fx2lib-devel — Mailing list for fx2lib development

Re: [Fx2lib-devel] Change sync delay macro (Daniel O'Connor)

[-t <tg...@ya...>]

Message: 3
Date: Sat, 26 Feb 2011 16:35:36 +1030
From: "Daniel O'Connor" <doc...@gs...>
Subject: [Fx2lib-devel] Change sync delay macro
To: fx2lib developer mailing list <fx2...@li...>
Message-ID: <238...@gs...>
Content-Type: text/plain; charset="us-ascii"

Hi,
I modified SYNCDELAY to match the Cypress version - this is useful because GPIF 
designer generates code which calls it and it's a pain to re-edit the file each 
time you change your design.


--->  Hi Daniel, sync delay is used in more places that just GPIF 
implemenations.  Various registers are gated by different clock domains.  When 
the CPU writes to these specific registers the contents may not be updated 
if the sync delay is not executed.  The amount of delay is dependent on the two 
main clock domains.  All the details of sync delay can be reviewed in the 
Cypress TRM.  This one originally had me scratching my head for several days...  
Hope this helps :)  <---

Re: [Fx2lib-devel] Sourcing slave FIFO packets

[Chris M. <fx...@m3...>]

Genius, thanks Dennis it works now. I've probably still got some superfluous 
stuff, but the following implementation of send_fifo() works.

Thanks again.

Chris


void fifo_send(void) {
    RESETFIFO(0x06);
    OUTPKTEND=0x86; SYNCDELAY();
    OUTPKTEND=0x86; SYNCDELAY();
    RESETFIFO(0x02);

    FIFORESET = 0x80; SYNCDELAY();
    FIFORESET = 0x06; SYNCDELAY();

    EP6FIFOBUF[0] = 0x00;
    EP6FIFOBUF[1] = 0x01;
    EP6FIFOBUF[2] = 0x00;
    EP6FIFOBUF[3] = 0x00;
    EP6FIFOBUF[4] = 0x00;
    EP6FIFOBUF[5] = sendCount++;
    SYNCDELAY(); EP6BCH = 0;
    SYNCDELAY(); EP6BCL = 6;

    OUTPKTEND = 0x86; SYNCDELAY();
    FIFORESET = 0; SYNCDELAY();
}

Re: [Fx2lib-devel] Sourcing slave FIFO packets

[Dennis M. <djm...@gm...>]

We sourced OUT packets to our FPGA.  Check out the UXN2030_firmware.
https://github.com/ubixum/UXN2030_firmware


On 02/27/2011 07:53 PM, Daniel O'Connor wrote:
>
> On 28/02/2011, at 24:49, Chris McClelland wrote:
>> Daniel, thanks for your reply. I wasn't clear. It's not the PC that should
>> be getting this data, it is the FPGA attached to the EP6 slave FIFO. Since
>> it's just a FIFO it has no notion of NAK. For example, in the code below I
>> have set AUTOOUT=0. Now whenever an EP6 packet arrives from the host,
>> fifo_send() is invoked which rewrites the packet to a dummy set of bytes. I
>> tested this and it works. So far so good. But, crucially, if I invoke
>> fifo_send() on some_other_condition() e.g a button press, the FPGA sees
>> nothing on the FIFO. Is it just not possible to manually prod the slave
>> FIFOs like this? Or am I missing something?
>
> Ahh hmm..
> I haven't used slave FIFO so I don't know :(
>
> I suspect you can't fake an out packet though so I think you may be out of luck..
>
> ie it's OK when you send to EP6 from the PC because everything is set up and so when you allow the packet through it works.
>
> AFAIK you can only generate IN packets (and sink OUT packets) using firmware, you can't generate an OUT packet.
>
> If you use GPIF you can manually trigger read&  write waveforms with data the micro chooses using XGPIFSGLDATL/XGPIFSGLDATH.
>
>>
>> // Called once at startup
>> //
>> void main_init(void) {
>>     FIFOPINPOLAR = 0x03;
>>     SYNCDELAY(); CPUCS = bmCLKSPD1;  // 48MHz
>>     SYNCDELAY(); IFCONFIG = (bmIFCLKSRC | bm3048MHZ | bmIFCLKOE |
>> bmSYNCFIFOS);
>>     SYNCDELAY(); REVCTL = (bmDYN_OUT | bmENH_PKT);
>>     SYNCDELAY(); EP6CFG = (bmVALID | bmBULK | bmDOUBLEBUFFERED);
>>     SYNCDELAY(); FIFORESET = bmNAKALL;
>>     SYNCDELAY(); FIFORESET = bmNAKALL | 6;  // Reset EP6
>>     SYNCDELAY(); FIFORESET = 0x00;
>>     SYNCDELAY(); EP6FIFOCFG = 0x00;
>>     SYNCDELAY(); OUTPKTEND = bmSKIP | 6;
>>     SYNCDELAY(); OUTPKTEND = bmSKIP | 6;
>>     sendCount = 0;
>> }
>>
>> // Compose a packet to send on the EP6 FIFO, and commit it.
>> //
>> void fifo_send(void) {
>>     EP6FIFOBUF[0] = 0x00;
>>     EP6FIFOBUF[1] = 0x01;
>>     EP6FIFOBUF[2] = 0x00;
>>     EP6FIFOBUF[3] = 0x00;
>>     EP6FIFOBUF[4] = 0x00;
>>     EP6FIFOBUF[5] = sendCount++;
>>     SYNCDELAY(); EP6BCH = 0;
>>     SYNCDELAY(); EP6BCL = 6;
>> }
>>
>> // Called repeatedly while the device is idle
>> //
>> void main_loop(void) {
>>     if ( !(EP2468STAT&  bmEP6EMPTY) ) {
>>         fifo_send();
>>     }
>>     if ( some_other_condition() ) {
>>         fifo_send();
>>     }
>> }
>>
>>
>> ------------------------------------------------------------------------------
>> Free Software Download: Index, Search&  Analyze Logs and other IT data in
>> Real-Time with Splunk. Collect, index and harness all the fast moving IT data
>> generated by your applications, servers and devices whether physical, virtual
>> or in the cloud. Deliver compliance at lower cost and gain new business
>> insights. http://p.sf.net/sfu/splunk-dev2dev
>> _______________________________________________
>> Fx2lib-devel mailing list
>> Fx2...@pu...
>> https://lists.sourceforge.net/lists/listinfo/fx2lib-devel
>>
>
> --
> Daniel O'Connor software and network engineer
> for Genesis Software - http://www.gsoft.com.au
> "The nice thing about standards is that there
> are so many of them to choose from."
>    -- Andrew Tanenbaum
> GPG Fingerprint - 5596 B766 97C0 0E94 4347 295E E593 DC20 7B3F CE8C
>
>
>
>
>
>
>
> ------------------------------------------------------------------------------
> Free Software Download: Index, Search&  Analyze Logs and other IT data in
> Real-Time with Splunk. Collect, index and harness all the fast moving IT data
> generated by your applications, servers and devices whether physical, virtual
> or in the cloud. Deliver compliance at lower cost and gain new business
> insights. http://p.sf.net/sfu/splunk-dev2dev

