Michael Misiewicz

Same issue here (Apple Silicon)

Modifying line 19 of the Makefile to `SHLIB_LINK = $(libpq) -L/opt/homebrew/Cellar/gettext/0.21.1/lib -lintl` got it to compile, though clearly a dirty hack

full boot log attached [r25361-a86ff06d2e boot.txt](https://github.com/openwrt/openwrt/files/14467454/r25361-a86ff06d2e.boot.txt)

full boot log from bpi stock [bpi-stock.txt](https://github.com/openwrt/openwrt/files/14467456/bpi-stock.txt)

Reading the lspci output between the bpi and snapshot images, it seems that one of the pci controllers is not being detected under 6.1 kernel. Specifically this one: ``` 0003:00:00.0...

Using the command `dtc -s -I fs /proc/device-tree -O dts` I have diffed the device trees between the sinovoip image and the snapshot device tree. Looking in the PCI section...

interesting idea about using the exponents. I'm expecting to have ~billions of rows so it sounds like keeping a separate list of index -> ID is the right way to...

Update on this. I have 768 dimension vectors, like this: [-0.018988762,0.050045867,0.027589366,-0.020255722, ... ] and I tried appended the ID (int cast to float) as a 769th dimensions [... 0.008151617,-0.037780758,-0.0036997749,-0.041787587,-0.011534975, 33931]...

Hey @var77 - `vec` is a `halfvec` from the latest pgvector. Running the indexing on a subset of 10,000, 100,000 and at least 1MM one time worked OK. Beyond that...

the issue also occurs when running lantern as a postgres background worker. ``` (39242) [local]:5432 mike@mike=# create index on bigtable using lantern_hnsw (cast(vec as vector(768)) dist_vec_cos_ops) with (m = 34,...