Any issue to be considered if ethernet send buffer is modified
Ashoka K <[email protected]>
| Newsgroups | gmane.linux.rt.user,gmane.linux.newbie |
|---|---|
| Message-ID | <CALJd4cb9aZF=jzbYiAf518KW-hS+QR4u0JDHefmR5Q4DipOuPw@mail.gmail.com> |
Hi, I use embedded system with 2.6.33, Preempt patch. To reduce the latency of sending data by our application, the ethernet send buffer is modified with setsockopt(). Before modification the send buffer size by default was shown as 16384 bytes. But on proc system value shown is different, both default and max are same: /proc/sys/net/core/wmem_default = 108544 Eventhough the above value is not modified, why the default is shown as 16384 instead 108544. Did kernel change these values during sending traffic ? If send buffer is set by setsockopt(), is this value modified by kernel during run time? AK -- To unsubscribe from this list: send the line "unsubscribe linux-rt-users" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html