# Ethernet TX ISR recursed

**URL:** <https://forums.digi.com/t/ethernet-tx-isr-recursed/4626>\
**Category:** NET+OS\
**Created:** [March 24, 2004, 3:05pm UTC](https://forums.digi.com/t/ethernet-tx-isr-recursed/4626 "2004-03-24T15:05:25Z")\
**Posts on this page:** 9\
**Page:** 1

<div class="post-metadata">

**Author:** ![llei\_at\_microhardcor](https://avatars.discourse-cdn.com/v4/letter/l/a183cd/32.png) [@llei\_at\_microhardcor](https://forums.digi.com/u/llei_at_microhardcor)\
**Post date:** [March 24, 2004, 3:05pm UTC](https://forums.digi.com/t/ethernet-tx-isr-recursed/4626/1 "2004-03-24T15:05:25Z")

</div>

We are using NET+50 and Altima AC101L phy in our application. We have seen TX ISR is called recursively so often, especially when we run the system through ICE debugger. Anyone has any idea? Your input will be greatly appreciated.

---

<div class="post-metadata">

**Author:** ![system](https://us1.discourse-cdn.com/flex016/uploads/digi/original/1X/11da31a6a1fbcd03ef21d3f785c317c824009ae2.png) [@system](https://forums.digi.com/u/system)\
**Post date:** [March 27, 2004, 1:26am UTC](https://forums.digi.com/t/ethernet-tx-isr-recursed/4626/2 "2004-03-27T01:26:36Z")

</div>

What I have done is removed support for nested intterupts. This is as simple as removing the calls to NASetIRQ() in NAIrqHandler(). I don’t know what version of software you are running, but I did report this bug a long time ago to Net Silicon. The later versions of software I received from them still didn’t have the fix in it either. It’s been a while since I messed with this but if IIRC the problem was that they never updated a particular variable with what was in the IER when masking lower level interrupts. At least something along those lines. You’re probably better off removing the call to NASetIRQ(). IMO the idea behind the whole nested interrupt scheme Net Silicon has is just stupid. The nested interrupts just introduce a whole lot of overhead to your interrupt routines. It takes more time to handle nested interrupts than it does to just complete the 20-30 instruction interrupt routine you have and then move onto the next short fast interrupt routine.

---

<div class="post-metadata">

**Author:** ![system](https://us1.discourse-cdn.com/flex016/uploads/digi/original/1X/11da31a6a1fbcd03ef21d3f785c317c824009ae2.png) [@system](https://forums.digi.com/u/system)\
**Post date:** [May 26, 2004, 7:29pm UTC](https://forums.digi.com/t/ethernet-tx-isr-recursed/4626/3 "2004-05-26T19:29:51Z")

</div>

We have an existing design with NetARM+40 and a Lucent LU3X31FT2-J80 PHY, which has reached end of life. I would like to upgrade to the AC101L PHY. Do you have the Mii.c that supports the AC101L? Thanks in advance

---

<div class="post-metadata">

**Author:** ![system](https://us1.discourse-cdn.com/flex016/uploads/digi/original/1X/11da31a6a1fbcd03ef21d3f785c317c824009ae2.png) [@system](https://forums.digi.com/u/system)\
**Post date:** [October 1, 2004, 7:48pm UTC](https://forums.digi.com/t/ethernet-tx-isr-recursed/4626/4 "2004-10-01T19:48:22Z")

</div>

Bill,Actually I have the same problem, and by chance I find your message in this forum. Which IC did you use to instead of LU3X31FT2-J80 PHY. Me also have a design based on LU3X31FT2-J80 PHY. Thanks in advance

---

<div class="post-metadata">

**Author:** ![system](https://us1.discourse-cdn.com/flex016/uploads/digi/original/1X/11da31a6a1fbcd03ef21d3f785c317c824009ae2.png) [@system](https://forums.digi.com/u/system)\
**Post date:** [November 3, 2004, 7:40pm UTC](https://forums.digi.com/t/ethernet-tx-isr-recursed/4626/5 "2004-11-03T19:40:59Z")

</div>

netarm+40 we may be able to help…please call for details keith 603 758 1760

---

<div class="post-metadata">

**Author:** ![system](https://us1.discourse-cdn.com/flex016/uploads/digi/original/1X/11da31a6a1fbcd03ef21d3f785c317c824009ae2.png) [@system](https://forums.digi.com/u/system)\
**Post date:** [November 3, 2004, 7:44pm UTC](https://forums.digi.com/t/ethernet-tx-isr-recursed/4626/6 "2004-11-03T19:44:20Z")

</div>

we have a large supply of netarm+40 and LU3X3IFT2-TE80-DB…we were forced to make a lifetie buy and as it would happen engineering made some changes…does anyone still use these…please advise 603 785 1760

---

<div class="post-metadata">

**Author:** ![system](https://us1.discourse-cdn.com/flex016/uploads/digi/original/1X/11da31a6a1fbcd03ef21d3f785c317c824009ae2.png) [@system](https://forums.digi.com/u/system)\
**Post date:** [March 2, 2005, 2:21pm UTC](https://forums.digi.com/t/ethernet-tx-isr-recursed/4626/7 "2005-03-02T14:21:46Z")

</div>

HI WE HAVE 10K UNITS OF THE LU3X31FT2-J80 IN STOCK 03 DC IF ANY INTEREST CALL ATLANTIS TECHNOLOGIES INC 407-366-9604 ASK FOR ERIC

---

<div class="post-metadata">

**Author:** ![system](https://us1.discourse-cdn.com/flex016/uploads/digi/original/1X/11da31a6a1fbcd03ef21d3f785c317c824009ae2.png) [@system](https://forums.digi.com/u/system)\
**Post date:** [October 2, 2004, 7:40pm UTC](https://forums.digi.com/t/ethernet-tx-isr-recursed/4626/8 "2004-10-02T19:40:27Z")

</div>

Hi David, AC101L works fine for me. mii.c even works with that chip.

---

<div class="post-metadata">

**Author:** ![system](https://us1.discourse-cdn.com/flex016/uploads/digi/original/1X/11da31a6a1fbcd03ef21d3f785c317c824009ae2.png) [@system](https://forums.digi.com/u/system)\
**Post date:** [February 28, 2005, 1:53pm UTC](https://forums.digi.com/t/ethernet-tx-isr-recursed/4626/9 "2005-02-28T13:53:27Z")

</div>

HI WONDERING HOW MANY OF THE LU3X31FT2-TE80 YOU STILL HAVE? HAYBE CAN HELP SELL THEM FOR YOU OR OTHER ITEMS YOU MAY WANT TO SELL THANKS ERIC PUGH ATLANTIS TECHNOLOGIES INC 407-366-9604
