Hi Guillermo, Interesting. So, if I understood your last few emails, you can read the whole bram fine if you do so one word at a time. But if you read 512 bytes or more things break.
The answer is somewhere in bulk_read in spifpga_user.c, I think. But I wonder, if you change MAX_BURST_SIZE <https://github.com/jack-h/katcp_devel/blob/5ad8df0ec6f28af97219d0046a9358dd9017d153/tcpborphserver3/spifpga_user.h#L8> (which I think reflects numbers of 32-bit reads which are commanded at once from the SPI driver) to 64 (so a max read per burst of 256 bytes) does that help? I'm not sure the tutorials were ever tested with a pi rev 1 B+, so maybe there was always a problem with large reads which was never noticed(?) Cheers Jack On Fri, 9 Apr 2021 at 16:09, Guillermo Gancio <[email protected]> wrote: > Hi Jack, > > I have been doing more tests, and I notice the following on the > raspberry pi tcpborphserver3 Log.... > > When I do Casperfpga -> fpga.read('even',512,0) on the Rpi tcpb3 log I got: > data added is 522, needed at least 10 + 513 > > Now if I do, fpga.read('even',1024*8,0), or any number of bytes larger > than 512, on the Rpi tcpb3 log I got: > can't send spi message! (error -1) > > Cheers, > > > El mié, 7 abr 2021 a las 11:24, Jack Hickish (<[email protected]>) > escribió: > > > > > > > > On Wed, 7 Apr 2021, 2:57 pm Guillermo Gancio, <[email protected]> > wrote: > >> > >> Hi, > >> > >> I'm using the tutorials from > >> > https://github.com/casper-astro/tutorials_devel/tree/master/snap/tut_spec > >> I just clone them and I got the same result, could be an issue with > >> the katcp installation on the snap-raspberry? > > > > > > You're using the prebuilt FPG file? It could be katcp I suppose - if you > read one word at a time, eg with CasperFpga.read_uint using an increasing > word_offset argument, how far can you read? > > > >> > >> I'm also checking the casperfpga installation... > >> > >> Cheers. > >> > >> > >> > >> > >> > >> El mié, 7 abr 2021 a las 10:15, Jack Hickish (<[email protected]>) > escribió: > >> > > >> > Hi Guillermo, > >> > > >> > If the values are 64 bits in the model, it is suspicious that the > reads are 1024*4 Bytes. I notice that in the latest tutorials_devel the > reads are 1024*8 bytes ( > https://github.com/casper-astro/tutorials_devel/blob/2982f9f14bffa065a80d9047ec5cd7ea1ca3ad6c/snap/tut_spec/snap_tut_spec.py#L19 > ) > >> > > >> > What version of the tutorial are you using -- it's possible that if > the version is old something else is wrong. > >> > > >> > Cheers > >> > Jack > >> > > >> > On Wed, 7 Apr 2021 at 14:09, Guillermo Gancio <[email protected]> > wrote: > >> >> > >> >> Hi Jack, > >> >> > >> >> What I understand from the tutorial, and correct me if I'm wrong, is: > >> >> Size of PFB and FFT, 2^12 i.e Re+Im 4096 channels, that's 2048 > >> >> frequency channels after obtaining power values from the complex > >> >> values. > >> >> Those 2048 channels end up in two registers 'even' and 'odd' of 2^10 > >> >> i.e. 1024 values each, each of this values is of 64bits, > >> >> So I have to read 1024 * 8 = 8192 bits, from each register (even and > odd). > >> >> > >> >> I compiled the project with those values and I got the same result as > >> >> before, (read ok with 128*4 bits, ,fpga.read('even',128*4,0) ) > >> >> > >> >> Cheers > >> >> > >> >> > >> >> > >> >> > >> >> > >> >> > >> >> > >> >> El mar, 6 abr 2021 a las 11:27, Jack Hickish (<[email protected]>) > escribió: > >> >> > > >> >> > Hi Guillermo, > >> >> > > >> >> > How many channels is that tutorial supposed to generate? It looks > like the RAM is smaller than expected by the software. Is it possible that > each even/odd RAM is supposed to be 512*4 (i.e., 512 words, each 32-bits > wide) for 1024 channels total? > >> >> > > >> >> > Cheers > >> >> > Jack > >> >> > > >> >> > On Mon, 5 Apr 2021 at 22:46, Guillermo Gancio <[email protected]> > wrote: > >> >> >> > >> >> >> Hi Casperities (?) > >> >> >> > >> >> >> I'm going thru the tutorial 3 on a snap board, and I have some > >> >> >> issue/error while reading the 'odd' and 'even' register, > >> >> >> when reading with > >> >> >> -> a_0=struct.unpack('>1024l',fpga.read('even',1024*4,0)) > >> >> >> > >> >> >> I end up with the error > >> >> >> -> Request: ?read odd 0 4096 > >> >> >> -> Reply: !read fail > >> >> >> > >> >> >> but if I change the number of bytes to read from 1024*4 to 128*4, > I > >> >> >> can read it with no