This fixes this build warning:
Configuring for qemu_mips64 - Board: qemu-mips64, Options: SYS_BIG_ENDIAN
text data bss dec hex filename
215344 13082 218720 447146 6d2aa qemu_mips64/u-boot
cfi_flash.c: In function 'flash_map':
cfi_flash.c:217:9: warning: cast to pointer from integer of different size [-Wint-to-pointer-cast]
Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Tom Rini <trini@ti.com>
This patch addjusted the write buffer size for M29EW devices those
are operated in 8-bit mode.
The M29EW devices seem to report the CFI information wrong when
it's in 8 bit mode.
There's an app note from Numonyx on this issue and there's a patch
in the open source as well for Linux, but it doesn't seem to be in mainline.
Signed-off-by: Jagannadha Sutradharudu Teki <jaganna@xilinx.com>
Tested-by: Jagannadha Sutradharudu Teki <jaganna@xilinx.com>
This commit is based on that patch from aaron.williams@caviumnetworks.com
with same commit title. pulled the same code changes into current u-boot tree.
http://patchwork.ozlabs.org/patch/140863/http://lists.denx.de/pipermail/u-boot/2011-April/089606.html
This patch corrects the addresses used when working with Spansion/AMD FLASH chips.
Addressing for 8 and 16 bits is almost identical except in the 16-bit case the
LSB of the address is always 0. The confusion arose because the addresses
in the datasheet for 16-bit mode are word addresses but this code assumed it was
byte addresses.
I have only been able to test this on our Octeon boards which use either an 8-bit
or 16-bit bus. I have not tested the case where there's an 8-bit part on a 16-bit
bus.
This patch also adds some delays as suggested by Spansion.
If a part can be both 8 and 16-bits, it forces it to work in 8-bit mode if an
8-bit bus is detected.
Apart from the pulled changes, fixed few minor code cleanups and tested
on 256M29EW, 512M29EW flashes.
Before this fix:
---------------
Bank # 1: CFI conformant flash (8 x 8) Size: 64 MB in 512 Sectors
AMD Standard command set, Manufacturer ID: 0xFF, Device ID: 0xFF
Erase timeout: 4096 ms, write timeout: 2 ms
Buffer write timeout: 5 ms, buffer size: 1024 bytes
After this fix:
--------------
Bank # 1: CFI conformant flash (8 x 8) Size: 64 MB in 512 Sectors
AMD Standard command set, Manufacturer ID: 0x89, Device ID: 0x7E2301
Erase timeout: 4096 ms, write timeout: 2 ms
Buffer write timeout: 5 ms, buffer size: 1024 bytes
Signed-off-by: Aaron Williams <aaron.williams@caviumnetworks.com>
Signed-off-by: Jagannadha Sutradharudu Teki <jaganna@xilinx.com>
Tested-by: Jagannadha Sutradharudu Teki <jaganna@xilinx.com>
Not only Spansion supports the Persistent Protection Bits (PPB) locking.
Other devices like the Micron JS28F512M29EWx also support this type
of locking/unlocking. Detection of support is done in the same way as
done for the Spansion chips - via the 0x49 CFI word.
This patch enables this PPB protection mechanism for all AMD type
(AMD commandset) chips.
Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Anatolij Gustschin <agust@denx.de>
Cc: Holger Brunck <holger.brunck@keymile.com>
Tested-by: Holger Brunck <holger.brunck@keymile.com>
Patch 66863b05 [cfi_flash: add support for Spansion flash PPB sector
protection] introduced the PPB (Persistent Protection Bit) locking for
Spansion chips. But right now the sector protection status (locked vs
unlocked) is set to unlocked for all sectors upon bootup. The real
sector protection status is ignored.
This patch now reads the current sector protection status and uses
it for these AMD/Spansion flash chips.
Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Anatolij Gustschin <agust@denx.de>
Cc: Holger Brunck <holger.brunck@keymile.com>
Tested-by: Holger Brunck <holger.brunck@keymile.com>
Consolidate manufacturer matching into the function manufact_match()
and use it.
Signed-off-by: Stefan Roese <sr@denx.de>
Tested-by: Holger Brunck <holger.brunck@keymile.com>
Add support for SST 4KB sector granularity.
Many recent SST flashes, i.e. SST39VF3201B and similar of this family
are declared CFI-conformant from SST. They support CFI query, but implement
2 different sector sizes in the same memory: a 64KB sector (they call it
"block", std AMD erase cmd=0x30), and a 4KB sector (they call it "sector",
erase cmd=0x50). Also, CFI query on these chips, reading from address 0x2dh
of cfi query struct, detects a number of secotrs for the 4KB granularity
(flinfo shows it).
For all other aspects, they are CFI compliant, so, as Linux do, i think
it's a good idea to handle these chips in the CFI driver, with a fixup
to allow 4KB granularity, as should be expected, instead of 64KB.
Signed-off-by: Angelo Dureghello <sysamfw@gmail.com>
Signed-off-by: Stefan Rose <sr@denx.de>
cfi_flash.c:756:43: warning: Using plain integer as NULL pointer
cfi_flash.c:1150:52: warning: cast to non-scalar
cfi_flash.c:1433:46: warning: cast truncates bits from constant value (890089 becomes 89)
cfi_flash.c:1490:61: warning: cast truncates bits from constant value (1f001f becomes 1f)
cfi_flash.c:1508:61: warning: cast truncates bits from constant value (10001 becomes 1)
cfi_flash.c:1738:63: warning: cast truncates bits from constant value (10001 becomes 1)
cfi_flash.c:1857:6: warning: symbol '__flash_cmd_reset' was not declared. Should it be static?
