"The Cutting-Off Way: Indigenous Warfare in Eastern North America, 1500-1800" by Wayne Lee is a good read that describes a similar kind of warfare. It doesn't necessarily prove (nor does it set out to prove) that all non-industrial warfare followed this model but it does provide decent evidence for population-limited societies generally preferring deadly ambush to pitched battle.
When running away is a big part of the tactic, it's not a "kill everyone" tactic. You don't have the advantage for long. And even if you did succeed, the next tribe over could just as easily do the same to you (and they probably would, seeing what you're like), so there's a strong reason to develop norms, rules of engagement limiting this sort of thing.
If you're talking cattle, you're talking a different time of history entirely. That allowed collecting and concentrating enough wealth that some could engage in that sort of thing without worrying about immediately falling victim to it themselves. In the paleolithic, that wasn't possible.
I think this makes the common mistake that all "primitive" cultures have the exact same behavior -- I suppose you're not speaking from PNG right? (although likewise I have no such specific knowledge) As the article cites, in this specific place (and in many other similar cultures), I don't think there's motive to wage large scale wars or take large amounts of territory, that would justify something like raiding and wiping out other groups. I suppose that could happen in the huge number of other cultures, ones that had more centralized leadership or competing more unified cultures. The conflicts in the case of PNG might have been localized dispute over borders and such.
In nonhuman animals, even in fierce disputes over territory and mating, killing is rare, because it's quite costly: you eliminate completely a healthy individual, and usually would run the risk of serious injury yourself, making both dying likely. Usually they e.g. resolve mating with displays of beauty/fitness, and territory with displays of force (chimpanzees for example have lots of display of force rituals, specially around selecting leadership, I believe) and sometimes non-lethal conflict (some species evolve apparatus just for this kind of conflict, e.g. antlers, so they can engage in conflict with low chance of injury).
Since we're referring to human cultures, something distinct we have is memory and communication -- information can get out on reputation, behavior, etc. of a group. I think if a group just goes out and murders another in their sleep (which TBF is not so easy), then I imagine for example word would also go out of their actions and may get into trouble with other groups, I'm sure spiritual traditions would also have a say about it, in favor of some sort of "spiritual revenge". But I suppose it's difficult to generalize.
I do think it's interesting to consider some forms of morality, or just metaphysical and logical understanding around ethics, as a kind of technology that enables people to live longer and enable more secure conditions, including preventing large scale conflict that can bring destruction on an industrial scale to entire societies (although misguided moral arguments ironically are also misused sometimes to enact destruction). Although I guess morality alone is not enough (because not everyone seems to internalize it well), and other mechanisms are also helpful to achieve more social stability.
(Also of note is that there can be such thing as too much stability -- for example, for some new more interesting and maybe more worthwhile forms of culture to emerge, old ones have to be replaced, so stability should be balanced with cautious openness to new ideas, technologies, methods, peoples (ethnicity), cultures, ..., and worldviews -- of course, as long as change is in the direction of improving our lives collectively. For example in some cases religious fundamentalism advocates for an absolutist cultural stagnation; we just shouldn't need one group to murder another to spread about new things)
Different cultures of course have different norms. There were even hunter-gatherer societies that practiced extreme forms of pacifism - Moriori would be one example (and their eventual fate shows why this isn't more common).
In general, though, pre-agricultural societies tend to have very high mortality rates from armed conflict with outgroups. In PNG in particular, as I recall, it gets to the point where >20% of all males die from enemy action. Similarly, pre-agricultural societies rarely take prisoners, and often target what we would consider non-combatants (i.e. there's no separation between "enemy warrior" and "enemy civilian", even in principle).
And this makes a lot of sense when you consider what motivates this warfare. Pre-agricultural societies are bounded by the output of the wildland that they forage on, and a single human requires a lot of land to live off in this manner. This is why those bands tend to be small - very few biomes can sustain a foraging band of even few hundred. But conversely it means that any other band of humans foraging off the same land is literally starving you out. It doesn't matter if they are otherwise hostile or not, and there's no potential compromise to be achieved even in theory if you have a plot of land that can sustain, say, 100 people, but now it has 150 feeding off it. The only way for one group to survive is to either kill or drive off the entirety of the remaining group, so it doesn't matter if it's women or children that you're killing. The point is to convince the other group that they are more likely to survive by going elsewhere by making it so precarious for them to remain, or if that is not possible, then to wipe them all entirely. Raids do exactly that.
This is also why prisoners are rarely taken and often ritually killed right away - every prisoner you take is an extra mouth to feed. In an agricultural society, an extra mouth can feed itself because it requires comparatively little land to do so, but for pre-agriculturalists an extra pair of working hands is worthless if the land is already at capacity wrt what can be extracted from it (which is usually the case since populations grow until they reach that point).
Stuff like ethics of warfare, spiritual tradition, reputation etc is derived from all this. Cultures that practice this kind of raid warfare generally glorify it. A war leader who leads a successful raid to massacre an unprotected enemy village is a hero, not just to his own people, but more broadly as a model for other aspiring warriors in the larger culture. So things that we might consider as "disreputable" are not necessarily seen as such there.
