On Sun, 01.02.15 14:28, Zbigniew Jędrzejewski-Szmek (zbys...@kemper.freedesktop.org) wrote:
> > if (!switch_root_init) { > - char sfd[16]; > + char sfd[DECIMAL_STR_MAX(int) + 1]; > > /* First try to spawn ourselves with the right > * path, and with full serialization. We do > @@ -1893,8 +1893,7 @@ finish: > assert(arg_serialization); > assert(fds); > > - snprintf(sfd, sizeof(sfd), "%i", > fileno(arg_serialization)); > - char_array_0(sfd); > + xsprintf(sfd, "%i", > - fileno(arg_serialization)); BTW, one minor comment on this: DECIMAL_STR_MAX() actually already includes space for the trailing NUL, the +1 for sizing this array are hence unnecesary. I like the xsprintf() concept! > diff --git a/src/timedate/timedatectl.c b/src/timedate/timedatectl.c > index 74f40b9..1f85536 100644 > --- a/src/timedate/timedatectl.c > +++ b/src/timedate/timedatectl.c > @@ -38,6 +38,8 @@ > #include "pager.h" > #include "time-dst.h" > > +#define xstrftime(buf, fmt, tm) assert_se(strftime(buf, sizeof(buf), fmt, > tm) > 0) > + Maybe move this to time-util.h, even though it currently has a single user only... > +typedef enum { > + CREATION_NORMAL, > + CREATION_EXISTING, > + CREATION_FORCE, > +} CreationMode; > + > +static const char* creation_verb(CreationMode mode) { > + switch(mode) { > + case CREATION_NORMAL: > + return "Created"; > + case CREATION_EXISTING: > + return "Found existing"; > + case CREATION_FORCE: > + return "Created replacement"; > + } > + assert_not_reached("Bad creation"); > +} Maybe a candidate for DEFINE_PRIVATE_STRING_TABLE_LOOKUP_TO_STRING()? I recently split up DEFINE_PRIVATE_STRING_TABLE_LOOKUP() so that one can also just male use of the to_string() part without the from_string() part... > --- a/units/systemd-backli...@.service.in > +++ b/units/systemd-backli...@.service.in > @@ -19,3 +19,4 @@ Type=oneshot > RemainAfterExit=yes > ExecStart=@rootlibexecdir@/systemd-backlight load %i > ExecStop=@rootlibexecdir@/systemd-backlight save %i > +TimeoutSec=2min Our usual default timeout is 90s so far (see def.h), we probably should stick to that here... Lennart -- Lennart Poettering, Red Hat _______________________________________________ systemd-devel mailing list systemd-devel@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/systemd-devel