PCA: batch internal calls

[alexytsu]

WIP for #1278

A preliminary attempt at batching some of the internal calls of PCA in order to reduce gas usage. The code factoring is in need of improvement but this is just an early attempt to see the impact of batching multiple calls into one. This PR introduces batchy versions of:

• Market::ActivateDeals
• Registry::ClaimAllocations

The initial traces show a ~7-10% reduction in gas usage when a batch of 8 sectors are proved together. Which can be seen here

The total message cost goes down even less ~5%. There's may be internal optimisations to make the batchy methods more efficient but simply batching the method calls seems to only have a smallish effect.

[alexytsu]

The branch on anorth/fvm-workbench used to profile this: filecoin-project/fvm-workbench#12

[alexytsu]

Collapsing the DataCap::Burn operations into one call is a more significant improvement on the total gas cost. With all three calls made batchy, the total saving is around 15%.

Span[Root, self: {sum=0, none=0}, total: {sum=279,543,986, OnVerifyAggregateSeals=115,956,010, OnBlockLink=83,177,016, wasm_exec=38,651,873, OnBlockOpenBase=27,366,240, OnSyscall=9,156,000, wasm_memory_init=3,145,728, OnMethodInvocation=525,000, OnActorUpdate=475,000, OnBlockOpenPerByte=359,610, OnGetRandomness=344,160, wasm_memory_grow=209,716, OnBlockCreate=147,270, OnBlockRead=14,942, OnHashing=9,422, OnValueTransfer=6,000, OnGetBuiltinActorType=0, none=0, OnBlockStat=0, OnNetworkContext=0, OnSelfBalance=0, OnMessageContext=0, OnGetActorCodeCid=0}]
Span[1-Call(102->f2gmwicvcjm37wudcg6giiha37thwguhswilekszy::26), self: {sum=174,815,498, OnVerifyAggregateSeals=115,956,010, OnBlockLink=22,199,335, wasm_exec=20,724,879, OnBlockOpenBase=11,246,400, OnSyscall=3,038,000, OnActorUpdate=475,000, wasm_memory_init=471,860, OnGetRandomness=344,160, OnBlockOpenPerByte=178,100, OnMethodInvocation=75,000, wasm_memory_grow=52,429, OnBlockCreate=46,790, OnBlockRead=7,425, OnHashing=112, OnGetActorCodeCid=0, OnSelfBalance=0, OnBlockStat=0, OnNetworkContext=0, OnMessageContext=0, OnGetBuiltinActorType=0}, total: {sum=279,543,986, OnVerifyAggregateSeals=115,956,010, OnBlockLink=83,177,016, wasm_exec=38,651,873, OnBlockOpenBase=27,366,240, OnSyscall=9,156,000, wasm_memory_init=3,145,728, OnMethodInvocation=525,000, OnActorUpdate=475,000, OnBlockOpenPerByte=359,610, OnGetRandomness=344,160, wasm_memory_grow=209,716, OnBlockCreate=147,270, OnBlockRead=14,942, OnHashing=9,422, OnValueTransfer=6,000, OnNetworkContext=0, OnMessageContext=0, OnSelfBalance=0, OnGetBuiltinActorType=0, OnGetActorCodeCid=0, OnBlockStat=0}]
Span[2-Call(105->f02::3), self: {sum=1,002,897, wasm_memory_init=445,645, OnBlockOpenBase=187,440, wasm_exec=154,751, OnSyscall=112,000, OnMethodInvocation=75,000, wasm_memory_grow=26,215, OnBlockOpenPerByte=1,170, OnBlockCreate=630, OnBlockRead=47, OnGetBuiltinActorType=0, OnGetActorCodeCid=0, OnMessageContext=0}, total: {sum=1,002,897, wasm_memory_init=445,645, OnBlockOpenBase=187,440, wasm_exec=154,751, OnSyscall=112,000, OnMethodInvocation=75,000, wasm_memory_grow=26,215, OnBlockOpenPerByte=1,170, OnBlockCreate=630, OnBlockRead=47, OnMessageContext=0, OnGetBuiltinActorType=0, OnGetActorCodeCid=0}]
Span[3-Call(105->f04::9), self: {sum=982,221, wasm_memory_init=445,645, OnBlockOpenBase=187,440, wasm_exec=133,812, OnSyscall=112,000, OnMethodInvocation=75,000, wasm_memory_grow=26,215, OnBlockOpenPerByte=1,500, OnBlockCreate=550, OnBlockRead=60, OnGetActorCodeCid=0, OnMessageContext=0, OnGetBuiltinActorType=0}, total: {sum=982,221, wasm_memory_init=445,645, OnBlockOpenBase=187,440, wasm_exec=133,812, OnSyscall=112,000, OnMethodInvocation=75,000, wasm_memory_grow=26,215, OnBlockOpenPerByte=1,500, OnBlockCreate=550, OnBlockRead=60, OnGetActorCodeCid=0, OnGetBuiltinActorType=0, OnMessageContext=0}]
