summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2025-05-25dev_table: Add placeholders for all devices in AN2606 rev 66HEADorigin/masterorigin/HEADmasterTormod Volden
Values taken from Table 209 (in Section 92 "Device-dependent bootloader parameters"). Reorder two table entries to maintain same order as in the document. Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2025-05-25stm32: Remove duplicate version fieldTormod Volden
It was only used locally so make it a local. In any case, this version, reported by the GVR command, should be the same as the bootloader version reported by the GET command. Print a warning if they ever happen to not be the same. Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2024-05-17dev_table: Add STM32C011xx (Device ID 0x443)Jasper de Zoete
Based on RM0490 rev 3. Most of the information was found in Section 2 and 3. https://sourceforge.net/p/stm32flash/code/merge-requests/23/
2024-05-11dev_table: Fix up option byte range for device 0x497Tormod Volden
Unlike all other ranges in the table, the option byte range is inclusive. Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2023-01-30dev_table: Fix typo in flash start for device 0x483Tormod Volden
Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2022-12-08dev_table: Tighten option bytes range for device ID 0x436 / 0x437Tormod Volden
These dual-bank devices have two ranges of option bytes, and we previously listed the start of the first and the end of the second. Since we use the range (only) to display the number of option bytes, this can get very wrong for some dual-bank devices where the two ranges are far apart. So for consistency ignore the second range, as long as our table only support one range. https://sourceforge.net/p/stm32flash/tickets/92/ Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2022-12-08main: Report as "Option bytes" and "System memory"Tormod Volden
These locations are not RAM. https://sourceforge.net/p/stm32flash/tickets/92/ Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2022-12-08dev_table.c: Add STM32WB10xx/15xx (Device ID 0x494)Tormod Volden
RM0478 rev 5: Table 3. Flash memory - Single bank organization (untested) Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2022-12-08dev_table.c: Add STM32L552xx/562xx (Device ID 0x472) dual-bankTormod Volden
RM0438 rev 7: Table 30. Flash module - 512 KB dual bank organization (64 bits read width) Section 6.3.1: "option bytes [...] not mapped to any memory address" (untested) Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2022-12-08dev_table.c: Add STM32L4P5xx/Q5xx (Device ID 0x471) dual-bankTormod Volden
RM0432 rev 9: Table 5. Memory mapping versus boot mode/physical remap footnote: "1 Mbyte for STM32L4P5xx and STM32L4Q5xx devices" Table 8. Flash module - 1 Mbyte dual-bank organization, DB1M = 1 (64 bits read width) Note the options bytes are split between 0x1FF00000-0x1FF0000F for bank 1 and in 0x1FF01000-0x1FF0100F for bank 2, and we only list the first in our table. (untested) Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2022-12-08dev_table.c: Add STM32WB30xx/50xx (Device ID 0x495)Tormod Volden
RM0471 rev 7: Table 3. Flash memory - Single bank organization (untested) Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2022-12-08dev_table.c: Add STM32U575xx/585xx (Device ID 0x482)Tormod Volden
RM0456 rev 3: Table 44. Flash module 2-Mbyte dual bank organization Section 7.3.1: "option bytes [...] not mapped to any memory address" (untested) Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2022-12-08dev_table.c: Add STM32H7A3xx/B3xx (Device ID 0x480)Tormod Volden
RM0455 rev 6: (STM32H7A3/7B3 and STM32H7B0 Value line) Table 12. Flash memory organization (STM32H7A3xI/7B3xI devices) Section 4.3.4: "The embedded Flash non-volatile memory is composed of: - For STM32H7B3xx and STM32H7A3xI devices: a 2-Mbyte main memory block, organized as two banks of 1 Mbyte each. - For STM32H7A3xG devices: a 1-Mbyte main memory block, organized as two banks of 512 Kbyte each." "Each bank is in turn divided in up to 128 sectors of 8 Kbytes each" (untested) Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2022-12-08dev_table.c: Add STM32H72xxx/73xxx (Device ID 0x483)Tormod Volden
RM0468 rev 3: (STM32H723/733, STM32H725/735 and STM32H730 Value line) Table 15. Flash memory organization (STM32H723/733 and STM32H725/735 devices) Section 4.5.2: "Any 128-Kbyte Flash sector can be independently write-protected or unprotected" (untested) Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2022-12-07dev_table: Add support for device 0x497 (STM32WLE5xx)Albertas Mickenas