Signed-off-by: Kim Phillips <kim.phillips@freescale.com>
Check for ctrlc() in operations that take time and loop over the flash
addresses.
In netconsole, tstc() is expensive. Only check once in a while to not
slow down the operation significantly.
Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>
Signed-off-by: Stefan Roese <sr@denx.de>
Added an optional check in the CFI driver to evaluate if the sector is
already blank before issuing an erase command. Improves erase time by
over a factor of 10 if already blank.
Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>
Signed-off-by: Stefan Roese <sr@denx.de>
commit 54652991
Work around bug in Numonyx P33/P30 256-Mbit 65nm flash chips
fixes a problem for Numonyx P33/P30 flashes for 256-Mbit, but this leads
to problems for smaller versions of this chip e.g. the 32Mbit version
with deviceid 0x16 on mgcoge. So move the code for this work around to
an own function and check previously manufacturer id and device id to
not break other flashes which don't need this work around.
Signed-off-by: Holger Brunck <holger.brunck@keymile.com>
Signed-off-by: Heiko Schocher <hs@denx.de>
cc: Stefan Roese <sr@denx.de>
cc: Philippe De Muyter <phdm@macqel.be>
cc: Gerlando Falauto <gerlando.falauto@keymile.com>
Signed-off-by: Stefan Roese <sr@denx.de>
Erasing flash sectors protected with persistent protection bit (PPB)
mechanism on Spansion flash chips doesn't work. Add sector protection
status checking and sector lock and unlock commands to fix this.
Signed-off-by: Anatolij Gustschin <agust@denx.de>
Cc: Stefan Roese <sr@denx.de>
Signed-off-by: Stefan Roese <sr@denx.de>
Without the fix, flash_write_cfibuffer will terminate the erased
status check once an all-0xFF word has been found instead of
continuing the erased status check utill the first non-0xFF word.
Signed-off-by: Tao Hou <hotforest@gmail.com>
Cc: Stefan Roese <sr@denx.de>
Signed-off-by: Stefan Roese <sr@denx.de>
The current implementation of cfi_flash uses udelay to reset watchdog.
If several blocks are blank (0xff filled) the watchdog timed out.
The reason is, udelay is never called, if flash device is ready to fast.
e.g.
mw.w $(copy_addr) FFFF 10000;cp.b $(copy_addr) 10880000 20000
-> watchdog timed out
mw.w $(copy_addr) 0000 10000;cp.b $(copy_addr) 10880000 20000
-> watchdog not timed out
This patch adds an extra watchdog reset in front of flash ready check.
Signed-off-by: Jens Scharsig (BuS Elektronik) <esw@bus-elektronik.de>
Signed-off-by: Stefan Roese <sr@denx.de>
cfi_flash.c: In function 'flash_protect_default':
cfi_flash.c:2152: warning: format '%08x' expects type 'unsigned int', but
argument 2 has type 'ulong'
cfi_flash.c:2152: warning: format '%08x' expects type 'unsigned int', but
argument 3 has type 'long unsigned int'
Signed-off-by: Marek Vasut <marek.vasut@gmail.com>
Cc: Wolfgang Denk <wd@denx.de>
Cc: Simon Glass <sjg@chromium.org>
Cc: Mike Frysinger <vapier@gentoo.org>
Cc: Scott Wood <scottwood@freescale.com>
For newer STM parts where CFI >= 1.1, there is a byte in the extended
structure that declares the flash layout type (just like the AMD parts),
so key off of that to find out when we need to reverse the geometry.
This can be seen with M29W640 parts where U-Boot does:
Bank # 1: CFI conformant FLASH (16 x 16) Size: 8 MB in 135 Sectors
AMD Standard command set, Manufacturer ID: 0x20, Device ID: 0x22ED
Erase timeout: 8192 ms, write timeout: 1 ms
Buffer write timeout: 1 ms, buffer size: 16 bytes
Sector Start Addresses:
20000000 RO 20002000 RO 20004000 RO 20006000 RO 20008000 RO
2000A000 RO 2000C000 RO 2000E000 RO 20010000 RO 20020000 RO
...
But Linux does:
physmap platform flash device: 00800000 at 20000000
physmap-flash.0: Found 1 x16 devices at 0x0 in 16-bit bank.
Manufacturer ID 0x000020 Chip ID 0x0022ed
physmap-flash.0: Swapping erase regions for top-boot CFI table.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Stefan Roese <sr@denx.de>
I ran into a problem where the reset was failing except when I enabled
debugging support. After talking with Garret Swalling at Spansion I
was told that the GL-N series of devices require a 500ns wait for the
reset to complete. The below patch adds a 1us delay after all reset
commands.
-Aaron Williams
Signed-off-by: Aaron Williams <aaron.williams@caviumnetworks.com>
Signed-off-by: Stefan Roese <sr@denx.de>
The M29W800DT parts also report their geometry with the sector layout
reversed. So add that ID to the flash_fixup_stm function.
