search

lowRISC / lowrisc-chip

The root repo for lowRISC project and FPGA demos.
http://www.lowrisc.org/
Other
596 stars 148 forks source link

Eth 8packets #102

Open clare7 opened 5 years ago

clare7 commented 5 years ago

Hi, I'm wondering what does the value {3'H7,{7{3'H5}}} indicates? It was found in Line 98

 ` if ((rx_nxt == {3'H7,{7{3'H5}}}) && (byte_sync == 0) && (sync == 0))`

Based on the codes given, it seems like is checking the end of 8th (collected) RX packets which each packet is 3 bits size. However, I couldnt figure out the how did it comes to value H7 and H5?

jrrk commented 5 years ago

In this case the coding is RMII, so it is checking two data bits and receive enable.

Sent from my iPhone

On 24 Dec 2018, at 05:39, clare7 notifications@github.com wrote:

Hi, I'm wondering what does the value {3'H7,{7{3'H5}}} indicates? It was found in Line 98

if ((rx_nxt == {3'H7,{7{3'H5}}}) && (byte_sync == 0) && (sync == 0)) Based on the codes given, it seems like is checking the end of 8th (collected) RX packets which each packet is 3 bits size. However, I couldnt figure out the how did it comes to value H7 and H5?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or mute the thread.

clare7 commented 5 years ago

In that case, why does it has to be 7 of ,{3'H5} ? In order to check for the SFD 1101 0101 isnt that 3 of {3'H5} will suffice for coding in RMII?

On Mon, Dec 24, 2018 at 3:52 PM jrrk notifications@github.com wrote:

In this case the coding is RMII, so it is checking two data bits and receive enable.

Sent from my iPhone

On 24 Dec 2018, at 05:39, clare7 notifications@github.com wrote:

Hi, I'm wondering what does the value {3'H7,{7{3'H5}}} indicates? It was found in Line 98

if ((rx_nxt == {3'H7,{7{3'H5}}}) && (byte_sync == 0) && (sync == 0)) Based on the codes given, it seems like is checking the end of 8th (collected) RX packets which each packet is 3 bits size. However, I couldnt figure out the how did it comes to value H7 and H5?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or mute the thread.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/lowRISC/lowrisc-chip/issues/102#issuecomment-449698888, or mute the thread https://github.com/notifications/unsubscribe-auth/AcHzHLYMMmp5XSBN3x3VVofBZyYui36Pks5u8Ie5gaJpZM4ZgJRm .

-- Regards, Ming

jrrk commented 5 years ago

You could certainly adapt the code to only check a single byte but thereby the chance of a false sync could be increased. Bearing in mind the chance of a bit error is greater early on in the packet due to the possibility of contention from multiple transmitters.

Sent from my iPhone

On 26 Dec 2018, at 04:56, clare7 notifications@github.com wrote:

In that case, why does it has to be 7 of ,{3'H5} ? In order to check for the SFD 1101 0101 isnt that 3 of {3'H5} will suffice for coding in RMII?

On Mon, Dec 24, 2018 at 3:52 PM jrrk notifications@github.com wrote:

In this case the coding is RMII, so it is checking two data bits and receive enable.

Sent from my iPhone

On 24 Dec 2018, at 05:39, clare7 notifications@github.com wrote:

Hi, I'm wondering what does the value {3'H7,{7{3'H5}}} indicates? It was found in Line 98

if ((rx_nxt == {3'H7,{7{3'H5}}}) && (byte_sync == 0) && (sync == 0)) Based on the codes given, it seems like is checking the end of 8th (collected) RX packets which each packet is 3 bits size. However, I couldnt figure out the how did it comes to value H7 and H5?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or mute the thread.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/lowRISC/lowrisc-chip/issues/102#issuecomment-449698888, or mute the thread https://github.com/notifications/unsubscribe-auth/AcHzHLYMMmp5XSBN3x3VVofBZyYui36Pks5u8Ie5gaJpZM4ZgJRm .

-- Regards, Ming — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or mute the thread.

clare7 commented 5 years ago

Noted. Thank you. I also notice that there is no signal that triggers tx_axis_tready . Did I miss out something?

if (tx_axis_tready) begin tx_frame_addr <= tx_frame_addr + 1; tx_axis_tlast <= (tx_frame_addr == tx_packet_length-2) & tx_axis_tvalid_dly; end

On Wed, Dec 26, 2018 at 8:36 PM jrrk notifications@github.com wrote:

You could certainly adapt the code to only check a single byte but thereby the chance of a false sync could be increased. Bearing in mind the chance of a bit error is greater early on in the packet due to the possibility of contention from multiple transmitters.

