From: NamJae Jeon <linkinjeon@gmail.com> To: Adrian Hunter <adrian.hunter@intel.com> Cc: cjb@laptop.org, linux-mmc@vger.kernel.org, linux-kernel@vger.kernel.org, awarkentin@vmware.com, linus.walleij@linaro.org, james_p_freyensee@linux.intel.com, sebras@gmail.com, Ulf.Hansson@stericsson.com, stefan.xk.nilsson@stericsson.com, per.forlin@stericsson.com, johan.rudholm@stericsson.com Subject: Re: [PATCH v10] mmc : general purpose partition support. Date: Wed, 5 Oct 2011 20:22:56 +0900 [thread overview] Message-ID: <CAKYAXd9iHW3bdeFkq=ywdOTPtzFdqK8Yb9fXxi8Jvoq6BRvHtQ@mail.gmail.com> (raw) In-Reply-To: <CAKYAXd_AoSgY6htSO5jJyD3QJ_xXFqgKLyNX49s3PuRW6AUmbg@mail.gmail.com> [-- Warning: decoded text below may be mangled, UTF-8 assumed --] [-- Attachment #1: Type: text/plain; charset=UTF-8, Size: 15004 bytes --] 2011/10/5 NamJae Jeon <linkinjeon@gmail.com>: > 2011/10/5 Adrian Hunter <adrian.hunter@intel.com>: >> On 05/10/11 13:21, NamJae Jeon wrote: >>> >>> 2011/10/5 Adrian Hunter<adrian.hunter@intel.com>: >>>> >>>> On 04/10/11 20:45, Namjae Jeon wrote: >>>>> >>>>> It allows gerneral purpose partitions in MMC Device. >>>>> And I try to simpliy make mmc_blk_alloc_parts using mmc_part structure >>>>> suggested by Andrei Warkentin. >>>>> After patching, we can see general purpose partitions like this. >>>>>> >>>>>> cat /proc/partitions >>>>> >>>>>      179 0 847872 mmcblk0 >>>>>      179 192 4096 mmcblk0gp3 >>>>>      179 160 4096 mmcblk0gp2 >>>>>      179 128 4096 mmcblk0gp1 >>>>>      179 96  1052672 mmcblk0gp0 >>>>>      179 64  1024 mmcblk0boot1 >>>>>      179 32  1024 mmcblk0boot0 >>>>> >>>>> Signed-off-by: Namjae Jeon<linkinjeon@gmail.com> >>>>> --- >>>>>  drivers/mmc/card/block.c |  34 +++++++++++++++++------------ >>>>>  drivers/mmc/core/mmc.c  |  52 >>>>> ++++++++++++++++++++++++++++++++++++++++++--- >>>>>  include/linux/mmc/card.h |  34 +++++++++++++++++++++++++++++- >>>>>  include/linux/mmc/mmc.h  |   5 +++- >>>>>  4 files changed, 105 insertions(+), 20 deletions(-) >>>>> >>>>> diff --git a/drivers/mmc/card/block.c b/drivers/mmc/card/block.c >>>>> index 782d5f8..fb6f1e9 100644 >>>>> --- a/drivers/mmc/card/block.c >>>>> +++ b/drivers/mmc/card/block.c >>>>> @@ -1470,26 +1470,32 @@ static int mmc_blk_alloc_part(struct mmc_card >>>>> *card, >>>>>     return 0; >>>>>  } >>>>> >>>>> +/* MMC Physical partition consist of two boot partitions and >>>>> + * up to four general purpose partitions. >>>>> + * For each partitions enabled in EXT_CSD >>>>> + * The block device will allocated to provide access to the partition. >>>>> + */ >>>>> + >>>>>  static int mmc_blk_alloc_parts(struct mmc_card *card, struct >>>>> mmc_blk_data >>>>> *md) >>>>>  { >>>>> -    int ret = 0; >>>>> +    int idx, ret = 0; >>>>> >>>>>     if (!mmc_card_mmc(card)) >>>>>         return 0; >>>>> >>>>> -    if (card->ext_csd.boot_size&& >>>>>  mmc_boot_partition_access(card->host)) { >>>>> -        ret = mmc_blk_alloc_part(card, md, >>>>> EXT_CSD_PART_CONFIG_ACC_BOOT0, >>>>> -                     card->ext_csd.boot_size>>   9, >>>>> -                     true, >>>>> -                     "boot0"); >>>>> -        if (ret) >>>>> -            return ret; >>>>> -        ret = mmc_blk_alloc_part(card, md, >>>>> EXT_CSD_PART_CONFIG_ACC_BOOT1, >>>>> -                     card->ext_csd.boot_size>>   9, >>>>> -                     true, >>>>> -                     "boot1"); >>>>> -        if (ret) >>>>> -            return ret; >>>>> +    for (idx = 0; idx<   card->nr_parts; idx++) { >>>>> +        if ((idx == 0 || idx == 1)&& >>>>> +            !mmc_boot_partition_access(card->host)) >>>>> +            continue; >>>> >>>> Seems a bit fragile.  