Otherwise, we get:
bfin> flinfo
Bank # 1: CFI conformant FLASH (16 x 16) Size: 1 MB in 19 Sectors
AMD Standard command set, Manufacturer ID: 0x20, Device ID: 0x22D7
Erase timeout: 8192 ms, write timeout: 1 ms
Sector Start Addresses:
20000000 20004000 20006000 20008000 20010000
20020000 20030000 20040000 20050000 20060000
20070000 20080000 20090000 200A0000 200B0000
200C0000 200D0000 200E0000 200F0000
Reported-by: Jianxi Fu <fujianxi@gmail.com>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Stefan Roese <sr@denx.de>
Commit 6ee1416e81 (mtd, cfi: introduce
void flash_protect_default(void)) introduced a bug which resulted in
boards that define CONFIG_SYS_FLASH_AUTOPROTECT_LIST not compiling with
the the following errors and warning:
ptyser@petert u-boot $ make -s xpedite520x
Configuring for xpedite520x board...
cfi_flash.c: In function 'flash_protect_default':
cfi_flash.c:2118: error: 'i' undeclared (first use in this function)
cfi_flash.c:2118: error: (Each undeclared identifier is reported only once
cfi_flash.c:2118: error: for each function it appears in.)
cfi_flash.c:2118: error: 'apl' undeclared (first use in this function)
cfi_flash.c:2118: error: invalid application of 'sizeof' to incomplete type 'struct apl_s'
cfi_flash.c: In function 'flash_init':
cfi_flash.c:2137: warning: unused variable 'apl'
Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Reported-by: Kumar Gala <galak@kernel.crashing.org>
Cc: Heiko Schocher <hs@denx.de>
collect code which protects default sectors in a function, called
flash_protect_default. So boardspecific code can call it too.
Signed-off-by: Heiko Schocher <hs@denx.de>
Signed-off-by: Stefan Roese <sr@denx.de>
The function find_sector() does not take into account if the flash bank
has changed since the last call. This could lead to illegal accesses inside
and beyond the flash_info_t info strcture. For example if the current
flash bank has less sectors than the last used flash bank.
This patch adds two cheks. One that insures, that the current sector does
not exceed the allowed maximum (which is always a good idea). And one that
checks if the current access is to the same flash bank as the last access.
If not, the search loop will start with sector 0.
Signed-off-by: Martin Krause <martin.krause@tqs.de>
Signed-off-by: Stefan Roese <sr@denx.de>
There's no compelling reason to have the output on bootup or the
"flinfo" command print "flash" in uppercase, so use the proper case
where appropriate.
Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
The flash_verbose logic is only used by the CFI MTD layer, so if we aren't
using that, disable the logic completely.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
This patch adds the possibility to (optinally) write to the
flash configuration register. The Intel style CFI chips support
such a register that can be used to configure the operation
mode to a non-default value.
This method will be used by the t3corp board, which needs to
configure the DS617 Xilinx flash for async read mode.
Signed-off-by: Stefan Roese <sr@denx.de>
The function sector_erased() is modified to not use pointer
access, but to use the correct accessor functions. This fixes a
problem on the t3corp board with the Xilinx DS617 flash chips. Here
a board specific accessor function is needed to read from flash
in 32bit mode. This patch enables such an operation mode.
Signed-off-by: Stefan Roese <sr@denx.de>
This patch adds some calls to set the flash chip in the read-status-
register- or read-id-mode before the corresponding register is
read back. This problem was detected while porting the common CFI
driver to support the Xilinx DS617 flash chips.
Signed-off-by: Stefan Roese <sr@denx.de>
commit ec50a8e389
"cfi_flash: handle 'chip size exceeds address window' situation"
added 3rd argument to flash_get_size() but didn't fix all the
function calls from the board specific code. Many boards have
their own flash_get_size() definitions in the board code and
use them there, but some boards (e.g. tqm834x, tqm85xx, pdm360ng)
use flash_get_size() from the cfi_flash.c driver.
The bug shows up if the value of the "max_size" argument (which
is not defined when calling the function with two arguments)
happens to be less than "info->size". In this case on the
affected boards we end up with a bank of reduced size and
in the worst case might even be not able to update U-Boot or
to boot the kernel from flash:
=> fli
Bank # 1: CFI conformant FLASH (32 x 16) Size: 0 kB in 1 Sectors
AMD Standard command set, Manufacturer ID: 0x01, Device ID: 0x227E
Erase timeout: 4096 ms, write timeout: 1 ms
Buffer write timeout: 3 ms, buffer size: 64 bytes
Sector Start Addresses:
F0000000 RO
Bank # 2: CFI conformant FLASH (32 x 16) Size: 128 MB in 512 Sectors
AMD Standard command set, Manufacturer ID: 0x01, Device ID: 0x227E
Erase timeout: 4096 ms, write timeout: 1 ms
Buffer write timeout: 3 ms, buffer size: 64 bytes
Sector Start Addresses:
F8000000 F8040000 F8080000 F80C0000 F8100000
F8140000 F8180000 F81C0000 F8200000 F8240000
...
E.g., updating U-Boot is not possible now:
=> protect off ${u-boot_addr} +${u-boot_size}
Error: end address (0xf007ffff) not in flash!
Bad address format
=> era ${u-boot_addr} +${u-boot_size}
Error: end address (0xf007ffff) not in flash!
Bad address format
This patch removes the 3rd argument of flash_get_size() again
and sets "max_size" in the function itself instead of passing
it as a function argument.
Signed-off-by: Anatolij Gustschin <agust@denx.de>
On some boards we have flash mapped high in the address space with
considerably small window (say 0xFE000000 and 32MB). When we install
bigger chip (say 64MB) on such a board strange things happen
(flash_write() doesn't work at all, for ex). That's because cfi_flash
driver doesn't care about window size at all.
Of course, cleanest solution would probably be to just extend address
window to be able to map the whole flash but for legacy/compatibility
reasons some people prefer just truncate the flash size and never use
the upper part.
This patch adds an option for cfi_flash driver to handle this situation
properly. To achieve this we add the new function cfi_flash_bank_size()
which can be provided by the board code and weak-aliased to default
implementation that returns value from the CONFIG_SYS_FLASH_BANKS_SIZES
array if it's defined or 0 otherwise (the last case is added for
compatibility).
If non-zero flash bank size is provided and detected chip size is bigger
than provided address window size the warning will be displayed and
flash chip will be truncated.
