Re: [HELP] BAD interrupts
Hi, We experimented something similar with a 8641D board. This seems to come from the gianfar and: - when we are binding the three interruptions enet_tx, enet_rx and enet_err of each interfaces on the same processor, the BAD interruptions do not increase any more. - when we are compiling the driver without the NAPI mode, there is no more BAD interruptions at all. Hope it will help even if we did not find a real fix. Regards. Régis. Kumar Gopalpet-B05799 wrote: -Original Message- From: gshan [mailto:gs...@alcatel-lucent.com] Sent: Thursday, December 24, 2009 5:16 PM To: mich...@ellerman.id.au Cc: Kumar Gopalpet-B05799; net...@vger.kernel.org; linuxppc-dev@lists.ozlabs.org Subject: Re: [HELP] BAD interrupts Michael Ellerman wrote: On Wed, 2009-12-23 at 19:41 +0530, Kumar Gopalpet-B05799 wrote: Hi all, I am trying linux 2.6.32-rc3 in SMP mode. I am seeing a lot of BAD interrupts when I do a cat /proc/interrupts. I am running a forwarding application b/w two ethernet ports (ethernet uses gianfar driver). Has any one observed these BAD interrupts ? If so, can someone help me in understanding why these interrupts come and how to resolve it ? You'll need to tell us what platform you're running on. The BAD interrupts come from ppc_spurious_interrupts. It's a count of the number of times we entered do_IRQ() (ie. took an external interrupt), but when we asked the interrupt controller which irq it was, the interrupt controller said there was no irq. Depending on your interrupt controller that might happen sometimes for valid reasons. Or it might indicate something is setup wrong. cheers I think it will help to show us your /proc/interrupts so that how you suffered from bad interrupts. I am running linux-2.6.29 in SMP mode on P2020RDB ( ruuning at 1200/600/667). The test scenario forwarding application and is a bidirectioanl flow between eth0 eth2. Attached is the complete log of the test scenario. I also noted another observation that when all the interrupts are mapped to a single core, the BAD interrupts stop coming. -- Thanks Sandeep tftp 100 etsec2/uImage_alpha Speed: 1000, full duplex Using eTSEC1 device TFTP from server 10.232.134.194; our IP address is 10.232.133.188 Filename 'etsec2/uImage_alpha'. Load address: 0x100 Loading: *# # ### done Bytes transferred = 2654090 (287f8a hex) = tftp 200 etsec2/rf ootfs_alpha Speed: 1000, full duplex Using eTSEC1 device TFTP from server 10.232.134.194; our IP address is 10.232.133.188 Filename 'etsec2/rootfs_alpha'. Load address: 0x200 Loading: *# # # # # # # # # # # # # # # # # # # # # # # # # # #
8641D, gianfar driver and BAD Interruptions
Hello, On a board based on 8641D, we have BAD interruptions (cat /proc/interrupts) which means spurious interruptions seen by the MPIC. The number of this BAD ITs are increasing during the life time of the system and depending on the ethernet activity. So this seems to come from the gianfar and we experimented that: - when we are binding the three interruptions enet_tx, enet_rx and enet_err of each interfaces on the same processor, the BAD interruptions do not increase any more. - when we are compiling the driver without the NAPI mode, there is no more BAD interruptions at all. We are playing with a 2.6.25 + some patches Linux kernel. I did not analyze the errata list of 8641D with this perspective but is there anybody who get some inputs about the phenomenon ? Regards. -- Régis ODEYE Kontron Modular Computers SA 150, rue M. Berthelot / ZI Toulon Est / BP 244 / Fr 83078 TOULON Cedex 9 Phone: (33) 4 98 16 34 86 Fax: (33) 4 98 16 34 01 E-mail: regis.od...@kontron.com Web : www.kontron.com ___ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
86xx: e600 and cputable.c
Hi, On our board based on 8641D, cat /proc/cpuinfo gives: -bash-3.2# cat /proc/cpuinfo processor : 0 cpu : 7448, altivec supported clock : 1000.00MHz revision: 0.2 (pvr 8004 0202) bogomips: 199.68 processor : 1 cpu : 7448, altivec supported clock : 1000.00MHz revision: 0.2 (pvr 8004 0202) bogomips: 199.68 total bogomips : 399.36 timebase: 1 platform: VM6250 Vendor : Kontron Modular Computers Machine : VM6250 SVR : 0x80900121 Memory : 1024 MB Is there a plan to deal with e600 core as it is done for e500 in cputable.c to be able to discriminate classic ppc against freescale cores ? Already a patch ? Regards. -- Régis ODEYE Kontron Modular Computers SA 150, rue M. Berthelot / ZI Toulon Est / BP 244 / Fr 83078 TOULON Cedex 9 Phone: (33) 4 98 16 34 86 Fax: (33) 4 98 16 34 01 E-mail: regis.od...@kontron.com Web : www.kontron.com ___ Linuxppc-dev mailing list Linuxppc-dev@ozlabs.org https://ozlabs.org/mailman/listinfo/linuxppc-dev