Sent from my iPhone

On 26 Dec 2018, at 04:56, clare7 notifications@github.com wrote:

In that case, why does it has to be 7 of ,{3'H5} ? In order to check for the SFD 1101 0101 isnt that 3 of {3'H5} will suffice for coding in RMII?

On Mon, Dec 24, 2018 at 3:52 PM jrrk notifications@github.com wrote:

In this case the coding is RMII, so it is checking two data bits and receive enable.

Sent from my iPhone

On 24 Dec 2018, at 05:39, clare7 notifications@github.com wrote:

Hi, I'm wondering what does the value {3'H7,{7{3'H5}}} indicates? It was found in Line 98

if ((rx_nxt == {3'H7,{7{3'H5}}}) && (byte_sync == 0) && (sync == 0)) Based on the codes given, it seems like is checking the end of 8th (collected) RX packets which each packet is 3 bits size. However, I couldnt figure out the how did it comes to value H7 and H5?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or mute the thread.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub < https://github.com/lowRISC/lowrisc-chip/issues/102#issuecomment-449698888 , or mute the thread < https://github.com/notifications/unsubscribe-auth/AcHzHLYMMmp5XSBN3x3VVofBZyYui36Pks5u8Ie5gaJpZM4ZgJRm

.

-- Regards, Ming — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or mute the thread.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/lowRISC/lowrisc-chip/issues/102#issuecomment-449960842, or mute the thread https://github.com/notifications/unsubscribe-auth/AcHzHCUjAyK87XaO5V1-ASXD8PRTIMpmks5u821egaJpZM4ZgJRm .

-- Regards, Ming

jrrk commented 5 years ago

You did. The port  input_axis_tready of axis_gmii_tx.v is actually an output despite its name.

This indicates in a convoluted way when the transmitter is ready for more input. If you have

issues with the coding style you could take it up with the original author but I would advise

against this because we have a phrase 'Don't look a gift horse in the mouth'.

This is entirely contradictory to the Trojan horse saying 'Beware of Greeks bearing gift [horses]'.

So you decide if you received a gift horse or trojan horse, that's what open source is all about!

In the same way you can judge this overall release to see if it more closely fits the description

'many hands make light work' or 'too many cooks spoil the broth'.

Either way, have a prosperous New Year.

On 28/12/2018 07:13, clare7 wrote:

Noted. Thank you. I also notice that there is no signal that triggers tx_axis_tready . Did I miss out something?

if (tx_axis_tready) begin tx_frame_addr <= tx_frame_addr + 1; tx_axis_tlast <= (tx_frame_addr == tx_packet_length-2) & tx_axis_tvalid_dly; end

On Wed, Dec 26, 2018 at 8:36 PM jrrk notifications@github.com wrote:

You could certainly adapt the code to only check a single byte but thereby the chance of a false sync could be increased. Bearing in mind the chance of a bit error is greater early on in the packet due to the possibility of contention from multiple transmitters.

Sent from my iPhone

On 26 Dec 2018, at 04:56, clare7 notifications@github.com wrote:

In that case, why does it has to be 7 of ,{3'H5} ? In order to check for the SFD 1101 0101 isnt that 3 of {3'H5} will suffice for coding in RMII?

On Mon, Dec 24, 2018 at 3:52 PM jrrk notifications@github.com wrote:

In this case the coding is RMII, so it is checking two data bits and receive enable.

Sent from my iPhone

On 24 Dec 2018, at 05:39, clare7 notifications@github.com wrote:

Hi, I'm wondering what does the value {3'H7,{7{3'H5}}} indicates? It was found in Line 98

if ((rx_nxt == {3'H7,{7{3'H5}}}) && (byte_sync == 0) && (sync == 0)) Based on the codes given, it seems like is checking the end of 8th (collected) RX packets which each packet is 3 bits size. However, I couldnt figure out the how did it comes to value H7 and H5?

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or mute the thread.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <

https://github.com/lowRISC/lowrisc-chip/issues/102#issuecomment-449698888

,

or mute the thread <

https://github.com/notifications/unsubscribe-auth/AcHzHLYMMmp5XSBN3x3VVofBZyYui36Pks5u8Ie5gaJpZM4ZgJRm

.

-- Regards, Ming — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or mute the thread.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub

https://github.com/lowRISC/lowrisc-chip/issues/102#issuecomment-449960842, or mute the thread

https://github.com/notifications/unsubscribe-auth/AcHzHCUjAyK87XaO5V1-ASXD8PRTIMpmks5u821egaJpZM4ZgJRm .

-- Regards, Ming

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/lowRISC/lowrisc-chip/issues/102#issuecomment-450305069, or mute the thread https://github.com/notifications/unsubscribe-auth/AAgF17qEA162oD37qy975gbuDXwRjt_yks5u9cSqgaJpZM4ZgJRm.