Signed-off-by: Ilya Yanok <yanok@emcraft.com>
Changed cfi_flash_bank_size() return type to unsigned long
to match caller function.
Signed-off-by: Wolfgang Denk <wd@denx.de>
When a CFI flash chip could not be detected an error message similar to
the following would be printed on bootup:
FLASH: ## Unknown FLASH on Bank 1 - Size = 0x01000000 = 0 MB
The printf incorrectly converted the flash size into megabytes. This
patch fixes the printing of the flash size in megabytes:
FLASH: ## Unknown FLASH on Bank 1 - Size = 0x01000000 = 16 MB
Signed-off-by: John Schmoller <jschmoller@xes-inc.com>
Signed-off-by: Peter Tyser <ptyser@xes-inc.com>
Signed-off-by: Stefan Roese <sr@denx.de>
This patch simplifies the use of CONFIG_SYS_MAX_FLASH_BANKS_DETECT. By
moving these optional variables and defines into the common code, board
specific code is minimized. Currently only the following board use
this feature:
APC405, IDS8247, TQM834x
And IDS8247 doesn't seem to really need this feature, since its not
updating the bank number variable at all. So this patch removes the
definition of CONFIG_SYS_MAX_FLASH_BANKS_DETECT from this board port.
This new framework will be used by the upcoming lwmon5 update as well.
Signed-off-by: Stefan Roese <sr@denx.de>
Acked-by: Heiko Schocher <hs@denx.de>
Cc: Matthias Fuchs <matthias.fuchs@esd.eu>
cfi_flash_bank_addr(int bank_nr) returns the base addresses of the
requested bank. Introducing this weak default enables boards to override
this functions with a board specific version when required.
This feature will be used in the lwmon5 board update, supporting runtime
detection of 2 board revisions with different flash layouts.
Signed-off-by: Stefan Roese <sr@denx.de>
This patch removes an unecessary check in the return statement. This is
not needed, since "info" is initializes to NULL. And "info" will not be
written to again, if the flash address is not found.
Additionally "info" is not initialized to "0" but to "NULL".
Signed-off-by: Stefan Roese <sr@denx.de>
I have "ported" U-boot to a in house made board with Numonyx Axcell P33/P30
256-Mbit 65nm flash chips.
After some time :( searching for bugs in our board or soft, we have
discovered that those chips have a small but annoying bug, documented in
"Numonyx Axcell P33/P30 256-Mbit Specification Update"
It states :
When customer uses [...] block unlock, the block lock status might be
altered inadvertently. Lock status might be set to either 01h or 03h
unexpectedly (00h as expected data), which leads to program/erase failure
on certain blocks.
A working workaround is given, which I have applied and tested with success :
Workaround: If the interval between 60h and its subsequent command
can be guaranteed within 20us, Option I is recommended,
otherwise Option II (involves hardware) should be selected.
Option I: The table below lists the detail command sequences:
Command
Data bus Address bus Remarks
Sequence
1 90h Block Address
Read Lock Status
2 Read Block Address + 02h
(2)(3) (1)
3 60h Block Address
(2)(3) (1) Lock/Unlock/RCR Configuration
4 D0h/01h/03h Block Address
Notes:
(1) Block Address refers to RCR configuration data only when the 60h
command sequence is used to set RCR register combined with 03h
subsequent command.
(2) For the third and fourth command sequences, the Block Address must
be the same.
(3) The interval between 60h command and its subsequent D0h/01h/2Fh/03h
commands should be less than 20us.
And here is a log comparison of a simple (destructive) flash test without
and with the workaround.
diff without-numonyx-workaround.log with-numonyx-workaround.log
-U-Boot 2010.06-00696-g22b002c-dirty (Aug 16 2010 - 15:07:47)
+U-Boot 2010.06-00696-g22b002c-dirty (Aug 16 2010 - 15:25:19)
CPU: Freescale MCF5484
CPU CLK 200 MHz BUS CLK 100 MHz
Board: Macq Electronique ME2060
I2C: ready
DRAM: 64 MiB
FLASH: 32 MiB
In: serial
Out: serial
Err: serial
Net: FEC0, FEC1
-> flinfo
Bank # 1: CFI conformant FLASH (16 x 16) Size: 32 MB in 259 Sectors
Intel Extended command set, Manufacturer ID: 0x89, Device ID: 0x8922
Erase timeout: 4096 ms, write timeout: 1 ms
Buffer write timeout: 5 ms, buffer size: 1024 bytes
Sector Start Addresses:
FE000000 RO FE008000 RO FE010000 RO FE018000 RO FE020000 RO
FE040000 RO FE060000 RO FE080000 RO FE0A0000 RO FE0C0000 RO
...
FFF80000 RO FFFA0000 RO FFFC0000 RO FFFE0000 RO
-> protect off all
Un-Protect Flash Bank # 1
................... done
-> erase all
Erase Flash Bank # 1
................... done
-> cp.b 1000000 fe000000 2000000
-Copy to Flash... Flash not Erased
+Copy to Flash... done
->
Signed-off-by: Philippe De Muyter <phdm@macqel.be>
Signed-off-by: Stefan Roese <sr@denx.de>
This patch does the following:
- Extract code to detect if sector is erased into function
sector_erased().
- Because of this, we don't have variable declarations inside the
sector loop in flash_print_info()
- Change "return" to "break" in the "if (ctrlc()) statement:
This fixes a problem with the resulting output. Before this
patch the output was:
Sector Start Addresses:
FC000000 FC020000 FC040000 =>
With this patch it is now:
Sector Start Addresses:
FC000000 FC020000 FC040000
=>
Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Kim Phillips <kim.phillips@freescale.com>
Cc: Wolfgang Denk <wd@denx.de>
Fix reading and printing of CFI flashes 16-bit devices identifiers
Nowadays CFI flashes have a 16-bit device identifier. U-boot still
print them and read them as if they were only 8-bit wide. Fix that.
