- Simplify PHY access API
- Move coexist initializing and deinitializing out from PHY API
to Wi-Fi and Bluetooth
- Remove coexist pause and resume for they are no longer needed.
Add following bugfixes
1. Station not able to connect when WPS pin is pressed first on AP.
2. PBC overlap getting detected for selected registrar PIN APs.
3. Station not considering authorised MACs for PIN method.
4. For PIN methodm If no AP is found, station will loop through
APs in its vicinity and try to do WPS with them one by one till
WPS timeout occurs. This is for some APs which do not set
selected registrar correctly.
1. fix the issue that no tbtt is generated in station+softap mode when maximum modem sleep is enabled
2. store PHY digital registers before disabling PHY and load them after enabling PHY
3. update libphy.a to V4670 719f9f6_20210218 for esp32
4. update libphy.a to V1500 20210225_0a54a2f for esp32s2
esp_wifi_mesh: fix the issue that layer2 node connect to lower-layer node when FIXED-ROOT root disappeared(backport_v4.2)
See merge request espressif/esp-idf!15730
gpio: Bugfix - Move esp_intr_free() out of the critical section in gpio_uninstall_isr_service() (backport v4.2)
See merge request espressif/esp-idf!16597
Closes https://github.com/espressif/esp-idf/issues/5571
Fix the bug that if the API was called from one core to free the interrupt source on the other core, it would trigger interrupt watchdog.
(cherry picked from commit 0e8286c57b)
Current issue: output reports sent by Windows not received.
The report characteristic declaration should also support write without
response as specified by HIDS profile:
See https://www.bluetooth.com/specifications/GATT/ (page 14)
ringbuf: Fix bug where comparision between a signed and unsigned operand resulted in incorrect free size for no-split/allow-split buffers (v4.2)
See merge request espressif/esp-idf!15883
Initial version of freemodbus master port files have been added to ESP-IDF based on https://github.com/armink/FreeModbus_Slave-Master-RTT-STM32.
The overall repository license, at the time of adding these files, has been BSD 3-clause. However at that time, several port files carried LGPL license headers. As the author of these files confirmed in https://github.com/armink/FreeModbus_Slave-Master-RTT-STM32/issues/61#issuecomment-977828450, this wasn't intentional. ESP-IDF version of modbus master port has been rewritten to target FreeRTOS instead of RT-Thread, but the license headers remained from the original version. This commit corrects this, replacing the license of these files with BSD 3-clause.
This commit fixes a bug in no-split and allow-split ring buffers free buffer size calculation.
When the free size available in the buffers less than the size of one item header,
the function prvGetCurMaxSizeNoSplit/AllowSplit() incorrectly returned the maxItemSize instead of 0.
This is due to the comparision between a negative and a positive value
where both operands are treated as unsigned during the comparision operation,
thereby treating the negative operand as a large integer.
Also added new unit tests to test buffer-full and almost-full conditions
where this scenario is likely to be hit.
Closes https://github.com/espressif/esp-idf/issues/7344
Closes https://github.com/espressif/esp-idf/pull/7371
The release of the semaphore indicating the item was successfully sent must be the last semaphore released. The receiver may be in another task and may delete the Ringbuffer (such as with a return code across tasks design pattern) if they are through with the Ringbuffer.
The function xRingbufferSendAcquire followed by xRingbufferSendComplete had the semaphores released in the proper order and that same pattern should have been used in xRingbufferSend and xRingbufferSendFromISR. This commit fixes this order.
Issue (IDFGH-6030) #7716 describes the problem in more detail.
Closes IDFGH-6030, https://github.com/espressif/esp-idf/issues/7716
Closes IDFGH-6036, https://github.com/espressif/esp-idf/pull/7721
1. fix the device can not receive IP_EVENT_STA_GOT_IP when the router restarted
2. fix the issue that layer2 node connect to lower-layer node when FIXED-ROOT root disappeared
This commit brings two esp-lwip fixes to IDF:
1) Add configuration to disable DHCP client identification
2195f7416f
This config could be used to disable option 61 in DHCP packets, so that
clients will be identified by their chaddr only.
(This is the lwip upstream original behaviour)
2) Fix server_idx increasing to DNS_MAX_SERVERS and trigger the LWIP_ASSERT
5a567d52f7
When lwip doesn't have DNS server and resolve a domain address, the server_idx
will increase to DNS_MAX_SERVERS, which will trigger the LWIP_ASSERT and make device crash.
Closes https://github.com/espressif/esp-idf/issues/7912
1. fix(pp): fix fragment plt loss when 2td pn compare with 1th pn
2. pmf allow keyindex in big endian format to workaround faulty APs
3. docs: update miswritten and abbreviated words and syntax errors for initialize WiFi section in esp_wifi.h header file
4. allow for minimal scope of wifi_init_config_t
1. Added config options to chose from protocom security.
It can be chosen 0/1 or custom.
Possible to set POP as well
2. Added support in `esp_local_ctrl.py` test script for sec_ver selection
Signed-off-by: Vikram Dattu <vikram.dattu@espressif.com>
1. Fix airkiss and esptouch find channel crash issue
2. Fix RF singal test without 9M rate issue
3. Fix issue of setting invalid size to nvs misc log even after erasing it
4. Fix WEP connection and crash issue
Current code does not correctly free all pins in wps_free_pins due to the
semicolon at the end of dl_list_for_each_safe(). Fix it.
Signed-off-by: Axel Lin <axel.lin@gmail.com>
The body of the bootloader_debug_buffer function was conditioned to
macros that were never defined, resulting in deactivated code.
Signed-off-by: Gustavo Henrique Nihei <gustavo.nihei@espressif.com>
When the build runner has a local git mirror configured via
LOCAL_GIT_MIRROR variable, use that mirror when cloning IDF insider
the docker build job. Follows similar logic for
LOCAL_GITLAB_HTTPS_HOST, which is used for geo nodes.
For the CID10564,10384,10280,10098,10038,The memory was released in other place.
For the CID10365,it release the memory in the function when sent successfully.
For the CID10268,10011, we need not change the code.
Increase reset assertion time from 100µs (as specified minimum in the datasheet) to 150µs.
Some specimen of the LAN8720 need the reset signal asserted longer than 100µs to initialise properly. Otherwise they are in a zombie state where they are establishing and loosing an Ethernet link once in a seconds interval.
The psram cache bug fix was also being applied to the bootloader binary (for cmake),
which doesnt do any psram access.
Applying this fix would increase the binary size, as much as 300 bytes in worst case scenarios
The issue is related to the non-sequential way of description when
such fields going together sequential.
Related to esp32h2 chip for eFuses: MAC_FACTORY and MAC_EXT.
The issue is in wrong indexes of MAC_EXT.