Re: [Fx2lib-devel] Change sync delay macro

[Dennis M. <djm...@gm...>]

Thanks Daniel,

Do you have a github repo set up with your changes?  If you do that, 
then you can create a pull request and I can simply merge the changes 
online.

-Dennis

On 02/25/2011 11:05 PM, Daniel O'Connor wrote:
> Hi,
> I modified SYNCDELAY to match the Cypress version - this is useful because GPIF designer generates code which calls it and it's a pain to re-edit the file each time you change your design.
>
>
>
>
>
>
>
> --
> Daniel O'Connor software and network engineer
> for Genesis Software - http://www.gsoft.com.au
> "The nice thing about standards is that there
> are so many of them to choose from."
>    -- Andrew Tanenbaum
> GPG Fingerprint - 5596 B766 97C0 0E94 4347 295E E593 DC20 7B3F CE8C
>
>
>
>
>
>
>
>
>
> ------------------------------------------------------------------------------
> Free Software Download: Index, Search&  Analyze Logs and other IT data in
> Real-Time with Splunk. Collect, index and harness all the fast moving IT data
> generated by your applications, servers and devices whether physical, virtual
> or in the cloud. Deliver compliance at lower cost and gain new business
> insights. http://p.sf.net/sfu/splunk-dev2dev
>
>
>
> _______________________________________________
> Fx2lib-devel mailing list
> Fx2...@pu...
> https://lists.sourceforge.net/lists/listinfo/fx2lib-devel

Re: [Fx2lib-devel] Sourcing slave FIFO packets

[Daniel O'C. <doc...@gs...>]

On 28/02/2011, at 24:49, Chris McClelland wrote:
> Daniel, thanks for your reply. I wasn't clear. It's not the PC that should 
> be getting this data, it is the FPGA attached to the EP6 slave FIFO. Since 
> it's just a FIFO it has no notion of NAK. For example, in the code below I 
> have set AUTOOUT=0. Now whenever an EP6 packet arrives from the host, 
> fifo_send() is invoked which rewrites the packet to a dummy set of bytes. I 
> tested this and it works. So far so good. But, crucially, if I invoke 
> fifo_send() on some_other_condition() e.g a button press, the FPGA sees 
> nothing on the FIFO. Is it just not possible to manually prod the slave 
> FIFOs like this? Or am I missing something?

Ahh hmm..
I haven't used slave FIFO so I don't know :(

I suspect you can't fake an out packet though so I think you may be out of luck..

ie it's OK when you send to EP6 from the PC because everything is set up and so when you allow the packet through it works.

AFAIK you can only generate IN packets (and sink OUT packets) using firmware, you can't generate an OUT packet.

If you use GPIF you can manually trigger read & write waveforms with data the micro chooses using XGPIFSGLDATL/XGPIFSGLDATH.

