From mboxrd@z Thu Jan 1 00:00:00 1970 Message-ID: <447EB707.7090605@domain.hid> Date: Thu, 01 Jun 2006 11:44:39 +0200 From: Jan Kiszka MIME-Version: 1.0 Subject: Re: [Xenomai-core] io apic and shared irq questions References: <447ED64F.7090902@domain.hid> In-Reply-To: <447ED64F.7090902@domain.hid> Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="------------enigFC5AD0F077DDAC672F9DF633" Sender: jan.kiszka@domain.hid List-Id: "Xenomai life and development \(bug reports, patches, discussions\)" List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: "s.a." Cc: xenomai@xenomai.org This is an OpenPGP/MIME signed message (RFC 2440 and 3156) --------------enigFC5AD0F077DDAC672F9DF633 Content-Type: text/plain; charset=ISO-8859-15 Content-Transfer-Encoding: quoted-printable s.a. wrote: > Hi, >=20 > Regarding an embedded realtime system, I have got a communication board= > on pci bus . >=20 > I would like it's interrupt not being shared (with usb) , because the > component receives an it every 100us and is realtime determinist. > I looked at how was routed IRQ's with the io apic , although it is able= > to use 24 irq's, linux always share my PCI boards IRQ with another > component and do not use all the 24 irq's range capability of ioapic. >=20 > Do you think it is possible to program the ioapic in order to have a > better irq mapping and avoid this problem and use all the range of > availbale vectors ? Logical IRQ numbers are one side, but you first have to find out if the PCI components do not share the same *physical* line. Dependent on the system design, there should be some way to flip those line assignment, either by moving to another extension slot or even by playing with jumpers (PC/104plus boards may provide this e.g.). Jan --------------enigFC5AD0F077DDAC672F9DF633 Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.2 (MingW32) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFEfrcHniDOoMHTA+kRArX+AJ4ybNAiQB4MSOdVKqQ6NUPKv7UKxQCfSekb a6t9bPUpfIG2d4SwvKUzdZc= =/zQk -----END PGP SIGNATURE----- --------------enigFC5AD0F077DDAC672F9DF633--