MAC_EXT got indexes like it is joined to MAC_FACTORY.
const esp_efuse_desc_t* ESP_EFUSE_MAC_FACTORY[] = {
&MAC_FACTORY[0],
&MAC_FACTORY[1],
&MAC_FACTORY[2],
&MAC_FACTORY[3],
&MAC_FACTORY[4],
&MAC_FACTORY[5],
NULL
};
const esp_efuse_desc_t* ESP_EFUSE_MAC_EXT[] = {
&MAC_EXT[6],
&MAC_EXT[7],
NULL
};
This commit fixed it to:
const esp_efuse_desc_t* ESP_EFUSE_MAC_EXT[] = {
&MAC_EXT[0],
&MAC_EXT[1],
NULL
};
esp_http_client: Fix HEAD request will affect the all next HTTP requests unless we close the HTTP request(backport v4.2)
See merge request espressif/esp-idf!15713
Large SD cards (16GB+) require significant amount of time for FS formatting.
Added FS mount checkpoint in example test python, timeout set to 60 sec
Closes IDFCI-706
It was when in the partition table there is a partition with type="data" and suptype=""(empty),
in this case type=1, suptype=0. It is similar to otadata partition.
This commit fixes it, now it will handle it as type=1, suptype=6 (ESP_PARTITION_SUBTYPE_DATA_UNDEFINED).
For https connection `ESP_TLS_ERR_SSL_WANT_READ` of esp_transport_read was getting treated as error.
Treated this as a timeout to fix connection abort issue!
Also handled http connection EAGAIN with `errno == EAGAIN` check.
Signed-off-by: Vikram Dattu <vikram.dattu@espressif.com>
Allow user to select specific ESP_TARGET while setting up ESD_IDF.
Only necessary tools for given target will be downloaded and installed.
Closes https://github.com/espressif/esp-idf/issues/5113
data cache is unicode. while we use bytes in RegEx expect. The index of
matched pattern is calculated with bytes, could be different from
unicode. Now we fix this issue by using unicode in expect.
Issue:
Missing newline between header and attachment
Issue seen with email clients viz., AppleMail and YahooMail. Attachment is not renderable.
Some clients like Google mail, Samsung mail tolerate the issue.
Fix:
Added newline between header and attachment.
Closes https://github.com/espressif/esp-idf/issues/6098
Signed-off-by: Vikram Dattu <vikram.dattu@espressif.com>
WPS strict disables workarounds with different APs and may cause
IOT issues. Remove this as default and introduce as a config option.
Also declare esp device as single band mobile device otherwise
WFA sniffer was not able to identify it in the certification setup.
When not compiling bootloader, a spinlock will be used for reading or writing
I2C internal devices/registers.
When compiling for bootloader, no need to use any lock.
* Patched longjmp to be context-switch safe
longjmp modifies the windowbase and windowstart
registers, which isn't safe if a context switch
occurs during the modification. After a context
switch, windowstart and windowbase will be
different, leading to a wrongly set windowstart
bit due to longjmp writing it based on the
windowbase before the context switch. This
corrupts the registers at the next window
overflow reaching that wrongly set bit.
The solution is to disable interrupts during
this code. It is only 6 instructions long,
the impact shouldn't be significant.
The fix is implemented as a wrapper which
replaces the original first instructions of
longjmp which are buggy. Then, it jumps back
to execute the rest of the original longjmp
function.
Added a comparably reliable test to the
test apps.
* Description of unregistering was incorrect
* Made clear that event loop arg mustn't be NULL
* Added parameter check in create function
Closes https://github.com/espressif/esp-idf/issues/6761
Closes IDFGH-4969
1. Fixed the crash when using scan parameter: window = 4 slots, interval = 5 slots
2. Fixed BLE ACL tx flush issue during reset
3. Added remove sniff when peer not accepted sniff mode
- Add `wifi_prov_scheme_ble_set_mfg_data` API to set custom manufacturer data
in BLE advertisements.
- Run format.sh script on modified files.
- Fix few typos in `protocomm_nimble.c`.
- Incorporate suggestion to remove extra check on protocomm_ble_mfg_data_len
- Remove few unnecessary comments.
This is mostly important on ESP32 ECO3 with the
ESP32_ECO3_CACHE_LOCK_FIX, because when we stall the other CPU core
before we disable the TG1 WDT then the first CPU can get stuck
in WDT ISR handle_livelock_int routine waiting for the other CPU.
After we have the performance dashboard, we have more data and no longer depend on the threshold to ensure performance.
Set looser performance thresholds to avoid CI failure.
The CPU might prefetch instructions, which means it in some cases
will try to fetch instruction located after the last instruction in
flash.text.
Add dummy bytes to ensure fetching these wont result in an error,
e.g. MMU exceptions
This makes the test faster and more robust in very a busy WiFi environment
(our CI) and still exercises the scenario of fragmenting the message on
both mqtt and ssl levels (binary size to send > mqtt buffer size)
This commit fixes an issue with gdbstub, where it would list threads
with TIDs 1 to N in qfThreadInfo/qsThreadInfo responses, and then
would tell GDB that the current TID is 0 in the qC response. This
caused an assertion failure in GDB, because it couldn't find the
thread structure corresponding to TID 0:
src/gdb/gdb/thread.c:93: internal-error: thread_info* inferior_thread(): Assertion `tp' failed.
The issue was caused by the logic of qfThreadInfo/qsThreadInfo.
If the "paniced" task index was 1, the code would report it in the
response to qfThreadInfo, and then mistakenly skip task with index 0
in qsThreadInfo, due to the use of pre-increment instead of a
post-increment.
With that issue fixed, GDB assertion doesn't happen anymore. However
the code contained a deeper problem, which manifested itself in the
fact that GDB would incorrectly show task index 0 as the current task,
after the above fix.
Previous version of the code assumed that when GDB requests the thread
list, it uses the first thread returned by the target as the "default"
thread, and subsequently shows the user that the program is stopped
in that thread. This assumption was incorrect. In fact, after
connecting to a remote target, GDB obtains information about the
"default" or "current" thread from two sources:
1. the 'thread' special register indicated in the status response
($T00thread;00000001#ee)
2. if the target has only sent the plain stop response ($T00#ee), GDB
would ask for the current thread using a qC packet.
With that in mind, it is not necessary to report the paniced task as
the first task in qfThreadInfo response. We can simply returns the
tasks in their natural order, and then indicate the current task in
the qS packet response.
However even that change does not fully resolve the issues with task
list. The previous version of this code also incorrectly interpreted
the meaning of GDB TIDs -1 and 0. When GDB sends an "Hg0" command
early in the connection process, it doesn't expect the server to set
task 0 as the current task, as the code assumed. Rather, it tells the
server to "set any (arbitrary) task as the current one", and the most
logical thing to do for the server that is already in "stopped" state
is to keep the current task selection.