> 
> // Called once at startup
> //
> void main_init(void) {
>    FIFOPINPOLAR = 0x03;
>    SYNCDELAY(); CPUCS = bmCLKSPD1;  // 48MHz
>    SYNCDELAY(); IFCONFIG = (bmIFCLKSRC | bm3048MHZ | bmIFCLKOE | 
> bmSYNCFIFOS);
>    SYNCDELAY(); REVCTL = (bmDYN_OUT | bmENH_PKT);
>    SYNCDELAY(); EP6CFG = (bmVALID | bmBULK | bmDOUBLEBUFFERED);
>    SYNCDELAY(); FIFORESET = bmNAKALL;
>    SYNCDELAY(); FIFORESET = bmNAKALL | 6;  // Reset EP6
>    SYNCDELAY(); FIFORESET = 0x00;
>    SYNCDELAY(); EP6FIFOCFG = 0x00;
>    SYNCDELAY(); OUTPKTEND = bmSKIP | 6;
>    SYNCDELAY(); OUTPKTEND = bmSKIP | 6;
>    sendCount = 0;
> }
> 
> // Compose a packet to send on the EP6 FIFO, and commit it.
> //
> void fifo_send(void) {
>    EP6FIFOBUF[0] = 0x00;
>    EP6FIFOBUF[1] = 0x01;
>    EP6FIFOBUF[2] = 0x00;
>    EP6FIFOBUF[3] = 0x00;
>    EP6FIFOBUF[4] = 0x00;
>    EP6FIFOBUF[5] = sendCount++;
>    SYNCDELAY(); EP6BCH = 0;
>    SYNCDELAY(); EP6BCL = 6;
> }
> 
> // Called repeatedly while the device is idle
> //
> void main_loop(void) {
>    if ( !(EP2468STAT & bmEP6EMPTY) ) {
>        fifo_send();
>    }
>    if ( some_other_condition() ) {
>        fifo_send();
>    }
> } 
> 
> 
> ------------------------------------------------------------------------------
> Free Software Download: Index, Search & Analyze Logs and other IT data in 
> Real-Time with Splunk. Collect, index and harness all the fast moving IT data 
> generated by your applications, servers and devices whether physical, virtual
> or in the cloud. Deliver compliance at lower cost and gain new business 
> insights. http://p.sf.net/sfu/splunk-dev2dev 
> _______________________________________________
> Fx2lib-devel mailing list
> Fx2...@li...
> https://lists.sourceforge.net/lists/listinfo/fx2lib-devel
> 

--
Daniel O'Connor software and network engineer
for Genesis Software - http://www.gsoft.com.au
"The nice thing about standards is that there
are so many of them to choose from."
  -- Andrew Tanenbaum
GPG Fingerprint - 5596 B766 97C0 0E94 4347 295E E593 DC20 7B3F CE8C

Re: [Fx2lib-devel] Sourcing slave FIFO packets

[Chris M. <fx...@m3...>]

Daniel, thanks for your reply. I wasn't clear. It's not the PC that should 
be getting this data, it is the FPGA attached to the EP6 slave FIFO. Since 
it's just a FIFO it has no notion of NAK. For example, in the code below I 
have set AUTOOUT=0. Now whenever an EP6 packet arrives from the host, 
fifo_send() is invoked which rewrites the packet to a dummy set of bytes. I 
tested this and it works. So far so good. But, crucially, if I invoke 
fifo_send() on some_other_condition() e.g a button press, the FPGA sees 
nothing on the FIFO. Is it just not possible to manually prod the slave 
FIFOs like this? Or am I missing something?

// Called once at startup
//
void main_init(void) {
    FIFOPINPOLAR = 0x03;
    SYNCDELAY(); CPUCS = bmCLKSPD1;  // 48MHz
    SYNCDELAY(); IFCONFIG = (bmIFCLKSRC | bm3048MHZ | bmIFCLKOE | 
bmSYNCFIFOS);
    SYNCDELAY(); REVCTL = (bmDYN_OUT | bmENH_PKT);
    SYNCDELAY(); EP6CFG = (bmVALID | bmBULK | bmDOUBLEBUFFERED);
    SYNCDELAY(); FIFORESET = bmNAKALL;
    SYNCDELAY(); FIFORESET = bmNAKALL | 6;  // Reset EP6
    SYNCDELAY(); FIFORESET = 0x00;
    SYNCDELAY(); EP6FIFOCFG = 0x00;
    SYNCDELAY(); OUTPKTEND = bmSKIP | 6;
    SYNCDELAY(); OUTPKTEND = bmSKIP | 6;
    sendCount = 0;
}

// Compose a packet to send on the EP6 FIFO, and commit it.
//
void fifo_send(void) {
    EP6FIFOBUF[0] = 0x00;
    EP6FIFOBUF[1] = 0x01;
    EP6FIFOBUF[2] = 0x00;
    EP6FIFOBUF[3] = 0x00;
    EP6FIFOBUF[4] = 0x00;
    EP6FIFOBUF[5] = sendCount++;
    SYNCDELAY(); EP6BCH = 0;
    SYNCDELAY(); EP6BCL = 6;
}

// Called repeatedly while the device is idle
//
void main_loop(void) {
    if ( !(EP2468STAT & bmEP6EMPTY) ) {
        fifo_send();
    }
    if ( some_other_condition() ) {
        fifo_send();
    }
} 

Re: [Fx2lib-devel] Sourcing slave FIFO packets

[Daniel O'C. <doc...@gs...>]

On 27/02/2011, at 5:12, Chris McClelland wrote:
> Has anyone managed to get an FX2 firmware to compose a packet from scratch 
> (i.e not originating from the USB host) and commit it to a slave FIFO? The 
> TRM seems to suggest it's only necessary to disable AUTOOUT, write the 
> packet to the endpoint buffer (e.g EP6FIFOBUF) and writing the packet length 
> to EP6BCH/EP6BCL, but I can't get it to work. See http://bit.ly/hQxIR1 for 
> example code (composes a packet to send to EP6's FIFO when a certain control 
> request is received).

In my code EP1 in/out and EP6/EP8 have data sent to/from the micro directly and EP2 sends data to the PC using GPIF.

I think your problem is that you set the NAK bit in the FIFO and then try and send it so I presume the PC is seeing a NAK reply from the FX2 instead of getting your data packet.

