xiaomi-sm8450/android_kernel_xiaomi_sm8450
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

[libata] support for > 512 byte sectors (e.g. 4K Native)

Browse Source 
This change enables my x86 machine to recognize and talk to a
"Native 4K" SATA device.

When I started working on this, I didn't know Matthew Wilcox had
posted a similar patch 2 years ago:
  http://git.kernel.org/?p=linux/kernel/git/willy/ata.git;a=shortlog;h=refs/heads/ata-large-sectors

Gwendal Grignou pointed me at the the above code and small portions of
this patch include Matthew's work. That's why Mathew is first on the
"Signed-off-by:". I've NOT included his use of a bitmap to determine
512 vs Native for ATA command block size - just used a simple table.
And bugs are almost certainly mine.

Lastly, the patch has been tested with a native 4K 'Engineering
Sample' drive provided by Hitachi GST.

Signed-off-by: Matthew Wilcox <matthew.r.wilcox@intel.com>
Signed-off-by: Grant Grundler <grundler@google.com>
Reviewed-by: Gwendal Grignou <gwendal@google.com>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
This commit is contained in:
Grant Grundler
2010-08-17 10:56:53 -07:00
committed by Jeff Garzik
parent 1aadf5c3bb
commit 295124dce4
2 changed files with 104 additions and 36 deletions
Download Patch File Download Diff File Expand all files Collapse all files

46
include/linux/ata.h
View File

@@ -89,6 +89,7 @@ enum {
ATA_ID_SPG = 98,
ATA_ID_LBA_CAPACITY_2 = 100,
ATA_ID_SECTOR_SIZE = 106,
ATA_ID_LOGICAL_SECTOR_SIZE = 117, /* and 118 */
ATA_ID_LAST_LUN = 126,
ATA_ID_DLF = 128,
ATA_ID_CSFO = 129,
@@ -640,16 +641,49 @@ static inline int ata_id_flush_ext_enabled(const u16 *id)
return (id[ATA_ID_CFS_ENABLE_2] & 0x2400) == 0x2400;
}
static inline int ata_id_has_large_logical_sectors(const u16 *id)
static inline u32 ata_id_logical_sector_size(const u16 *id)
{
if ((id[ATA_ID_SECTOR_SIZE] & 0xc000) != 0x4000)
return 0;
return id[ATA_ID_SECTOR_SIZE] & (1 << 13);
/* T13/1699-D Revision 6a, Sep 6, 2008. Page 128.
* IDENTIFY DEVICE data, word 117-118.
* 0xd000 ignores bit 13 (logical:physical > 1)
*/
if ((id[ATA_ID_SECTOR_SIZE] & 0xd000) == 0x5000)
return (((id[ATA_ID_LOGICAL_SECTOR_SIZE+1] << 16)
+ id[ATA_ID_LOGICAL_SECTOR_SIZE]) * sizeof(u16)) ;
return ATA_SECT_SIZE;
}
static inline u16 ata_id_logical_per_physical_sectors(const u16 *id)
static inline u8 ata_id_log2_per_physical_sector(const u16 *id)
{
return 1 << (id[ATA_ID_SECTOR_SIZE] & 0xf);
/* T13/1699-D Revision 6a, Sep 6, 2008. Page 128.
* IDENTIFY DEVICE data, word 106.
* 0xe000 ignores bit 12 (logical sector > 512 bytes)
*/
if ((id[ATA_ID_SECTOR_SIZE] & 0xe000) == 0x6000)
return (id[ATA_ID_SECTOR_SIZE] & 0xf);
return 0;
}
/* Offset of logical sectors relative to physical sectors.
*
* If device has more than one logical sector per physical sector
* (aka 512 byte emulation), vendors might offset the "sector 0" address
* so sector 63 is "naturally aligned" - e.g. FAT partition table.
* This avoids Read/Mod/Write penalties when using FAT partition table
* and updating "well aligned" (FS perspective) physical sectors on every
* transaction.
*/
static inline u16 ata_id_logical_sector_offset(const u16 *id,
u8 log2_per_phys)
{
u16 word_209 = id[209];
if ((log2_per_phys > 1) && (word_209 & 0xc000) == 0x4000) {
u16 first = word_209 & 0x3fff;
if (first > 0)
return (1 << log2_per_phys) - first;
}
return 0;
}
static inline int ata_id_has_lba48(const u16 *id)