On 2019-09-05, Peter Zijlstra wrote: > On Thu, Sep 05, 2019 at 07:26:22PM +1000, Aleksa Sarai wrote: > > On 2019-09-05, Peter Zijlstra wrote: > > > On Thu, Sep 05, 2019 at 06:19:22AM +1000, Aleksa Sarai wrote: > > > > + > > > > + while (rest > 0) { > > > > + size_t bufsize = min(rest, sizeof(buffer)); > > > > + > > > > + if (__copy_from_user(buffer, addr, bufsize)) > > > > + return -EFAULT; > > > > + if (memchr_inv(buffer, 0, bufsize)) > > > > + return -E2BIG; > > > > + > > > > + addr += bufsize; > > > > + rest -= bufsize; > > > > + } > > > > > > The perf implementation uses get_user(); but if that is too slow, surely > > > we can do something with uaccess_try() here? > > > > Is there a non-x86-specific way to do that (unless I'm mistaken only x86 > > has uaccess_try() or the other *_try() wrappers)? The main "performance > > improvement" (if you can even call it that) is that we use memchr_inv() > > which finds non-matching characters more efficiently than just doing a > > loop. > > Oh, you're right, that's x86 only :/ Though, I just had an idea -- am I wrong to think that the following would work just as well (without the need for an intermediate buffer)? if (memchr_inv((const char __force *) src + size, 0, rest)) return -E2BIG; Or is this type of thing very much frowned upon? What if it was a separate memchr_inv_user() instead -- I feel as though there's not a strong argument for needing to use a buffer when we're single-passing the __user buffer and doing a basic boolean check. -- Aleksa Sarai Senior Software Engineer (Containers) SUSE Linux GmbH