--
Daniel O'Connor software and network engineer
for Genesis Software - http://www.gsoft.com.au
"The nice thing about standards is that there
are so many of them to choose from."
  -- Andrew Tanenbaum
GPG Fingerprint - 5596 B766 97C0 0E94 4347 295E E593 DC20 7B3F CE8C

[Fx2lib-devel] Sourcing slave FIFO packets

[Chris M. <fx...@m3...>]

Has anyone managed to get an FX2 firmware to compose a packet from scratch 
(i.e not originating from the USB host) and commit it to a slave FIFO? The 
TRM seems to suggest it's only necessary to disable AUTOOUT, write the 
packet to the endpoint buffer (e.g EP6FIFOBUF) and writing the packet length 
to EP6BCH/EP6BCL, but I can't get it to work. See http://bit.ly/hQxIR1 for 
example code (composes a packet to send to EP6's FIFO when a certain control 
request is received).

Chris 

[Fx2lib-devel] Get 'make' and 'make clean' to work from the top of the tree

[Daniel O'C. <doc...@gs...>]

This lets you build or clean everything from the top of the tree.

[Fx2lib-devel] Change sync delay macro

[Daniel O'C. <doc...@gs...>]

Hi,
I modified SYNCDELAY to match the Cypress version - this is useful because GPIF designer generates code which calls it and it's a pain to re-edit the file each time you change your design.


 

[Fx2lib-devel] Toggle reset on clearing stall

[Daniel O'C. <doc...@gs...>]

Hi,
I found a bug in handle_clear_feature - it doesn't reset the data toggle which can cause confusion in some cases.

The following diff appears to fix it for me

diff --git a/lib/setupdat.c b/lib/setupdat.c
index f4c217a..c3d7c43 100644
--- a/lib/setupdat.c
+++ b/lib/setupdat.c
@@ -216,6 +216,7 @@ BOOL handle_clear_feature() {
         __xdata BYTE* pep=ep_addr(SETUPDAT[4]);
         printf ( "unstall endpoint %02X\n" , SETUPDAT[4] );
         *pep &= ~bmEPSTALL;        
+       RESETTOGGLE(SETUPDAT[4]);
     } else {
         printf ( "unsupported ep feature %02x", SETUPDAT[2] );
         return FALSE;

--
Daniel O'Connor software and network engineer
for Genesis Software - http://www.gsoft.com.au
"The nice thing about standards is that there
are so many of them to choose from."
  -- Andrew Tanenbaum
GPG Fingerprint - 5596 B766 97C0 0E94 4347 295E E593 DC20 7B3F CE8C

Re: [Fx2lib-devel] Issues with I2C

[Dennis M. <djm...@gm...>]

I guess I never did upload a build to sourceforge.  I do recommend that 
everyone uses 0.2 though.

On 01/05/2011 01:18 PM, Kevin Johnson wrote:
> Thanks. I was looking for a confirmation there were alot of improvements
> to the I2C before going through the process of figuring out how to
> access git. I am sure it isn't hard I just had never done it before and
> was trying to avoid getting side tracked if it didn't make sense.
>
> Kevin
>
>
> --- On *Wed, 1/5/11, Dennis Muhlestein
> /<djm...@pu...>/* wrote:
>
>
>     From: Dennis Muhlestein
>     <djm...@pu...>
>     Subject: Re: [Fx2lib-devel] Issues with I2C
>     To: fx2...@li...
>     Date: Wednesday, January 5, 2011, 5:51 PM
>
>     Try the latest sources from git. Github has an 0.2 build if I recall.
>     Or at least an 0.2 download. I think there were a lot of i2c fixes
>     between 0.1 and 0.2.
>
>     -Dennis
>
>     On 01/05/2011 10:10 AM, Kevin Johnson wrote:
>      > I am using a Cypress 68013 and a 16 kB 3.3 V EEPROM. The fx2lib I2C
>      > functions to read the EEPROM isn't working, but if I manually
>     send the
>      > I2C read and write functions as shown in the read and write
>     EEPROM code,
>      > but with a delay in between it works fine.
>      >
>      > Has anyone else seen this? I am only running at 100kHz.
>      >
>      > I am using fx2lib-src-0.1.tgz which looks like the only released
>      > version. I am using the newest version of SDCC so I had to go through
>      > and update everything. Didn't know about the github until already had
>      > done it.
>      >
>      > I am just interested if there were any other fixes to I2C that I
>     might
>      > not have roled in or if I should just add a SYNCDELAY in my
>     fx2lib and
>      > rebuild it.
>      >
>      > When I check with a scope, what I notice is that without the
>     delay the
>      > first 1 to 3 reads are NACKed. It retries and eventually gets a
>     ACK when
>      > the part is ready but the randomness of when it gets acked means that
>      > the data is random. If I add the delay the first read is ALWAYS acked
>      > and the data is always good.
>      >
>      > Thanks
>      >
>      > Kevin
>      >
>      >
>      >
>      >
>      >
>     ------------------------------------------------------------------------------
>      > Learn how Oracle Real Application Clusters (RAC) One Node allows
>     customers
>      > to consolidate database storage, standardize their database
>     environment, and,
>      > should the need arise, upgrade to a full multi-node Oracle RAC
>     database
>      > without downtime or disruption
>      > http://p.sf.net/sfu/oracle-sfdevnl
>      >
>      >
>      >
>      > _______________________________________________
>      > Fx2lib-devel mailing list
>      >
>     Fx2lib-...@pu...
>     </mc/compose?to=LV9...@pu...>
>      > https://lists.sourceforge.net/lists/listinfo/fx2lib-devel
>
>
>     ------------------------------------------------------------------------------
>     Learn how Oracle Real Application Clusters (RAC) One Node allows
>     customers
>     to consolidate database storage, standardize their database
>     environment, and,
>     should the need arise, upgrade to a full multi-node Oracle RAC database
>     without downtime or disruption
>     http://p.sf.net/sfu/oracle-sfdevnl
>     _______________________________________________
>     Fx2lib-devel mailing list
>     Fx2...@li...
>     </mc/compose?to=Fx2...@pu...>
>     https://lists.sourceforge.net/lists/listinfo/fx2lib-devel
>
>
>
>
> ------------------------------------------------------------------------------
> Learn how Oracle Real Application Clusters (RAC) One Node allows customers
> to consolidate database storage, standardize their database environment, and,
> should the need arise, upgrade to a full multi-node Oracle RAC database
> without downtime or disruption
> http://p.sf.net/sfu/oracle-sfdevnl
>
>
>
> _______________________________________________
> Fx2lib-devel mailing list
> Fx2...@pu...
> https://lists.sourceforge.net/lists/listinfo/fx2lib-devel

