From mboxrd@z Thu Jan 1 00:00:00 1970 Message-ID: <512D1356.4060004@web.de> Date: Tue, 26 Feb 2013 20:56:06 +0100 From: Jan Kiszka MIME-Version: 1.0 References: <5123BCE7.2050609@xenomai.org> <224444793.158669.1360843957336.JavaMail.ngmail@webmail16.arcor-online.net> <511CD5C2.50701@arcor.de> <511CD673.2030400@xenomai.org> <511CDF65.1020907@arcor.de> <511D3A22.40307@xenomai.org> <511DF30B.6040601@arcor.de> <511E26AE.1010203@xenomai.org> <511E3A5C.4000209@arcor.de> <51237586.6030201@arcor.de> <963432115.61822.1361872899385.JavaMail.ngmail@webmail10.arcor-online.net> <512CA697.5060405@xenomai.org> <512CA7FA.60807@siemens.com> <512D014D.8020208@xenomai.org> In-Reply-To: <512D014D.8020208@xenomai.org> Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Subject: Re: [Xenomai] kernel configuration problems with ACPI List-Id: Discussions about the Xenomai project List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Gilles Chanteperdrix Cc: xenomai@xenomai.org On 2013-02-26 19:39, Gilles Chanteperdrix wrote: > On 02/26/2013 01:18 PM, Jan Kiszka wrote: > = >> On 2013-02-26 13:12, Gilles Chanteperdrix wrote: >>> On 02/26/2013 11:01 AM, hahnchristopher@arcor.de wrote: >>> >>>> On 02/19/2013 6:56 PM, Gilles Chanteperdrix wrote: >>>>> On 02/19/2013 01:52 PM, Christopher Hahn wrote: >>>>> >>>>>> Do you have any idea how to solve the problem? >>>>>> Even if the system receives interrupts at IRQ 4, the program doesn't = >>>>>> receive any message from the serial port with this kernel configurat= ion. >>>>> >>>>> >>>>> Trying to register the irq handler as level instead of edge? >>>>> >>>>> -- = >>>>> Gille= s. >>>>> >>>> >>>> I removed the 'RTDM_IRQTYPE_EDGE' from ksrc/drivers/serial/16550A.c at= line 123. >>>> >>>> Now I get 'Xenomai: xnintr_shirq_handler: IRQ4 not handled. Disabling = IRQ line.' >>> >>> >>> Jan, any idea? >> >> Not really. If there is an IRQ and the UART says "not for me", either >> the UART is buggy or someone else injects on that line. > = > = > Or we are handling the irq as edge triggered whereas it is level > triggered. Is not there an RTDM_IRQTYPE_LEVEL instead of RTDM_IRQTYPE_EDG= E? Level is default, ie. the absence of EDGE will choose it. I noticed that the kernel considers this line level triggered. It's unusual. Still, the change above should have solved it then if that was the problem. I-pipe picks up the type from Linux, thus should mask&ack the line properly. RTDM_IRQTYPE_EDGE just avoids loosing edge events by selecting the shirq_handler. Jan -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 263 bytes Desc: OpenPGP digital signature URL: