[Toybox] [PATCH] route: support iface names > 9 char

2016-10-02 Thread David R. Hedges
83ac0c2 80 00 fe806512d4c76293bf99 0001 0002 01050003 rmnet_data0 toys/pending/route.c | 10 +++--- 1 file changed, 7 insertions(+), 3 deletions(-) >From 2992772f4362febd5ad223586000d0407ced684d Mon Sep 17 00:00:00 2001 From: "David R. Hedges"

Re: [Toybox] [PATCH] route: support iface names > 9 char

2016-10-03 Thread David R. Hedges
On 2016-10-03 14:34 (-0500), Rob Landley wrote: On 09/30/2016 04:54 PM, David R. Hedges wrote: Also, try to more gracefully handle unexpectedly long lines from /proc/net/ipv6_route in fscanf by matching (but not assigning) any trailing characters before the newline, otherwise they'r

Re: [Toybox] fun with vfork

2016-10-12 Thread David R. Hedges
For what it's worth, your example (and the resulting symptom) seem to fall squarely in this undefined territory from the vfork manpage / POSIX description: the behavior is undefined if the process created by vfork() [..] calls any other function before successfully calling [..] one of the exec(3