kernel/linux.git/drivers/usb/serial/generic.c, branch v4.4.235

USB: serial: fix unthrottle races

2019-05-16T17:45:16+00:00

[ Upstream commit 3f5edd58d040bfa4b74fb89bc02f0bc6b9cd06ab ] Fix two long-standing bugs which could potentially lead to memory corruption or leave the port throttled until it is reopened (on weakly ordered systems), respectively, when read-URB completion races with unthrottle(). First, the URB must not be marked as free before processing is complete to prevent it from being submitted by unthrottle() on another CPU. CPU 1 CPU 2 ================ ================ complete() unthrottle() process_urb(); smp_mb__before_atomic(); set_bit(i, free); if (test_and_clear_bit(i, free)) submit_urb(); Second, the URB must be marked as free before checking the throttled flag to prevent unthrottle() on another CPU from failing to observe that the URB needs to be submitted if complete() sees that the throttled flag is set. CPU 1 CPU 2 ================ ================ complete() unthrottle() set_bit(i, free); throttled = 0; smp_mb__after_atomic(); smp_mb(); if (throttled) if (test_and_clear_bit(i, free)) return; submit_urb(); Note that test_and_clear_bit() only implies barriers when the test is successful. To handle the case where the URB is still in use an explicit barrier needs to be added to unthrottle() for the second race condition. Fixes: d83b405383c9 ("USB: serial: add support for multiple read urbs") Signed-off-by: Johan Hovold Signed-off-by: Sasha Levin

USB: serial: use variable for status

2019-05-16T17:45:16+00:00

[ Upstream commit 3161da970d38cd6ed2ba8cadec93874d1d06e11e ] This patch turns status in a variable read once from the URB. The long term plan is to deliver status to the callback. In addition it makes the code a bit more elegant. Signed-off-by: Oliver Neukum Signed-off-by: Johan Hovold Signed-off-by: Sasha Levin

USB: serial: fix infinite wait_until_sent timeout

2015-03-07T02:44:14+00:00

Make sure to handle an infinite timeout (0). Note that wait_until_sent is currently never called with a 0-timeout argument due to a bug in tty_wait_until_sent. Fixes: dcf010503966 ("USB: serial: add generic wait_until_sent implementation") Cc: stable # v3.10 Signed-off-by: Johan Hovold Signed-off-by: Greg Kroah-Hartman

usb: serial: handle -ENODEV quietly in generic_submit_read_urb

2015-01-12T09:23:54+00:00

If a USB serial device (e.g. /dev/ttyUSB0) with an active program is unplugged, an -ENODEV (19) error will be produced after it gives up trying to resubmit a read. usb_serial_generic_submit_read_urb - usb_submit_urb failed: -19 Add -ENODEV as one of the permanent errors along with -EPERM that usb_serial_generic_submit_read_urb() handles quietly without an error. Signed-off-by: Jeremiah Mahler Signed-off-by: Johan Hovold

usb: serial: silence all non-critical read errors

2015-01-12T09:23:35+00:00

If a USB serial device is unplugged while there is an active program using the device it may spam the logs with -EPROTO (71) messages as it attempts to retry. Most serial usb drivers (metro-usb, pl2303, mos7840, ...) only output these messages for debugging. The generic driver treats these as errors. Change the default output for the generic serial driver from error to debug to silence these non-critical errors. Signed-off-by: Jeremiah Mahler Signed-off-by: Johan Hovold

USB: serial: add missing braces

2014-03-12T19:44:50+00:00

Add missing braces to conditional branches and one loop in usb-serial core and generic implementation. Signed-off-by: Johan Hovold Signed-off-by: Greg Kroah-Hartman

USB: serial: continue to write on errors

2014-03-12T19:44:49+00:00

Do not discard buffered data and make sure to try to resubmit the write urbs on errors. Currently a recoverable error would lead to more data than necessary being dropped. Also upgrade error messages from debug to error log level. Signed-off-by: Johan Hovold Signed-off-by: Greg Kroah-Hartman

USB: serial: continue to read on errors

2014-03-12T19:44:49+00:00

Make sure to try to resubmit the read urb on errors. Currently a recoverable error would lead to reduced throughput as only one urb will be used until the port is closed and reopened (or resumed or unthrottled). Also upgrade error messages from debug to error log level. Signed-off-by: Johan Hovold Signed-off-by: Greg Kroah-Hartman

USB: serial: fix write memory-allocation flag

2013-11-25T16:39:40+00:00

Fix regression introduced by commit 818f60365a29 ("USB: serial: add memory flags to usb_serial_generic_write_start"), which incorrectly used GFP_KERNEL in write(), which must not not sleep. Reported-by: Dave Jones Tested-by: Dave Jones Cc: Dave Jones Signed-off-by: Johan Hovold Signed-off-by: Greg Kroah-Hartman

USB: serial: fix race in generic write

2013-11-25T16:39:40+00:00

Fix race in generic write implementation, which could lead to temporarily degraded throughput. The current generic write implementation introduced by commit 27c7acf22047 ("USB: serial: reimplement generic fifo-based writes") has always had this bug, although it's fairly hard to trigger and the consequences are not likely to be noticed. Specifically, a write() on one CPU while the completion handler is running on another could result in only one of the two write urbs being utilised to empty the remainder of the write fifo (unless there is a second write() that doesn't race during that time). Cc: stable # 2.6.35 Signed-off-by: Johan Hovold Signed-off-by: Greg Kroah-Hartman