Before:
Intel Extended command set, Manufacturer ID: 0x89, Device ID: 0x1B
After:
Intel Extended command set, Manufacturer ID: 0x89, Device ID: 0x881B
Signed-off-by: Philippe De Muyter <phdm@macqel.be>
Signed-off-by: Stefan Roese <sr@denx.de>
While running from flash, i. e. before relocation, we have only a
limited C runtime environment without writable data segment. In this
phase, some configurations (for example with environment in EEPROM)
must not use the normal getenv(), but a special function. This
function had been called getenv_r(), with the idea that the "_r"
suffix would mean the same as in the _r_eentrant versions of some of
the C library functions (for example getdate vs. getdate_r, getgrent
vs. getgrent_r, etc.).
Unfortunately this was a misleading name, as in U-Boot the "_r"
generally means "running from RAM", i. e. _after_ relocation.
To avoid confusion, rename into getenv_f() [as "running from flash"]
Signed-off-by: Wolfgang Denk <wd@denx.de>
Acked-by: Detlev Zundel <dzu@denx.de>
This patch adds reset_timer() before the flash status check
waiting loop.
Since the timer is basically running asynchronous to the cfi
code, it is possible to call get_timer(0), then only a few
_SYSCLK_ cycles later an interrupt is generated. This causes
timeout even though much less time has elapsed. So the timer
period registers should be reset before get_timer(0) is
called.
There is similar usage in nand_base.c.
Signed-off-by: Thomas Chou <thomas@wytron.com.tw>
Signed-off-by: Stefan Roese <sr@denx.de>
This patch adds status polling method to offer an alternative to
data toggle method for amd flash chips.
This patch is needed for nios2 cfi flash interface, where the bus
controller performs 4 bytes read cycles for a single byte read
instruction. The data toggle method can not detect chip busy
status correctly. So we have to poll DQ7, which will be inverted
when the chip is busy.
This feature is enabled with the config def,
CONFIG_SYS_CFI_FLASH_STATUS_POLL
Signed-off-by: Thomas Chou <thomas@wytron.com.tw>
Signed-off-by: Stefan Roese <sr@denx.de>
With old configuration it could happen tout=0 if CONFIG_SYS_HZ<1000.
Signed-off-by: Renato Andreola <renato.andreola@imagos.it>
Signed-off-by: Alessandro Rubini <rubini@gnudd.com>
Signed-off-by: Thomas Chou <thomas@wytron.com.tw>
Signed-off-by: Stefan Roese <sr@denx.de>
For platforms with flash below ram addresses, the current check to
activate monitor protection is wrong/insufficient. This patch fixes
CONFIG_MONITOR_IS_IN_RAM for these systems by adding a check for
this configuration.
Signed-off-by: Wolfgang Wegner <w.wegner@astro-kom.de>
Signed-off-by: Stefan Roese <sr@denx.de>
Currently the CFI driver issues both AMD and Intel reset commands.
This is because the driver doesn't know yet which chips are connected.
This dual reset seems to cause problems with the M29W128G chips as
reported by Richard Retanubun. This patch now introduces a weak default
function for the CFI reset command, still with both resets. This can
be overridden by a board specific version if necessary.
Signed-off-by: Stefan Roese <sr@denx.de>
Cc: Richard Retanubun <RichardRetanubun@ruggedcom.com>
see http://www.jedec.org/download/search/jep106Z.pdf
Add some second source legacy flash chips 256x8.
Signed-off-by: Niklaus Giger <niklaus.giger@member.fsf.org>
Signed-off-by: Stefan Roese <sr@denx.de>
Fix bug introduced by 9c048b5234.
The cfi_flash.c driver cast the flash buffer size to a uchar in
flash_write_cfibuffer(). On some flash parts, (tested on Numonyx
part PC32F512M29EWH), the buffer size is 1KB. Remove the cast to
uchar to enable buffer sizes to be larger.
Signed-off-by: John Schmoller <jschmoller@xes-inc.com>
Signed-off-by: Stefan Roese <sr@denx.de>
For JEDEC flash, we should issue word programming command relative to
base address rather than sector base address. Original source makes
SST Flash fails to program sectors which are not on the 0x10000 boundaries.
e.g.
SST39LF040 uses addr1=0x5555 and addr2=0x2AAA, however, each sector
is 0x1000 bytes.
Thus, if we issue command to "sector base (0x41000) + offset(0x5555)",
it sends to 0x46555 and the chip fails to recognize that address.
This patch is tested with SST39LF040.
Signed-off-by: Po-Yu Chuang <ratbert@faraday-tech.com>
Signed-off-by: Stefan Roese <sr@denx.de>
Several boards used different ways to specify the size of the
protected area when enabling flash write protection for the sectors
holding the environment variables: some used CONFIG_ENV_SIZE and
CONFIG_ENV_SIZE_REDUND, some used CONFIG_ENV_SECT_SIZE, and some even
a mix of both for the "normal" and the "redundant" areas.
Normally, this makes no difference at all. However, things are
different when you have to deal with boards that can come with
different types of flash chips, which may have different sector
sizes.
Here we may have to chose CONFIG_ENV_SECT_SIZE such that it fits the
biggest sector size, which may include several sectors on boards using
the smaller sector flash types. In such a case, using CONFIG_ENV_SIZE
or CONFIG_ENV_SIZE_REDUND to enable the protection may lead to the
case that only the first of these sectors get protected, while the
following ones aren't.
