source: mainline/uspace/srv/vfs/vfs_pager.c@ 3c5b86c

/*
 * Copyright (c) 2016 Jakub Jermar
 * All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 *
 * - Redistributions of source code must retain the above copyright
 *   notice, this list of conditions and the following disclaimer.
 * - Redistributions in binary form must reproduce the above copyright
 *   notice, this list of conditions and the following disclaimer in the
 *   documentation and/or other materials provided with the distribution.
 * - The name of the author may not be used to endorse or promote products
 *   derived from this software without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
 * IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
 * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
 * IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
 * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
 * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
 * DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
 * THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
 * THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 */

/** @addtogroup fs
 * @{
 */

/**
 * @file vfs_pager.c
 * @brief VFS pager operations.
 */

#include "vfs.h"
#include <async.h>
#include <fibril_synch.h>
#include <errno.h>
#include <as.h>

void vfs_page_in(ipc_callid_t rid, ipc_call_t *request)
{
        aoff64_t offset = IPC_GET_ARG1(*request);
        size_t page_size = IPC_GET_ARG2(*request);
        int fd = IPC_GET_ARG3(*request);
        void *page;
        int rc;

        vfs_file_t *file = vfs_file_get(fd);
        if (!file) {
                async_answer_0(rid, ENOENT);
                return;
        }

        page = as_area_create(AS_AREA_ANY, page_size,
            AS_AREA_READ | AS_AREA_WRITE | AS_AREA_CACHEABLE,
            AS_AREA_UNPAGED);

        if (page == AS_MAP_FAILED) {
                vfs_file_put(file);
                async_answer_0(rid, ENOMEM);
                return;
        }

        rdwr_io_chunk_t chunk = {
                .buffer = page,
                .size = page_size
        };

        fibril_mutex_lock(&file->lock);
        file->pos = offset;
        fibril_mutex_unlock(&file->lock);

        size_t total = 0;
        do {
                rc = vfs_rdwr_internal(fd, true, &chunk);
                if (rc != EOK)
                        break;
                if (chunk.size == 0)
                        break;
                total += chunk.size;
                chunk.buffer += chunk.size;
                chunk.size = page_size - total;
        } while (total < page_size);

        vfs_file_put(file);

        async_answer_1(rid, rc, (sysarg_t) page);

        /*
         * FIXME: 
         * This is just for now until we implement proper page cache
         * management.  Not keeping the pages around in a cache results in
         * inherently non-coherent private mappings.
         */
        as_area_destroy(page);
}

/**
 * @}
 */