Maintain and observe a ZBUF_FLAG_IMMUTABLE flag on zero-copy BPF

buffer kernel descriptors, which is used to allow the buffer
currently in the BPF "store" position to be assigned to userspace
when it fills, even if userspace hasn't acknowledged the buffer
in the "hold" position yet.  To implement this, notify the buffer
model when a buffer becomes full, and check that the store buffer
is writable, not just for it being full, before trying to append
new packet data.  Shared memory buffers will be assigned to
userspace at most once per fill, be it in the store or in the
hold position.

This removes the restriction that at most one shared memory can
by owned by userspace, reducing the chances that userspace will
need to call select() after acknowledging one buffer in order to
wait for the next buffer when under high load.  This more fully
realizes the goal of zero system calls in order to process a
high-speed packet stream from BPF.

Update bpf.4 to reflect that both buffers may be owned by userspace
at once; caution against assuming this.

1 files changed, 8 insertions, 8 deletions

diff --git a/share/man/man4/bpf.4 b/share/man/man4/bpf.4
index 9116b2d..39e1399 100644
--- a/share/man/man4/bpf.4
+++ b/share/man/man4/bpf.4
@@ -259,14 +259,14 @@ may be used to sleep awaiting the availbility of a completed buffer.
 They will return a readable file descriptor when ownership of the next buffer
 is assigned to user space.
 .Pp
-In the current implementation, the kernel will assign ownership of at most
-one buffer at a time to the user process.
-The user processes must acknowledge the current buffer in order to be
-notified that the next buffer is ready for processing.
-Programs should not rely on this as an invariant, as it may change in future
-versions; in particular, they must maintain their own notion of which buffer
-is "next" so that if both buffers are owned by userspace, it can process them
-in the correct order.
+In the current implementation, the kernel may assign zero, one, or both
+buffers to the user process; however, an earlier implementation maintained
+the invariant that at most one buffer could be assigned to the user process
+at a time.
+In order to both ensure progress and high performance, user processes should
+acknowledge a completely processed buffer as quickly as possible, returning
+it for reuse, and not block waiting on a second buffer while holding another
+buffer.
 .Sh IOCTLS
 The
 .Xr ioctl 2