What if someone decided to put the gp >>>> partitions before the boot partitions.  What about: >>> >>> There are the default two boot partition in mmc spec. I think that >>> this case can not be happened. >>> And boot partition size of EXT CSD is read-only. >>> Can you explain which is case put to gp partitions before the boot >>> partition ? >> >> Relying on the partition order in the partition array is fragile. >> You cannot predict how someone may want to change it in the future, >> and that person will not expect that certain positions have certain meaning. 0,1 of Boot partition is fixed if spec is not changed over. but gp partition is variable in 2,3,4,5,6 of array. If How someone change code, problem occur ? >> >>> >>>> >>>>         if (mmc_is_boot_partition(&part[idx])&& >>>>           !mmc_boot_partition_access(card->host) >>>>             continue; >>>> >>>> And: >>>> >>>> int mmc_is_boot_partition(struct mmc_part *part) >>>> { >>>>     return part->part_cfg>= EXT_CSD_PART_CONFIG_ACC_BOOT0&& >>>>        part->part_cfg<= EXT_CSD_PART_CONFIG_ACC_BOOT0 + >>>>                 MMC_NUM_BOOT_PARTITION; >>>> } >>>> >>>>> +        if (card->part[idx].size) { >>>>> +            ret = mmc_blk_alloc_part(card, md, >>>>> +                card->part[idx].part_cfg, >>>>> +                card->part[idx].size>>   9, >>>>> +                card->part[idx].force_ro, >>>>> +                card->part[idx].name); >>>>> +            if (ret) >>>>> +                return ret; >>>>> +        } >>>>>     } >>>>> >>>>>     return ret; >>>>> diff --git a/drivers/mmc/core/mmc.c b/drivers/mmc/core/mmc.c >>>>> index c632b1f..2f458a0 100644 >>>>> --- a/drivers/mmc/core/mmc.c >>>>> +++ b/drivers/mmc/core/mmc.c >>>>> @@ -239,7 +239,9 @@ static int mmc_get_ext_csd(struct mmc_card *card, u8 >>>>> **new_ext_csd) >>>>>  */ >>>>>  static int mmc_read_ext_csd(struct mmc_card *card, u8 *ext_csd) >>>>>  { >>>>> -    int err = 0; >>>>> +    int err = 0, idx; >>>>> +    unsigned int part_size; >>>>> +    u8 hc_erase_grp_sz = 0, hc_wp_grp_sz = 0; >>>>> >>>>>     BUG_ON(!card); >>>>> >>>>> @@ -340,7 +342,14 @@ static int mmc_read_ext_csd(struct mmc_card *card, >>>>> u8 >>>>> *ext_csd) >>>>>         * There are two boot regions of equal size, defined in >>>>>         * multiples of 128K. >>>>>         */ >>>>> -        card->ext_csd.boot_size = ext_csd[EXT_CSD_BOOT_MULT]<< >>>>>  17; >>>>> +        if (ext_csd[EXT_CSD_BOOT_MULT]) { >>>>> +            for (idx = 0; idx<   MMC_NUM_BOOT_PARTITION; >>>>> idx++) >>>>> { >>>>> +                part_size = ext_csd[EXT_CSD_BOOT_MULT]<< >>>>>  17; >>>>> +                mmc_part_add(card, part_size, >>>>> +                    EXT_CSD_PART_CONFIG_ACC_BOOT0 + >>>>> idx, >>>>> +                    "boot%d", idx, true); >>>>> +            } >>>>> +        } >>>>>     } >>>>> >>>>>     card->ext_csd.raw_hc_erase_gap_size = >>>>> @@ -362,9 +371,9 @@ static int mmc_read_ext_csd(struct mmc_card *card, >>>>> u8 >>>>> *ext_csd) >>>>>         card->ext_csd.raw_partition_support = >>>>> ext_csd[EXT_CSD_PARTITION_SUPPORT]; >>>>>         if ((ext_csd[EXT_CSD_PARTITION_SUPPORT]&   0x2)&& >>>>>         (ext_csd[EXT_CSD_PARTITION_ATTRIBUTE]&   0x1)) { >>>>> -            u8 hc_erase_grp_sz = >>>>> +            hc_erase_grp_sz = >>>>>                 ext_csd[EXT_CSD_HC_ERASE_GRP_SIZE]; >>>>> -            u8 hc_wp_grp_sz = >>>>> +            hc_wp_grp_sz = >>>>>                 ext_csd[EXT_CSD_HC_WP_GRP_SIZE]; >>>>> >>>>>             card->ext_csd.enhanced_area_en = 1; >>>>> @@ -393,6 +402,41 @@ static int mmc_read_ext_csd(struct mmc_card *card, >>>>> u8 >>>>> *ext_csd) >>>>>             card->ext_csd.enhanced_area_offset = -EINVAL; >>>>>             card->ext_csd.enhanced_area_size = -EINVAL; >>>>>         } >>>>> + >>>>> +        /* >>>>> +         * General purpose partition feature support -- >>>>> +         * If ext_csd have the size of general purpose >>>>> partitions, >>>>> +         * set size, part_cfg, partition name in mmc_part. >>>>> +         */ >>>>> +        if (ext_csd[EXT_CSD_PARTITION_SUPPORT]& >>>>> +            EXT_CSD_PART_SUPPORT_PART_EN) { >>>>> +            if (card->ext_csd.enhanced_area_en != 1) { >>>>> +                hc_erase_grp_sz = >>>>> + >>>>> ext_csd[EXT_CSD_HC_ERASE_GRP_SIZE]; >>>>> +                hc_wp_grp_sz = >>>>> +                    ext_csd[EXT_CSD_HC_WP_GRP_SIZE]; >>>>> + >>>>> +                card->ext_csd.enhanced_area_en = 1; >>>>> +            } >>>>> + >>>>> +            for (idx = 0; idx<   MMC_NUM_GP_PARTITION; >>>>> idx++) { >>>>> +                if (!ext_csd[EXT_CSD_GP_SIZE_MULT + idx >>>>> * >>>>> 3]&& >>>>> +                !ext_csd[EXT_CSD_GP_SIZE_MULT + idx * 3 >>>>> + >>>>> 1]&& >>>>> +                !ext_csd[EXT_CSD_GP_SIZE_MULT + idx * 3 >>>>> + >>>>> 2]) >>>>> +                    continue; >>>>> +                part_size = >>>>> +                (ext_csd[EXT_CSD_GP_SIZE_MULT + idx * 3 >>>>> + >>>>> 2] >>>>> +<<   16) + >>>>> +                (ext_csd[EXT_CSD_GP_SIZE_MULT + idx * 3 >>>>> + >>>>> 1] >>>>> +<<   8) + >>>>> +                ext_csd[EXT_CSD_GP_SIZE_MULT + idx * 3]; >>>>> +                part_size *= (size_t)(hc_erase_grp_sz * >>>>> +                    hc_wp_grp_sz); >>>>> +                mmc_part_add(card, part_size<<   19, >>>>> +                    EXT_CSD_PART_CONFIG_ACC_GP0 + >>>>> idx, >>>>> +                    "gp%d", idx, false); >>>>> +            } >>>>> +        } >>>>>         card->ext_csd.sec_trim_mult = >>>>>             ext_csd[EXT_CSD_SEC_TRIM_MULT]; >>>>>         card->ext_csd.sec_erase_mult = >>>>> diff --git a/include/linux/mmc/card.h b/include/linux/mmc/card.h >>>>> index 5294ddf..0522627 100644 >>>>> --- a/include/linux/mmc/card.h >>>>> +++ b/include/linux/mmc/card.h >>>>> @@ -12,6 +12,7 @@ >>>>> >>>>>  #include<linux/mmc/core.h> >>>>>  #include<linux/mod_devicetable.h> >>>>> +#include<linux/genhd.h> >>>>> >>>>>  struct mmc_cid { >>>>>     unsigned