On Mon, Nov 07, 2016 at 04:14:41PM +0100, Simon Horman wrote: > From: Takatoshi Akiyama <takatoshi.akiyama.kj@xxxxxxxxxxxxxxxxxxxxxxxx> > > While spin is already locked, serial output request causes the deadlock, > because serial output process also tries to lock the spin. > This patch removes serial output with spin locked. > > Signed-off-by: Takatoshi Akiyama <takatoshi.akiyama.kj@xxxxxxxxxxxxxxxxxxxxxxxx> > Signed-off-by: Takeshi Kihara <takeshi.kihara.df@xxxxxxxxxxx> > Signed-off-by: Simon Horman <horms+renesas@xxxxxxxxxxxx> > --- > This is a patch from the Gen3 3.3.2 BSP. > Geert, please consider it for mainline. > > Two approaches are mixed here: simply removing log messages that would > cause a deadlock; and moving log messages to outside the lock. Some > consideration was given (by Simon) to a more consistent approach of > simply removing all log messages in question. However this was taken as > logging ouside of the lock seems consistent with other logging in > rx_timer_fn() and sci_dma_rx_complete(). Acked-by: Wolfram Sang <wsa+renesas@xxxxxxxxxxxxxxxxxxxx> Simon, could you use the 'wsa+renesas' email address in the future? That would make things a tad easier for me. Thanks!
Attachment:
signature.asc
Description: PGP signature