Re: [Fx2lib-devel] Issues with I2C

[Kevin J. <kjp...@ya...>]

Thanks.  I was looking for a confirmation there were  alot of improvements to the I2C before going through the process of figuring out how to access git.  I am sure it isn't hard I just had never done it before and was trying to avoid getting side tracked if it didn't make sense.

Kevin


--- On Wed, 1/5/11, Dennis Muhlestein <djm...@gm...> wrote:

From: Dennis Muhlestein <djm...@gm...>
Subject: Re: [Fx2lib-devel] Issues with I2C
To: fx2...@li...
Date: Wednesday, January 5, 2011, 5:51 PM

Try the latest sources from git.  Github has an 0.2 build if I recall. 
Or at least an 0.2 download.  I think there were a lot of i2c fixes 
between 0.1 and 0.2.

-Dennis

On 01/05/2011 10:10 AM, Kevin Johnson wrote:
> I am using a Cypress 68013 and a 16 kB 3.3 V EEPROM. The fx2lib I2C
> functions to read the EEPROM isn't working, but if I manually send the
> I2C read and write functions as shown in the read and write EEPROM code,
> but with a delay in between it works fine.
>
> Has anyone else seen this? I am only running at 100kHz.
>
> I am using fx2lib-src-0.1.tgz which looks like the only released
> version. I am using the newest version of SDCC so I had to go through
> and update everything. Didn't know about the github until already had
> done it.
>
> I am just interested if there were any other fixes to I2C that I might
> not have roled in or if I should just add a SYNCDELAY in my fx2lib and
> rebuild it.
>
> When I check with a scope, what I notice is that without the delay the
> first 1 to 3 reads are NACKed. It retries and eventually gets a ACK when
> the part is ready but the randomness of when it gets acked means that
> the data is random. If I add the delay the first read is ALWAYS acked
> and the data is always good.
>
> Thanks
>
> Kevin
>
>
>
>
> ------------------------------------------------------------------------------
> Learn how Oracle Real Application Clusters (RAC) One Node allows customers
> to consolidate database storage, standardize their database environment, and,
> should the need arise, upgrade to a full multi-node Oracle RAC database
> without downtime or disruption
> http://p.sf.net/sfu/oracle-sfdevnl
>
>
>
> _______________________________________________
> Fx2lib-devel mailing list
> Fx2...@pu...
> https://lists.sourceforge.net/lists/listinfo/fx2lib-devel


------------------------------------------------------------------------------
Learn how Oracle Real Application Clusters (RAC) One Node allows customers
to consolidate database storage, standardize their database environment, and, 
should the need arise, upgrade to a full multi-node Oracle RAC database 
without downtime or disruption
http://p.sf.net/sfu/oracle-sfdevnl
_______________________________________________
Fx2lib-devel mailing list
Fx2...@li...
https://lists.sourceforge.net/lists/listinfo/fx2lib-devel



      

Re: [Fx2lib-devel] Issues with I2C

[Dennis M. <djm...@gm...>]

Try the latest sources from git.  Github has an 0.2 build if I recall. 
Or at least an 0.2 download.  I think there were a lot of i2c fixes 
between 0.1 and 0.2.

-Dennis

On 01/05/2011 10:10 AM, Kevin Johnson wrote:
> I am using a Cypress 68013 and a 16 kB 3.3 V EEPROM. The fx2lib I2C
> functions to read the EEPROM isn't working, but if I manually send the
> I2C read and write functions as shown in the read and write EEPROM code,
> but with a delay in between it works fine.
>
> Has anyone else seen this? I am only running at 100kHz.
>
> I am using fx2lib-src-0.1.tgz which looks like the only released
> version. I am using the newest version of SDCC so I had to go through
> and update everything. Didn't know about the github until already had
> done it.
>
> I am just interested if there were any other fixes to I2C that I might
> not have roled in or if I should just add a SYNCDELAY in my fx2lib and
> rebuild it.
>
> When I check with a scope, what I notice is that without the delay the
> first 1 to 3 reads are NACKed. It retries and eventually gets a ACK when
> the part is ready but the randomness of when it gets acked means that
> the data is random. If I add the delay the first read is ALWAYS acked
> and the data is always good.
>
> Thanks
>
> Kevin
>
>
>
>
> ------------------------------------------------------------------------------
> Learn how Oracle Real Application Clusters (RAC) One Node allows customers
> to consolidate database storage, standardize their database environment, and,
> should the need arise, upgrade to a full multi-node Oracle RAC database
> without downtime or disruption
> http://p.sf.net/sfu/oracle-sfdevnl
>
>
>
> _______________________________________________
> Fx2lib-devel mailing list
> Fx2...@pu...
> https://lists.sourceforge.net/lists/listinfo/fx2lib-devel

[Fx2lib-devel] Issues with I2C

[Kevin J. <kjp...@ya...>]

I am using a Cypress 68013 and a 16 kB 3.3 V EEPROM.  The fx2lib I2C functions to read the EEPROM isn't working, but if I manually send the I2C read and write functions as shown in the read and write EEPROM code, but with a delay in between it works fine.