int       manfid; >>>>> @@ -64,7 +65,6 @@ struct mmc_ext_csd { >>>>>     bool           enhanced_area_en;    /* enable bit */ >>>>>     unsigned long long    enhanced_area_offset;  /* Units: Byte */ >>>>>     unsigned int       enhanced_area_size;   /* Units: KB */ >>>>> -    unsigned int       boot_size;        /* in bytes */ >>>>>     u8            raw_partition_support;  /* 160 */ >>>>>     u8            raw_erased_mem_count;  /* 181 */ >>>>>     u8            raw_ext_csd_structure;  /* 194 */ >>>>> @@ -158,6 +158,23 @@ struct sdio_func_tuple; >>>>> >>>>>  #define SDIO_MAX_FUNCS         7 >>>>> >>>>> +/* The number of MMC physical partitions >>>>> + * It consist of boot partitions(2), general purpose partitions(4) in >>>>> MMC >>>>> v4.4 >>>>> + */ >>>>> +#define MMC_NUM_BOOT_PARTITION 2 >>>>> +#define MMC_NUM_GP_PARTITION  4 >>>>> +#define MMC_NUM_PHY_PARTITION  6 >>>>> + >>>>> +/* >>>>> + * MMC Physical partitions >>>>> + */ >>>>> +struct mmc_part { >>>>> +    unsigned int   size;  /* partition size (in bytes) */ >>>>> +    unsigned int   part_cfg;    /* it used to part_type */ >>>>> +    char   name[DISK_NAME_LEN]; >>>>> +    bool   force_ro;    /* to make boot parts RO by default */ >>>>> +}; >>>>> + >>>>>  /* >>>>>  * MMC device >>>>>  */ >>>>> @@ -219,9 +236,24 @@ struct mmc_card { >>>>>     unsigned int       sd_bus_speed;  /* Bus Speed Mode set for >>>>> the card */ >>>>> >>>>>     struct dentry      *debugfs_root; >>>>> +    struct mmc_part part[MMC_NUM_PHY_PARTITION]; /* physical >>>>> partitions */ >>>>> +    unsigned int   nr_parts; >>>>>  }; >>>>> >>>>>  /* >>>>> + * This function fill contents in mmc_part. >>>>> + */ >>>>> +static inline void mmc_part_add(struct mmc_card *card, unsigned int >>>>> size, >>>>> +            unsigned int part_cfg, char *name, int idx, bool >>>>> ro) >>>>> +{ >>>>> +    card->part[card->nr_parts].size = size; >>>>> +    card->part[card->nr_parts].part_cfg = part_cfg; >>>>> +    sprintf(card->part[card->nr_parts].name, name, idx); >>>>> +    card->part[card->nr_parts].force_ro = ro; >>>>> +    card->nr_parts++; >>>>> +} >>>>> + >>>>> +/* >>>>>  *  The world is not perfect and supplies us with broken mmc/sdio >>>>> devices. >>>>>  *  For at least some of these bugs we need a work-around. >>>>>  */ >>>>> diff --git a/include/linux/mmc/mmc.h b/include/linux/mmc/mmc.h >>>>> index 50af227..ea558eb 100644 >>>>> --- a/include/linux/mmc/mmc.h >>>>> +++ b/include/linux/mmc/mmc.h >>>>> @@ -270,6 +270,7 @@ struct _mmc_csd { >>>>>  * EXT_CSD fields >>>>>  */ >>>>> >>>>> +#define EXT_CSD_GP_SIZE_MULT      143   /* R/W */ >>>>>  #define EXT_CSD_PARTITION_ATTRIBUTE  156   /* R/W */ >>>>>  #define EXT_CSD_PARTITION_SUPPORT   160   /* RO */ >>>>>  #define EXT_CSD_RST_N_FUNCTION         162   /* R/W */ >>>>> @@ -313,7 +314,9 @@ struct _mmc_csd { >>>>> >>>>>  #define EXT_CSD_PART_CONFIG_ACC_MASK  (0x7) >>>>>  #define EXT_CSD_PART_CONFIG_ACC_BOOT0 (0x1) >>>>> -#define EXT_CSD_PART_CONFIG_ACC_BOOT1  (0x2) >>>>> +#define EXT_CSD_PART_CONFIG_ACC_GP0   (0x4) >>>>> + >>>>> +#define EXT_CSD_PART_SUPPORT_PART_EN  (0x1) >>>>> >>>>>  #define EXT_CSD_CMD_SET_NORMAL         (1<<0) >>>>>  #define EXT_CSD_CMD_SET_SECURE         (1<<1) >>>> >>>> >> >> > ÿôèº{.nÇ+·®+%Ëÿ±éݶ\x17¥wÿº{.nÇ+·¥{±þG«éÿ{ayº\x1dÊÚë,j\a¢f£¢·hïêÿêçz_è®\x03(éÝ¢j"ú\x1a¶^[m§ÿÿ¾\a«þG«éÿ¢¸?¨èÚ&£ø§~á¶iOæ¬z·vØ^\x14\x04\x1a¶^[m§ÿÿÃ\fÿ¶ìÿ¢¸?I¥
