> Yea. I'm not a huge fan of LLMs, but I'm finding they sometimes catch > things I miss, particularly small details.
I have almost the same feelings as you about LLMs. Indeed it can locate things I missed, particularly typos, nits or likewise easily and quickly, but with lots of noisy as side-effect. Pan -----Original Message----- From: Jeffrey Law <[email protected]> Sent: Wednesday, April 1, 2026 11:26 AM To: Li, Pan2 <[email protected]>; Jeffrey Law <[email protected]>; [email protected] Cc: [email protected]; [email protected]; [email protected]; Chen, Ken <[email protected]>; Liu, Hongtao <[email protected]> Subject: Re: [PATCH v1 1/3] RISC-V: Allow RVV reg overlap of v[sz]ext.vf2 for m2 On 3/31/2026 6:29 PM, Li, Pan2 wrote: > WARNING: This email originated from outside of Qualcomm. Please be wary of > any links or attachments, and do not enable macros. > > Never mind, Jeff. But to be honest it is difficult to tell which part comes > from the AI up to a point. Yea. I'm not a huge fan of LLMs, but I'm finding they sometimes catch things I miss, particularly small details. So naturally I'm playing with them. Often the things aren't real issues because LLMs often don't have enough context to know something is safe. We've also got access to multiple models, the one I prefer is harder to seamlessly integrate into my review flow ;( Jeff
