Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1642725
| From | Bjørn Mork <bjorn@mork.no> |
|---|---|
| Newsgroups | linux.kernel |
| Subject | Re: [RFC V1 1/1] net: cdc_ncm: Reduce memory use when kernel memory low |
| Date | 2017-05-16 20:30 +0200 |
| Message-ID | <tHPom-Bv-3@gated-at.bofh.it> (permalink) |
| References | <tHOLD-9u-3@gated-at.bofh.it> <tHOLD-9u-1@gated-at.bofh.it> |
| Organization | m |
Jim Baxter <jim_baxter@mentor.com> writes: > The CDC-NCM driver can require large amounts of memory to create > skb's and this can be a problem when the memory becomes fragmented. > > This especially affects embedded systems that have constrained > resources but wish to maximise the throughput of CDC-NCM with 16KiB > NTB's. > > The issue is after running for a while the kernel memory can become > fragmented and it needs compacting. > If the NTB allocation is needed before the memory has been compacted > the atomic allocation can fail which can cause increased latency, > large re-transmissions or disconnections depending upon the data > being transmitted at the time. > This situation occurs for less than a second until the kernel has > compacted the memory but the failed devices can take a lot longer to > recover from the failed TX packets. > > To ease this temporary situation I modified the CDC-NCM TX path to > temporarily switch into a reduced memory mode which allocates an NTB > that will fit into a USB_CDC_NCM_NTB_MIN_OUT_SIZE (default 2048 Bytes) > sized memory block and only transmit NTB's with a single network frame > until the memory situation is resolved. > Once the memory is compacted the CDC-NCM data can resume transmitting > at the normal tx_max rate once again. I must say that I don't like the additional complexity added here. If there are memory issues and you can reduce the buffer size to USB_CDC_NCM_NTB_MIN_OUT_SIZE, then why don't you just set a lower tx_max buffer size in the first place? echo 2048 > /sys/class/net/wwan0/cdc_ncm/tx_max Bjørn
Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread
[RFC V1 1/1] net: cdc_ncm: Reduce memory use when kernel memory low Jim Baxter <jim_baxter@mentor.com> - 2017-05-16 19:50 +0200
Re: [RFC V1 1/1] net: cdc_ncm: Reduce memory use when kernel memory low Bjørn Mork <bjorn@mork.no> - 2017-05-16 20:30 +0200
Re: [RFC V1 1/1] net: cdc_ncm: Reduce memory use when kernel memory low Oliver Neukum <oneukum@suse.com> - 2017-05-17 09:50 +0200
Re: [RFC V1 1/1] net: cdc_ncm: Reduce memory use when kernel memory low "Baxter, Jim" <jim_baxter@mentor.com> - 2017-05-17 13:00 +0200
Re: [RFC V1 1/1] net: cdc_ncm: Reduce memory use when kernel memory low David Miller <davem@davemloft.net> - 2017-05-17 20:20 +0200
Re: [RFC V1 1/1] net: cdc_ncm: Reduce memory use when kernel memory low Oliver Neukum <oneukum@suse.com> - 2017-05-18 12:10 +0200
Re: [RFC V1 1/1] net: cdc_ncm: Reduce memory use when kernel memory low "Baxter, Jim" <jim_baxter@mentor.com> - 2017-05-22 17:50 +0200
Re: [RFC V1 1/1] net: cdc_ncm: Reduce memory use when kernel memory low David Miller <davem@davemloft.net> - 2017-05-22 18:00 +0200
Re: [RFC V1 1/1] net: cdc_ncm: Reduce memory use when kernel memory low Oliver Neukum <oneukum@suse.com> - 2017-05-23 10:50 +0200
Re: [RFC V1 1/1] net: cdc_ncm: Reduce memory use when kernel memory low David Miller <davem@davemloft.net> - 2017-05-23 17:30 +0200
Re: [RFC V1 1/1] net: cdc_ncm: Reduce memory use when kernel memory low "Baxter, Jim" <jim_baxter@mentor.com> - 2017-05-23 21:10 +0200
RE: [RFC V1 1/1] net: cdc_ncm: Reduce memory use when kernel memory low David Laight <David.Laight@ACULAB.COM> - 2017-05-19 13:20 +0200
Re: [RFC V1 1/1] net: cdc_ncm: Reduce memory use when kernel memory low Bjørn Mork <bjorn@mork.no> - 2017-05-19 16:00 +0200
RE: [RFC V1 1/1] net: cdc_ncm: Reduce memory use when kernel memory low David Laight <David.Laight@ACULAB.COM> - 2017-05-19 16:50 +0200
Re: [RFC V1 1/1] net: cdc_ncm: Reduce memory use when kernel memory low Oliver Neukum <oneukum@suse.com> - 2017-05-22 15:30 +0200
csiph-web