
|    Cc: Russell King <>
|    Acked-by: Aristeu Sergio Rozanski Filho <>
|    Signed-off-by: Andrew Morton <>
|    Signed-off-by: Linus Torvalds <>

AFAICS this patch tries to handle the same problem, and added reasonable
framework with serial8250_backup_timeout, etc.  And now we have UART_BUG
_THRE for this issue.  Then it might be better to use existing codes as
much as possible.

Just random thoughts,

- could we dynamically probe this bug in serial8250_startup as well?

- is there any possibility this hardware works with an existing
  serial8250_backup_timeout, not doing this below:

@@ -1714,7 +1722,7 @@ static void serial8250_timeout(unsigned long data)
        unsigned int iir;
iir = serial_in(up, UART_IIR);
-       if (!(iir & UART_IIR_NO_INT))
+       if (!(iir & UART_IIR_NO_INT) || (up->bugs & UART_BUG_TIMEOUT))
        mod_timer(&up->timer, jiffies + poll_timeout(up->port.timeout));

I'm not an UART expert, any feedback is highly appreciated.

... All our boards are now working with no extra bug flags needed. So I will simplify this patch set by getting rid of UART_BUG_TIMEOUT.


David Daney

<Prev in Thread] Current Thread [Next in Thread>