Since TID 0 has a special meaning in GDB remote protocol, gdbstub code
is now modified to map task indices (which start from 0) to GDB TIDs.
GDB TIDs are arbitrary, and for simplicity we keep the same order and
start counting them from 1.
The summary of all the above changes is:
1. Use "task index + 1" as the TID reported to GDB
2. Report the tasks in natural order; don't complicate the code to
make the paniced task first in the list.
3. Centralize modification of 'current_task_index' and 'regfile'
in the new 'set_active_task' function, to improve encapsulation.
Currently IDF_VERSION_* variables are not available to the component
CMakeLists.txt files at the requirements expansion stage. This makes
it harder to write component CMakeLists files compatible with
different IDF versions.
Include version.cmake from the requirements expansion script, add a
build system test.
The previous location was the return from the first ets_printf call
that prints ROM sign-on message. Since the main function was patched
in ECO3, the new address no longer works — there is no instruction at
0x40007901 in ECO3 ROM. This could be solved by setting two
breakpoints (one would work for ECO <=2, the other for ECO3), but we
would need to remove the unused breakpoint later.
Fix this by setting the breakpoint at ets_printf. This means that when
debugging a loadable ELF the ROM sign-on message will no longer be
shown, but this doesn't seem to be an issue.
Possible for a joined task to be deleted at the moment it is logging,
meaning it might hold the stdout lock. In that case the lock isn't
released and the next task to try and take it (i.e. call printf)
will block indefinitely.
At -O2 optimization level, GCC seems to optimize out the copying of the
first byte of the checksum, assuming it is zero. This "miscompilation"
happens because the esp_app_desc struct is declared const, but then modified
post-compilation. Casting to volatile disables the optimization.
Closes: https://github.com/espressif/esp-idf/pull/6389
Fixes an issure where the first part of an object file name is not
included, due to matching the rule for a section entry previously.
Reduce depedency on matching literal strings in sections which might
change depending on toolchain (ex. matching 'elf32-xtensa-le')
Make sure parsing rule succeeds for the entirety of the sections info
string by adding 'parseAll=True'.
Add test for sections info parsing.
1. fix TypeError when running with python3
2. fix throughput chart x/y axis label error
3. make test case compatibile with iperf bin on earlier release branches
* Better comment of nvs_get_used_entry_count()
* Mention C++ example in API reference
* WIP: Used target instead of hard code ESP32
* Note that strings can only span one page
* Reflect that item types have been moved
* Some clarification about nvs_commit()
* Improved reference to the ESP Partition API
* fixed little mistake in documenting-code.rst
* Change of nvs_open_from_part() to
nvs_open_from_partition() reflected in docs
* Corrected documentation of
NVSHandle::get_string(), NVSHandle::get_blob()
and NVSHandle::get_item_size().
* Closes IDF-1563
* Closes IDF-859
* Closes https://github.com/espressif/esp-idf/issues/6123
Additionally, always enable the partition MD5 check if flash encryption is on in
Release mode. This ensures the partition table ciphertext has not been modified
(CVE-2021-27926).
The exception is pre-V3.1 ESP-IDF bootloaders and partition tables, which
don't have support for the MD5 entry.
ROM will erase the region a partition is in as soon as it receives the
first bit of the data that is in the partition. For large partitions it
takes more than 5 seconds to erase which is a hard-coded limit in
dfu-utils.
This splits large binaries and adds them by chunks which should avoid
timing-out during flashing.
Closes https://github.com/espressif/esp-idf/issues/6999
Esptool submodule got updated which changed the function signature for EncryptFlashDataArgs.
Update example test script to reflect the new function signature
Since dd849ffc, _rodata_start label has been moved to a different
linker output section from where the TLS templates (.tdata, .tbss)
are located. Since link-time addresses of thread-local variables are
calculated relative to the section start address, this resulted in
incorrect calculation of THREADPTR/$tp registers.
Fix by introducing new linker label, _flash_rodata_start, which points
to the .flash.rodata output section where TLS variables are located,
and use it when calculating THREADPTR/$tp.
Also remove the hardcoded rodata section alignment for Xtensa targets.
Alignment of rodata can be affected by the user application, which is
the issue dd849ffc was fixing. To accommodate any possible alignment,
save it in a linker label (_flash_rodata_align) and then use when
calculating THREADPTR. Note that this is not required on RISC-V, since
this target doesn't use TPOFF.
Newer versions of doxygen will give warnings for comments in
INPUT lists
Delete all comment lines to fix these warnings, our folder structure
stil gives an OK overview of what we are including
mbedtls: fix mbedtls dynamic resource memory leaks and mbedtls_ssl_send_alert_message crash due to ssl->out_iv is NULL[backport v4.2]
See merge request espressif/esp-idf!13301
RX process caches the session information in "ssl->in_ctr" not in "ssl->in_buf".
So when freeing the SSL, can't free the "ssl->in_ctr", because the "ssl->in_buf"
is empty.
Make the RX process like TX process, and cache the session information in
"ssl->in_buf", so that the cache buffer can be freed when freeing the SSL.
Closes https://github.com/espressif/esp-idf/issues/6104
When user forgot to add git.exe or python to the PATH, there was a not
very helpful error message.
This improves the error with which tool is misssing and shows a link to
the espressif installer tool.
mdns resolver didn't correctly resolved queries when host name wasn't
assigned. Fixed by allowing processing also if some answer present
(non-strict mode)
Closes https://github.com/espressif/esp-idf/issues/6598
The resolver was able to respond correctly, but would also resolve its
own queries and cause issues with BCT 1.5.2, specifically
* MULTIPLE QUESTIONS - DUPLICATE SUPPRESSION
* MULTIPLE QUESTIONS - DISTRIBUTED DUPLICATE SUPPRESSION
tests failed.
Strict mode was hardcoded in private header file, but it's useful for
users to enable/disable it depending on the mdns library they are using.
e.g. Avahi might not resolve the non-strict answers.
Current code stopped inquiry if major class is not Phone. Modified the
condition to consider for both Phone and Audio to cover majority of real
world devices.
Closes https://github.com/espressif/esp-idf/issues/6548
1. Unlock mutex for esp_wifi_80211_tx when out of memory.
2. Fix issue that esp_wifi_80211_tx can't be used in ap mode.
3. Clear hotspot bss when auth timeout happens.
4. Fix PMK not saved in NVS for SoftAP for PSK mode.
5. Fix the bug for esptouch find hidden AP fail.
6. Fix not connect to open AP when WPA3 auth fail.
7. Fix issue with hidden AP scans after connecting AP.
8. Fix the watchdog issue when receiving action frame.
9. Fix the issue that the parameters obtained form RAM cannot be saved to NVS.
10.Modify not to store the default value in NVS.
