elenril

I can reproduce this as well, bisecting points at commit e7e0c52db9093a9ecd9dcaa0766e66515a546a75 as the culprit.

Quoting Patrick Totzke (2020-05-08 21:36:39) > Again, I'm quite sure that it is down to the notmuch bindings... Most likely. The memory safety of the python binding is highly suspect....

Quoting Frank LENORMAND (2020-05-09 07:49:30) > @elenril do you have a patch of your fix I could try, please? > My branch has diverged from master quite considerably, so testing...

Quoting Frank LENORMAND (2020-05-09 10:48:33) > > Could it be a better move to have a new `alot` branch that uses > `notmuch2` bindings, which allegedly don't have the crash...

Quoting Frank LENORMAND (2020-05-09 12:45:09) > I ran `gdb` to try to get a more precise stacktrace, this is what I got: > > ``` > #0 0x00007ffff788ace5 in raise...

Quoting Frank LENORMAND (2020-05-09 13:14:41) > Correct, but in any case this is a `notmuch` issue, and doesn't need to be handled here. Not necessarily. The current python bindings are...

Quoting Frank LENORMAND (2020-05-18 12:05:13) > @elenril it doesn't look like the `Xapian::DatabaseModifiedError` abortion was reported [upstream](https://nmbug.notmuchmail.org/status/#Bugs), have you notified them about it? Yes, I have discussed it with developers...

First of all, thank you for the reply. It's so hard to find any reliable information on this - it seems very few people use jumbo frames, even though in...

From a brief look, I see a large number of clear violations. At least - mpp/codec/dec/av1/av1d_codec.h - mpp/codec/dec/av1/av1d_cbs.c - mpp/codec/dec/av1/av1d_cbs.h - mpp/codec/dec/av1/av1d_parser2_syntax.c - mpp/codec/dec/h265/h265d_codec.h - mpp/codec/dec/h265/h265d_parser.c - mpp/codec/dec/h265/h265d_ps.c - mpp/codec/dec/vp9/vp9d_codec.h...