Span[4-Call(105->f05::295558876), self: {sum=43,417,838, OnBlockLink=22,402,344, OnBlockOpenBase=8,997,120, wasm_exec=8,617,718, OnSyscall=2,688,000, wasm_memory_init=445,645, OnBlockOpenPerByte=116,230, OnMethodInvocation=75,000, OnBlockCreate=35,980, wasm_memory_grow=26,215, OnHashing=8,904, OnBlockRead=4,682, OnNetworkContext=0, OnGetBuiltinActorType=0, OnMessageContext=0, OnBlockStat=0, OnGetActorCodeCid=0}, total: {sum=43,417,838, OnBlockLink=22,402,344, OnBlockOpenBase=8,997,120, wasm_exec=8,617,718, OnSyscall=2,688,000, wasm_memory_init=445,645, OnBlockOpenPerByte=116,230, OnMethodInvocation=75,000, OnBlockCreate=35,980, wasm_memory_grow=26,215, OnHashing=8,904, OnBlockRead=4,682, OnGetBuiltinActorType=0, OnMessageContext=0, OnBlockStat=0, OnGetActorCodeCid=0, OnNetworkContext=0}]
Span[5-Call(105->f06::1455686749), self: {sum=53,901,018, OnBlockLink=36,085,142, wasm_exec=8,390,062, OnBlockOpenBase=5,998,080, OnSyscall=2,758,000, wasm_memory_init=445,645, OnMethodInvocation=75,000, OnBlockCreate=60,400, OnBlockOpenPerByte=59,490, wasm_memory_grow=26,215, OnBlockRead=2,593, OnHashing=392, OnGetBuiltinActorType=0, OnGetActorCodeCid=0, OnMessageContext=0, OnBlockStat=0, OnNetworkContext=0}, total: {sum=56,837,168, OnBlockLink=37,532,954, wasm_exec=8,743,786, OnBlockOpenBase=6,372,960, OnSyscall=2,968,000, wasm_memory_init=891,290, OnMethodInvocation=150,000, OnBlockCreate=61,720, OnBlockOpenPerByte=60,980, wasm_memory_grow=52,429, OnBlockRead=2,659, OnHashing=392, OnGetBuiltinActorType=0, OnGetActorCodeCid=0, OnBlockStat=0, OnMessageContext=0, OnNetworkContext=0}]
Span[6-Call(6->f07::1434719642), self: {sum=2,936,151, OnBlockLink=1,447,812, wasm_memory_init=445,645, OnBlockOpenBase=374,880, wasm_exec=353,724, OnSyscall=210,000, OnMethodInvocation=75,000, wasm_memory_grow=26,215, OnBlockOpenPerByte=1,490, OnBlockCreate=1,320, OnBlockRead=66, OnBlockStat=0, OnMessageContext=0}, total: {sum=2,936,151, OnBlockLink=1,447,812, wasm_memory_init=445,645, OnBlockOpenBase=374,880, wasm_exec=353,724, OnSyscall=210,000, OnMethodInvocation=75,000, wasm_memory_grow=26,215, OnBlockOpenPerByte=1,490, OnBlockCreate=1,320, OnBlockRead=66, OnBlockStat=0, OnMessageContext=0}]
Span[7-Call(105->f04::6), self: {sum=2,482,366, OnBlockLink=1,042,384, wasm_memory_init=445,645, OnBlockOpenBase=374,880, wasm_exec=276,930, OnSyscall=238,000, OnMethodInvocation=75,000, wasm_memory_grow=26,215, OnBlockOpenPerByte=1,630, OnBlockCreate=1,600, OnBlockRead=70, OnHashing=14, OnGetActorCodeCid=0, OnGetBuiltinActorType=0, OnBlockStat=0, OnMessageContext=0}, total: {sum=2,482,366, OnBlockLink=1,042,384, wasm_memory_init=445,645, OnBlockOpenBase=374,880, wasm_exec=276,930, OnSyscall=238,000, OnMethodInvocation=75,000, wasm_memory_grow=26,215, OnBlockOpenPerByte=1,630, OnBlockCreate=1,600, OnBlockRead=70, OnHashing=14, OnGetActorCodeCid=0, OnMessageContext=0, OnBlockStat=0, OnGetBuiltinActorType=0}]
Span[8-Call(105->f099::0), self: {sum=6,000, OnValueTransfer=6,000}, total: {sum=6,000, OnValueTransfer=6,000}]

The larger saving here might be expected as batching the burns collapses 8 state operations into one. On the other hand, batching Market::ActivateDeals and Registry::ClaimAllocations reduces message overheads but doesn't necessarily save much on actual state operations without a more invasive refactor.