problem, ( but the data plot does not seem to > be > >> >> >> ok ). Also the accumulation value increments ok. > >> >> >> > >> >> >> Any thoughts? Thanks! > >> >> >> > >> >> >> > >> >> >> > >> >> >> > >> >> >> > >> >> >> -- > >> >> >> Instituto Argentino de Radioastronomia > >> >> >> [Argentine Institute of Radioastronomy] > >> >> >> > >> >> >> Guillermo M. Gancio > >> >> >> Responsable Área Observatorio > >> >> >> [Head of Observatory] > >> >> >> > >> >> >> Tel: (0054-0221) 482-4903 Int: 106 > >> >> >> Mail laboral [email protected] > >> >> >> > >> >> >> -- > >> >> >> You received this message because you are subscribed to the > Google Groups "[email protected]" group. > >> >> >> To unsubscribe from this group and stop receiving emails from it, > send an email to [email protected]. > >> >> >> To view this discussion on the web visit > https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CA%2BEePfTjY%2BZzCUWr_ksaGfNRSoVfeF251VRbLP4S5dkFSnddcA%40mail.gmail.com > . > >> >> > > >> >> > -- > >> >> > You received this message because you are subscribed to the Google > Groups "[email protected]" group. > >> >> > To unsubscribe from this group and stop receiving emails from it, > send an email to [email protected]. > >> >> > To view this discussion on the web visit > https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CAG1GKSk2tv25XinAy%3D0mt_WKY1s61p2qBOYLVsJ-%2B5j1twueQw%40mail.gmail.com > . > >> >> > >> >> > >> >> > >> >> -- > >> >> Instituto Argentino de Radioastronomia > >> >> [Argentine Institute of Radioastronomy] > >> >> > >> >> Guillermo M. Gancio > >> >> Responsable Área Observatorio > >> >> [Head of Observatory] > >> >> > >> >> Tel: (0054-0221) 482-4903 Int: 106 > >> >> Mail laboral [email protected] > >> >> > >> >> -- > >> >> You received this message because you are subscribed to the Google > Groups "[email protected]" group. > >> >> To unsubscribe from this group and stop receiving emails from it, > send an email to [email protected]. > >> >> To view this discussion on the web visit > https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CA%2BEePfQWfV6jG1pMSNeqxbh0z5J0jn%2BBOSLuiY7PWpdcdDDu5Q%40mail.gmail.com > . > >> > > >> > -- > >> > You received this message because you are subscribed to the Google > Groups "[email protected]" group. > >> > To unsubscribe from this group and stop receiving emails from it, > send an email to [email protected]. > >> > To view this discussion on the web visit > https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CAG1GKSkK_YPV3Pk11skAn49WEqG6pXs75q6QJfiH%3DTMqKpx6yw%40mail.gmail.com > . > >> > >> > >> > >> -- > >> Instituto Argentino de Radioastronomia > >> [Argentine Institute of Radioastronomy] > >> > >> Guillermo M. Gancio > >> Responsable Área Observatorio > >> [Head of Observatory] > >> > >> Tel: (0054-0221) 482-4903 Int: 106 > >> Mail laboral [email protected] > >> > >> -- > >> You received this message because you are subscribed to the Google > Groups "[email protected]" group. > >> To unsubscribe from this group and stop receiving emails from it, send > an email to [email protected]. > >> To view this discussion on the web visit > https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CA%2BEePfQp5uK4cz9Hm2_JpaLErw0rGy8LqyOYq%2Bi_XysyrP97cA%40mail.gmail.com > . > > > > -- > > You received this message because you are subscribed to the Google > Groups "[email protected]" group. > > To unsubscribe from this group and stop receiving emails from it, send > an email to [email protected]. > > To view this discussion on the web visit > https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CAG1GKSmuJwRMt_NOCz8XBbWrF0uR%2Bi7A7k5wBNq477V6vLqQww%40mail.gmail.com > . > > > > -- > Instituto Argentino de Radioastronomia > [Argentine Institute of Radioastronomy] > > Guillermo M. Gancio > Responsable Área Observatorio > [Head of Observatory] > > Tel: (0054-0221) 482-4903 Int: 106 > Mail laboral [email protected] > > -- > You received this message because you are subscribed to the Google Groups " > [email protected]" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CA%2BEePfSwkJYG7e%3DarWymU4Rm9TPSKSB-NOzZxFSOPoW9LEKAng%40mail.gmail.com > . > -- You received this message because you are subscribed to the Google Groups "[email protected]" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/a/lists.berkeley.edu/d/msgid/casper/CAG1GKSnuZOfboGDmubcjiBg1-N5HxmkLpETgG%3DGWH1kGrvxY0A%40mail.gmail.com.
