| Message ID | 1345478416-23900-5-git-send-email-richard.genoud@gmail.com |
|---|---|
| State | New, archived |
| Headers | show |
Amended the commit subject, On Mon, 2012-08-20 at 18:00 +0200, Richard Genoud wrote: > This patch provides the possibility to adjust the "maximum expected number of > bad blocks per 1024 blocks" (max_beb_per1024) for each mtd device. > > The majority of NAND devices have their max_beb_per1024 equal to 20, but > sometimes it's more. > Now, we can adjust that via a kernel parameter: > ubi.mtd=<name|num|path>[,<vid_hdr_offs>[,max_beb_per1024]] > > Signed-off-by: Richard Genoud <richard.genoud@gmail.com> Did few minor amendments and applied this one. And pushed out. > + pr_err("UBI error: bad value for max_beb_per1024 parameter: %s", > + tokens[2]); Changed this to 'printk(KERN_ERR' and wrapped the line. The former is because the entire function uses 'printk(KERN_ERR', and I am a fan of consistency. We can change them all to 'pr_err()' and 'pr_warn()' in a separate patch. Similar for wrapping - UBI wraps all messages. If we want, we can unwrap them all in a separate patch. Feel free to submit such a patch, I won't have objections. > -MODULE_PARM_DESC(mtd, "MTD devices to attach. Parameter format: " > - "mtd=<name|num|path>[,<vid_hdr_offs>].\n" You unwrapped this one and I pushed that as is, because it is just much more readable, and it is the only place where we describe a module, so no inconsistencies. :-) Thank you!
2012/8/21 Artem Bityutskiy <dedekind1@gmail.com>: > Changed this to 'printk(KERN_ERR' and wrapped the line. > > The former is because the entire function uses 'printk(KERN_ERR', and I > am a fan of consistency. We can change them all to 'pr_err()' and > 'pr_warn()' in a separate patch. > > Similar for wrapping - UBI wraps all messages. If we want, we can unwrap > them all in a separate patch. > > Feel free to submit such a patch, I won't have objections. > >> -MODULE_PARM_DESC(mtd, "MTD devices to attach. Parameter format: " >> - "mtd=<name|num|path>[,<vid_hdr_offs>].\n" > > You unwrapped this one and I pushed that as is, because it is just much > more readable, and it is the only place where we describe a module, so > no inconsistencies. :-) > > Thank you! > > -- > Best Regards, > Artem Bityutskiy Looking for consistency, I made a quick git grep "\<printk( *KERN_" | wc -l the answer is > 32000 Looks like I found something to do if I'm getting bored with openstreetmap :) I double checked linux-ubi/master tree with mine, it's all right. I'll resend the mtd-utils part updated with the new kernel header ubi-user.h Best regards Richard.
On Wed, 2012-08-22 at 10:32 +0200, Richard Genoud wrote: > Looking for consistency, I made a quick git grep "\<printk( *KERN_" | > wc -l > the answer is > 32000 > Looks like I found something to do if I'm getting bored with > openstreetmap :) Well, now I am talking about consistency within UBI only, withing the project. I think that function is the only one which uses plain printk in UBI. I do not remember the details, but there were some issues using the UBI wrappers like 'ubi_msg' in this function because it is called very early during boot time (when UBI is compiled-in). And I just did not know about pr_err() that time, or pr_err() did not exist. So you can just amend this single function if you wish. >
2012/8/22 Artem Bityutskiy <dedekind1@gmail.com>: > Well, now I am talking about consistency within UBI only, withing the > project. I think that function is the only one which uses plain printk > in UBI. I do not remember the details, but there were some issues using > the UBI wrappers like 'ubi_msg' in this function because it is called > very early during boot time (when UBI is compiled-in). And I just did > not know about pr_err() that time, or pr_err() did not exist. So you can > just amend this single function if you wish. (for the record, pr_err didn't exists at the time UBI got into the kernel) I tested this function with pr_err and ubi_err, and both are working. (by using ubi.mtd= ubi.mtd=2,0,toto ubi.mtd=1,2,3,4) It should be better to use ubi_err don't you think ? (or I missed a corner case ?) Richard.
On Wed, 2012-08-22 at 14:09 +0200, Richard Genoud wrote: > 2012/8/22 Artem Bityutskiy <dedekind1@gmail.com>: > > Well, now I am talking about consistency within UBI only, withing the > > project. I think that function is the only one which uses plain printk > > in UBI. I do not remember the details, but there were some issues using > > the UBI wrappers like 'ubi_msg' in this function because it is called > > very early during boot time (when UBI is compiled-in). And I just did > > not know about pr_err() that time, or pr_err() did not exist. So you can > > just amend this single function if you wish. > > (for the record, pr_err didn't exists at the time UBI got into the kernel) > I tested this function with pr_err and ubi_err, and both are working. > (by using ubi.mtd= ubi.mtd=2,0,toto ubi.mtd=1,2,3,4) > It should be better to use ubi_err don't you think ? > (or I missed a corner case ?) I guess it is safe to change than, thank you!
