Revert of net: merge two versions of SetTCPNoDelay() function into one

net/socket/tcp_socket.h

Index: net/socket/tcp_socket.h
diff --git a/net/socket/tcp_socket.h b/net/socket/tcp_socket.h
index 319a6b261d02f78961b3d950648f230dd0733857..58797bb0851ca59cab11cd6ab7aa6c0d5c90dbaf 100644
--- a/net/socket/tcp_socket.h
+++ b/net/socket/tcp_socket.h
@@ -7,7 +7,6 @@
 
 #include "build/build_config.h"
 #include "net/base/net_export.h"
-#include "net/socket/socket_descriptor.h"
 
 #if defined(OS_WIN)
 #include "net/socket/tcp_socket_win.h"
@@ -40,39 +39,6 @@
 // Not thread safe.  Must be called during initialization/startup only.
 NET_EXPORT void CheckSupportAndMaybeEnableTCPFastOpen(bool user_enabled);
 
-// This function enables/disables buffering in the kernel. By default, on Linux,
-// TCP sockets will wait up to 200ms for more data to complete a packet before
-// transmitting. After calling this function, the kernel will not wait. See
-// TCP_NODELAY in `man 7 tcp`.
-//
-// For Windows:
-//
-// The Nagle implementation on Windows is governed by RFC 896.  The idea
-// behind Nagle is to reduce small packets on the network.  When Nagle is
-// enabled, if a partial packet has been sent, the TCP stack will disallow
-// further *partial* packets until an ACK has been received from the other
-// side.  Good applications should always strive to send as much data as
-// possible and avoid partial-packet sends.  However, in most real world
-// applications, there are edge cases where this does not happen, and two
-// partial packets may be sent back to back.  For a browser, it is NEVER
-// a benefit to delay for an RTT before the second packet is sent.
-//
-// As a practical example in Chromium today, consider the case of a small
-// POST.  I have verified this:
-//     Client writes 649 bytes of header  (partial packet #1)
-//     Client writes 50 bytes of POST data (partial packet #2)
-// In the above example, with Nagle, a RTT delay is inserted between these
-// two sends due to nagle.  RTTs can easily be 100ms or more.  The best
-// fix is to make sure that for POSTing data, we write as much data as
-// possible and minimize partial packets.  We will fix that.  But disabling
-// Nagle also ensure we don't run into this delay in other edge cases.
-// See also:
-//    http://technet.microsoft.com/en-us/library/bb726981.aspx
-//
-// This function returns true if it succeeds to set the TCP_NODELAY option,
-// otherwise returns false.
-NET_EXPORT_PRIVATE bool SetTCPNoDelay(SocketDescriptor socket, bool no_delay);
-
 }  // namespace net
 
 #endif  // NET_SOCKET_TCP_SOCKET_H_