-----Jason Gunthorpe <jgg@xxxxxxxx> wrote: ----- >To: Steve Wise <swise@xxxxxxxxxxxxxxxxxxxxx> >From: Jason Gunthorpe <jgg@xxxxxxxx> >Date: 01/23/2018 06:28PM >Cc: "'Bernard Metzler'" <bmt@xxxxxxxxxxxxxx>, >linux-rdma@xxxxxxxxxxxxxxx >Subject: Re: [PATCH v3 03/13] Attach/detach SoftiWarp to/from network >and RDMA subsystem > >On Tue, Jan 23, 2018 at 11:24:24AM -0600, Steve Wise wrote: >> > >> > On Tue, Jan 23, 2018 at 10:58:01AM -0600, Steve Wise wrote: >> > >> > > From: >https://urldefense.proofpoint.com/v2/url?u=https-3A__tools.ietf.org_h >tml_rfc5041-23section-2D5.2&d=DwIBAg&c=jf_iaSHvJObTbx-siA1ZOg&r=2TaYX >Q0T-r8ZO1PP1alNwU_QJcRRLfmYTAgd3QCvqSc&m=bZXyEf2ir9_OhkryeDGP_iwh7sDL >DExD5C6xOEAgGAE&s=N-AUE48HDPRCcFKzdGiM59z8OSmoZKD9AkS1WBpnfLY&e= >> > > >> > > "At the Data Source, the DDP layer MUST segment the data >contained in >> > > a ULP message into a series of DDP Segments, where each DDP >Segment >> > > contains a DDP Header and ULP Payload, and MUST be no larger >than the >> > > MULPDU value Advertised by the LLP." >> > > >> > > Where MULDPDU is the maximum ULP PDU that will fit in the TCP >MSS... >> > >> > But exceeding the MULPDU has nothing to do with the netstack GSO >> > function.. right? GSO is entirely a local node optimization that >> > should not be detectable on the wire. >> >> It is not detectable by TCP on the wire, however the iWARP >protocols that >> impose message boundaries, among other things, require that the >iWARP PDU >> fits in a single TCP segment. Since softiwarp is building the >iwarp PDU, if >> it builds one based on a 64K GSO advertised MSS, then the resulting >wire >> packets will have man TCP segments all containing parts of a single >iWARP >> PDU, which violates the spec I quoted. > >But that still has nothing to do with GSO, can't you GSS up to >MULPDU? > >Isn't the issue here more that, as Bernard says, siw is totally >broken >since it can't control the TCP layer segmentation boundaries? :( > totally broken. hmm. Not integrating siw with the TCP kernel code makes it impossible to avoid these things. Can kernel iSCSI avoid fragmentation and mis-alignment? No. I wanted to avoid tweaking kernel TCP code. That would make siw non acceptable. -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