11.Delete interface STATIONAP_IF.
12.Modify the _do_wifi_start and _do_wifi_stop parameters to mode.
13.Fixed issue of reason code change from 15 to 204 when provide wrong password.
14.Fix set config return value error.
15.Fix ampdu age timer memory leak.
Fixes issue with DPORT init task, this task uses minimum stack size and may not be
enough if stack smashing detection is set to Overall mode.
Also reworks the way we calculate minimum stack to allow for adding multiple
contributing factors.
Closes https://github.com/espressif/esp-idf/issues/6403
1. Change wifi scan duration from 120ms to 100ms
2. Using deport reg instead of ahb
3. Revert to report specific reason code when receiving deauth during 4-way-handshark
4. Fix the bug that tx ampdu parameter is not from peer device
5. Faster WiFi station connect improvement, avoid 100ms passive scan
6. Add FCS failed packets filter
7. Fix country code last byte to space instead of NULL
8. Fix softap cannot forward A-MSDU
9. Fix max tx power to 20dBm
10. Fix the issue that the esp_wifi_sta_get_ap_info can't get country
11. Update esp32 phy lib to v4660
12. Update esp32s2 phy lib to v1300
13. Fix the bug that esp32s2 can't get CSI info
On S2 the brown out detector would occasionally trigger erroneously during deep sleep.
Disable it before sleeping to circumvent this issue.
Closes https://github.com/espressif/esp-idf/issues/6179
Sometimes the flash size read from bootloader is not correct. This may
forbid SPI Flash driver from reading the the area larger than the size
in bootloader header.
When the new config option is enabled, the latest configured
ESPTOOLPY_FLAHSIZE in the app header will be used to override the value
read from bootloader header.
Added BD_ADDR comparison in l2cu_lcp_allocate and removed check from
security connection request handler as it's handled in l2cu_lcp_allocate
for both connection request and create connection.
There was no need for this function as there is already a way present to
fetch own mac address. Own mac addr can be fetched by calling
controller_get_interface().
With the introduction of cookies to track a session, it is possible that the
clients restart the provisioning on the same session, specifically when a user
cancels a current provisioning attempt. This can result in an error as the state
on the device side and client side will go out of sync.
This has now been changed such that if SESSION_STATE_CMD0 is received on
an existing session, the state is reset and flow allowed to continue.
Earlier, a "socket connection" was being considered as a protocomm "session".
However, for HTTP, we cannot expect clients to necessarily use the same
socket for all requests of a given transaction. This was indeed causing
some failures for Wi-Fi provisioning from Android phones.
So, we have introduced http cookies to track a session even if the socket
connection does not stay intact throughout the transaction.
This change is backward compatible. If the client does not use cookies, the
socket connection will be considered as the session.
Closes: https://github.com/espressif/esp-idf/issues/6316
Reference: https://github.com/espressif/esp-idf/pull/6330
Not necessary in these cases, and prevents parens from expanding into the
assembly code added in the parent commit - a pattern which is accepted
by GCC assembler but illegal syntax for LLVM assembler.
Fix for issues where RTC FAST memory is updated as part of going into deep
sleep. Very high risk if heaps are in RTC memory - in particular task stacks
may be in RTC memory, but also other variables.
Also fixes potential concurrency problems as RTC FAST memory is not accessible
by CPU during the CRC calculation itself.
Method:
- Disable interrupts (currently for single core only, will need update for S3)
- Load all registers before calculating CRC or going to sleep
This reverts commit bf35ef1ce7.
It has been noticed that there are scenarios where even though firmware is not enabled
with flash encryption config feature, it should be able to write to encrypted partitions.
This revert adds the feature back which was removed around v4.0 timelines, and same
change will be backported to all releases (upto v4.0) for consistency.
esp_http_server: Add flag in sock_db to identify httpd_sess_close is called from httpd_session_close_lru (v4.2)
See merge request espressif/esp-idf!12112
esp_modem_stop_ppp() stops both ppp netif and switches the modem back to
command mode. IF these two actions are not synchronised, we might
experience issues of
* active PPP session trying to send/receive uart-data
* command mode already active before modem switched to it
both resulting in crashes.
Fixed by introducing the transition mode and running these actions in sequence
* set esp-modem to transition mode
* enter command mode, wait for the reply or re-sync
* close the PPP netif
* wait until the netif closes
Other fixes include ignoring certain events if modem component not ready
or not in appropriate mode:
* ignoring all UART events comming from DTE with no DCE attached
* ignore pattern detection in PPP mode
- line was truncated because 64 characters were not sufficient
- length passed to snprintf should be full buffer length, not -1
- make the width of lock name field fixed
- fix alignment of lock type column
To be in line with other interfaces, when deleting the driver and it's glue layer to esp-netif, we DO NOT delete the related esp-netif (since is was allocated and created outside of the glue code and it works the same way in esp-eth and esp-wifi driver/glue layers).
Closes https://github.com/espressif/esp-idf/issues/5486
Previously if starting/stopping BLE advertising when BLE Mesh
is enabled, the corresponding events will be notified through
the callback registered by esp_ble_mesh_register_prov_callback().
With this commit, the func esp_ble_mesh_register_ble_callback()
needs to be invoked for resgitering the callback used for BLE
coexistence related functionalities (i.e. ADV/SCAN).
the mDNS responder should not repeat questions when replying, however resolvers
must ignore these questions field if they are present. esp-idf mDNS
library does include questions in answering packets (thus not strictly
following the RFC6762) so the resolver did not correctly resolved
another instance host name.
Closes https://github.com/espressif/esp-idf/issues/6190
This commit fixes 'Impersonation in Passkey entry protocol'
(CVE-2020-26558) and suggests fixes for other vulnerabilites like
'Impersonation in the Pin Pairing Protocol' (CVE-2020-26555) and
'Authentication of the LE Legacy Pairing Protocol'
CVE-2020-26558 can be easily implemented if the peer device can
impersonate our public key. This commit adds a check by comparing our
and received public key and returns failed pairing if keys are same.
This commit also adds comments suggesting to use secure connection when
supported by all devices.
* When deinitializing or erasing a partition,
nvs used to close all handles instead of only
the current partition's handles.
This is fixed now
* Added a unit test for that case
* Unit tests clean up after each test case now
Closes FCS-533
Flag ESP32_WIFI_ENABLE_WPA3_SAE from Menuconfig is used to
control the stack size required by WPA3. Use the same flag
to compile out the WPA3 code and control the code size.
With this flag disabled code footprint reduces by about 7.7kB
in libwpa_supplicant.a
Update WiFi libs with added checks to cleanly handle disabling
of WPA3 callbacks.
