| To: | Andi Kleen <ak@xxxxxx> |
|---|---|
| Subject: | Re: Mystery packet killing tg3 |
| From: | Peter Buckingham <peter@xxxxxxxxxxxx> |
| Date: | Thu, 05 May 2005 11:21:26 -0700 |
| Cc: | "David S. Miller" <davem@xxxxxxxxxxxxx>, jgarzik@xxxxxxxxx, netdev@xxxxxxxxxxx |
| In-reply-to: | <20050505180609.GB24386@xxxxxx> |
| References: | <20050502162405.65dfb4a9@xxxxxxxxxxxxxxxxxxxxx> <20050502200251.38271b61.davem@xxxxxxxxxxxxx> <m14qdiyhcn.fsf@xxxxxx> <42791825.2080204@xxxxxxxxxxxx> <20050505114327.GA51761@xxxxxx> <427A5363.2080703@xxxxxxxxxxxx> <20050505180609.GB24386@xxxxxx> |
| Sender: | netdev-bounce@xxxxxxxxxxx |
| User-agent: | Debian Thunderbird 1.0.2 (X11/20050331) |
Andi Kleen wrote: In what initialization exactly? YOu mean changing the pci_set_dma_mask() in the e1000 driver? exactly. Kernel panic - not syncing: Cannot allocate iommu bitmapHmm, that must be a different problem in the IOMMU code not related to the e1000. Probably it got confused somehow. Why did you get the idea the ethernet driver has anything to do with it? i don't think it has anything to do with the e1000 actually. we only see this when the serial console redirection is enabled. It was the only dmesg I had lying around that had the IOMMU not forced. We run our systems diskless so it's not easy to get the boot log when the ethernet device is not working correctly. I can try to get a system with a hard drive and give you the dmesg from that instead (probably more useful ;-) peter |
| Previous by Date: | Re: (diet-)FIB alternative fib_hlist.c, Andi Kleen |
|---|---|
| Next by Date: | Re: Mystery packet killing tg3, Andi Kleen |
| Previous by Thread: | Re: Mystery packet killing tg3, Andi Kleen |
| Next by Thread: | Re: Mystery packet killing tg3, Andi Kleen |
| Indexes: | [Date] [Thread] [Top] [All Lists] |