Has anyone else seen this?  I am only running at 100kHz.

I am using fx2lib-src-0.1.tgz which looks like the only released version.  I am using the newest version of SDCC so I had to go through and update everything.  Didn't know about the github until already had done it.

I am just interested if there were any other fixes to I2C that I might not have roled in or if I should just add a SYNCDELAY in my fx2lib and rebuild it.

When I check with a scope, what I notice is that without the delay the first 1 to 3 reads are NACKed.  It retries and eventually gets a ACK when the part is ready but the randomness of when it gets acked means that the data is random.  If I add the delay the first read is ALWAYS acked and the data is always good.

Thanks 

Kevin




      

Re: [Fx2lib-devel] Fwd: mini box

[Dennis M. <de...@ub...>]

>
> Unfortunately that doesn't give you anything about how it's  
> connected to the FX2 :(
>
> That said the datasheet example circuit doesn't look too complex as  
> the control of the chip is via I2C.
>

If it is connected on the i2c bus then it should be pretty trivial to  
read/write i2c data/commands with the fx2lib i2c functions.  Good luck!

Re: [Fx2lib-devel] Fwd: mini box

[Daniel O'C. <doc...@gs...>]

On 21/09/2010, at 6:30, Novice Virupa wrote:
> about talking to the Cypress chip on the Mini Box TV ( see Linux TV
> org link to picture ) system um how do you find out information about
> the A/D converter or other chips that may be on the board to continue
> the conversation :)

Google the part numbers :)
http://www.alldatasheet.com/datasheet-pdf/pdf/110117/ZARLINK/ZL10353.html

Unfortunately that doesn't give you anything about how it's connected to the FX2 :(

That said the datasheet example circuit doesn't look too complex as the control of the chip is via I2C.

--
Daniel O'Connor software and network engineer
for Genesis Software - http://www.gsoft.com.au
"The nice thing about standards is that there
are so many of them to choose from."
  -- Andrew Tanenbaum
GPG Fingerprint - 5596 B766 97C0 0E94 4347 295E E593 DC20 7B3F CE8C

[Fx2lib-devel] Fwd: mini box

[Novice V. <ven...@gm...>]

Here is my page with some shots of some communication with cycfx2prog, lsusb etc
https://sites.google.com/site/venvirupa/tvusb

 about talking to the Cypress chip on the Mini Box TV ( see Linux TV
org link to picture ) system um how do you find out information about
the A/D converter or other chips that may be on the board to continue
the conversation :)

Thanks

[Fx2lib-devel] mini box

[Novice V. <ven...@gm...>]

Here is my page with some shots of some communication with cycfx2prog, lsusb etc
https://sites.google.com/site/venvirupa/tvusb

 about talking to the Cypress chip on the Mini Box TV ( see Linux TV
org link to picture ) system um how do you find out information about
the A/D converter or other chips that may be on the board to continue
the conversation :)

Thanks

Re: [Fx2lib-devel] fx2lib @ gmane.org

[Trygve L. <tr...@in...>]

On 8/26/10 4:30 AM, Dennis Muhlestein wrote:
> OK, you can subscribe on gmane now:
> gmane.comp.embedded.fx2lib.devel

Sweet, works like a charm!

--
Trygve

> On Aug 25, 2010, at 8:25 PM, Dennis Muhlestein wrote:
>
>> I got a notification back that the group would be created as soon as a
>> message is posted.  Guess we'll see...
>>
>> On Aug 23, 2010, at 6:29 AM, Dennis Muhlestein wrote:
>>
>>> Good idea.  I made the request.  Once accepted/added I'll fill out
>>> the
>>> archive request too.
>>>
>>> On Aug 23, 2010, at 6:12 AM, Trygve Laugstøl wrote:
>>>
>>>> Hi Dennis and list
>>>>
>>>> I was wondering if you could request this list to be added to
>>>> gmane.org,
>>>> and ideally the archives too. If you want I can do the request for
>>>> you
>>>> but they prefer if the list owner performs the request.
>>>>
>>>> Gmane.org is a mail to news gateway which makes it easier for those
>>>> of
>>>> us who like to like read lists with news. Their web archive reader
>>>> is
>>>> also way better and much faster than Sourceforge's own stuff.
>>>>
>>>> See http://gmane.org/subscribe.php and http://gmane.org/import.php
>>>>
>>>> --
>>>> Trygve
>>>>
>>>> ------------------------------------------------------------------------------
>>>> This SF.net email is sponsored by
>>>>
>>>> Make an app they can't live without
>>>> Enter the BlackBerry Developer Challenge
>>>> http://p.sf.net/sfu/RIM-dev2dev
>>>> _______________________________________________
>>>> Fx2lib-devel mailing list
>>>> Fx2...@li...
>>>> https://lists.sourceforge.net/lists/listinfo/fx2lib-devel
>>>
>>>
>>> ------------------------------------------------------------------------------
>>> This SF.net email is sponsored by
>>>
>>> Make an app they can't live without
>>> Enter the BlackBerry Developer Challenge
>>> http://p.sf.net/sfu/RIM-dev2dev
>>> _______________________________________________
>>> Fx2lib-devel mailing list
>>> Fx2...@li...
>>> https://lists.sourceforge.net/lists/listinfo/fx2lib-devel
>>
>>
>> ------------------------------------------------------------------------------
>> Sell apps to millions through the Intel(R) Atom(Tm) Developer Program
>> Be part of this innovative community and reach millions of netbook
>> users
>> worldwide. Take advantage of special opportunities to increase
>> revenue and
>> speed time-to-market. Join now, and jumpstart your future.
>> http://p.sf.net/sfu/intel-atom-d2d
>> _______________________________________________
>> Fx2lib-devel mailing list
>> Fx2...@li...
>> https://lists.sourceforge.net/lists/listinfo/fx2lib-devel
>
>
> ------------------------------------------------------------------------------
> Sell apps to millions through the Intel(R) Atom(Tm) Developer Program
> Be part of this innovative community and reach millions of netbook users
> worldwide. Take advantage of special opportunities to increase revenue and
> speed time-to-market. Join now, and jumpstart your future.
> http://p.sf.net/sfu/intel-atom-d2d
> _______________________________________________
> Fx2lib-devel mailing list
> Fx2...@li...
> https://lists.sourceforge.net/lists/listinfo/fx2lib-devel

