In 2.1.3/2.3 sdk , observed that if I trigger a ecall multiple times by external application ( 20 times or so), "Segmentation fault (core dumped)" happens. Following is the signature of the ecall that I tried to trigger multiple times by external program.
public void ecall_test([in, size=20480] char *in1, int len, [out, size=20480] char *out1, [out] int *out2)
{
return;
}
Noticed that in/out size has dependency on number of times ecall can be triggered before the Segmentation fault.
The user guide says "automatically allocates space on the trusted stack to hold a copy of the structure" , Dose not the bridge routine that allocates memory frees up the memory when ecall ends/returns? If this is a problem ,then what is the solution?
Used following config for enclave
If I reduce the size from 20480 to 2048,
<EnclaveConfiguration>
<ProdID>0</ProdID>
<ISVSVN>0</ISVSVN>
<StackMaxSize>0x40000</StackMaxSize>
<HeapMaxSize>0x100000</HeapMaxSize>
<TCSNum>10</TCSNum>
<TCSPolicy>1</TCSPolicy>
<!-- Recommend changing 'DisableDebug' to 1 to make the enclave undebuggable for enclave release -->
<DisableDebug>0</DisableDebug>
<MiscSelect>0</MiscSelect>
<MiscMask>0xFFFFFFFF</MiscMask>
</EnclaveConfiguration>
tcs_num 10, tcs_max_num 10, tcs_min_pool 1
The required memory is 4173824B.
