Allow LwIP TCP retransmissions to be configured and tune those smaller.

[SeppoTakalo]

Description

Allow LwIP TCP retransmissions to be configured and tune those smaller.

Currently, LwIP segment retransmission time is 12, which is very long
time as each timeout doubles the retransmission timeout.
Make that to 6 as that is same what we use in Nanostack.

LwIP TCP engine uses 500 ms slow timer and each retransmit will double the previous timeout.

          /* Double retransmission time-out unless we are trying to
           * connect to somebody (i.e., we are in SYN_SENT). */
          if (pcb->state != SYN_SENT) {
            u8_t backoff_idx = LWIP_MIN(pcb->nrtx, sizeof(tcp_backoff)-1);
            pcb->rto = ((pcb->sa >> 3) + pcb->sv) << tcp_backoff[backoff_idx];
          }

TCP timeouts come from table:

static const u8_t tcp_backoff[13] =
    { 1, 2, 3, 4, 5, 6, 7, 7, 7, 7, 7, 7, 7};

And to my best understanding, this pcb->sv is 6. so 6 << tcp_backoff[7] will already lead to 6 minutes. Going through all timeouts will eventually take something like 40 minutes, unless I'm mistaken by values of pcb->sa and pcb->sv

This will fix the issue where Client+HTTP update stops into DNS queries not going through because timeouted TCP sessions have already eaten the full heap.

Also change the number of Freescale EMAC driver ring buffers down to 3. Earlier it was using 16+8 buffers, each 1500 bytes. This was the biggest consumer of LwIP heap.

Pull request type

[X] Fix
[ ] Refactor
[ ] Target update
[ ] Functionality change
[ ] Docs update
[ ] Test update
[ ] Breaking change

Reviewers

@kjbracey-arm
@teetak01
@yogpan01

[ciarmcom]

@SeppoTakalo, thank you for your changes.
@yogpan01 @teetak01 @kjbracey-arm@ARMmbed/mbed-os-ipcore @ARMmbed/mbed-os-maintainers please review.

[0xc0170]

Please review reviewers

[kjbracey]

Doing this without correspondingly reducing the lwIP mem config is a little "sneaky" - effectively you've giving this specific platform a huge memory pool for lwIP, because it was set big in a target override apparently to make room for these buffers.

I'd like you to either

a) reduce the target-override memory pool for lwIP by the corresponding amount - if you can (but I assume you do need the extra space, so you can't).
b) reduce by less, and raise the default memory pool (and other target overrides?) by the difference - so that all platforms get the extra memory you've found you need. Maybe the other targets with big pools can also have their buffer count reduced, so they don't need to get bigger.

[SeppoTakalo]

I'll lower the lwip.mem-size value for some amount..
This 16, was actually causing memory_manager->alloc_heap(ENET_ETH_MAX_FLEN, ENET_BUFF_ALIGNMENT); called 16 times.. Where ENET_ETH_MAX_FLEN is 1522 bytes and alignment is 16. I'm unsure of how much the allocator overhead is.

Total memory usage is something between 24352 bytes 24608 bytes.

As a result, I lowered the lwip.mem-size for Freescale boards from 36560 to 16384 (~20 kB less), which still seems to pass the Cloud Client FW update test.

[kjbracey]

If 16K is required with 2 RX buffers, that does suggest cloud wants about 12K of lwIP mem workspace, so maybe that should be the default mem-size? (I believe we assume that target defaults should be good for cloud client). And then other targets should be doing 12K + their buffer size?

I guess the issue is that any upward increase like that might not be appropriate for a patch release, so should be done separately. Something to follow up.

[SeppoTakalo]

Updated the PR based on review feedback.

Should now be ready for testing.

[cmonr]

@kjbracey-arm @michalpasztamobica @yogpan01 @teetak01 When y'all get a chance.

[SeppoTakalo]

@0xc0170 Please start the tests for this.

[0xc0170]

We will do as soon as 5.11.1 RC completes (currently in CI, should be completed in 1-2h),

[0xc0170]

Test started

[0xc0170]

Exporters failures are related to the CI packages, we are investigating what has changed, will restart once the root cause found cc @ARMmbed/mbed-os-test

[0xc0170]

Exporter restarted

[mbed-ci]

Test run: FAILED

Summary: 2 of 11 test jobs failed
Build number : 1
Build artifacts

Failed test jobs:

• jenkins-ci/mbed-os-ci_exporter
• jenkins-ci/mbed-os-ci_greentea-test

[cmonr]

CI job restarted: jenkins-ci/mbed-os-ci_greentea-test