Changeset View
Standalone View
share/man/man4/sume.4
Property | Old Value | New Value |
---|---|---|
svn:eol-style | null | native \ No newline at end of property |
svn:keywords | null | FreeBSD=%H \ No newline at end of property |
svn:mime-type | null | text/plain \ No newline at end of property |
.\" Manpage for NetFPGA SUME device driver. | |||||
rgrimes: This file is missing copyright, license and $FreeBSD$. | |||||
.\" Contact denissal@freebsd.org to correct errors or typos. | |||||
.Dd July 25, 2020 | |||||
.Dt SUME 4 | |||||
.Os | |||||
.Sh NAME | |||||
Done Inline ActionsSince you just wrote this man page how is it possible to have a copyright from Cambridge or BZ, and especially from 2015? Or was some of this text from a linux file? rgrimes: Since you just wrote this man page how is it possible to have a copyright from Cambridge or BZ… | |||||
.Nm sume | |||||
Done Inline ActionsThe "All rights reserved" formula is obsolete and can and could be ommited, especially in freshly created files zec: The "All rights reserved" formula is obsolete and can and could be ommited, especially in… | |||||
.Nd "NetFPGA SUME 4x10Gb Ethernet driver" | |||||
.Sh SYNOPSIS | |||||
To compile this driver into the kernel, | |||||
place the following lines in your | |||||
kernel configuration file: | |||||
.Bd -ragged -offset indent | |||||
.Cd "device sume" | |||||
.Ed | |||||
.Pp | |||||
Alternatively, to load the driver as a | |||||
module at boot time, place the following line in | |||||
.Xr loader.conf 5 : | |||||
.Bd -literal -offset indent | |||||
if_sume_load="YES" | |||||
.Ed | |||||
.Sh DESCRIPTION | |||||
The | |||||
.Nm | |||||
driver provides support for NetFPGA SUME Virtex-7 FPGA Development Board with | |||||
the reference NIC bitstream loaded onto it. The HDL design for the reference | |||||
NIC project uses the RIFFA based DMA engine to communicate with the host | |||||
machine over PCIe. Every packet is transmitted to / from the board via one DMA | |||||
transaction, taking up to two or three interrupts per one transaction which | |||||
yields in low performance. | |||||
rgrimesUnsubmitted Done Inline Actionsman pages usually have a line break at full stops per style guide (full stop == period), This is done to ease future edits. rgrimes: man pages usually have a line break at full stops per style guide (full stop == period), This… | |||||
rgrimesUnsubmitted Done Inline Actions
This has not been done, the text is still flowed to fill margins, which is not the normal form for man pages. rgrimes: > man pages usually have a line break at full stops per style guide (full stop == period)… | |||||
denissalAuthorUnsubmitted Done Inline ActionsIs this fixed now? denissal: Is this fixed now? | |||||
rgrimesUnsubmitted Done Inline ActionsNot really, but this is not a blocking issue. rgrimes: Not really, but this is not a blocking issue. | |||||
There is no support for Jumbo frames as the hardware is capable of dealing only | |||||
with frames with maximum size of 1514 bytes. | |||||
Done Inline ActionsNeither Stanford nor UCAM have authored this manpage, so this paragraph should be removed. zec: Neither Stanford nor UCAM have authored this manpage, so this paragraph should be removed. | |||||
The hardware provides no checksums and offers no other offloading. | |||||
.Sh BUGS | |||||
For me, the pre-built bitstream from Cambridge was not working correctly - at | |||||
rgrimesUnsubmitted Done Inline Actions"me" should probably not be in a man page. rgrimes: "me" should probably not be in a man page. | |||||
higher RX packet rates, the newly incoming packets would overwrite the ones in | |||||
some internal FIFO so the packets would come out of the board scrambled until | |||||
physical reset of the board. | |||||
There is also a strange behaviour in the TX path: when sending "small" packets, | |||||
TX works as it should, but when sending packets above some size limit (above | |||||
1072 bytes on my computer) right after loading the driver, SUME doesn't send | |||||
the TX_DONE interrupt and refuses to send packets until the driver is reloaded | |||||
/ board is reset. This would mean that the driver would get stuck in a non-IDLE | |||||
TX state but a watchdog-like function is implemented in the driver to check | |||||
whether the non-IDLE state lasts more then 3 seconds and it resets the board | |||||
automatically. For more details, visit the NetFPGA SUME project site. | |||||
.Sh SEE ALSO | |||||
.Xr arp 4 , | |||||
.Xr ifconfig 8 , | |||||
.Xr netgraph 4 , | |||||
.Xr netintro 4 , | |||||
.Xr ng_ether 4 , | |||||
Done Inline ActionsInsert "single" before "DMA transaction", to clarify the inefficiency of the reference NIC DMA design and the huge PIO PCI traffic overhead (to set up the DMA) associated with each packet transfer. zec: Insert "single" before "DMA transaction", to clarify the inefficiency of the reference NIC DMA… | |||||
.Xr vlan 4 | |||||
Done Inline ActionsDrop "in" after "yields", I don't think it is necessary. zec: Drop "in" after "yields", I don't think it is necessary. | |||||
.Sh AUTHOR | |||||
The original Linux NetFPGA SUME driver was originally written by Bjoern A. | |||||
rgrimesUnsubmitted Done Inline ActionsDouble use of original does not read well here. I would just drop the first one, as it is implied by the second one. rgrimes: Double use of original does not read well here. I would just drop the first one, as it is… | |||||
Zeeb. The FreeBSD version is written by Denis Salopek as a GSoC project. More | |||||
information about the project can be found here: | |||||
.Pa https://wiki.freebsd.org/SummerOfCode2020Projects/NetFPGA_SUME_Driver | |||||
Done Inline ActionsChange in tense from past to present (was in last sentence, is in this sentence), the "is" should be changed to a "was". rgrimes: Change in tense from past to present (was in last sentence, is in this sentence), the "is"… | |||||
Done Inline Actionss/Cambridge/the NetFPGA project/ bz: s/Cambridge/the NetFPGA project/ | |||||
Done Inline ActionsHaving played with the driver over the past few days, it would always successfully recover from rare glitches (presumably in FPGA logic(. Hence, I'm curious whether this comment is still valid and needed, and if not, I'd vote for removing it, or at least to trim it down and simply say that . I think it would be more important to notify users about the current FPGA design's inability to stop physical ports which have been configured as DOWN from consuming PCI / memory bandwidth, interrupts and CPU cycles, and that there's absolutely noting we can do about it in the driver. zec: Having played with the driver over the past few days, it would always successfully recover from… | |||||
Done Inline ActionsThere's also no support for multicast filters... zec: There's also no support for multicast filters... |
This file is missing copyright, license and $FreeBSD$.