Rafi, Albert I have to run now. But that dump tells us that the mmap reads are returning garbage. I assume you are running this clean out of power up and NOT after any failed attempt of running loadBitstream? If that is true; it is as if you somehow have a defective *default* bitstream. (Where did it come from?) I may not be able to debug this with you side by side until I'm back in my lab Thursday. Everything looks fine; except for the garbage coming back. Adding the list back so others may offer a debug suggestion. -Shep
On Mon, Jun 13, 2011 at 12:09 PM, Rafi Rubin <[email protected]> wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > On 06/13/11 12:01, Albert Kwon wrote: > > Hi Rafi, > > > > Can you run the following commands as sudo for me? Shep wants me to > > test this out by running things actually as sudo, not what we've been > doing. > > > > Environment variable that needs to be set: > > export > > XIL_PAR_DESIGN_CHECK_VERBOSE=1 > > export XIL_TIMING_ALLOW_IMPOSSIBLE=1 > > export PATH=$PATH:/scratch/ic3/ocpi/bin/ > > > > Then inside /scratch/ic3/ocpi/script, run > > sudo -E ./ocfrp_check, > > Error during probe for OCFRP: Magic numbers do not match in region/bar 0 > Did not find any OpenCPI FPGA reference platform boards. > > > sudo -E *swctl **0xFA000000 0xFB9F0000 admin* > > OCCP Admin Space > Open: 0xffffffff "ÿÿÿÿ" > CPI: 0xffffffff "ÿÿÿÿ" > revision: 0xffffffff > birthday: 0xffffffff Sun Feb 7 06:28:15 2106 > workerMask: 0xffffffff workers 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 > 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 exist > pci_dev_id: 0xffffffff > attention: 0xffffffff > cpStatus: 0xffffffff > scratch20: 0xffffffff > scratch24: 0xffffffff > cpControl: 0xffffffff > timeStatus: 0xffffffff ppsLostSticky gpsInSticky ppsInSticky > timeSetSticky ppsOK ppsLost > timeControl: 0xffffffff > gpsTimeMS: 0xffffffff (4294967295) Sun Feb 7 06:28:15 2106 > gpsTimeLS: 0xffffffff (4294967295) > deltaTimeMS: 0xffffffff > deltaTimeLS: 0xffffffff > refPerPPS: 0xffffffff (-1) > numDPMemReg: 0xffffffff (-1) > UUID[ 3: 0]: 0xffffffff 0xffffffff 0xffffffff 0xffffffff > UUID[ 7: 4]: 0xffffffff 0xffffffff 0xffffffff 0xffffffff > UUID[11: 8]: 0xffffffff 0xffffffff 0xffffffff 0xffffffff > UUID[15:12]: 0xffffffff 0xffffffff 0xffffffff 0xffffffff > > > > sudo lspci -v > > $ sudo lspci -vs 05:00.0 > /scratch/ic3/ocpi/scripts 12:06PM hex > 05:00.0 RAM memory: Xilinx Corporation Device 4243 (rev 02) > Subsystem: Xilinx Corporation Device 0007 > Flags: fast devsel, IRQ 10 > [virtual] Memory at fa000000 (32-bit, non-prefetchable) [size=16M] > [virtual] Memory at fb9f0000 (32-bit, non-prefetchable) [size=64K] > Capabilities: [40] Power Management version 3 > Capabilities: [48] MSI: Enable- Count=1/1 Maskable- 64bit+ > Capabilities: [60] Express Endpoint, MSI 01 > Capabilities: [100] Device Serial Number 00-00-00-01-01-00-0a-35 > > $ sudo lspci -vvs 05:00.0 > /scratch/ic3/ocpi/scripts 12:06PM hex > 05:00.0 RAM memory: Xilinx Corporation Device 4243 (rev 02) > Subsystem: Xilinx Corporation Device 