2012/11/29 Stefan Tauner <[email protected]>: > serialport_write could loop endlessly when used with a seemingly valid port > that does always return 0 on writes instead of an error. > Give up after about 125 ms i.e. 250 tries with a period of 500 us. > > Signed-off-by: Stefan Tauner <[email protected]>
Note that this introduces a minor execution slowdown when used with -o/--output. Acked-by: Idwer Vollering <[email protected]> > --- > serial.c | 14 +++++++++++--- > 1 file changed, 11 insertions(+), 3 deletions(-) > > diff --git a/serial.c b/serial.c > index 7e47dcc..794763e 100644 > --- a/serial.c > +++ b/serial.c > @@ -262,6 +262,7 @@ int serialport_write(unsigned char *buf, unsigned int > writecnt) > #else > ssize_t tmp = 0; > #endif > + unsigned int empty_writes = 250; /* results in a ca. 125ms timeout */ > > while (writecnt > 0) { > #ifdef _WIN32 > @@ -273,9 +274,16 @@ int serialport_write(unsigned char *buf, unsigned int > writecnt) > msg_perr("Serial port write error!\n"); > return 1; > } > - if (!tmp) > - msg_pdbg("Empty write\n"); > - writecnt -= tmp; > + if (!tmp) { > + msg_pdbg2("Empty write\n"); > + empty_writes--; > + programmer_delay(500); > + if (empty_writes == 0) { > + msg_perr("Serial port seems dead!\n"); > + return 1; > + } > + } > + writecnt -= tmp; > buf += tmp; > } > > -- > Kind regards, Stefan Tauner > > > _______________________________________________ > flashrom mailing list > [email protected] > http://www.flashrom.org/mailman/listinfo/flashrom _______________________________________________ flashrom mailing list [email protected] http://www.flashrom.org/mailman/listinfo/flashrom