This is no real problem, but it can be confusing for the user -
especially on boards that use CONFIG_ENV_SECT_SIZE to protect the
"normal" areas, while using CONFIG_ENV_SIZE_REDUND for the
"redundant" area.
To avoid such inconsistencies, I changed all sucn boards that I found
to consistently use CONFIG_ENV_SECT_SIZE for protection. This should
not cause any functional changes to the code.
Signed-off-by: Wolfgang Denk <wd@denx.de>
Cc: Paul Ruhland
Cc: Pantelis Antoniou <panto@intracom.gr>
Cc: Stefan Roese <sr@denx.de>
Cc: Gary Jennejohn <garyj@denx.de>
Cc: Dave Ellis <DGE@sixnetio.com>
Acked-by: Stefan Roese <sr@denx.de>
The AT91RM9200-EK Evaluation Board supports the AT91RM9200
ARM9-based 32-bit RISC microcontroller and enables real-time code development
and evaluation.
Here is the chip page on Atmel website:
http://www.atmel.com/dyn/products/tools_card.asp?tool_id=3507
with
- NOR (cfi driver)
- DataFlash
- USB OHCI
- Net
- I2C (hard)
Signed-off-by: Ulf Samuelsson <ulf@atmel.com>
Signed-off-by: Jean-Christophe PLAGNIOL-VILLARD <plagnioj@jcrosoft.com>
On the pcm030 the environment is located in the onboard EEPROM. But we want
to handle flash sector protection in a safe manner. So we must read the
unlock environment variable from EEPROM instead from flash.
This patch is required as long the evironment is saved into the EEPROM.
Stefan: Additional change as suggested by Wolfgang, use bigger char array
(instead of 4).
Signed-off-by: Eric Schumann <E.Schumann@phytec.de>
Signed-off-by: Sascha Hauer <s.hauer@pengutronix.de>
Acked-by: Wolfgang Denk <wd@denx.de>
Signed-off-by: Stefan Roese <sr@denx.de>
Follow up to the flash_fixup_stm to fix geometry reversal
on STMicro M29W320ET flash chip. The M29W320DT has 4 erase region.
Signed-off-by: Richard Retanubun <RichardRetanubun@RuggedCom.com>
Signed-off-by: Stefan Roese <sr@denx.de>
If on your board is more than one flash, you must know
the size of every single flash, for example, for updating
the DTS before booting Linux. So make this function
flash_get_info() extern, and you can have all info
about your flashes.
Signed-off-by: Heiko Schocher <hs@denx.de>
Signed-off-by: Stefan Roese <sr@denx.de>
Patch "flash/cfi_flash: Use virtual sector start address, not phys"
introduced a small typo and compilation warning for systems with CFI
legacy support (e.g. hcu4). This patch fixes it.
Signed-off-by: Stefan Roese <sr@denx.de>
Patch "flash/cfi_flash: Use virtual sector start address, not phys"
introduced a small compilation warning. This patch fixes it.
Signed-off-by: Stefan Roese <sr@denx.de>
include/flash.h was commented to say that the address in
flash_info->start was a physical address. However, from u-boot's
point of view, and looking at most flash code, it makes more
sense for this to be a virtual address. So I corrected the
comment to indicate that this was a virtual address.
The only flash driver that was actually treating the address
as physical was the mtd/cfi_flash driver. However, this code
was using it inconsistently as it actually directly dereferenced
the "start" element, while it used map_physmem to get a
virtual address in other places. I changed this driver so
that the code which initializes the info->start field calls
map_physmem to get a virtual address, eliminating the need for
further map_physmem calls. The code is now consistent.
The *only* place a physical address should be used is when defining the
flash banks list that is used to initialize the flash_info struct,
usually found in the board config file.
Signed-off-by: Becky Bruce <beckyb@kernel.crashing.org>
Signed-off-by: Stefan Roese <sr@denx.de>
Added flash_fixup_stm to fix geometry reversal on STMicro M29W320ET flash chip.
Modeled after flash_fixup_amd, this patch handles the geometry reversal
or erase sectors that exist for ST Micro (now Numonyx) M29W320ET flash.
Since I cannot test all STM's chips, the detection is implemented as
narrow as possible for now.
Signed-off-by: Richard Retanubun <RichardRetanubun@RuggedCom.com>
Signed-off-by: Stefan Roese <sr@denx.de>
The function find_sector() doesn't need to be called twice in
the case of AMD command set.
Tested on TQM5200S-BD with Samsung K8P2815UQB.
Signed-off-by: Jens Gehrlein <sew_s@tqs.de>
Signed-off-by: Stefan Roese <sr@denx.de>
Add cfi-mtd driver, which exports CFI flash to MTD layer.
This allows CFI flash devices to be used from MTD layer.
Building of the new driver is controlled by CONFIG_FLASH_CFI_MTD
option. Initialization is done by calling cfi_mtd_init() from
flash_init().
Signed-off-by: Piotr Ziecik <kosmo@semihalf.com>
Signed-off-by: Stefan Roese <sr@denx.de>
Add interface for flash verbosity control. It allows
to disable output from low-level flash API. It is useful
when calling these low-level functions from context other
than flash commands (for example the MTD/CFI interface
implmentation).
Signed-off-by: Piotr Ziecik <kosmo@semihalf.com>
Signed-off-by: Stefan Roese <sr@denx.de>
Export flash_sector_size() function from drivers/mtd/cfi_flash.c,
so that it can be used in the upcoming cfi-mtd driver.