0007 > Control: I/O- Mem- BusMaster- SpecCycle- MemWINV- VGASnoop- > ParErr- Stepping- SERR- FastB2B- DisINTx- > Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- > <TAbort- <MAbort- >SERR- <PERR- INTx- > Interrupt: pin A routed to IRQ 10 > Region 0: [virtual] Memory at fa000000 (32-bit, > non-prefetchable) [size=16M] > Region 1: [virtual] Memory at fb9f0000 (32-bit, > non-prefetchable) [size=64K] > Capabilities: [40] Power Management version 3 > Flags: PMEClk- DSI- D1- D2- AuxCurrent=0mA > PME(D0+,D1+,D2+,D3hot+,D3cold-) > Status: D0 NoSoftRst+ PME-Enable- DSel=0 DScale=0 PME- > Capabilities: [48] MSI: Enable- Count=1/1 Maskable- 64bit+ > Address: 0000000000000000 Data: 0000 > Capabilities: [60] Express (v2) Endpoint, MSI 01 > DevCap: MaxPayload 512 bytes, PhantFunc 0, Latency L0s > <64ns, L1 unlimited > ExtTag- AttnBtn- AttnInd- PwrInd- RBE+ FLReset- > DevCtl: Report errors: Correctable- Non-Fatal- Fatal- > Unsupported- > RlxdOrd+ ExtTag- PhantFunc- AuxPwr- NoSnoop+ > MaxPayload 128 bytes, MaxReadReq 512 bytes > DevSta: CorrErr- UncorrErr- FatalErr- UnsuppReq- AuxPwr- > TransPend- > LnkCap: Port #0, Speed 5GT/s, Width x4, ASPM L0s, > Latency L0 unlimited, L1 unlimited > ClockPM- Surprise- LLActRep- BwNot- > LnkCtl: ASPM Disabled; RCB 64 bytes Disabled- Retrain- > CommClk- > ExtSynch- ClockPM- AutWidDis- BWInt- AutBWInt- > LnkSta: Speed 5GT/s, Width x4, TrErr- Train- SlotClk- > DLActive- BWMgmt- ABWMgmt- > DevCap2: Completion Timeout: Range B, TimeoutDis- > DevCtl2: Completion Timeout: 50us to 50ms, TimeoutDis- > LnkCtl2: Target Link Speed: 5GT/s, EnterCompliance- > SpeedDis-, Selectable De-emphasis: -6dB > Transmit Margin: Normal Operating Range, > EnterModifiedCompliance- ComplianceSOS- > Compliance De-emphasis: -6dB > LnkSta2: Current De-emphasis Level: -6dB > Capabilities: [100 v1] Device Serial Number 00-00-00-01-01-00-0a-35 > > > > > > and if you can send me the outputs of each, that would be great > > (only the output of Xilinx for lspci -v). > > > > And after the three commands above, if you don't mind, please try to > run > > sudo ./loadBitStreamML605_sudo > > ../build/ml605-20110609_1313/fpgaTop.bit 0000:05:00.0 auto > > inside the same script folder. > > > > Thanks a lot. > > Albert > > > > > > -------- Original Message -------- > > Subject: Re: OCPI Question > > Date: Mon, 13 Jun 2011 11:52:18 -0400 > > From: Shepard Siegel <[email protected]> > > To: Albert Kwon <[email protected]> > > > > > > > > right. good > > so (with Rafi's help) try from $OCPI_DIR/bin > > > > *$ sudo -E ./swctl 0xFA000000 0xFB9F0000 admin* > > > > This should dump out a bunch of registers with sensible values, similar > > to this > > > > [shep@core960 bin]$ sudo ./swctl 0xF1000000 0xF2FF0000 admin > > > > OCCP Admin Space > > Open: 0x4f70656e "nepO" > > CPI: 0x43504900 "" > > revision: 0x00000001 > > birthday: 0x4b2d41ab Sat Dec 19 21:12:11 2009 > > workerMask: 0x0000601f workers 0 1 2 3 4 13 14 exist > > pci_dev_id: 0x00000300 > > attention: 0x00000000 > > cpStatus: 0x00000381 > > scratch20: 0x00000000 > > scratch24: 