Adding MTB_MTS_XDOT as a new target #5975
Conversation
|
@theotherjimmy, @studavekar, thoughts on providing the cmsis ROM/RAM info as defines so tests could filter based on available ROM/RAM? It looks like this would be needed for devices with <64KB RAM for the filesystem tests due to the reliance on the heap block device. |
|
@geky I would like to discuss that in an issue :D. For now, I think turning just this target off for that one test is a welcome workaround. |
|
As it's just IAR, this is again related to non dynamic heap - would IAR 8 update fix this?
Those 2 tests that are failing for this device, do they need more than 32kB RAM devices? I think assuming 32kB limit should be fine. Having a discussion about ROM/RAM would be good. There is still opened PR about minimum memory requirements test that is waiting for IAR 8 to get in. Back to this target addition, @geky - can we make those 2 tests run without decreasing the functionality on a target with 32kB? |
|
@theotherjimmy, are you thinking ifdefs in the test based on the TARGET_MTB_MTS_XDOT? Technically we can, but it seems like a point solution. @0xc0170, are you ok with this? |
|
@geky Yeah, I was thinking that might be an okay workaround for now. |
Yes, fine but not as long term fix. Still would like to understand the test requirements and if this device can fulfill those |
We currently don't have a mechanism for selecting tests based on the available ram/heap, so the best solution right now is to disable these tests specifically for this target.
|
Should be good to go now |
|
@0xc0170 ..over to you for review and CI..thanks! |
|
@cmonr is also managing the queue I believe |
|
@cmonr .. could you please review / CI..? Thank you... |
|
/morph build |
|
Man, @geky, that hack makes me uncomfortable, but if it's what's needed and this needs to be pushed in quickly, then so be it. @geky @ashok-rao I would appreciate seeing an issue opened and reference to this PR so that we can track that hack, and revert it when a proper solution is make. |
Build : FAILUREBuild number : 1051 |
|
@ashok-rao, hope you don't mind, but I added a second LED definition that was apparently needed for tests, since it would be nice if we had CI results by the start of next week. /morph build |
Build : SUCCESSBuild number : 1055 Triggering tests/morph test |
Test : SUCCESSBuild number : 861 |
Exporter Build : SUCCESSBuild number : 734 |
|
@0xc0170 ..will this go into 5.7.5 release? |
|
Yes, will be |
Description
Adding MTS xDOT (MTB) as a new target.
Status
READY
Todos
@0xc0170 , attaching GT logs below. GCC and ARMCC are Passing while IAR is failing due to low RAM (32K) on this target. Tagging @geky for pointers on this.
MTB_xDOT_GT_Log_ARMCC_Pass.txt
MTB_xDOT_GT_Log_GCC_Pass.txt
MTB_xDOT_GT_Log_IAR_Fail.txt
cc @screamerbg @chris-styles @studavekar