Signed-off-by: Piotr Ziecik <kosmo@semihalf.com>
Signed-off-by: Stefan Roese <sr@denx.de>
This patch defines all flash access functions as weak so that
they can be overridden by board specific versions.
This will be used by the upcoming VCTH board support where the NOR
FLASH unfortunately can't be accessed memory-mapped. Special
accessor functions are needed here.
To enable this weak functions you need to define
CONFIG_CFI_FLASH_USE_WEAK_ACCESSORS in your board config header.
Otherwise the "old" default functions will be used resulting
in smaller code.
Signed-off-by: Stefan Roese <sr@denx.de>
Acked-by: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
This patch fixes a problem on systems where the NOR flash is attached
to a 64 bit bus. The toggle bit detection in flash_toggle() is based
on the assumption that the same flash address is read twice without
any other interjacent flash accesses. However, on 32 bit systems the
function flash_read64() [as currently implemented] does not perform
an atomic 64 bit read - instead, this is broken down into two 32 bit
read accesses on addresses "addr" and "addr + 4". So instead of
reading a 64 bit value twice from "addr", we see a sequence of 4 32
bit reads from "addr", "addr + 4", "addr", and "addr + 4". The
consequence is that flash_toggle() fails to work.
This patch implements a simple, but somewhat ugly solution, as it
avoids the use of flash_read64() in this critical place (by breaking
it down manually into 32 bit read operations) instead of rewriting
flash_read64() such to perform atomic 64 bit reads as one could
expect. However, such a rewrite would require the use of floating
point load operations, which becomes pretty complex:
save MSR;
set Floating Point Enable bit in MSR;
use "lfd" instruction to perform atomic 64 bit read;
use "stfd" to store value to temporary variable on stack;
load u64 value from temporary variable;
restore saved MSR;
return u64 value;
The benefit-cost ratio of such an implementation was considered too
bad to actually attempt this, especially as we can expect that such
an implementation would not only have a bigger memory footprint but
also cause a performance degradation.
Signed-off-by: Wolfgang Denk <wd@denx.de>
Signed-off-by: Stefan Roese <sr@denx.de>
The flash_unlock_seq requires a sector for AMD_LEGACY.
Fix a retcode check typeo.
Signed-off-by: Ed Swarthout <Ed.Swarthout@freescale.com>
Signed-off-by: Stefan Roese <sr@denx.de>
The CFI flash driver starts at flash_init() which calls down into
flash_get_size(). This starts by calling flash_detect_cfi(). If said
function fails, flash_get_size() finishes by attempting to reset the
flash. Unfortunately, it does this with an info->portwidth set to 0x10
which filters down into flash_make_cmd() and that happily smashes the
stack by sticking info->portwidth bytes into a cfiword_t variable that
lives on the stack. On a 64bit system you probably won't notice, but
killing the last 8 bytes on a 32bit system usually leads to a corrupt
return address. Which is what happens on a Blackfin system.
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Signed-off-by: Stefan Roese <sr@denx.de>
Fix printf format-string/arg mismatches under -DDEBUG.
These warnings occur with DEBUG defined for a platform using
cpu/mpc85xx. Users of other architectures can unearth similar
problems by adding the line "CFLAGS += -DDEBUG=1" in config.mk right
after "CFLAGS += $(call cc-option,-fno-stack-protector)".
Signed-off-by: Andrew Klossner <andrew@cesa.opbu.xerox.com>
Signed-off-by: Andy Fleming <afleming@freescale.com>
This patch fixes a missing vendor code in the flash_real_protect() function.
Signed-off-by: Nick Spence <nick.spence@freescale.com>
Signed-off-by: Stefan Roese <sr@denx.de>
New implement sector lock and unlock or softlock commands
do not exist in AMD legacy flash. Thus, causing issue
when erasing AMD legacy flash (such as lv040)
Signed-off-by: TsiChung Liew <Tsi-Chung.Liew@freescale.com>
Signed-off-by: Stefan Roese <sr@denx.de>
The CFI driver allowed only for static initializers in the
CFG_FLASH_BANKS_LIST definition, i. e. it did not allow to map
several flash banks contiguously if the bank sizes were not known in
advance, which kind of violates U-Boot's design philosophy.
(will be used for example by the TQM8xxL boards)
Signed-off-by: Wolfgang Denk <wd@denx.de>
Some of the flash memories produced by ATMEL start in read-only mode.
We need to unprotect it. This patch allows the AT49BV6416 to work with
cfi_flash memories. Tested in the at91rm9200ek board.
Signed-off-by: Rafael Campos Las Heras <rafael.campos@hanscan.com>
Signed-off-by: Stefan Roese <sr@denx.de>
This got broken by commits 93c56f212c
[cfi_flash: support of long cmd in U-boot.]
That command needs to be in little endian format on BE machines
with CFG_WRITE_SWAPPED_DATA. Without this patch, the command 0xf0
gets saved on stack as 0x00 00 00 f0 and 0x00 gets written into
the cmdbuf in case portwidth = chipwidth = 8bit.
Cc: Alexey Korolev <akorolev@infradead.org>
Cc: Vasiliy Leonenko <vasiliy.leonenko@mail.ru>
Signed-off-by: Sebastian Siewior <bigeasy@linutronix.de>
This got changed by commit 93c56f212c
[cfi_flash: support of long cmd in U-boot.]
Long is the wrong type because it will behave differently on 64bit
machines in a way that is probably not expected. u32 should be
enough.
