Re: [PATCH v2 2/2] svcrdma: Use contiguous pages for RDMA Read sink buffers

["Chuck Lever" <cel@kernel.org>]

On Fri, Mar 13, 2026, at 4:51 AM, kernel test robot wrote:
> Hi Chuck,
>
> kernel test robot noticed the following build errors:
>
> [auto build test ERROR on v7.0-rc1]
> [also build test ERROR on next-20260312]
> [If your patch is applied to the wrong git tree, kindly drop us a note.
> And when submitting patch, we suggest to use '--base' as documented in
> https://git-scm.com/docs/git-format-patch#_base_tree_information]
>
> url:    
> https://github.com/intel-lab-lkp/linux/commits/Chuck-Lever/RDMA-rw-Fix-MR-pool-exhaustion-in-bvec-RDMA-READ-path/20260313-085521
> base:   v7.0-rc1
> patch link:    
> https://lore.kernel.org/r/20260312134008.7387-3-cel%40kernel.org
> patch subject: [PATCH v2 2/2] svcrdma: Use contiguous pages for RDMA 
> Read sink buffers
> config: x86_64-rhel-9.4 
> (https://download.01.org/0day-ci/archive/20260313/202603130922.uCz0Ofwx-lkp@intel.com/config)
> compiler: gcc-14 (Debian 14.2.0-19) 14.2.0
> reproduce (this is a W=1 build): 
> (https://download.01.org/0day-ci/archive/20260313/202603130922.uCz0Ofwx-lkp@intel.com/reproduce)
>
> If you fix the issue in a separate patch/commit (i.e. not just a new 
> version of
> the same patch/commit), kindly add following tags
> | Reported-by: kernel test robot <lkp@intel.com>
> | Closes: 
> https://lore.kernel.org/oe-kbuild-all/202603130922.uCz0Ofwx-lkp@intel.com/
>
> All errors (new ones prefixed by >>):
>
>    net/sunrpc/xprtrdma/svc_rdma_rw.c: In function 'svc_rdma_fill_contig_bvec':
>>> net/sunrpc/xprtrdma/svc_rdma_rw.c:813:17: error: implicit declaration of function 'svc_rqst_page_release'; did you mean 'svc_rdma_cc_release'? [-Wimplicit-function-declaration]
>      813 |                 svc_rqst_page_release(rqstp,
>          |                 ^~~~~~~~~~~~~~~~~~~~~
>          |                 svc_rdma_cc_release

The current version of the series is missing a patch that provides
this helper function. Next version (if one is needed) will include
it.


> vim +813 net/sunrpc/xprtrdma/svc_rdma_rw.c
>
>    779	
>    780	/*
>    781	 * svc_rdma_fill_contig_bvec - Replace rq_pages with a 
> contiguous allocation
>    782	 * @rqstp: RPC transaction context
>    783	 * @head: context for ongoing I/O
>    784	 * @bv: bvec entry to fill
>    785	 * @pages_left: number of data pages remaining in the segment
>    786	 * @len_left: bytes remaining in the segment
>    787	 *
>    788	 * On success, fills @bv with a bvec spanning the contiguous 
> range and
>    789	 * advances rc_curpage/rc_page_count. Returns the byte length 
> covered,
>    790	 * or zero if the allocation failed or would overrun rq_maxpages.
>    791	 */
>    792	static unsigned int
>    793	svc_rdma_fill_contig_bvec(struct svc_rqst *rqstp,
>    794				  struct svc_rdma_recv_ctxt *head,
>    795				  struct bio_vec *bv, unsigned int pages_left,
>    796				  unsigned int len_left)
>    797	{
>    798		unsigned int order, alloc_nr, chunk_pages, chunk_len, i;
>    799		struct page *page;
>    800	
>    801		page = svc_rdma_alloc_read_pages(pages_left, &order);
>    802		if (!page)
>    803			return 0;
>    804		alloc_nr = 1 << order;
>    805	
>    806		if (head->rc_curpage + alloc_nr > rqstp->rq_maxpages) {
>    807			for (i = 0; i < alloc_nr; i++)
>    808				__free_page(page + i);
>    809			return 0;
>    810		}
>    811	
>    812		for (i = 0; i < alloc_nr; i++) {
>  > 813			svc_rqst_page_release(rqstp,
>    814					      rqstp->rq_pages[head->rc_curpage + i]);
>    815			rqstp->rq_pages[head->rc_curpage + i] = page + i;
>    816		}
>    817	
>    818		chunk_pages = min(alloc_nr, pages_left);
>    819		chunk_len = min_t(unsigned int, chunk_pages << PAGE_SHIFT, 
> len_left);
>    820		bvec_set_page(bv, page, chunk_len, 0);
>    821		head->rc_page_count += chunk_pages;
>    822		head->rc_curpage += chunk_pages;
>    823		return chunk_len;
>    824	}
>    825	
>
> -- 
> 0-DAY CI Kernel Test Service
> https://github.com/intel/lkp-tests/wiki

-- 
Chuck Lever