From mboxrd@z Thu Jan 1 00:00:00 1970 From: Mark Jonas Date: Tue, 13 Jul 2004 09:47:40 +0200 Subject: AW: [U-Boot-Users] LITE5200 board: Bus Fault with 2 network cards In-Reply-To: References: Message-ID: <40F3939C.2080300@freescale.com> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: u-boot@lists.denx.de Martin Krause wrote: >>Is there a known issue with the PCI on the MPC5200, and more specifically >>with PCI network cards ? >> >> > >I also have problems with some PCI cards. Some are recognized, some not. I haven't found an explanation, yet. I guess it has something to do with some "3.3 V compatible" cards are not really compatible with 3.3 V only systems. It could as well be an errata in the MPC5200 described in the MPC5200 Errata Listing: > >ID 322: Violation of PCI Tval min time > >The PCI Timing Speci?cation de?nes the timing parameter Tval (CLK to Signal Valid Delay): > Tval min 2 ns, max 11 ns (33MHz) > Tval min 2 ns, max 6 ns (66MHz) > When M66EN is asserted, the minimum speci?cation for Tval(min) may be reduced to 1 ns if a mechanism is provided to guarantee a minimum value of 2 ns when M66EN is deasserted. >MPC5200 provides Tval min of 1 ns (instead of 2 ns), which is a violation of the >Timing Speci?cation for 33MHz. > > Yes, some cards are not detected because of this errata. Please note that this only happens if the card / chipset used is also not PCI timing spec compliant. A compliant card / chipset needs to be able to handle 0 ns Tval(min). And some cards say "3.3 V compatible" but they aren't because either 3.3 V is simply not enough or they do nasty things like shortening the 5 V and 3.3 V power supply. That's always worth a check before inserting the card in a 3.3 V / 5 V system.