https://sourceforge.net/p/stm32flash/tickets/151/ Data sources: * RM0453 Rev 4 Table 11 * AN2606 Rev 55 Table 159
2022-11-28dev_table: Add placeholders for all devices in AN2606 rev 55Tormod Volden
Somebody needs to look up the datasheets and fill in the missing numbers. Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2022-11-28serial_posix: Fix setup from odd parity or two stop bitsFrancesco Cusolito
https://sourceforge.net/p/stm32flash/tickets/146/
2022-11-28hex parser: Warn about large gaps and detect allocation failureTormod Volden
https://sourceforge.net/p/stm32flash/tickets/150/ Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2022-03-27Release 0.7v0.7Tormod Volden
Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2022-03-27serial_posix: Warn about failed settingsTormod Volden
At least on macOS this can happen if the baud rate is not supported on the serial port. Unfortunately we cannot tell more specifically what went wrong in this case. Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2022-03-26I2C.txt: Mention some I2C-USB bridgesTormod Volden
https://sourceforge.net/p/stm32flash/tickets/141/ Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2022-03-26Allow skipping serial port configuration with -b 0Tormod Volden
If the baud rate is specified as zero with -b 0, stm32flash will not do any configuration at all of the port. This can be useful for tunnelling over various transports, or for e.g. separately setting up baud rates in platform-specific ways that stm32flash doesn't handle. Example: stty -F /dev/ttyUSB0 raw parenb -parodd -cstopb cs8 115200 time 5 min 0 \ line 0 -brkint -icrnl -imaxbel -opost -onlcr -isig -icanon -iexten \ -echo -echok -echoctl -echoke stm32flash -b 0 /dev/ttyUSB0 Example UDP transport with socat: hostA$ socat /dev/ttyUSB0,nonblock,rawer UDP-LISTEN:3334 hostB$ socat UDP:hostA:3334 PTY,link=/tmp/ttyremote,rawer hostB$ stm32flash -b 0 /tmp/ttyremote This has only been tested on Linux. https://sourceforge.net/p/stm32flash/code/merge-requests/15/ https://sourceforge.net/p/stm32flash/tickets/94/ https://sourceforge.net/p/stm32flash/tickets/105/ Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2022-01-13Fix error message for page erase initializationTormod Volden
Fixup of commit 666dde2e Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2021-09-06Fix binary file read from stdinRenaud Fivet
Test case: echo "Short data chunk" | stm32flash /dev/ttyUSB0 -S 0x20008000 -w - stm32flash -c /dev/ttyUSB0 -S 0x20008000:0x20 -r - | hd https://sourceforge.net/p/stm32flash/tickets/129/
2021-09-05hex parser: Get base location from fileRenaud Fivet
Also fixes unnecessary padding when start at offset != 0 https://sourceforge.net/p/stm32flash/tickets/121/
2021-09-05dev_table: add STM32L412xx/422xxAndy Little
AN2606 rev 49 table 152 says 0x20002100-0x20008000, but testing shows it only should be from 0x20003100. This is the 32KB in SRAM1, without the 8KB in SRAM2. The STM32L412xx and STM32L422xx datasheets confirm 128KB flash. https://sourceforge.net/p/stm32flash/tickets/114/ Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2021-09-02Handle STM32L412/L422 replying with double ACK on initTormod Volden
As per AN2606 the USART bootloader (ID V13.1) on STM32L412xx/422xx devices "replies with two ACK bytes (0x79) instead of only one". Thanks to Andy Little for reporting, initial patch, and testing. https://sourceforge.net/p/stm32flash/tickets/114/ Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2021-09-02serial_posix: More details on baud setting errorsTormod Volden
Also error out earlier in this case. This won't change anything in practice according to my tests, because the flag setting would also fail if the speed setting failed. However, the error messages can now help distinguish different failure modes. Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2021-09-02serial: Warn explicitly if baud rate is not supportedTormod Volden
We were just returning an unknown error which is not very helpful to the user. This will help for some cases. Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2021-09-02Allow higher baud rates on BSD flavours (including macOS)Tormod Volden
It could be a lot more straightforward in the BSD case but with our current baud handling code this is what we get for cheap. https://sourceforge.net/p/stm32flash/tickets/127/ Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2021-09-02serial: Add baud rates up to 4000000Tormod Volden
