From mboxrd@z Thu Jan 1 00:00:00 1970 Message-ID: <51829CF1.2090600@bollue.de> Date: Thu, 02 May 2013 19:05:53 +0200 From: Kai Bollue MIME-Version: 1.0 References: <51757858.6040604@bollue.de> <51767C09.3070502@siemens.com> <51769250.7050504@bollue.de> <5176950F.3040702@siemens.com> <5176AB28.7000305@bollue.de> <5176AC37.1040405@siemens.com> <5176C108.5050102@bollue.de> <5176C24B.3000300@siemens.com> <5176E0D2.2080101@bollue.de> <51778E9A.7090206@siemens.com> <5177D2EA.7060306@bollue.de> <517956F9.3010404@web.de> <517EBD2D.1020009@bollue.de> <517EBF13.1020602@web.de> In-Reply-To: <517EBF13.1020602@web.de> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Subject: Re: [Xenomai] Sporadic freeze using RTnet List-Id: Discussions about the Xenomai project List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Jan Kiszka Cc: Xenomai On 29.04.2013 20:42, Jan Kiszka wrote: > On 2013-04-29 20:34, Kai Bollue wrote: >> On 25.04.2013 18:16, Jan Kiszka wrote: >>> On 2013-04-24 14:41, Kai Bollue wrote: >>>> On 24.04.2013 09:49, Jan Kiszka wrote: >>>>> Yep, that's it, thank you very much. >>>>> >>>>> Does the rt_8139too driver support pre-mapping? >>>>> Nope, only the two I mentioned (simply because they are more common on >>>>> platforms that have >4G and/or an IOMMU). But - wait - the 8139too >>>>> has a >>>>> fixed DMA window (due to some older chips) and therefore does copying >>>>> to/from a pre-allocated bounce buffer on its own. That makes it work >>>>> with IOMMUs as well. >>>> OK, this means that cards using the RTL 8139 are currently the only >>>> non-Intel cards that work correctly with RTnet and > 4 GB of RAM, right? >>>> Maybe this information should be put somewhere googleable, as modern >>>> systems tend to have much RAM - and often still PCI slots. >>> True. There is an RTnet-related section on the Xenomai Wiki, and I'm >>> always open for patches against the in-tree doc of RTnet. >> Maybe like this? >> >> diff --git a/README b/README >> index b98e6fd..35e1cee 100644 >> --- a/README >> +++ b/README >> @@ -26,6 +26,9 @@ What are the requirements? >> e1000e), DEC 21x4x-based (tulip), AMD PCnet32/PCnetPCI, VIA Rhine, >> NatSemi >> DP8381x, RTL8169 network adapter, MPC8260 FCC, MPC8xx SCC/FEC, or >> MPC5200 >> on-chip controllers, SMSC LAN91C111 embedded controller, Intel 82575 >> (igb) >> +- CAUTION: On systems with > 4 GB RAM, currently only the following >> chipsets >> + will work: RealTek 8139, Intel PRO/1000 PCI-E (e1000e, NOT e1000), >> + Intel 82575 (igb) >> - For Ethernet-over-FireWire see rtfirewire.berlios.de >> - For RT-WLAN: Ralink 2500 >> > Looks ok. Can you send as full patch (subject, description, signed-off)? > Here it is: Subject: [PATCH] Added note in README on lacking support for several devices on systems with > 4 GB RAM Signed-off-by: Kai Bollue --- README | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/README b/README index b98e6fd..9010ad7 100644 --- a/README +++ b/README @@ -26,6 +26,10 @@ What are the requirements? e1000e), DEC 21x4x-based (tulip), AMD PCnet32/PCnetPCI, VIA Rhine, NatSemi DP8381x, RTL8169 network adapter, MPC8260 FCC, MPC8xx SCC/FEC, or MPC5200 on-chip controllers, SMSC LAN91C111 embedded controller, Intel 82575 (igb) +- CAUTION: On systems with > 4 GB RAM, currently only the following chipsets + will work: RealTek 8139, Intel PRO/1000 PCI-E (e1000e, NOT e1000), + Intel 82575 (igb). For other drivers, as a workaround, RAM usage can be + limited to 4 GB by giving mem=4096M as boot parameter. - For Ethernet-over-FireWire see rtfirewire.berlios.de - For RT-WLAN: Ralink 2500 -- 1.8.2.2