Re: [Fx2lib-devel] fx2lib @ gmane.org

[Dennis M. <de...@ub...>]

OK, you can subscribe on gmane now:
gmane.comp.embedded.fx2lib.devel

On Aug 25, 2010, at 8:25 PM, Dennis Muhlestein wrote:

> I got a notification back that the group would be created as soon as a
> message is posted.  Guess we'll see...
>
> On Aug 23, 2010, at 6:29 AM, Dennis Muhlestein wrote:
>
>> Good idea.  I made the request.  Once accepted/added I'll fill out  
>> the
>> archive request too.
>>
>> On Aug 23, 2010, at 6:12 AM, Trygve Laugstøl wrote:
>>
>>> Hi Dennis and list
>>>
>>> I was wondering if you could request this list to be added to
>>> gmane.org,
>>> and ideally the archives too. If you want I can do the request for
>>> you
>>> but they prefer if the list owner performs the request.
>>>
>>> Gmane.org is a mail to news gateway which makes it easier for those
>>> of
>>> us who like to like read lists with news. Their web archive reader  
>>> is
>>> also way better and much faster than Sourceforge's own stuff.
>>>
>>> See http://gmane.org/subscribe.php and http://gmane.org/import.php
>>>
>>> --
>>> Trygve
>>>
>>> ------------------------------------------------------------------------------
>>> This SF.net email is sponsored by
>>>
>>> Make an app they can't live without
>>> Enter the BlackBerry Developer Challenge
>>> http://p.sf.net/sfu/RIM-dev2dev
>>> _______________________________________________
>>> Fx2lib-devel mailing list
>>> Fx2...@li...
>>> https://lists.sourceforge.net/lists/listinfo/fx2lib-devel
>>
>>
>> ------------------------------------------------------------------------------
>> This SF.net email is sponsored by
>>
>> Make an app they can't live without
>> Enter the BlackBerry Developer Challenge
>> http://p.sf.net/sfu/RIM-dev2dev
>> _______________________________________________
>> Fx2lib-devel mailing list
>> Fx2...@li...
>> https://lists.sourceforge.net/lists/listinfo/fx2lib-devel
>
>
> ------------------------------------------------------------------------------
> Sell apps to millions through the Intel(R) Atom(Tm) Developer Program
> Be part of this innovative community and reach millions of netbook  
> users
> worldwide. Take advantage of special opportunities to increase  
> revenue and
> speed time-to-market. Join now, and jumpstart your future.
> http://p.sf.net/sfu/intel-atom-d2d
> _______________________________________________
> Fx2lib-devel mailing list
> Fx2...@li...
> https://lists.sourceforge.net/lists/listinfo/fx2lib-devel

Re: [Fx2lib-devel] fx2lib @ gmane.org

[Dennis M. <de...@ub...>]

I got a notification back that the group would be created as soon as a  
message is posted.  Guess we'll see...

On Aug 23, 2010, at 6:29 AM, Dennis Muhlestein wrote:

> Good idea.  I made the request.  Once accepted/added I'll fill out the
> archive request too.
>
> On Aug 23, 2010, at 6:12 AM, Trygve Laugstøl wrote:
>
>> Hi Dennis and list
>>
>> I was wondering if you could request this list to be added to
>> gmane.org,
>> and ideally the archives too. If you want I can do the request for  
>> you
>> but they prefer if the list owner performs the request.
>>
>> Gmane.org is a mail to news gateway which makes it easier for those  
>> of
>> us who like to like read lists with news. Their web archive reader is
>> also way better and much faster than Sourceforge's own stuff.
>>
>> See http://gmane.org/subscribe.php and http://gmane.org/import.php
>>
>> --
>> Trygve
>>
>> ------------------------------------------------------------------------------
>> This SF.net email is sponsored by
>>
>> Make an app they can't live without
>> Enter the BlackBerry Developer Challenge
>> http://p.sf.net/sfu/RIM-dev2dev
>> _______________________________________________
>> Fx2lib-devel mailing list
>> Fx2...@li...
>> https://lists.sourceforge.net/lists/listinfo/fx2lib-devel
>
>
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by
>
> Make an app they can't live without
> Enter the BlackBerry Developer Challenge
> http://p.sf.net/sfu/RIM-dev2dev
> _______________________________________________
> Fx2lib-devel mailing list
> Fx2...@li...
> https://lists.sourceforge.net/lists/listinfo/fx2lib-devel

Re: [Fx2lib-devel] fx2lib @ gmane.org

[Dennis M. <de...@ub...>]

Good idea.  I made the request.  Once accepted/added I'll fill out the  
archive request too.