diff --git a/drivers/mtd/ubi/Kconfig b/drivers/mtd/ubi/Kconfig index dcbaae3..37e070c 100644 --- a/drivers/mtd/ubi/Kconfig +++ b/drivers/mtd/ubi/Kconfig @@ -51,6 +51,8 @@ config MTD_UBI_BEB_LIMIT MTD partitions of the same size, UBI will reserve 40 eraseblocks when attaching a partition. + This option can be overridden by the kernel parameter ubi.mtd. + Leave the default value if unsure. config MTD_UBI_GLUEBI diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c index 6a749d6..5b987cb 100644 --- a/drivers/mtd/ubi/build.c +++ b/drivers/mtd/ubi/build.c @@ -47,7 +47,7 @@ #define MTD_PARAM_LEN_MAX 64 /* Maximum number of comma-separated items in the 'mtd=' parameter */ -#define MTD_PARAM_MAX_COUNT 2 +#define MTD_PARAM_MAX_COUNT 3 /* Maximum value for the number of bad eraseblocks per 1024 limit */ #define MAX_MTD_UBI_BEB_LIMIT 768 @@ -63,10 +63,12 @@ * @name: MTD character device node path, MTD device name, or MTD device number * string * @vid_hdr_offs: VID header offset + * @max_beb_per1024: maximum expected number of bad blocks per 1024 erase blocks */ struct mtd_dev_param { char name[MTD_PARAM_LEN_MAX]; int vid_hdr_offs; + int max_beb_per1024; }; /* Numbers of elements set in the @mtd_dev_param array */ @@ -1224,8 +1226,7 @@ static int __init ubi_init(void) mutex_lock(&ubi_devices_mutex); err = ubi_attach_mtd_dev(mtd, UBI_DEV_NUM_AUTO, - p->vid_hdr_offs, - CONFIG_MTD_UBI_BEB_LIMIT); + p->vid_hdr_offs, p->max_beb_per1024); mutex_unlock(&ubi_devices_mutex); if (err < 0) { ubi_err("cannot attach mtd%d", mtd->index); @@ -1345,6 +1346,7 @@ static int __init ubi_mtd_param_parse(const char *val, struct kernel_param *kp) char buf[MTD_PARAM_LEN_MAX]; char *pbuf = &buf[0]; char *tokens[MTD_PARAM_MAX_COUNT]; + int err; if (!val) return -EINVAL; @@ -1392,23 +1394,31 @@ static int __init ubi_mtd_param_parse(const char *val, struct kernel_param *kp) if (p->vid_hdr_offs < 0) return p->vid_hdr_offs; + if (tokens[2]) { + err = kstrtoint(tokens[2], 10, &p->max_beb_per1024); + if (err) { + pr_err("UBI error: bad value for max_beb_per1024 parameter: %s", + tokens[2]); + return -EINVAL; + } + } + mtd_devs += 1; return 0; } module_param_call(mtd, ubi_mtd_param_parse, NULL, NULL, 000); -MODULE_PARM_DESC(mtd, "MTD devices to attach. Parameter format: " - "mtd=<name|num|path>[,<vid_hdr_offs>].\n" +MODULE_PARM_DESC(mtd, "MTD devices to attach. Parameter format: mtd=<name|num|path>[,<vid_hdr_offs>[,max_beb_per1024]].\n" "Multiple \"mtd\" parameters may be specified.\n" - "MTD devices may be specified by their number, name, or " - "path to the MTD character device node.\n" - "Optional \"vid_hdr_offs\" parameter specifies UBI VID " - "header position to be used by UBI.\n" - "Example 1: mtd=/dev/mtd0 - attach MTD device " - "/dev/mtd0.\n" - "Example 2: mtd=content,1984 mtd=4 - attach MTD device " - "with name \"content\" using VID header offset 1984, and " - "MTD device number 4 with default VID header offset."); + "MTD devices may be specified by their number, name, or path to the MTD character device node.\n" + "Optional \"vid_hdr_offs\" parameter specifies UBI VID header position to be used by UBI. (default value if 0)\n" + "Optional \"max_beb_per1024\" parameter specifies the maximum expected bad eraseblock per 1024 eraseblocks. (default value (" + __stringify(CONFIG_MTD_UBI_BEB_LIMIT) ") if 0)\n" + "\n" + "Example 1: mtd=/dev/mtd0 - attach MTD device /dev/mtd0.\n" + "Example 2: mtd=content,1984 mtd=4 - attach MTD device with name \"content\" using VID header offset 1984, and MTD device number 4 with default VID header offset.\n" + "Example 3: mtd=/dev/mtd1,0,25 - attach MTD device /dev/mtd1 using default VID header offset and reserve 25*nand_size_in_blocks/1024 erase blocks for bad block handling.\n" + "\t(e.g. if the NAND *chipset* has 4096 PEB, 100 will be reserved for this UBI device)."); MODULE_VERSION(__stringify(UBI_VERSION)); MODULE_DESCRIPTION("UBI - Unsorted Block Images");
This patch provides the possibility to adjust the "maximum expected number of bad blocks per 1024 blocks" (max_beb_per1024) for each mtd device. The majority of NAND devices have their max_beb_per1024 equal to 20, but sometimes it's more. Now, we can adjust that via a kernel parameter: ubi.mtd=<name|num|path>[,<vid_hdr_offs>[,max_beb_per1024]] Signed-off-by: Richard Genoud <richard.genoud@gmail.com> --- drivers/mtd/ubi/Kconfig | 2 ++ drivers/mtd/ubi/build.c | 38 ++++++++++++++++++++++++-------------- 2 files changed, 26 insertions(+), 14 deletions(-)