xdr_count_write_chunks(9)

🎀 NAME 🎀

okay so bestie, this is xdr_count_write_chunks(9) — yes, section 9, from the Kernel API Manual — and it does exactly one iconic job: Count number of Write chunks in Write list. That's the whole slay. 💅

✨ SYNOPSIS ✨

bool xdr_count_write_chunks (struct svc_rdma_recv_ctxt *rctxt , __be32
       *p );

🫶 ARGUMENTS 🫶

• rctxt — Received header and decoding state. She's carrying the context, okay? 💖
• p — start of an un-decoded Write list. Literally the starting point of a Write list that hasn't been decoded yet. 🦋

💅 DESCRIPTION 💅

Real talk: before allocating anything, ensure the ingress Write list is safe to use. Safety check first, allocation later, no cap. ✨

Return values, bestie:

• true: Write list is valid. rctxt's xdr_stream is updated to point to the first byte past the Write list, and the number of Write chunks is in rc_write_pcl.cl_count. Slay. 👑
• false: Write list is corrupt. rctxt's xdr_stream is left in an indeterminate state. Not cute, very much a "handle with care" moment. 😭

🚨 That false case is the vibe check failing: the Write list is corrupt, and the xdr_stream is left in an indeterminate state. Keep that in mind, fr fr.

🔥 SEE ALSO 🔥

• Kernel file: ./net/sunrpc/xprtrdma/svc_rdma_recvfrom.c
• svc_rdma_recv_ctxts_destroy(9)
• svc_rdma_recv_ctxt_get(9)
• svc_rdma_recv_ctxt_put(9)
• svc_rdma_release_ctxt(9)
• svc_rdma_post_recvs(9)
• svc_rdma_wc_receive(9)
• svc_rdma_flush_recv_queues(9)
• xdr_count_read_segments(9)
• svc_rdma_xdr_decode_req(9)
• svc_rdma_recvfrom(9)

and that's the whole page, bestie. source/version: ./net/sunrpc/xprtrdma; date: October 2026; page: xdr_count_write_chunks(9). mwah 💋✨