On Fri, Oct 20, 2023 at 10:40:12AM -0700, Justin Stitt wrote: > On Thu, Oct 19, 2023 at 9:46 PM Christoph Hellwig <h...@lst.de> wrote: > > > > On Wed, Oct 18, 2023 at 11:01:54PM -0700, Kees Cook wrote: > > > Almost all of the remaining strncpy() usage is just string to string > > > copying, but the corner cases that are being spun out that aren't > > > strscpy() or strscpy_pad() are covered by strtomem(), kmemdup_nul(), > > > and memcpy(). Each of these are a clear improvement since they remove > > > the ambiguity of the intended behavior. Using seq_buf ends up being way > > > more overhead than is needed. > > > > I'm really not sure strscpy is much of an improvement. In this particular > > case in most other places we simply use a snprintf for nqns, which seems > > useful here to if we don't want the full buf. > > > > But switching to a completely undocumented helper like strscpy seems not > > useful at all.
I'm curious where you looked and didn't find documentation -- perhaps there is an improvement to be made to aim one to where the existing documentation lives? > > There's some docs at [1]. Perhaps there could be more? > > [1]: > https://elixir.bootlin.com/linux/v6.6-rc6/source/include/linux/fortify-string.h#L292 Right, And it's even valid kern-doc, which gets rendered in the kernel API docs, along with all the other string functions: https://docs.kernel.org/core-api/kernel-api.html#c.strscpy -- Kees Cook