Re: [RFC/example] powerpc: add the mpic global timer support
Thanks for this. Regards Kumar Gala wrote: From: Dave Liu dave...@freescale.com The mpic timer works as wake up source for power management, the max timer period is 336 seconds when the CCB freq is 400MHz. to setup timer, type echo 30 /sys/devices/ffe0.soc8572/ffe41100.timer/timeout before the system enter to sleep mode. Signed-off-by: Dave Liu dave...@freescale.com --- Posting this both as a example of timer code for Regis as well as code for partial review.. need to clean up a number of things. - k -- Régis ODEYE Kontron Modular Computers SA 150, rue M. Berthelot / ZI Toulon Est / BP 244 / Fr 83078 TOULON Cedex 9 Phone: (33) 4 98 16 34 86 Fax: (33) 4 98 16 34 01 E-mail: regis.od...@kontron.com Web : www.kontron.com ___ Linuxppc-dev mailing list Linuxppc-dev@ozlabs.org https://ozlabs.org/mailman/listinfo/linuxppc-dev
MPIC Timers as clocksource
Hi everybody, Is there anyone who implemented a clocksource driver (drivers/clocksource) to deal with the MPIC timers especially with 86xx and 85xx soc's ? Regards. -- Régis ODEYE Kontron Modular Computers SA 150, rue M. Berthelot / ZI Toulon Est / BP 244 / Fr 83078 TOULON Cedex 9 Phone: (33) 4 98 16 34 86 Fax: (33) 4 98 16 34 01 E-mail: regis.od...@kontron.com Web : www.kontron.com ___ Linuxppc-dev mailing list Linuxppc-dev@ozlabs.org https://ozlabs.org/mailman/listinfo/linuxppc-dev
Re: Extended Addressing Mode
Kumar Gala wrote: On Oct 22, 2008, at 3:59 AM, Régis Odeyé wrote: Hello Everyone, I'm looking for some information about the Extended Addressing Mode (XAEN bit of HID0 register) of PPC32 support in Linux. I do not see anything in the main kernel tree but there may be some patches available ? Any information will be appreciate. Regards. There are patches from Becky Bruce that are going into 2.6.28. What are you needing 32-bit for? There are some SW IO MMU changes that are still pending to complete this work. - k We are developing a board based on Freescale 8641D which can get 4GB of ram. So I need 4GB+IOs (~1GB) of physical addressing space. My plan is to put a part of this ram above of 4GB to keep accesses to the IOs below the 4GB limit. It means non-contiguous ram addressing and XAEN features to be working. Where can I glance through Becky patches ? -- Régis ODEYE Kontron Modular Computers SA 150, rue M. Berthelot / ZI Toulon Est / BP 244 / Fr 83078 TOULON Cedex 9 Phone: (33) 4 98 16 34 86 Fax: (33) 4 98 16 34 01 E-mail: [EMAIL PROTECTED] Web : www.kontron.com ___ Linuxppc-dev mailing list Linuxppc-dev@ozlabs.org https://ozlabs.org/mailman/listinfo/linuxppc-dev
Re: Extended Addressing Mode
Kumar Gala wrote: So we have XAEN support in the tree.. however non-contiguous is something you'll have to work on yourself. Patches are welcome for this OK. I will let you know about this. Where can I glance through Becky patches ? This is the bulk: http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=4ee7084eb11e00eb02dc8435fd18273a61ffa9bf Thank you very much, Sir. -- Régis ODEYE Kontron Modular Computers SA 150, rue M. Berthelot / ZI Toulon Est / BP 244 / Fr 83078 TOULON Cedex 9 Phone: (33) 4 98 16 34 86 Fax: (33) 4 98 16 34 01 E-mail: [EMAIL PROTECTED] Web : www.kontron.com ___ Linuxppc-dev mailing list Linuxppc-dev@ozlabs.org https://ozlabs.org/mailman/listinfo/linuxppc-dev
Re: Extended Addressing Mode
Matt Sealey wrote: I'd also be interested in any work done to enable non-contiguous memory areas. Reading the docs for the MPC8641D though I am not sure you can set up LAWs for it? One thing I wanted to try was installing 4GB in a system and overlapping IO (since there is very little of it on a stock MPC8641DHPCN) in the top ~256MB-512MB, but I am fairly sure this is NOT supported because of the way the LAWs work, and also the alignment of the LAWs means it is not fine enough granularity to map between 2GB and 4GB into a window (you can have 2GB or 4GB but not some more arbitrary value?) On my point of view, LAWs sizes are power of 2 and should be aligned to the window size. But there is 10 LAWs you can combine to achieve quite a lot of different mappings. Regards. -- Régis ODEYE Kontron Modular Computers SA 150, rue M. Berthelot / ZI Toulon Est / BP 244 / Fr 83078 TOULON Cedex 9 Phone: (33) 4 98 16 34 86 Fax: (33) 4 98 16 34 01 E-mail: [EMAIL PROTECTED] Web : www.kontron.com ___ Linuxppc-dev mailing list Linuxppc-dev@ozlabs.org https://ozlabs.org/mailman/listinfo/linuxppc-dev