Date: Mon, 14 Sep 2015 08:36:23 +0000 (UTC) From: Hans Petter Selasky <hselasky@FreeBSD.org> To: src-committers@freebsd.org, svn-src-all@freebsd.org, svn-src-head@freebsd.org Subject: svn commit: r287775 - in head/sys: net netinet Message-ID: <201509140836.t8E8aNXW051334@repo.freebsd.org>
next in thread | raw e-mail | index | archive | help
Author: hselasky Date: Mon Sep 14 08:36:22 2015 New Revision: 287775 URL: https://svnweb.freebsd.org/changeset/base/287775 Log: Update TSO limits to include all headers. To make driver programming easier the TSO limits are changed to reflect the values used in the BUSDMA tag a network adapter driver is using. The TCP/IP network stack will subtract space for all linklevel and protocol level headers and ensure that the full mbuf chain passed to the network adapter fits within the given limits. Implementation notes: If a network adapter driver needs to fixup the first mbuf in order to support VLAN tag insertion, the size of the VLAN tag should be subtracted from the TSO limit. Else not. Network adapters which typically inline the complete header mbuf could technically transmit one more segment. This patch does not implement a mechanism to recover the last segment for data transmission. It is believed when sufficiently large mbuf clusters are used, the segment limit will not be reached and recovering the last segment will not have any effect. The current TSO algorithm tries to send MTU-sized packets, where the MTU typically is 1500 bytes, which gives 1448 bytes of TCP data payload per packet for IPv4. That means if the TSO length limitiation is set to 65536 bytes, there will be a data payload remainder of (65536 - 1500) mod 1448 bytes which is equal to 324 bytes. Trying to recover total TSO length due to inlining mbuf header data will not have any effect, because adding or removing the ETH/IP/TCP headers to or from 324 bytes will not cause more or less TCP payload to be TSO'ed. Existing network adapter limits will be updated separately. Differential Revision: https://reviews.freebsd.org/D3458 Reviewed by: rmacklem MFC after: 2 weeks Modified: head/sys/net/if_var.h head/sys/netinet/tcp_output.c Modified: head/sys/net/if_var.h ============================================================================== --- head/sys/net/if_var.h Mon Sep 14 08:01:05 2015 (r287774) +++ head/sys/net/if_var.h Mon Sep 14 08:36:22 2015 (r287775) @@ -243,11 +243,12 @@ struct ifnet { * count limit does not apply. If all three fields are zero, * there is no TSO limit. * - * NOTE: The TSO limits only apply to the data payload part of - * a TCP/IP packet. That means there is no need to subtract - * space for ethernet-, vlan-, IP- or TCP- headers from the - * TSO limits unless the hardware driver in question requires - * so. + * NOTE: The TSO limits should reflect the values used in the + * BUSDMA tag a network adapter is using to load a mbuf chain + * for transmission. The TCP/IP network stack will subtract + * space for all linklevel and protocol level headers and + * ensure that the full mbuf chain passed to the network + * adapter fits within the given limits. */ u_int if_hw_tsomax; /* TSO maximum size in bytes */ u_int if_hw_tsomaxsegcount; /* TSO maximum segment count */ Modified: head/sys/netinet/tcp_output.c ============================================================================== --- head/sys/netinet/tcp_output.c Mon Sep 14 08:01:05 2015 (r287774) +++ head/sys/netinet/tcp_output.c Mon Sep 14 08:36:22 2015 (r287775) @@ -811,7 +811,8 @@ send: */ if (if_hw_tsomax != 0) { /* compute maximum TSO length */ - max_len = (if_hw_tsomax - hdrlen); + max_len = (if_hw_tsomax - hdrlen - + max_linkhdr); if (max_len <= 0) { len = 0; } else if (len > max_len) { @@ -826,6 +827,15 @@ send: */ if (if_hw_tsomaxsegcount != 0 && if_hw_tsomaxsegsize != 0) { + /* + * Subtract one segment for the LINK + * and TCP/IP headers mbuf that will + * be prepended to this mbuf chain + * after the code in this section + * limits the number of mbufs in the + * chain to if_hw_tsomaxsegcount. + */ + if_hw_tsomaxsegcount -= 1; max_len = 0; mb = sbsndmbuf(&so->so_snd, off, &moff);
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201509140836.t8E8aNXW051334>