All of lore.kernel.org
 help / color / mirror / Atom feed
From: Artem Bityutskiy <dedekind@infradead.org>
To: Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Cc: Frank Haverkamp <haver@vnet.ibm.com>,
	David Woodhouse <dwmw2@infradead.org>,
	Josh Boyer <jwboyer@linux.vnet.ibm.com>,
	Artem Bityutskiy <dedekind@infradead.org>
Subject: [PATCH 03/20 take 4] UBI: kernel-space API header
Date: Fri, 23 Mar 2007 18:43:12 +0200	[thread overview]
Message-ID: <20070323164312.8958.4162.sendpatchset@localhost.localdomain> (raw)
In-Reply-To: <20070323164256.8958.58377.sendpatchset@localhost.localdomain>

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 7468 bytes --]

diff -auNrp tmp-from/include/linux/mtd/ubi.h tmp-to/include/linux/mtd/ubi.h
--- tmp-from/include/linux/mtd/ubi.h	1970-01-01 02:00:00.000000000 +0200
+++ tmp-to/include/linux/mtd/ubi.h	2007-03-23 18:20:01.000000000 +0200
@@ -0,0 +1,191 @@
+/*
+ * Copyright (c) International Business Machines Corp., 2006
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See
+ * the GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
+ *
+ * Author: Artem Bityutskiy (Битюцкий Артём)
+ */
+
+#ifndef __LINUX_UBI_H__
+#define __LINUX_UBI_H__
+
+#include <asm/ioctl.h>
+#include <linux/types.h>
+#include <mtd/ubi-user.h>
+
+/*
+ * enum ubi_data_type - UBI data type hint constants.
+ *
+ * UBI_DATA_LONGTERM: long-term data
+ * UBI_DATA_SHORTTERM: short-term data
+ * UBI_DATA_UNKNOWN: data persistence is unknown
+ *
+ * These constants are used when data is written to UBI volumes in order to
+ * help the UBI wear-leveling unit to find more appropriate physical
+ * eraseblocks.
+ */
+enum ubi_data_type {
+	UBI_DATA_LONGTERM = 1,
+	UBI_DATA_SHORTTERM,
+	UBI_DATA_UNKNOWN
+};
+
+/*
+ * enum ubi_open_mode - UBI volume open mode constants.
+ *
+ * UBI_READONLY: read-only mode
+ * UBI_READWRITE: read-write mode
+ * UBI_EXCLUSIVE: exclusive mode
+ */
+enum ubi_open_mode {
+	UBI_READONLY = 1,
+	UBI_READWRITE,
+	UBI_EXCLUSIVE
+};
+
+/**
+ * struct ubi_volume_info - UBI volume description data structure.
+ * @vol_id: volume ID
+ * @ubi_num: UBI device number this volume belongs to
+ * @size: how many physical eraseblocks are reserved for this volume
+ * @used_bytes: how many bytes of data this volume contains
+ * @used_ebs: how many physical eraseblocks of this volume actually contain any
+ * data
+ * @vol_type: volume type (%UBI_DYNAMIC_VOLUME or %UBI_STATIC_VOLUME)
+ * @corrupted: non-zero if the volume is corrupted (static volumes only)
+ * @upd_marker: non-zero if the volume has update marker set
+ * @alignment: volume alignment
+ * @usable_leb_size: how many bytes are available in logical eraseblocks of
+ * this volume
+ * @name_len: volume name length
+ * @name: volume name
+ * @cdev: UBI volume character device major and minor numbers
+ *
+ * The @corrupted flag is only relevant to static volumes and is always zero
+ * for dynamic ones. This is because UBI does not care about dynamic volume
+ * data protection and only cares about protecting static volume data.
+ *
+ * The @upd_marker flag is set if the volume update operation was interrupted.
+ * Before touching the volume data during the update operation, UBI first sets
+ * the update marker flag for this volume. If the volume update operation was
+ * further interrupted, the update marker indicates this. If the update marker
+ * is set, the contents of the volume is certainly damaged and a new volume
+ * update operation has to be started.
+ *
+ * To put it differently, @corrupted and @upd_marker fields have different
+ * semantics:
+ *     o the @corrupted flag means that this static volume is corrupted for some
+ *       reasons, but not because an interrupted volume update
+ *     o the @upd_marker field means that the volume is damaged because of an
+ *       interrupted update operation.
+ *
+ * I.e., the @corrupted flag is never set if the @upd_marker flag is set.
+ *
+ * The @used_bytes and @used_ebs fields are only really needed for static
+ * volumes and contain the number of bytes stored in this static volume and how
+ * many eraseblock this data occupies. In case of dynamic volumes, the
+ * @used_bytes field is equivalent to @size*@usable_leb_size, and the @used_ebs
+ * field is equivalent to @size.
+ *
+ * In general, logical eraseblock size is a property of the UBI device, not
+ * of the UBI volume. Indeed, the logical eraseblock size depends on the
+ * physical eraseblock size and on how much bytes UBI headers consume. But
+ * because of the volume alignment (@alignment), the usable size of logical
+ * eraseblocks if a volume may be less. The following equation is true:
+ * 	@usable_leb_size = LEB size - (LEB size mod @alignment),
+ * where LEB size is the logical eraseblock size defined by the UBI device.
+ *
+ * The alignment is multiple to the minimal flash input/output unit size or %1
+ * if all the available space is used.
+ *
+ * To put this differently, alignment may be considered is a way to change
+ * volume logical eraseblock sizes.
+ */
+struct ubi_volume_info {
+	int ubi_num;
+	int vol_id;
+	int size;
+	long long used_bytes;
+	int used_ebs;
+	int vol_type;
+	int corrupted;
+	int upd_marker;
+	int alignment;
+	int usable_leb_size;
+	int name_len;
+	const char *name;
+	dev_t cdev;
+};
+
+/**
+ * struct ubi_device_info - UBI device description data structure.
+ * @ubi_num: ubi device number
+ * @leb_size: logical eraseblock size on this UBI device
+ * @min_io_size: minimal I/O unit size
+ * @ro_mode: if this device is in read-only mode
+ * @cdev: UBI character device major and minor numbers
+ *
+ * Note, @leb_size is the logical eraseblock size offered by the UBI device.
+ * Volumes of this UBI device may have smaller logical eraseblock size if their
+ * alignment is not equivalent to %1.
+ */
+struct ubi_device_info {
+	int ubi_num;
+	int leb_size;
+	int min_io_size;
+	int ro_mode;
+	dev_t cdev;
+};
+
+/* UBI descriptor given to users when they open UBI volumes */
+struct ubi_volume_desc;
+
+int ubi_get_device_info(int ubi_num, struct ubi_device_info *di);
+void ubi_get_volume_info(struct ubi_volume_desc *desc,
+			 struct ubi_volume_info *vi);
+struct ubi_volume_desc *ubi_open_volume(int ubi_num, int vol_id, int mode);
+struct ubi_volume_desc *ubi_open_volume_nm(int ubi_num, const char *name,
+					int mode);
+void ubi_close_volume(struct ubi_volume_desc *desc);
+int ubi_eraseblock_read(struct ubi_volume_desc *desc, int lnum, char *buf,
+			int offset, int len, int check);
+int ubi_eraseblock_write(struct ubi_volume_desc *desc, int lnum,
+			 const void *buf, int offset, int len, int dtype);
+int ubi_eraseblock_erase(struct ubi_volume_desc *desc, int lnum);
+int ubi_eraseblock_unmap(struct ubi_volume_desc *desc, int lnum);
+int ubi_eraseblock_is_mapped(struct ubi_volume_desc *desc, int lnum);
+
+/*
+ * This function is the same as the 'ubi_eraseblock_read()' function, but it
+ * does not provide the checking capability.
+ */
+static inline int ubi_read(struct ubi_volume_desc *desc, int lnum, char *buf,
+			   int offset, int len)
+{
+	return ubi_eraseblock_read(desc, lnum, buf, offset, len, 0);
+}
+
+/*
+ * This function is the same as the 'ubi_eraseblock_write()' functions, but it
+ * does not have the data type argument.
+ */
+static inline int ubi_write(struct ubi_volume_desc *desc, int lnum,
+			    const void *buf, int offset, int len)
+{
+	return ubi_eraseblock_write(desc, lnum, buf, offset, len,
+				    UBI_DATA_UNKNOWN);
+}
+
+#endif /* !__LINUX_UBI_H__ */