For Windows also fill in all baud rates defined in WinBase.h. For Posix platforms the higher rates will available if defined in the system headers, which is the case for Linux. Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2021-09-02serial_w32: Flush input buffer on Windows tooTormod Volden
From the beginning the actual flush was commented out and not adapted to the Windows API since at the time there was no apparent need to flush. Especially since the GPIO sequences were added flushing can be needed, for instance after resetting into the bootloader from a talkative application. Thanks to Jan Belohoubek for reporting, fix, and testing. https://sourceforge.net/p/stm32flash/tickets/132/ Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2021-09-02stm32: Fix typo from recent commitTormod Volden
Oops, I got this wrong while amending commit b079cd09. Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2021-08-28I2C.txt: Use waybackmachine for lost web pageTormod Volden
Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2021-08-28stm32: Improve clock stretching warning messageTormod Volden
Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2021-08-28stm32: More detailed init error messagesTormod Volden
Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2021-08-28stm32: Consistent and unique erase error messagesTormod Volden
Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2021-08-28Follow bootloader protocol for page-by-page erase over I2CYann Sionneau
For I2C the protocol is slighly different than for USART, requiring a checksum after the number of pages. https://sourceforge.net/p/stm32flash/tickets/98/ Signed-off-by: Yann Sionneau <ysionneau@kalray.eu> [Tormod: Add port flag, no wait, amend messages] Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2021-08-16dev_table: add STM32G491xx/A1xxTormod Volden
Add values from RM0440 rev 6 sections 2.4 and 4.3.1 and 5.4.1 (untested) Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2021-08-16dev_table: really add STM32G05xxx/061xxTormod Volden
(The previous commit actually added STM32G0B0/B1/C1xx) Filled in values from RM0444 Rev 5 table 4 (untested) Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2021-08-08dev_table: add STM32G05xxx/061xxLuca Amati
Based on RM0444 Rev5 (chapter 3) and AN2606 Rev 49 (chapter 70, table 152).
2021-07-20dev_table: Add placeholders for all devices in AN2606 rev 49Tormod Volden
We only miss 11 devices from AN2606 rev 49 (July 2021). Fill in the list with the information from Table 152 in the hope that missing device parameters and testing will be contributed. AN2606 has conflicting information for instance for STM32WB30xx in Table 148 (20 KB for bootloader) and Table 152 (16 KB), and for STM32WB10xx Table 146 (16 KB) vs Table 152 (20 KB). Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2021-07-20dev_table: Reorder F7 section according to AN2606 rev 49Tormod Volden
Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2021-02-12Release 0.6v0.6Tormod Volden
Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2021-02-12man page: Minor adjustments in GPIO examplesTormod Volden
Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2021-02-12man page: Explain what gets erased if -e is not specifiedTormod Volden
The effect of the -S parameters is old, but that the file content size (for binary files) is used was only recently introduced. Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2021-02-01Minor documentation updatesTormod Volden
Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2021-02-01Check if the user provided a start addressTormod Volden
This allows setting a length but an empty address. Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2021-02-01Always use flash start as default start addressTormod Volden
Also if a write length is known from the file size. Commit 5dbb767 "Do not erase more than file contents indicate" caused a regression if no start address was given, in which case the start address was set to be 0. Also improve handling of the case where the user specifies start address 0. Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
2021-01-31automake: Add header files to source listsTormod Volden
The posix or w32 files should ideally be picked by a conditional but this has to do for now. Also ship all documentation files and Android build files in the generated dist tarball. Also allow to ship our handmade makefiles in the tarball made by the autogenerated makefiles. Signed-off-by: Tormod Volden <debian.tormod@gmail.com>
generated by cgit v1.2.3 (git 2.46.0) at 2025-08-06 21:07:25 +0000