1. Add description for esp_wifi_set_inactive_time()
2. Add documentation to avoid using WiFi modem sleep for WiFi/BT coexistence
3. Remove description of unusable API esp_wifi_restart()
4. Update esp_now_fetch_peer discription
5. Update table format prblm for esp_wifi_set_max_tx_power()
6. Update description for ssid and password
Curretly wpa_supplicant uses internal APIs for RSA operations
which internally uses lots of big num operations.
Big num operations are CPU expensive and can take a lot of time
which can cause watchdog timer to tigger.
This can be optimize by using mbedtls APIs which uses
hardware blocks for big num operations.
To fix this, write new crypto_mbedtls-rsa.c which has APIs
similar to crypto_internal-rsa.c but uses mbedtls APIs.
By default adds original queries to responses in order to be resolved by some resolvers, such as lwIP mdns library. This functionality however is discouraged by the RFC6762, so it could be disabled in menuconfig if MDNS_STRICT_MODE configured
Closes https://github.com/espressif/esp-idf/issues/5521
esp_http_client: Skip check for redirection counter if status code is success, fix issue with digest auth, configurable user agent string. (v4.2)
See merge request espressif/esp-idf!11597
tools/ldgen: Parse paths with Unicode characters in archives & fix parsing of section names on Windows (v4.2)
See merge request espressif/esp-idf!10964
* Fix the issue that deinit node with "erase_flash"
set to true, but info is not erased from nvs
* Reuse bt_mesh_cfg_reset() when deinit node
* Optimize Provisioner related erase operations
* No store pending timeout will be used when Node
is not provisioned OR Provisioner is disabled
and erase operation is performed
* Change the default timeout for settings operation
to 0, and rpl store rate to 0
This commit adds the feature where the TWAI ISR will continue to
run even if the cache is disabled. Whilst cache is disabled, any
received messages will go into the RX queue, and any pending TX
messages in the TX queue will be transmitted. This feature should
be enabled using the CONFIG_TWAI_ISR_IN_IRAM option.
This commit simplifies the defines made in the _caps.h header. Kconfig
option dependencies were moved into the LL, and the check for a
valid BRP has bee simplified.
esp-modem netif glue layer implements esp-netif attach callback to setup corresponding driver functions and start the modem. The error code of esp_modem_start_ppp() was ignored and ESP_OK returned in all cases. Fixed by passing esp_modem_start_ppp()'s error code to post_attach callback.
Closes https://github.com/espressif/esp-idf/issues/5430
The struct definition of ip6_addr_t in lwip and esp_ip6_addr_t
differs since zone could be possibly disabled in lwip. Using memcpy to copy the
address will cause wrong source address. Copy the entries manually
instead.
Merges https://github.com/espressif/esp-idf/pull/6055
1. will post twice disconnect event when read error
2. will block `timeout` times when set disable_auto_connect
3. When `esp_websocket_client_stop` before `esp_websocket_client_send*`,
if the `esp_websocket_client_send*` fails, the status will change to
'WEBSOCKET_STATE_WAIT_TIMEOUT', and the next `esp_websocket_client_start` will fail forever
When `pthread_mutex_destroy` is used to release mutex, `pthread_mutex_lock_internal` is used,
which results in the increase of `uxmutexehold` and no recovery base priority
1.Fix supervision timeout in LLC DISC busy statue can not disconnect issue.
2.Solve the problem that the ble terminate ack packet may lose with a high probability in coex mode.
For WPA3 connection nearing PMK lifetime expiry, PMK Cache
needs a re-authentication or the cache will expire. After
current expiry of 12 hours Station ends up sending a deauth
to the AP. An SAE re-authentication also cannot occur without
a disconnection with current implementation. So increase the
PMK lifetime to 100 days for now.
use eventgroup to sync spp_vfs_write
each connection has a switch_delay_timer not sharing a one
revert functions like spp_find_slot_by_xxx
fix vfs read bug when peer close
1. Fix FIX-ROOT does not reconnect to router when disconnect reason is too many
2. Add API esp_mesh_print_scan_result
3. Modify not to reset mesh_xonseq of self and children when flush_upstream_packets
4. Fix not switch to a parent candidate which has the same layer and assoc as current parent
5. Fix not arm parent monitor when a parent candidate is cleared without sending an event to mesh layer
6. Fix the new voted root does not reconnect to router if rssi is weak
1. Add STA checks during STA PMF operations
2. Fix WPA2-Ent issue with Open AP
3. Skip WPA-TKIP profile if PMF is required
4. Skip & clear Supplicant PMK Cache with mismatching AP config
Exclude paths that were specified with slash as a path separator were compared
as strings. This would fail on Windows which uses backslash as a path separator.
Also add some explanation about configuring the project in the README
Previously with the default build-esp32.sh script, the build directory was
deleted and recreated so the project would always be built with default config.
When reset the rx info of transport layer, the
rpl list will always cleared, and rpl stored
in the nvs will only be erased when erase flag
is true and BLE_MESH_SETTINGS is enabled.
Compared with the previous solution, it should
be more clear.
Commit 5e8795eebe has changed the partition table offset, which
has resulted in the ciphertext not matching the one expected in the
example test.
Fix by calculating the ciphertext using espsecure.py.
When the flash encryption is enabled then we do not need to change the partition table.
The partition_example.csv should not have fixed offsets for partitions because we want to move the whole table.
The fixed offsets in the table were cleared.
When idf.py gdb starts, it expects openocd was started in the background
and creates a thread to watch for openocd errors. when gdb target exits,
the debug_ext.py aims to cleanup all threads and processes, but fails
with traceback if openocd-watch thread not available, which could happen
if openocd started separately.
Builds on previous commit.
Note: Getting the project version still pases --tags so still works with plain
tags, to keep compatibility for existing projects
For getting secure_version field in anti rollback case, bootloader tries
to map whole firmware partition but fails for cases where partition size
is beyond available MMU free pages capacity.
Fix here ensures to map only required length upto application descriptor
size in firmware partition.
Closes https://github.com/espressif/esp-idf/issues/5911
Rewrite ble dynamic prio to fix ble disconn in conn_param_update/channel_map_update
Rewrite ble dynamic prio in connection establishment
Fix ble dynamic prio with latency
Fix status bit set error when conn fail
Since we have provided separate functions for deleting node
information with node's unicast address, device uuid, etc.
So we update the behavior of this function, which will only
be used to delete device information which is not provisioned
or just under provisioning.
