Re: [000/136] 2.6.31.2-stable review



Greg KH <gregkh@xxxxxxx> writes:

On Thu, Oct 01, 2009 at 10:01:34PM -0700, Eric W. Biederman wrote:
Greg KH <gregkh@xxxxxxx> writes:

Note from the -stable maintainer on this release:

This release is big. Yeah, really big. There are a number of
areas that needed some rework in order to get things back to
working order. Like the tty layer. Hopefully everyone can now
use their usb to serial devices again without oopsing the
kernel

The ftdi_sio driver still crashes and burns.

What were you doing when it crashed? I tested the pl2303 device and it
worked fine for me here, I don't have a ftdi_sio one :(

The patch that floated on the list earlier does fix the worst of the
symptoms.

Eric

From: Johan Hovold <jhovold@xxxxxxxxx>
Subject: [PATCH] USB: ftdi_sio: Remove tty->low_latency.

Fixes tty_flip_buffer_push being called from hard interrupt context with
low_latency set.

Signed-off-by: Johan Hovold <jhovold@xxxxxxxxx>
---
drivers/usb/serial/ftdi_sio.c | 4 ----
1 files changed, 0 insertions(+), 4 deletions(-)

diff --git a/drivers/usb/serial/ftdi_sio.c b/drivers/usb/serial/ftdi_sio.c
index 8fec5d4..22b72e8 100644
--- a/drivers/usb/serial/ftdi_sio.c
+++ b/drivers/usb/serial/ftdi_sio.c
@@ -1228,7 +1228,6 @@ static int set_serial_info(struct tty_struct *tty,
(new_serial.flags & ASYNC_FLAGS));
priv->custom_divisor = new_serial.custom_divisor;

- tty->low_latency = (priv->flags & ASYNC_LOW_LATENCY) ? 1 : 0;
write_latency_timer(port);

check_and_exit:
@@ -1699,9 +1698,6 @@ static int ftdi_open(struct tty_struct *tty,
priv->rx_bytes = 0;
spin_unlock_irqrestore(&priv->rx_lock, flags);

- if (tty)
- tty->low_latency = (priv->flags & ASYNC_LOW_LATENCY) ? 1 : 0;
-
write_latency_timer(port);

/* No error checking for this (will get errors later anyway) */
--
1.6.2.5
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/



Relevant Pages

  • [GIT PATCH] USB bugfixes for 2.6.12-rc3
    ... Here are a number of USB bugfixes for 2.6.12-rc3. ... tree 72abe7e0dd4b4135ca8b2d6f4e7cb81bd173a437 ... parent f3fae6ed6aafe71826e03876ad3d4e1d3f238ec8 ... This patch recognizes that pseudo-class by the GUIDs in those descriptors, ...
    (Linux-Kernel)
  • [PATCH] USB: Fix USB suspend/resume crasher (#2)
    ... USB: ... This patch closes the IRQ race and makes various other OHCI & EHCI code ... path safer vs. suspend/resume. ...
    (Linux-Kernel)
  • [PATCH RFC] maxSize option for usb-serial to increase max endpoint buffer size
    ... I don't know if this patch is 1) really necessary, ... Add a module option to increase the USB endpoint buffer size. ... The affected modems are PCMCIA cards that use a USB host controller ... the maximum packet size on the inbound bulk endpoint. ...
    (Linux-Kernel)
  • Re: Update: "nobody cared" on Toshiba Satellite A100
    ... the problematic IRQ has switched from 18 ... that they all share the USB handler. ... I have compiled 2.6.27.4 with the spurious interrupt patch you provided, ...
    (Linux-Kernel)
  • Re: 2.6.11, USB: High latency?
    ... > The issue with IN transfers is that microframe scheduling is ... ... > the relevant EHCI data structures are almost as irregular as the USB trees ... > enough to repost them against current BK (they include the patch above). ... send the line "unsubscribe linux-kernel" in ...
    (Linux-Kernel)