ok op@, stsp@

op@ pointed out a problem in my initial patch where I forgot to call unveil(2) with a path before unveil(NULL, NULL). ok op, jamsek

The listen process now communicates the client UID/GID to the parent, and the auth process verifies this on behalf of the parent. This allows us to remove the "unix" pledge promise from the parent, removing parent access to syscalls such as listen() and accept() in the AF_UNIX domain. ok tracey@ op@

The authentication timeout was accidentally overriden by the request timeout. Fix this and set both timeouts in the same place for clarity. ok op@

ok op@

ok op, jamsek

errno may not be set to something interesting so switch to fatalx, and simplify the error message (knowing the failed function, which is also wrong, doesn't buy much here.) ok jamsek

in no specific order: - define the "aaa..." string as variable to avoid long lines (same for the 255 "A") - drop the comment about the "OpenBSD cmp(1) offset extension": it's quite popular actually. - consistently use "cmp foo.expected foo" instead of "cmp foo foo.expected". It makes easier to copy-paste it if we want to diff(1) the files. - diff files when cmp(1) fails if feasible (i.e. when we're not cmp(1) at offsets) - use printf when the string to print contains "\n" - add some missing blank lines between functions

Irrespective of the active view, always echo to the bottom left of the parent view. ok op@

Ironically, the typo "blocks" log search when in hsplit because we block input on the wrong view! Pass the proper "v" view to nodelay() calls. ok op@

And rename req_wrong_commit.sh to request_bad.sh to reflect the broader scope of this test script.

ok tracey@

the latter is empty when serving from a directory-like path prefix with an httpd.conf(5) as per gotwebd(8). ok tracey@

Patch by Mikhail

ok tracey@

ok tracey@

gotwebd uses sprintf to update index_page_str and page_str if they were negative. While this specific usage is safe, let's drop it. Since the index_page_str and page_str fields are never read from, remove them. This saves two allocations per request too. ok tracey@

ok tracey@

ok tracey@

ok tracey@

In the future, gotd will fork+exec new processes for each client connection. Using unveil instead of chroot avoids having to start such processes as root. The -portable version could use chroot(2) where no equivalent to unveil(2) exists. A future component which starts new processes will be isolated as a separate process, which could run as root in the -portable version. ok op@