1. Modify LR for esp32 and esp32s2
2. Fix esp32s2 RTS rate
3. Optimize some debug logs for probe response
4. Optimize phy and soc files
5. Add IPv6 DSCP
6. Fix classify AC to TID
7. Fix the bug that no event for connect
Commit 891eb3b0 was fixing an issue with PS and EPC1 not being
preserved after the window spill procedure. It did so by saving PS in
a2 and EPC1 in a4. However the a4 register may be a live register of
another window in the call stack, and if it is overwritten and then
spilled to the stack, then the corresponding register value will end
up being corrupted. In practice the problem would show up as an
IllegalInstruction exception, when trying to return from a function
when a0 value was 0x40020.
Fix by using a0 register instead of a4 as scratch. Also fix a comment
about xthal_save_extra_nw, as this function in fact doesn't clobber
a4 or a5 because XCHAL_NCP_NUM_ATMPS is defined as 1.
Closes https://github.com/espressif/esp-idf/issues/5758
Can still be enabled by passing --cmake-warn-uninitialized on the command line
Prevents CMake warnings printed by default if IDF_PATH is underneath the CMake
project directory.
The reason for this is that CMake --warn-uninitialized only enables checks
inside the project directory (ie top-level CMakeLists.txt directory and
subdirectories), it doesn't enable for files included from other directories.
(The only way to enable warnings in other directories is to pass
--check-system-dirs and this looks like it's only useful for CMake's own
developers as it prints a lot of warnings from inside CMake otherwise - see
https://gitlab.kitware.com/cmake/cmake/-/issues/19645 )
Plan to follow up with a later commit to clean up most of the warnings (which
aren't problems for CMake execution), but we'll also disable this option by
default to avoid this unexpected triggering of IDF warnings.
Due to eager_exclude, Sphinx jobs cannot use the doc cache (doctrees).
PDF and HTML building are now seperate CI jobs to avoid this issue.
PDF building will only be ran on deploy branches and when started
with build_docs tag to reduce CI load
1. Add g_av_with_rc to avrc_tg_init function
2. Remove g_av_open_fail
3. Add comment in API files
4. Add a2dp init & deinit state variable to inidcate a2dp statement
1. Add reaction for a2dp snk & src disconn req when no link up
2. Change state machine set before EVT upto APP layer in some cases
3. Add bool open_fail to choose whether post disconnection evt to app layer
Fixup more for PSRAM issue;
Allow GDB work correctly with Privileged Registers;
Based on GCC 8.4.0;
Update a fix of C++ exception crashs from the GCC mainline;
Fixup strptime_l() and some locale in libc;
If Secure Boot is enabled, there's no bootloader flasher arguments in the
project (to encourage careful manual flashing), so this command would
fail with a KeyError.
Per WiFi library requirement, SSID can be non-null terminated string
if its length goes to 32 bytes (maximum). Use of strncpy in this case,
along with compiler optimization level -O2 results in some warnings
for potential use of non-null terminated strings.
Fix here ensures use of memcpy to copy SSID string upto appropriate
desired length. This helps to avoid compiler specific workaround
flags added earlier.
Closes https://github.com/espressif/esp-idf/issues/5866
Closes IDFGH-3983
bootloader_support: don't check signature when JTAG is attached, fix signed partition table generation (v4.2)
See merge request espressif/esp-idf!10378
Bootloader DRAM now ends at 0x3FFEAB00 which is the start of ROM
static RAM (reclaimable after app is running).
IRAM loader segment increased by 8KB.
Available total static RAM for the app is now reduced by 16KB.
Provisioner should always uses the latest IV Index for provisioning.
For example, if the current IV Index is 0x00000001, but prov->iv_index
is still initialized with 0x00000000, and if Provisioner uses prov->
iv_index for provisioning, this will cause the Provisioner failing to
control the node.
So here bt_mesh.iv_index is used instead of prov->iv_index.
Fixes the issue that signed partition table was not generated when
CONFIG_SECURE_BOOT_BUILD_SIGNED_BINARIES was on, because
partition_table_bin depended on unsigned_partition_bin twice.
Regression from acb7a211.
Also use final_partition_bin variable in esptool_py_flash_target_image
arguments, to avoid issues in the future if final_partition_bin is
changed.
- Previously when a model is initialized or deinitialized, in the
access layer, we need to check the model id with the ids in the
table in order to find the proper model operation function.
- Currently all the operation functions of each model will be set
during the mesh initialization. When the model is found, we can
directly use the corresponding callback for different operations.
- Currently only init/deinit operations are registered, later we
will add more operations.
{esp32,esp32s2}.inc files contain most of the bits and code snippets
which are different between targets. These bits are included from the
rest of the source files.
Also included is an update to the section about OpenOCD configuration
variables and files. This section wasn't updated when some of the
files and variables got renamed.
CONFIG_FREERTOS_ISR_STACKSIZE was set to 2100 when ELF core dump was
enabled, which resulted in a non-16-byte-aligned interrupt stack
offset. This triggered "is SP corrupted" check in the backtrace,
terminating the backtrace early.
Fix the default value, and make sure that the stack is always aligned,
regardless of the value of CONFIG_FREERTOS_ISR_STACKSIZE.
InstrFetchProhibited usually occurs because of a jump to an invalid
pointer. In this case, PC in the exception frame is the address of
the jump destination. 'esp_ptr_executable' check in print_backtrace
function recognizes the first frame as invalid, and the backtrace is
interrupted. This prevents the user from finding the location where
the invalid pointer is dereferenced.
Bypass the 'esp_ptr_executable' check if the exception cause is
InstrFetchProhibited. Update the test case to no longer ignore this
issue.
1. Add logs for SPP not enabled successfully
2. Add information to Applicatioin layer that the A2DP is connecting to peer device
Regular BTC and BTU Task name & Add BTC Init protection
1. Regular the BTC Task Name to Close a Github issue
Closes https://github.com/espressif/esp-idf/issues/4966
2. Add BTC Init failure protection
3. Regular BTU Task Name with BTC
4. Correct the comment of
a. SDP_ServiceSearchRequest and SDP_ServiceSearchRequest2
b. l2cu_create_conn_after_switch
5. Correct comment of esp_bt_hf_vra();
6. Fix cpp compilier region
Add logic to set `FIN` flag automatically for transmitted WS frames, but
if `fragmented` option set indicating an expert/manual mode, then the
`FIN` flag is set according to the `final` option.
H/W decryption of Mgmt frames was disabled for PMF and done through
S/W. If ESPNOW packets go through this path, it affects backward
compatibility since method of decrypting Mgmt packets is different in H/W.
To address PMF + ESPNOW Co-existance, CCMP decryption method is modified
for ESPNOW packets so that they can be decrypted correctly. Since Tx
of ESPNOW packets can still be done in H/W alongside PMF, no change
required in encryption method in S/W.
