userfaultfd: reset err to be 0 when move_pages_ptes succeeded

From 2599b924548175bef41ca6a601de58236a55a998 Mon Sep 17 00:00:00 2001
From: Foxie Flakey <[email protected]>
Date: Sun, 16 Aug 2026 23:03:46 +0700
Subject: [PATCH] userfaultfd: reset err to be 0 when move_pages_ptes succeeded

During move_pages() operation, when move_pages_ptes() returns EAGAIN, the error code is not cleared even after we processed it. This leads to a successful retry but then the same pages are retried again due to the stale error code. This time move fails because pages are already moved, loop is terminated and move_pages() reports a failure. Clear the error code once we processes EAGAIN.

---
Changes in v2:
- Updated on description to be easier to understood
- Reset err before retrying, instead of fixing right at the move_pages_ptes branch.
- Added Assisted-by, received an answer few days ago from linux-newbies IRC that even I just used AI/LLM as "fast path" of git bisect. I still need to add it following
https://docs.kernel.org/process/coding-assistants.html doc.

---
 mm/userfaultfd.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

index c3adedaaf7d5..ec5d511bd179 100644
@@ -2085,8 +2085,10 @@ static ssize_t move_pages(struct userfaultfd_ctx *ctx, unsigned long dst_start,
 		}
 
 		if (err) {
-			if (err == -EAGAIN)
+			if (err == -EAGAIN) {
+				err = 0;
 				continue;
+			}
 			break;
 		}
 

base-commit: 62cc90241548d5570ee68e01aaba6506964e9811

2.55.0

Not sure where I put text that is not the patch description and not in changes. I put this on seperate mail reply.

I'll delay figuring what to do with Signed-off-by, I'd rather keep Foxie Flakey because that what is my "known identity" on internet at various places like GitHub, Alpine's GitLab and now here in Linux kernel development. Would be nice if I can keep Foxie Flakey, like again policy is policy not a suggestion if Foxie Flakey is not allowed.

Would be nice to not have to add this Assisted-by. Mainly because I just used ChatGPT as "fast path" of git bisect to avoid needing recompiling kernels during bisect if ChatGPT found result first before I finished git bisect. I feel like my effort trying to understand how to mail, write patches is undermined, but policy is policy so I don't mind if it has to stay.