linux-kernel - Re: [PATCH for-next v6 5/7] RDMA/rxe: Allow registering MRs for On-Demand Paging

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <ZPsuscPwNslStltB@ziepe.ca>
Date:   Fri, 8 Sep 2023 11:24:49 -0300
From:   Jason Gunthorpe <jgg@...pe.ca>
To:     Daisuke Matsuda <matsuda-daisuke@...itsu.com>
Cc:     linux-rdma@...r.kernel.org, leon@...nel.org, zyjzyj2000@...il.com,
        linux-kernel@...r.kernel.org, rpearsonhpe@...il.com,
        yangx.jy@...itsu.com, lizhijian@...itsu.com, y-goto@...itsu.com
Subject: Re: [PATCH for-next v6 5/7] RDMA/rxe: Allow registering MRs for
 On-Demand Paging

On Fri, Sep 08, 2023 at 03:26:46PM +0900, Daisuke Matsuda wrote:
> diff --git a/drivers/infiniband/sw/rxe/rxe_odp.c b/drivers/infiniband/sw/rxe/rxe_odp.c
> index 834fb1a84800..713bef9161e3 100644
> --- a/drivers/infiniband/sw/rxe/rxe_odp.c
> +++ b/drivers/infiniband/sw/rxe/rxe_odp.c
> @@ -32,6 +32,31 @@ static void rxe_mr_unset_xarray(struct rxe_mr *mr, unsigned long start,
>  	xas_unlock(&xas);
>  }
>  
> +static void rxe_mr_set_xarray(struct rxe_mr *mr, unsigned long start,
> +			      unsigned long end, unsigned long *pfn_list)
> +{
> +	unsigned long lower = rxe_mr_iova_to_index(mr, start);
> +	unsigned long upper = rxe_mr_iova_to_index(mr, end - 1);
> +	struct page *page;
> +	void *entry;
> +
> +	XA_STATE(xas, &mr->page_list, lower);
> +
> +	/* ib_umem_odp_unmap_dma_pages() ensures pages are HMM_PFN_VALID */
> +	xas_lock(&xas);
> +	while (true) {
> +		page = hmm_pfn_to_page(pfn_list[xas.xa_index]);
> +		xas_store(&xas, page);
> +
> +		entry = xas_next(&xas);
> +		if (xas_retry(&xas, entry) || (xas.xa_index <= upper))
> +			continue;
> +
> +		break;
> +	}

while (xas.xa_index <= upper) {
      xas_store(&xas, hmm_pfn_to_page(pfn_list[xas.xa_index]));
      xas_next(&xas);
}

Again no need for retries

Jason