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 <alex.williamson@xxxxxx>
|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 <alex.williamson@xxxxxx>
| Cc: "H. Peter Anvin" <hpa@xxxxxxxxx>
| Cc: Russell King <rmk@xxxxxxxxxxxxxxxx>
| Acked-by: Aristeu Sergio Rozanski Filho <aris@xxxxxxxxxxxxxxxxx>
| Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
| Signed-off-by: Linus Torvalds <torvalds@xxxxxxxxxxxxxxxxxxxx>
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))
serial8250_handle_port(up);
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.
Thanks,
David Daney
