Hi Joe, On Tue, Aug 25, 2015 at 3:22 PM, Bin Meng <bmeng...@gmail.com> wrote: > Testing either pch_gbe or e1000 driver via tftp command on Intel > Crown Bay board, shows the following failure. > > TFTP error: 'Unsupported option(s) requested' (8) > > It turns out commit 620776d causes this. Revert this commit for now.
Please check http://lists.denx.de/pipermail/u-boot/2015-August/225187.html on why this commit should be reverted. Let me know if you have different thoughts (eg: I can respin a v2 to explicitly mention in the commit message that commit 620776d is a spec violation to RTC 2349 thus we need revert it) > > Signed-off-by: Bin Meng <bmeng...@gmail.com> > --- > > net/tftp.c | 10 +++++----- > 1 file changed, 5 insertions(+), 5 deletions(-) > > diff --git a/net/tftp.c b/net/tftp.c > index 18ce84c..89be32a 100644 > --- a/net/tftp.c > +++ b/net/tftp.c > @@ -19,10 +19,10 @@ > /* Well known TFTP port # */ > #define WELL_KNOWN_PORT 69 > /* Millisecs to timeout for lost pkt */ > -#define TIMEOUT 100UL > +#define TIMEOUT 5000UL > #ifndef CONFIG_NET_RETRY_COUNT > /* # of timeouts before giving up */ > -# define TIMEOUT_COUNT 1000 > +# define TIMEOUT_COUNT 10 > #else > # define TIMEOUT_COUNT (CONFIG_NET_RETRY_COUNT * 2) > #endif > @@ -711,10 +711,10 @@ void tftp_start(enum proto_t protocol) > if (ep != NULL) > timeout_ms = simple_strtol(ep, NULL, 10); > > - if (timeout_ms < 10) { > - printf("TFTP timeout (%ld ms) too low, set min = 10 ms\n", > + if (timeout_ms < 1000) { > + printf("TFTP timeout (%ld ms) too low, set min = 1000 ms\n", > timeout_ms); > - timeout_ms = 10; > + timeout_ms = 1000; > } > > debug("TFTP blocksize = %i, timeout = %ld ms\n", > -- Regards, Bin _______________________________________________ U-Boot mailing list U-Boot@lists.denx.de http://lists.denx.de/mailman/listinfo/u-boot