Co-Authored-By: Nachiket Kukade <nachiket.kukade@espressif.com>
Co-Authored-By: zhangyanjiao <zhangyanjiao@espressif.com>
Co-Authored-By: kapil.gupta <kapil.gupta@espressif.com>
wpa_supplicant: Deinit wpa2 states in wifi deinit
See merge request espressif/esp-idf!10103
(cherry picked from commit 4dcaa0e6f4)
06bc7f1d wpa_supplicant: Deinit wpa2 states in wifi deinit
ESP32 lets the user choose from using Vref, Lookup Table, and Two-Point Calibration. In ESP32S2 only two-point calibration is supported. This commit support these changes in idf.
Closes https://github.com/espressif/esp-idf/issues/5455
1.Fix WiFi fragment issue
2.fix the bug for softAP power save buffer issue
3.fix softap mode event issue
4.fix connect hidden AP doesn't update information
5.update esp32 phy v4370
If esp_restart_noos() is run and the stack address points to external memory (SPIRAM)
then Cache_Read_Disable() raises up the error "Cache disabled but cached memory region accessed"
to fix this we switch stack to internal RAM before disable cache.
Added unit tests.
Closes: https://github.com/espressif/esp-idf/issues/5107
1. If Device is connected to AP in WPA3-PSK mode, AP switching
security to WPA2-PSK causes connection failures even after reset.
Fix is to not store WPA3's PMK in NVS for caching.
2. AP switching back to WPA3 causes even more connection failures.
This is due to device not clearing Supplicant level PMK Cache when
it is no longer valid. Fix is to clear the Cache when 4-way handshake
fails and to check Key Mgmt of Cache before using.
3. When AP switches from WPA3 to WPA2, device's PMF config in
Supplicant remains enabled. This may cause failures during
4-way handshake. So clear PMF config in when PMF is no longer used.
If the olddelta argument is not a null pointer, the adjtime function returns information
about any previous time adjustment that has not yet completed.
Closes: https://github.com/espressif/esp-idf/issues/5194
Replace periph_module_enable/disable by periph_wifi_bt_common_module_enable which are in IRAM.
AddIRAM_ATTR periph_ll_wifi_bt_module_enable_clk_clear_rstandIRAM_ATTR periph_ll_wifi_bt_module_disable_clk_set_rstto fit O0 optimization level.
Delete duplicated spinlock and counter.
Extended test cases for DHCP server and DHCP client state transitions to include also manual transitions using esp_netif_dhcps_start/stop() esp_netif_dhcpc_start/stop()
This commit adds TWAI driver support for the
ESP32-S2. The following features were added:
- Expanded BRP support
- Expanded CLKOUT Divider Support
- Updated example READMEs
In some cases gen_version_specific_includes.py would fail to find a branch/tag name for a commit
and would fall back to using commit SHA-1.
This should never happen in docs that should be published as we use the branch/tag name to generate
example commands for our users. E.g. "git clone -b $branch_name"
Closes https://github.com/espressif/esp-idf/issues/5657
SD cards don't support CMD7 (select_card) in SPI mode. Highspeed probe
of sdspi will fail in this step and stop working in highspeed mode.
Remove the CMD7 in enable_hs_mode_and_check to fix this issue.
Please note that, on ESP32, you have to use the IOMUX pins to use sdspi
in 40MHz, otherwise the initialization process will report reading issue
and fail.
WPS can send multiple AP credentials, while existing implementation
will only use the first credentials which could be for the 5G band.
Fix this by passing these credentials to the App and attempting
to connect using each of those. Older Apps will remain compatible
without breaking WPS, but the issue will remain.
Add following changes as part of this:
1. EAP client will crash during validation of key size when CA
certs and keys not present. Add changes to validate it first.
2. Free memory allocated in TLS context
Allows booting in QIO/QOUT mode or with PSRAM on ESP32-PICO-V3 and
ESP32-PICO-V3-O2 without any config changes.
Custom WP pins (needed for fully custom circuit boards) should still be compatible.
Currently only keep func pointer for the followings:
- Invalid parameter (mesh btc & mesh stack)
- Out of memory (mesh btc & mesh stack)
- Unknown act (mesh btc)
- Invalid model user data (mesh stack)
- BT_DBG("%s", __func__) (mesh btc & mesh stack)
- A few other specific situations (buf ref debug, send status check)
Different bluetooth host has different behaviors, so it's better
to maintain a scan check mechanism of BLE Mesh itself.
Fixes an issue when only PB-GATT is enabled for node, which will
output a scan error log when the device is provisioned.
mbedTLS stack does not keep TLS renegotiation enabled even if
relevant config option is turned on, it needs explicit API call
`mbedtls_ssl_conf_renegotiation` to do so.
This issue was observed in case of Azure IoT, where keys needs to
be refreshed periodically to keep TLS connection intact.
Same logic as for the ESP32, except two changes:
* need to set UART_SW_FLOW_CON_EN bit for UART_FORCE_XOFF to take
effect
* need to check if the peripheral is not clockgated and out of reset
Previously both tasks had equal priority, possible for write task and another
internal task to be scheduled at the same time - starving read task and
causing data loss.
Related to IDFCI-59
Some errors (for example the errors caused by macro values, as fixed in previous
commit) get logged to stderr even if WARN_LOGFILE is set, but Doxygen succeeds.
Update send_ttl mainly for server models. When a server model
receives a message, and the status is required to be replied
by the application, we need to set send_ttl to the msg context.
If send_ttl is not updated in btc, and the applcation does not
set the TTL either, then the status will be replied with TTL=0,
which may cause the client side (e.g. the phone App) failed to
receive the status.
Closes https://github.com/espressif/esp-idf/issues/5300
1. Avoid parsing RSN/WPA IE's after connection
2. Remove AUTH_CHANGED reason code logic
3. fix the bug when calculating the PMK, the local rx beacon time increment is 0
4. mesh: don't disconnect when tbtt update
Also remove some redundant CONFIG_BLE_MESH_PROXY checks, because
when the following options are satisfied, the CONFIG_BLE_MESH_PROXY
option will be selected mandatorily.
The issue is caused by:
1. The disable_qio_mode inside read_id may have side effects.
2. read_id twice may have side effects.
Fix this issue by moving disable_qio_mode out of read_id and only do it
once before read_id. And retry read_id only when the first one is
failed.
Issue introduced in 3ecbb59c15.
Prevent a remote device from doing a Bluetooth Impersonation Attack
(BIAS) by:
- Preventing remote device to downgrade secure connection
feature mask. Secure connection feature mask should remain same or
increase to enabled in link key generation and authentication.
- Doing a mutual authentication during Legacy Authentication.
Signed-off-by: Chinmay Chhajed <chinmay.chhajed@espressif.com>
1. remove warning log when mesh is started:
Error! Should use default passive scan time parameter for WiFi scan when Bluetooth is enabled!!!!!!