0x00000000 > > cpControl: 0x2202f798 > > timeStatus: 0x00000000 > > timeControl: 0x00000000 > > gpsTimeMS: 0x0000fd6c (64876) Thu Jan 1 18:01:16 1970 > > gpsTimeLS: 0x06ba5820 (112874480) > > deltaTimeMS: 0x00000000 > > deltaTimeLS: 0x00000000 > > refPerPPS: 0x4b4a4948 (1263159624) > > numDPMemReg: 0xdeadc0de (-559038242) > > UUID[ 3: 0]: 0x4f4e4d4c 0x4b4a4948 0x00000000 0x00000000 > > UUID[ 7: 4]: 0xdeadc0de 0xdeadc0de 0xdeadc0de 0xdeadc0de > > UUID[11: 8]: 0xdeadc0de 0xdeadc0de 0xdeadc0de 0xdeadc0de > > UUID[15:12]: 0xdeadc0de 0xdeadc0de 0xdeadc0de 0xdeadc0de > > > > > > > > > > On Mon, Jun 13, 2011 at 11:47 AM, Albert Kwon <[email protected] > > <mailto:[email protected]>> wrote: > > > > I will ask Rafi to run those things in sudo, and see what happens. > > This is the output of lspci -v: > > > > > > 05:00.0 RAM memory: Xilinx Corporation Device 4243 (rev 02) > > Subsystem: Xilinx Corporation Device 0007 > > Flags: fast devsel, IRQ 10 > > [virtual] Memory at fa000000 (32-bit, non-prefetchable) > [size=16M] > > [virtual] Memory at fb9f0000 (32-bit, non-prefetchable) > [size=64K] > > Capabilities: <access denied> > > > > It seems like to get the full output, I need to run this as sudo as > > well. Also, are the bar0 and bar1 in this case 0xFA000000 and > > 0xFB9F0000 respectively? > > > > Thanks, > > Albert > > > > > > > > On 06/13/2011 11:32 AM, Shepard Siegel wrote: > >> Albert, > >> It's highly unusual for lspci to see the board just fine; and > >> ocfrp_check not find it. Especially with the default bitstream. > >> If you can, do it as root with sudo -E, not Rafi's alternative. > >> Also, can you post the output from lspci -v (but just for the > >> Xlinix device) > >> There is a compiled version of swctl in the $OCPI_DIR/bin directory > >> Or you can go the /src dir, say $gcc swctl.c; mv a.out ../bin/swctl > >> In any event, you will need to know the BAR0 and BAR1 addresses as > >> reported by lspci -v and say > >> > >> $ sudo -E ./swctl 0xDE000000 0xDFFF0000 admin > >> > >> (I just made up the Bar addresses, you must use the correct ones) > >> > >> And post the output you get here with the default flashed bitstream. > >> > >> Thanks. > >> _Shep > >> > >> > >> On Mon, Jun 13, 2011 at 11:22 AM, Albert Kwon > >> <[email protected] <mailto:[email protected]>> wrote: > >> > >> ocfrp_check gives the same result: > >> > >> > >> Error during probe for OCFRP: Magic numbers do not match in > >> region/bar 0 > >> Did not find any OpenCPI FPGA reference platform boards. > >> > >> And I can't find the command/script swctl. Do I need to do > >> something to get this program/script? > >> > >> Thanks so much for the help. > >> -Albert > >> > >> > >> On 06/13/2011 11:14 AM, Shepard Siegel wrote: > >>> Albert, > >>> I've never done it that way but Rafi must know what he is > doing. > >>> > >>> Let's back up one more time before we move ahead. Before we > >>> try to load your own bitstream with loadBitstreamML605, do > >>> these programs work as expected on the default flash bitstream: > >>> > >>> ocfrp_check (to see the