WARNING: multiple messages have this Message-ID (diff)
From: NamJae Jeon <linkinjeon@gmail.com> To: Adrian Hunter <adrian.hunter@intel.com> Cc: cjb@laptop.org, linux-mmc@vger.kernel.org, linux-kernel@vger.kernel.org, awarkentin@vmware.com, linus.walleij@linaro.org, james_p_freyensee@linux.intel.com, sebras@gmail.com, Ulf.Hansson@stericsson.com, stefan.xk.nilsson@stericsson.com, per.forlin@stericsson.com, johan.rudholm@stericsson.com Subject: Re: [PATCH v10] mmc : general purpose partition support. Date: Wed, 5 Oct 2011 20:22:56 +0900 [thread overview] Message-ID: <CAKYAXd9iHW3bdeFkq=ywdOTPtzFdqK8Yb9fXxi8Jvoq6BRvHtQ@mail.gmail.com> (raw) In-Reply-To: <CAKYAXd_AoSgY6htSO5jJyD3QJ_xXFqgKLyNX49s3PuRW6AUmbg@mail.gmail.com> 2011/10/5 NamJae Jeon <linkinjeon@gmail.com>: > 2011/10/5 Adrian Hunter <adrian.hunter@intel.com>: >> On 05/10/11 13:21, NamJae Jeon wrote: >>> >>> 2011/10/5 Adrian Hunter<adrian.hunter@intel.com>: >>>> >>>> On 04/10/11 20:45, Namjae Jeon wrote: >>>>> >>>>> It allows gerneral purpose partitions in MMC Device. >>>>> And I try to simpliy make mmc_blk_alloc_parts using mmc_part structure >>>>> suggested by Andrei Warkentin. >>>>> After patching, we can see general purpose partitions like this. >>>>>> >>>>>> cat /proc/partitions >>>>> >>>>> 179 0 847872 mmcblk0 >>>>> 179 192 4096 mmcblk0gp3 >>>>> 179 160 4096 mmcblk0gp2 >>>>> 179 128 4096 mmcblk0gp1 >>>>> 179 96 1052672 mmcblk0gp0 >>>>> 179 64 1024 mmcblk0boot1 >>>>> 179 32 1024 mmcblk0boot0 >>>>> >>>>> Signed-off-by: Namjae Jeon<linkinjeon@gmail.com> >>>>> --- >>>>> drivers/mmc/card/block.c | 34 +++++++++++++++++------------ >>>>> drivers/mmc/core/mmc.c | 52 >>>>> ++++++++++++++++++++++++++++++++++++++++++--- >>>>> include/linux/mmc/card.h | 34 +++++++++++++++++++++++++++++- >>>>> include/linux/mmc/mmc.h | 5 +++- >>>>> 4 files changed, 105 insertions(+), 20 deletions(-) >>>>> >>>>> diff --git a/drivers/mmc/card/block.c b/drivers/mmc/card/block.c >>>>> index 782d5f8..fb6f1e9 100644 >>>>> --- a/drivers/mmc/card/block.c >>>>> +++ b/drivers/mmc/card/block.c >>>>> @@ -1470,26 +1470,32 @@ static int mmc_blk_alloc_part(struct mmc_card >>>>> *card, >>>>> return 0; >>>>> } >>>>> >>>>> +/* MMC Physical partition consist of two boot partitions and >>>>> + * up to four general purpose partitions. >>>>> + * For each partitions enabled in EXT_CSD >>>>> + * The block device will allocated to provide access to the partition. >>>>> + */ >>>>> + >>>>> static int mmc_blk_alloc_parts(struct mmc_card *card, struct >>>>> mmc_blk_data >>>>> *md) >>>>> { >>>>> - int ret = 0; >>>>> + int idx, ret = 0; >>>>> >>>>> if (!mmc_card_mmc(card)) >>>>> return 0; >>>>> >>>>> - if (card->ext_csd.boot_size&& >>>>> mmc_boot_partition_access(card->host)) { >>>>> - ret = mmc_blk_alloc_part(card, md, >>>>> EXT_CSD_PART_CONFIG_ACC_BOOT0, >>>>> - card->ext_csd.boot_size>> 9, >>>>> - true, >>>>> - "boot0"); >>>>> - if (ret) >>>>> - return ret; >>>>> - ret = mmc_blk_alloc_part(card, md, >>>>> EXT_CSD_PART_CONFIG_ACC_BOOT1, >>>>> - card->ext_csd.boot_size>> 