On Aug 23, 2010, at 6:12 AM, Trygve Laugstøl wrote:

> Hi Dennis and list
>
> I was wondering if you could request this list to be added to  
> gmane.org,
> and ideally the archives too. If you want I can do the request for you
> but they prefer if the list owner performs the request.
>
> Gmane.org is a mail to news gateway which makes it easier for those of
> us who like to like read lists with news. Their web archive reader is
> also way better and much faster than Sourceforge's own stuff.
>
> See http://gmane.org/subscribe.php and http://gmane.org/import.php
>
> --
> Trygve
>
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by
>
> Make an app they can't live without
> Enter the BlackBerry Developer Challenge
> http://p.sf.net/sfu/RIM-dev2dev
> _______________________________________________
> Fx2lib-devel mailing list
> Fx2...@li...
> https://lists.sourceforge.net/lists/listinfo/fx2lib-devel

Re: [Fx2lib-devel] Patches for fx2lib

[Trygve L. <tr...@in...>]

On 8/23/10 3:12 PM, Dennis Muhlestein wrote:
> Thanks, I'll try and pull some of this in when I get a moment.  Daniel
> O'Connor also made some changes for fixing the deprecations which I
> haven't pulled in yet.

Yeah, I didn't find that mail before I suddenly discovered this list. I 
noticed that the change log claims that the "new" __sfr form should work 
with the previous releases so it should be safe to pull in the changes 
and it should still build properly with 2.9.0.

I would like some comments on my stuff for writing the descriptors in C 
when you have some time though.

My javax.usb firmware needs to create a whole lot of different 
descriptors so it would be nice to make the descriptors use pointers etc.

--
Trygve

> -Dennis
>
> On Aug 23, 2010, at 7:06 AM, Trygve Laugstøl wrote:
>
>> Howdy
>>
>> I've made a two patches for fx2lib and I'm working on another one but
>> I'd like some feedback before going all the way.
>>
>> The first patch fixes a small issue when building fx2lib on Solaris
>> which doesn't have GNU Make as "make".
>>
>> The second patch fixes compatibility the upcoming 3.0 release of SDCC.
>> SDCC 2.9.0+ also give a lot of warnings as fx2lib is currently using a
>> lot of deprecated symbols when building with 2.9.7.
>>
>> The third patch is a work in progress, but should make it even
>> easier to
>> use fx2lib as I think most people enjoy writing C code over assembly.
>> I'm willing to do all of the work if it sounds useful and would be
>> accepted upstream. The code I'm using fx2lib which uses the changes
>> I've
>> outlined with is available on [1].
>>
>> All the patches are available on github, but I've detailed the patches
>> on my own wiki [2].
>>
>> [1]: http://github.com/trygvis/javax-usb-libusb1/tree/master/firmwares
>> [2]: http://cobain.arktekk.no/~trygvis/wiki/Fx2lib
>>
>> --
>> Trygve
>>
>> ------------------------------------------------------------------------------
>> This SF.net email is sponsored by
>>
>> Make an app they can't live without
>> Enter the BlackBerry Developer Challenge
>> http://p.sf.net/sfu/RIM-dev2dev
>> _______________________________________________
>> Fx2lib-devel mailing list
>> Fx2...@li...
>> https://lists.sourceforge.net/lists/listinfo/fx2lib-devel
>
>
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by
>
> Make an app they can't live without
> Enter the BlackBerry Developer Challenge
> http://p.sf.net/sfu/RIM-dev2dev
> _______________________________________________
> Fx2lib-devel mailing list
> Fx2...@li...
> https://lists.sourceforge.net/lists/listinfo/fx2lib-devel

Re: [Fx2lib-devel] Patches for fx2lib

[Dennis M. <de...@ub...>]

Thanks, I'll try and pull some of this in when I get a moment.  Daniel  
O'Connor also made some changes for fixing the deprecations which I  
haven't pulled in yet.

-Dennis

On Aug 23, 2010, at 7:06 AM, Trygve Laugstøl wrote:

> Howdy
>
> I've made a two patches for fx2lib and I'm working on another one but
> I'd like some feedback before going all the way.
>
> The first patch fixes a small issue when building fx2lib on Solaris
> which doesn't have GNU Make as "make".
>
> The second patch fixes compatibility the upcoming 3.0 release of SDCC.
> SDCC 2.9.0+ also give a lot of warnings as fx2lib is currently using a
> lot of deprecated symbols when building with 2.9.7.
>
> The third patch is a work in progress, but should make it even  
> easier to
> use fx2lib as I think most people enjoy writing C code over assembly.
> I'm willing to do all of the work if it sounds useful and would be
> accepted upstream. The code I'm using fx2lib which uses the changes  
> I've
> outlined with is available on [1].
>
> All the patches are available on github, but I've detailed the patches
> on my own wiki [2].
>
> [1]: http://github.com/trygvis/javax-usb-libusb1/tree/master/firmwares
> [2]: http://cobain.arktekk.no/~trygvis/wiki/Fx2lib
>
> --
> Trygve
>
> ------------------------------------------------------------------------------
> This SF.net email is sponsored by
>
> Make an app they can't live without
> Enter the BlackBerry Developer Challenge
> http://p.sf.net/sfu/RIM-dev2dev
> _______________________________________________
> Fx2lib-devel mailing list
> Fx2...@li...
> https://lists.sourceforge.net/lists/listinfo/fx2lib-devel