board) > >>> swctl <bar0> <bar1> admin > >>> > >>> If not, then before we debug loadBitstreamXXX, we should get > >>> ocfrp_check and swctl working. > >>> > >>> -Shep > >>> > >>> note: sharing with opencpi_dev for others to chime in. > >>> > >>> > >>> > >>> > >>> On Mon, Jun 13, 2011 at 10:59 AM, Albert Kwon > >>> <[email protected] <mailto:[email protected]>> > wrote: > >>> > >>> Instead of having to run these things as sudo (since it's > >>> Rafi's machine that we are running this on), Rafi set the > >>> suid on ocfrp_check, and put it under /opt/opencpi/bin. > >>> So on the script, instead of doing sudo ./ocfrp_check > >>> $device, I do /opt/opencpi/bit/ocfrp_check $device. I've > >>> also asked him to change permission on > >>> /sys/bus/pci/devices/$device/config, so that I can write > >>> to it. I've also changed the location of $tmp for > >>> loadBitStreamML605 script to my home folder's tmp. So > >>> currently, I don't think there are any commands that > >>> require sudo. > >>> > >>> environment OCPI_DIR is currently set to be > >>> /scratch/ic3/ocpi. > >>> > >>> Is there something wrong with that we are doing here? > >>> > >>> -Albert > >>> > >>> > >>> > >>> On Sat, Jun 11, 2011 at 7:58 PM, Albert Kwon > >>> <[email protected] <mailto:[email protected]>> > wrote: > >>> > >>> Hi Shep, > >>> > >>> The bitstream upload failed once due to some sudo > >>> privilege issues, and now when I run loadBitStream > >>> script, I get this error: > >>> > >>> Checking existing bitstream.. > >>> Error during probe for OCFRP: Magic numbers do not match > >>> in region/bar 0 > >>> Did not find an OpenCPI FPGA reference platform board > >>> named "0000:05:00.0". > >>> Existing bitstream does not appear to be functional. > >>> There is no config snapshot in > >>> /opt/cpi/ocfrp/pci-config-snapshots/0000:05:00.0 to > restore. > >>> > >>> Do you have any suggestion as to what to try? > >>> > >>> Thanks, > >>> Albert > >>> > >>> > >>> > >>> snip > >> > >> > > > > > > > > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.11 (GNU/Linux) > Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ > > iQIcBAEBAgAGBQJN9jZQAAoJEPILXytRLnK2SNkP/if3msS7afaug/E2ZKeERJ0A > QSmYuVH6/bjVAxH0+kzoAWAj+g2Niy2XA0y3EapksQdaEgZtAqLTxMgxLo8Qiq8z > gNYn/AeubfgTuZSu1eLHCgCOb/1hMvc6cC/daRdhCBqrUxWTDfEKh6HtGAYRIkef > R0V7SRdfiOjDZGur0sbLpPUQqpfBrxfzVFxabkfR2XIxM2UXG+aUislSmkb6qx8R > qQfqPsF+P9vK9YN4YNe/5zrojGct6/fTGkjHv9BTCtUCh596QxUD085MJKNQ5nMu > DJ0Gxq9bDf+n+5PlFFJtlhE7g1nE2lWZ/EcWYtVe2PkVo6+vauiWF8U3tstgKiRA > c8yIWS2bqk/GOEGHtedPjzAoD35Ulx/DkstYFu1/NfJvpg8eWjve+JEIPuu0frF2 > /jNQXMC1cHwBX5UJh7l0hTeBQ0I0WXIIv13/6mIjTT7t0IPRzpVB5VARPab6iYr8 > uiAKN53MxiPOyPYXPCbt1k2Yt9/k+QaY/4fWSOUC34EftFxGaCRgYLOnP5Mh8GGS > YBQUdri7Cs5wpeWo7vsgSVgHo/xF1so6yPGYTWO791Wy0qpqCTuV6oJQTNK1d+hs > rh/CHR5KWXBmTdC8kJTMaWI4t9TrvyoR6pA0RZGJl2W5Azv59AFgK9RpXCLezSQA > ZCOXRN0NouoKGGRnd6zO > =sy9Z > -----END PGP SIGNATURE----- >
_______________________________________________ opencpi_dev mailing list [email protected] http://lists.opencpi.org/listinfo.cgi/opencpi_dev-opencpi.org
