Sivaram

LLVM installation: https://apt.llvm.org/ I used the automatic script for installing llvm 11, as shown below. ``` wget https://apt.llvm.org/llvm.sh chmod +x llvm.sh sudo ./llvm.sh 11 ```

> * It needs LLVM **v11.0.1**. #137 changes the required version to **v11**

Sorry for the delay. I took a look at `scratch_alloc` done by `pippenger_batch` and `strauss_batch`. A shared format for the scratch space (allocated with scalars and points), which the `ecmult_mulit_var`...

In `fe_sqrt`: https://github.com/bitcoin-core/secp256k1/blob/a1102b12196ea27f44d6201de4d25926a2ae9640/src/field_impl.h#L132-L135 Here, the `secp256k1_fe_sqr` function ensures that `t1->magnitude = 1` but there is no guarantee for the magnitude of `a` to be one since it is passed into...

> It's only used in a single var-time caller Oh, Okay. Then, creating a new function for `secp256k1_fe_sqrt_var` (while renaming the old one to `_fe_sqrt_const`) might be overkill. > we...

addressed https://github.com/bitcoin-core/secp256k1/pull/1062#discussion_r778917127 and https://github.com/bitcoin-core/secp256k1/pull/1062#pullrequestreview-844742940 (detailed explanation in 08186e8).

I also benchmarked the `tests.c` file (using `time ./tests`) since a lot of `secp256k1_fe_verify` calls were added in this PR. These are the results (on 64-bit, i7-8750H) for three iterations:...

addressed #1061

Okay, I will remove `fe_equal_var` in a new commit (this PR).

- Removed `fe_equal_var` and documented the reason in `fe_equal` - Replaced the `fe_equal_var` calls with `fe_equal`