9, >>>>> - true, >>>>> - "boot1"); >>>>> - if (ret) >>>>> - return ret; >>>>> + for (idx = 0; idx< card->nr_parts; idx++) { >>>>> + if ((idx == 0 || idx == 1)&& >>>>> + !mmc_boot_partition_access(card->host)) >>>>> + continue; >>>> >>>> Seems a bit fragile. What if someone decided to put the gp >>>> partitions before the boot partitions. What about: >>> >>> There are the default two boot partition in mmc spec. I think that >>> this case can not be happened. >>> And boot partition size of EXT CSD is read-only. >>> Can you explain which is case put to gp partitions before the boot >>> partition ? >> >> Relying on the partition order in the partition array is fragile. >> You cannot predict how someone may want to change it in the future, >> and that person will not expect that certain positions have certain meaning. 0,1 of Boot partition is fixed if spec is not changed over. but gp partition is variable in 2,3,4,5,6 of array. If How someone change code, problem occur ? >> >>> >>>> >>>> if (mmc_is_boot_partition(&part[idx])&& >>>> !mmc_boot_partition_access(card->host) >>>> continue; >>>> >>>> And: >>>> >>>> int mmc_is_boot_partition(struct mmc_part *part) >>>> { >>>> return part->part_cfg>= EXT_CSD_PART_CONFIG_ACC_BOOT0&& >>>> part->part_cfg<= EXT_CSD_PART_CONFIG_ACC_BOOT0 + >>>> MMC_NUM_BOOT_PARTITION; >>>> } >>>> >>>>> + if (card->part[idx].size) { >>>>> + ret = mmc_blk_alloc_part(card, md, >>>>> + card->part[idx].part_cfg, >>>>> + card->part[idx].size>> 9, >>>>> + card->part[idx].force_ro, >>>>> + card->part[idx].name); >>>>> + if (ret) >>>>> + return ret; >>>>> + } >>>>> } >>>>> >>>>> return ret; >>>>> diff --git a/drivers/mmc/core/mmc.c b/drivers/mmc/core/mmc.c >>>>> index c632b1f..2f458a0 100644 >>>>> --- a/drivers/mmc/core/mmc.c >>>>> +++ b/drivers/mmc/core/mmc.c >>>>> @@ -239,7 +239,9 @@ static int mmc_get_ext_csd(struct mmc_card *card, u8 >>>>> **new_ext_csd) >>>>> */ >>>>> static int mmc_read_ext_csd(struct mmc_card *card, u8 *ext_csd) >>>>> { >>>>> - int err = 0; >>>>> + int err = 0, idx; >>>>> + unsigned int part_size; >>>>> + u8 hc_erase_grp_sz = 0, hc_wp_grp_sz = 0; >>>>> >>>>> BUG_ON(!card); >>>>> >>>>> @@ -340,7 +342,14 @@ static int mmc_read_ext_csd(struct mmc_card *card, >>>>> u8 >>>>> *ext_csd) >>>>> * There are two boot regions of equal size, defined in >>>>> * multiples of 128K. >>>>> */ >>>>> - card->ext_csd.boot_size = ext_csd[EXT_CSD_BOOT_MULT]<< >>>>> 17; >>>>> + if (ext_csd[EXT_CSD_BOOT_MULT]) { >>>>> + for (idx = 0; idx< MMC_NUM_BOOT_PARTITION; >>>>> idx++) >>>>> { >>>>> + part_size = ext_csd[EXT_CSD_BOOT_MULT]<< >>>>> 17; >>>>> + mmc_part_add(card, part_size, >>>>> + EXT_CSD_PART_CONFIG_ACC_BOOT0 + >>>>> idx, >>>>> + "boot%d", idx, true); >>>>> + } >>>>> + } >>>>> } >>>>> >>>>> card->ext_csd.raw_hc_erase_gap_size = >>>>> @@ -362,9 +371,9 @@ static int mmc_read_ext_csd(struct mmc_card *card, >>>>> u8 >>>>> *ext_csd) >>>>> card->ext_csd.raw_partition_support = >>>>> ext_csd[EXT_CSD_PARTITION_SUPPORT]; >>>>> if ((ext_csd[EXT_CSD_PARTITION_SUPPORT]& 0x2)&& >>>>> (ext_csd[EXT_CSD_PARTITION_ATTRIBUTE]& 0x1)) { >>>>> - u8 hc_erase_grp_sz = >>>>> + hc_erase_grp_sz = >>>>> ext_csd[EXT_CSD_HC_ERASE_GRP_SIZE]; >>>>> - u8 hc_wp_grp_sz = >>>>> + hc_wp_grp_sz = >>>>> ext_csd[EXT_CSD_HC_WP_GRP_SIZE]; >>>>> >>>>> card->ext_csd.enhanced_area_en = 1; >>>>> @@ -393,6 +402,41 @@ static int mmc_read_ext_csd(struct mmc_card *card, >>>>> u8 >>>>> *ext_csd) >>>>> card->ext_csd.enhanced_area_offset = -EINVAL; >>>>> card->ext_csd.enhanced_area_size = -EINVAL; >>>>> } >>>>> + >>>>> + /* >>>>> + * General purpose partition feature support -- >>>>> + * If ext_csd have the size of general purpose >>>>> partitions, >>>>> + * set size, part_cfg, partition name in mmc_part. >>>>> + */ >>>>> + if (ext_csd[EXT_CSD_PARTITION_SUPPORT]& >>>>> + EXT_CSD_PART_SUPPORT_PART_EN) { >>>>> + if (card->ext_csd.enhanced_area_en != 1) { >>>>> + hc_erase_grp_sz = >>>>> + >>>>> ext_csd[EXT_CSD_HC_ERASE_GRP_SIZE]; >>>>> + hc_wp_grp_sz = >>>>> + ext_csd[EXT_CSD_HC_WP_GRP_SIZE]; >>>>> + >>>>> + card->ext_csd.enhanced_area_en = 1; >>>>> + } >>>>> + >>>>> + for (idx = 0; idx< MMC_NUM_GP_PARTITION; >>>>> idx++) { >>>>> + if (!ext_csd[EXT_CSD_GP_SIZE_MULT + idx >>>>> * >>>>> 3]&& >>>>> + !ext_csd[EXT_CSD_GP_SIZE_MULT + idx * 3 >>>>> + >>>>> 1]&& >>>>> + !ext_csd[EXT_CSD_GP_SIZE_MULT + idx * 3 >>>>> + >>>>> 2]) >>>>> + continue; >>>>> + part_size = >>>>> + (ext_csd[EXT_CSD_GP_SIZE_MULT + idx * 3 >>>>> + >>>>> 2] >>>>> +<< 16) + >>>>> + (ext_csd[EXT_CSD_GP_SIZE_MULT + idx * 3 >>>>> + >>>>> 1] >>>>> +<< 8) + >>>>> + ext_csd[EXT_CSD_GP_SIZE_MULT + idx * 3]; >>>>> + part_size *= (size_t)(hc_erase_grp_sz * >>>>> + hc_wp_grp_sz); >>>>> + mmc_part_add(card, part_size<< 19, >>>>> + EXT_CSD_PART_CONFIG_ACC_GP0 + >>>>> idx, >>>>> + "gp%d", idx, false); >>>>> + } >>>>> + } >>>>> card->ext_csd.sec_trim_mult = >>>>> ext_csd[EXT_CSD_SEC_TRIM_MULT]; >>>>> card->ext_csd.sec_erase_mult = >>>>> diff --git a/include/linux/mmc/card.h b/include/linux/mmc/card.h >>>>> index 5294ddf..0522627 100644 >>>>> --- a/include/linux/mmc/card.h >>>>> +++ b/include/linux/mmc/card.h >>>>> @@ -12,6 +12,7 @@ >>>>> >>>>> #include<linux/mmc/core.h> >>>>> #include<linux/mod_devicetable.h> >>>>> +#include<linux/genhd.h> >>>>> >>>>> struct mmc_cid { >>>>> unsigned int manfid; >>>>> @@ -64,7 +65,6 @@ struct mmc_ext_csd { >>>>> bool enhanced_area_en; /* enable bit */ >>>>> unsigned long long enhanced_area_offset; /* Units: Byte */ >>>>> unsigned int enhanced_area_size; /* Units: KB */ >>>>> - unsigned int boot_size; /* in bytes */ >>>>> u8 raw_partition_support; /* 160 */ >>>>> u8 raw_erased_mem_count; /* 181 */ >>>>> u8 raw_ext_csd_structure; /* 194 */ >>>>> @@ -158,6 +158,23 @@ struct sdio_func_tuple; >>>>> >>>>> #define SDIO_MAX_FUNCS 7 >>>>> >>>>> +/* The number of MMC physical partitions >>>>> + * It consist of boot partitions(2), general purpose partitions(4) in >>>>> MMC >>>>> v4.4 >>>>> + */ >>>>> +#define MMC_NUM_BOOT_PARTITION 2 >>>>> +#define MMC_NUM_GP_PARTITION 4 >>>>> +#define MMC_NUM_PHY_PARTITION 6 >>>>> + >>>>> +/* >>>>> + * MMC Physical partitions >>>>> + */ >>>>> +struct mmc_part { >>>>> + unsigned int size; /* partition size (in bytes) */ >>>>> + unsigned int part_cfg; /* it used to part_type */ >>>>> + char name[DISK_NAME_LEN]; >>>>> + bool force_ro; /* to make boot parts RO by default */ >>>>> +}; >>>>> + >>>>> /* >>>>> * MMC device >>>>> */ >>>>> @@ -219,9 +236,24 @@ struct mmc_card { >>>>> unsigned int sd_bus_speed; /* Bus Speed Mode set for >>>>> the card */ >>>>> >>>>> struct dentry *debugfs_root; >>>>> + struct mmc_part part[MMC_NUM_PHY_PARTITION]; /* physical >>>>> partitions */ >>>>> + unsigned int nr_parts; >>>>> }; >>>>> >>>>> /* >>>>> + * This function fill contents in mmc_part. >>>>> + */ >>>>> +static inline void mmc_part_add(struct mmc_card *card, unsigned int >>>>> size, >>>>> + unsigned int part_cfg, char *name, int idx, bool >>>>> ro) >>>>> +{ >>>>> + card->part[card->nr_parts].size = size; >>>>> + card->part[card->nr_parts].part_cfg = part_cfg; >>>>> + sprintf(card->part[card->nr_parts].name, name, idx); >>>>> + card->part[card->nr_parts].force_ro = ro; >>>>> + card->nr_parts++; >>>>> +} >>>>> + >>>>> +/* >>>>> * The world is not perfect and supplies us with broken mmc/sdio >>>>> devices. >>>>> * For at least some of these bugs we need a work-around. >>>>> */ >>>>> diff --git a/include/linux/mmc/mmc.h b/include/linux/mmc/mmc.h >>>>> index 50af227..ea558eb 100644 >>>>> --- a/include/linux/mmc/mmc.h >>>>> +++ b/include/linux/mmc/mmc.h >>>>> @@ -270,6 +270,7 @@ struct _mmc_csd { >>>>> * EXT_CSD fields >>>>> */ >>>>> >>>>> +#define EXT_CSD_GP_SIZE_MULT 143 /* R/W */ >>>>> #define EXT_CSD_PARTITION_ATTRIBUTE 156 /* R/W */ >>>>> #define EXT_CSD_PARTITION_SUPPORT 160 /* RO */ >>>>> #define EXT_CSD_RST_N_FUNCTION 162 /* R/W */ >>>>> @@ -313,7 +314,9 @@ struct _mmc_csd { >>>>> >>>>> #define EXT_CSD_PART_CONFIG_ACC_MASK (0x7) >>>>> #define EXT_CSD_PART_CONFIG_ACC_BOOT0 (0x1) >>>>> -#define EXT_CSD_PART_CONFIG_ACC_BOOT1 (0x2) >>>>> +#define EXT_CSD_PART_CONFIG_ACC_GP0 (0x4) >>>>> + >>>>> +#define EXT_CSD_PART_SUPPORT_PART_EN (0x1) >>>>> >>>>> #define EXT_CSD_CMD_SET_NORMAL (1<<0) >>>>> #define EXT_CSD_CMD_SET_SECURE (1<<1) >>>> >>>> >> >> >
next prev parent reply other threads:[~2011-10-05 11:23 UTC|newest] Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top 2011-10-04 17:45 [PATCH v10] mmc : general purpose partition support Namjae Jeon 2011-10-04 18:19 ` S, Venkatraman 2011-10-04 18:25 ` Chris Ball 2011-10-04 18:25 ` Chris Ball 2011-10-04 18:27 ` S, Venkatraman 2011-10-04 20:20 ` Andrei Warkentin 2011-10-05 9:12 ` Adrian Hunter 2011-10-05 10:21 ` NamJae Jeon 2011-10-05 10:21 ` NamJae Jeon 2011-10-05 10:53 ` Adrian Hunter 2011-10-05 11:21 ` NamJae Jeon 2011-10-05 11:21 ` NamJae Jeon 2011-10-05 11:22 ` NamJae Jeon [this message] 2011-10-05 11:22 ` NamJae Jeon 2011-10-05 11:37 ` Adrian Hunter 2011-10-05 11:49 ` NamJae Jeon
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='CAKYAXd9iHW3bdeFkq=ywdOTPtzFdqK8Yb9fXxi8Jvoq6BRvHtQ@mail.gmail.com' \ --to=linkinjeon@gmail.com \ --cc=Ulf.Hansson@stericsson.com \ --cc=adrian.hunter@intel.com \ --cc=awarkentin@vmware.com \ --cc=cjb@laptop.org \ --cc=james_p_freyensee@linux.intel.com \ --cc=johan.rudholm@stericsson.com \ --cc=linus.walleij@linaro.org \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-mmc@vger.kernel.org \ --cc=per.forlin@stericsson.com \ --cc=sebras@gmail.com \ --cc=stefan.xk.nilsson@stericsson.com \ /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: linkBe 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.