  parent reply	other threads:[~2007-03-23 16:49 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-03-23 16:42 [PATCH 00/20 take 4] UBI: Unsorted Block Images Artem Bityutskiy
2007-03-23 16:43 ` [PATCH 01/20 take 4] UBI: on-flash data structures header Artem Bityutskiy
2007-03-23 16:43 ` [PATCH 02/20 take 4] UBI: user-space API header Artem Bityutskiy
2007-03-23 16:43 ` Artem Bityutskiy [this message]
2007-03-23 16:43 ` [PATCH 04/20 take 4] UBI: internal header Artem Bityutskiy
2007-03-23 16:43 ` [PATCH 05/20 take 4] UBI: startup code Artem Bityutskiy
2007-03-23 16:43 ` [PATCH 06/20 take 4] UBI: scanning unit Artem Bityutskiy
2007-03-23 16:43 ` [PATCH 07/20 take 4] UBI: I/O unit Artem Bityutskiy
2007-03-23 16:43 ` [PATCH 08/20 take 4] UBI: volume table unit Artem Bityutskiy
2007-03-23 16:43 ` [PATCH 09/20 take 4] UBI: wear-leveling unit Artem Bityutskiy
2007-03-26 12:34   ` Frank Haverkamp
2007-03-23 16:43 ` [PATCH 10/20 take 4] UBI: EBA unit Artem Bityutskiy
2007-03-23 16:43 ` [PATCH 11/20 take 4] UBI: kernel API functions Artem Bityutskiy
2007-03-23 16:43 ` [PATCH 12/20 take 4] UBI: update functionality Artem Bityutskiy
2007-03-23 16:44 ` [PATCH 13/20 take 4] UBI: volume management functionality Artem Bityutskiy
2007-03-23 16:44 ` [PATCH 14/20 take 4] UBI: character devices functionality Artem Bityutskiy
2007-03-23 16:44 ` [PATCH 15/20 take 4] UBI: gluebi functionality Artem Bityutskiy
2007-03-23 16:44 ` [PATCH 16/20 take 4] UBI: misc stuff Artem Bityutskiy
2007-03-23 16:44 ` [PATCH 17/20 take 4] UBI: debugging stuff Artem Bityutskiy
2007-03-23 16:44 ` [PATCH 18/20 take 4] UBI: JFFS2 UBI support Artem Bityutskiy
2007-03-23 16:44 ` [PATCH 19/20 take 4] UBI: update MAINTAINERS Artem Bityutskiy
2007-03-23 16:44 ` [PATCH 20/20 take 4] UBI: linux build integration Artem Bityutskiy

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20070323164312.8958.4162.sendpatchset@localhost.localdomain \
    --to=dedekind@infradead.org \
    --cc=dwmw2@infradead.org \
    --cc=haver@vnet.ibm.com \
    --cc=jwboyer@linux.vnet.ibm.com \
    --cc=linux-kernel@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.