Cc: Alexey Korolev <akorolev@infradead.org>
Cc: Vasiliy Leonenko <vasiliy.leonenko@mail.ru>
Signed-off-by: Sebastian Siewior <bigeasy@linutronix.de>
This patch simplifies flash_toggle() (AMD commandset), which is used to
detect if a FLASH device is still busy with erase/program operations. On
800MHz Canyonlands/Glacier boards (460EX/GT) the current implementation
did not detect the busy state reliably, resulting in non erased sectors
etc. This patch now simplifies this function by "just" comparing the
complete data-word instead of ANDing it with the command-word (0x40)
before the compatison. It is done the same way in the Linux implementation
chip_ready() in cfi_cmdset_0002.c.
Signed-off-by: Stefan Roese <sr@denx.de>
Added new command set ID. Buffered write command processing is changed
in order to support M18 flash chips family.
Signed-off-by: Alexey Korolev <akorolev@infradead.org>
Signed-off-by: Vasiliy Leonenko <vasiliy.leonenko@mail.ru>
Some NOR flash chips needs support of commands with length grether than max
value size of uchar. For example all M18 family chips use 0x1ff command in
buffered write mode as value of program loops count.
Signed-off-by: Alexey Korolev <akorolev@infradead.org>
Signed-off-by: Vasiliy Leonenko <vasiliy.leonenko@mail.ru>
* The cfi_flash.c memset fix actual allows the board to boot so there is
a bit more going on here than just resolving warnings associated with
uninitialized variables.
* include/asm/bitops.h:302: warning: '__swab32p' is static but used in
inline function 'ext2_find_next_zero_bit' which is not static
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
This patch adds a configurable flash auto protection list that can be used
to make U-Boot protect flash regions in flash_init().
The idea has been discussed on the u-boot mailing list starting
on Nov 18th, 2007.
Even this patch brings a new feature it is used as a bugfix for 4xx
platforms where flash_init() does not completely protect the
monitor's flash range in all situations.
U-Boot protects the flash range from CFG_MONITOR_BASE to
(CFG_MONITOR_BASE + monitor_flash_len - 1) by default. This does not
include the reset vector at 0xfffffffc.
Example:
#define CFG_FLASH_AUTOPROTECT_LIST {{0xfff80000, 0x80000}}
This config option will auto protect the last 512k of flash that
contains the bootloader on board like APC405 and PMC405.
Signed-off-by: Matthias Fuchs <matthias.fuchs@esd-electronics.com>
Some NOR flash chip from Spansion, for example, the s29ws-n MirrorBit
series require different addresses for buffered write commands. Define a
configuration option to support buffered writes on those chips. A more
elegant solution would be to automatically detect those chips by parsing
their CFI records, but that would require introduction of a fixup table
into the cfi_flash driver.
Signed-off-by: Guennadi Liakhovetski <lg@denx.de>
SPARC has implemented __raw_readq, it reads 64-bit from any 32-bit address.
SPARC CPUs implement flash_read64 which calls __raw_readq.
For current SPARC architectures (LEON2 and LEON3) each read from the
FLASH must lead to a cache miss. This is because FLASH can not be set
non-cacheable since program code resides there, and alternatively disabling
cache is poor from performance view, or doing a cache flush between each
read is even poorer.
Forcing a cache miss on a SPARC is done by a special instruction "lda" -
load alternative space, the alternative space number (ASI) is processor
implementation spcific and can be found by including <asm/processor.h>.
Signed-off-by: Daniel Hellstrom <daniel@gaisler.com>
Add entry for 512Kx16 AMD flash to jedec_table.
Read out 16bit device id if chipwidth is 16bit.
Fixed coding style after Stefans feedback
Signed-off-by: Tor Krill <tor@excito.com>
With this patch we don't need that many #ifdef's in the code. It moves
the subtraction into the macro and defines a NOP-macro when
CONFIG_FLASH_SHOW_PROGRESS is not defined.
Signed-off-by: Stefan Roese <sr@denx.de>
Acked-by: Gerald Van Baren <vanbaren@cideas.com>
Do not use uninitialized cmd_reset; issue both AMD and Intel reset
commands instead
From a short test, it looks like AMD-style flash roms treat *any* unknown
command write as a reset, at least when in CFI Query mode, so issuing the
Intel reset command to AMD-style flashs seems safe (from the small sample I
have), plus the 3-cycle magic sequence should kick the state machine into
the right state even without a reset command. Since the AMD-style flashs
require the unlock sequence for real operation, I chose to try the AMD reset
command first, so that Intel flashs do no see an invalid command prior to
the CFI query.
I have tested the patch on AM29LV320-style flashs from Fujitsu and Macronix,
plus Intel StrataFlash.
Signed-off-by: Michael Schwingen <michael@schwingen.org>
Signed-off-by: Stefan Roese <sr@denx.de>
This bug was detected on the LWMON5 target which has 2 Intel 16bit wide
flash chips connected to a 32bit wide port.
Signed-off-by: Stefan Roese <sr@denx.de>
The check for an sufficiently erased destination was missing in the
buffered write function of the cfi flash driver (when
CFG_FLASH_USE_BUFFER_WRITE is defined). This patch adds this check to that
writing to such a region will fail with the currect error message.
Signed-off-by: Stefan Roese <sr@denx.de>
Run fixups based on the JEDEC manufacturer ID independent of the
command set ID.
This changes current behaviour: Previously, geometry reversal for AMD
chips were done based on the command set ID, while they are now done
based on the JEDEC manufacturer and device ID.
Also add fixup for top-boot Atmel chips. A fixup is needed for
AT49BV6416(T) too, but since u-boot currently only reads the low byte
of the device ID, there's no way to tell it apart from AT49BV642D,
which should not have this fixup. Since AT49BV642D support is
necessary to get ATNGW100 board support into mainline, I've commented
out the fixup for now.
Signed-off-by: Haavard Skinnemoen <hskinnemoen@atmel.com>
