Ludovic Henry

@martin-frbg what you're proposing are very valid workarounds. However, do you think it could be also possible to change OpenBLAS so that it doesn't allocate unnecessarily on the stack? Here,...

Hi @fisk, thanks for the amazing cleanup! We've had a bunch of issues with the code sweeper and AsyncGetCallTrace with trying to unwind from dead code blobs (or what looked...

> I see the LKML link mentions this: "Just remove the check altogether and let PROT_WRITE imply PROT_READ as is the case on other architectures.". Is this change still necessary...

/integrate

/integrate

@magicus thanks! I committed your change.

/integrate

Hi @dlwh, let me try to reproduce that locally, it's first time I see it.

I can reproduce with OpenBLAS, but not with Intel MKL. I also can only reproduce if `OPENBLAS_NUM_THREADS` is greater than 1. I'm now looking at how `dgetrf_parallel` (the function in...

Here is what I'm observing. When calling `dgetrf_` from Java, we have `$rsp = 0x7ffff599d0e8`. It then calls `dgetrf_parallel` a first time, which allocates arrays on the stack and changes...