On Thu, Aug 27, 2009 at 01:48:53PM +0200, Paolo Bonzini wrote: > On 08/20/2009 05:57 PM, Michael S. Tsirkin wrote: >> +static inline int eventfd_writeable(struct eventfd_ctx *ctx, u64 n) >> +{ >> + return ULLONG_MAX - n> ctx->count; >> +} >> + >> +static inline void eventfd_dowrite(struct eventfd_ctx *ctx, u64 ucnt) >> +{ >> + if (eventfd_writeable(ctx, ucnt)) >> + ucnt = ULLONG_MAX - ctx->count; >> + >> + ctx->count += ucnt; > > In any case, this usage of eventfd_writeable is wrong: the code was like > this: > > - if (ULLONG_MAX - ctx->count < n) > - n = (int) (ULLONG_MAX - ctx->count); > - ctx->count += n; > + eventfd_dowrite(ctx, n); > > and so it should be !eventfd_writable. (This smelled when I was editing > patch 2, so I went back and checked patch 1). > > Paolo Right, patch 2 actually fixed this back. The best thing is to open-code it actually, since we handle it in a way that only applies to a counter, anyway. Will post a fixed v1. -- MST -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html