2. remove abort when mesh is started:
Error! Should enable WiFi modem sleep when both WiFi and Bluetooth are enabled!!!!!!
- Before updating the "beacon_sent" of a subnet, we need to check
if the subnet still exists, especially for a Provisioner.
- Fix a bug which will cause Provisioner failed to send Secure
Network Beacon if no device is provisioned
ctx values to cb_params for the mesh stack.
recv_rssi was not copied.
This means the rssi could not be read when receiving generic server messages using ble_mesh.
1.In the description of the ESP_API header file, it should be the event that the application layer needs to handle, not the BTA layer.
2.Fix name typo of BTA_GATTS_CREATE_SRVC_EVT event
Closes https://github.com/espressif/esp-idf/issues/5446
In commit 309376f51a, it seems like regression
was added to use ROM level API for disabling flash write protection. This
started random firmware crashes (on specific modules) with exception
`IllegalInstruction` during encrypted flash writes.
Fix here removes relevant ROM API call, since disabling flash write protection
is already ensured by caller of this API.
Closes https://github.com/espressif/esp-idf/issues/5467
- This issue was introduced in the commit: a788e7cd3d
which updated the bt_hex() function incorrectly.
- And in bt_hex(), we use 2 for the two-dimensional array, because currently
at most two bt_hex() will be used at the same time. Also this will save
some DRAM compared with using 4.
Move the TAG definition of each example to the corresponding
source file (previously it is defined in the common example
intialization header file), which can avoid the redefinition
of TAG in some situations.
If IDF_PYTHON_ENV_PATH was not set, an exception occurred:
Traceback (most recent call last):
File "/home/user/esp/esp-idf/tools/check_python_dependencies.py", line 108, in <module>
if idf_python_env_path not in sys.executable:
TypeError: 'in <string>' requires string as left operand, not NoneType
and the final line in the diagnostic message was not printed.
Fix to print the PATH if IDF_PYTHON_ENV_PATH is not set.
esp32/esp32s2: Reduce using ESP_EARLY_LOGx and move some code after the stdout initialization in startup code (v4.2)
See merge request espressif/esp-idf!9202
1. Fix WPA3 Auth timeout during BLE coex
2. fix addba ssn and send bar for first ampdu
3. fix the bug that airkiss find hidden AP fail
4. update mesh lib
5. fix interrupt handle param variable error
When using DPort to read fifo, fifo_cnt is not credible, we need to calculate the real cnt based on the fifo read and write pointer. When using AHB to read FIFO, we can use fifo_cnt to indicate the data length in fifo.
check_public_headers.py would fail when reading non-ascii chars on systems where
the default encoding was ascii.
Also fixes error handling issues, as any uncaught exeception would cause the program
to run indefinitely.
Added following as part of this change
1. Removed auth_changed event
2. Updated doc regarding usage of WPA/WEP secuiry modes
3. Fixed WPA2 enterprise authmode threshold.
Previously the model recv operation is a littl fuzzy.
With the changes, the model recv operation is splitted into:
- find op
- check app_idx
- check dst
- check length
- update info & handle message
@@ -24,7 +24,7 @@ If the issue cannot be solved after the steps before, please follow these instru
1. Fill in all the fields under **Environment** marked with [ ] by picking the correct option for you in each case and deleting the others.
2. Describe your problem.
3. Include [debug logs from the "monitor" tool](https://docs.espressif.com/projects/esp-idf/en/latest/api-guides/tools/idf-monitor.html), or [coredumps](https://docs.espressif.com/projects/esp-idf/en/latest/api-guides/core_dump.html).
3. Include [debug logs from the "monitor" tool](https://docs.espressif.com/projects/esp-idf/en/latest/get-started/idf-monitor.html#automatically-decoding-addresses), or [coredumps](https://docs.espressif.com/projects/esp-idf/en/latest/api-guides/core_dump.html).
4. Providing as much information as possible under **Other items if possible** will help us locate and fix the problem.
5. Use [Markdown](https://guides.github.com/features/mastering-markdown/) (see formatting buttons above) and the Preview tab to check what the issue will look like.
6. Delete these instructions from the above to the below marker lines before submitting this issue.
@@ -38,7 +38,7 @@ If the issue cannot be solved after the steps before, please follow these instru
- Development Kit: [ESP32-Wrover-Kit|ESP32-DevKitC|ESP32-PICO-Kit|ESP32-LyraT|ESP32-LyraTD-MSC|none]
- Kit version (for WroverKit/PicoKit/DevKitC): [v1|v2|v3|v4]
- Module or chip used: [ESP32-WROOM-32|ESP32-WROOM-32D|ESP32-WROOM-32U|ESP32-WROVER|ESP32-WROVER-I|ESP32-WROVER-B|ESP32-WROVER-IB|ESP32-SOLO-1|ESP32-PICO-D4|ESP32]
- IDF version (run ``git describe --tags`` to find it):
- IDF version (run ``git describe`` to find it):
// v3.2-dev-1148-g96cd3b75c
- Build System: [Make|CMake|idf.py]
- Compiler version (run ``xtensa-esp32-elf-gcc --version`` to find it):
- [Manifest File to Control the Build/Test apps](#manifest-file-to-control-the-buildtest-apps)
- [Grammar](#grammar)
## General Workflow
1. Push to a remote branch
2. Create an MR, choose related labels (not required)
3. A `detached` pipeline will be created.
4. if you push a new commit, a new pipeline will be created automatically.
## What if Expected Jobs ARE NOT Created?
1. check the file patterns
If you found a job that is not running as expected with some file changes, a git commit to improve the `pattern` will be appreciated.
2. please add MR labels to run additional tests, currently we have to do this only for `target-test` jobs, please use it as few as possible. Our final goal is to remove all the labels and let the file changes decide everything!
## MR labels for additional jobs
### Supported MR Labels
-`build`
-`build_docs`
-`component_ut[_esp32/esp32s2/...]`
-`custom_test[_esp32/esp32s2/...]`
-`docker`
-`docs`
-`docs_fast`, triggers a fast docs build, not a full build which is the CI default. This skips PDF build as well as doxygen APIs, reducing the build time by 90+%.
-`example_test[_esp32/esp32s2/...]`
-`fuzzer_test`
-`host_test`
-`integration_test`
-`iperf_stress_test`
-`macos`
-`macos_test`
-`nvs_coverage`
-`submodule`
-`unit_test[_esp32/esp32s2/...]`
-`weekend_test`
-`windows`
There are two general labels (not recommended since these two labels will trigger a lot of jobs)
-`target_test`: includes all target for `example_test`, `custom_test`, `component_ut`, `unit_test`, `integration_test`
-`all_test`: includes all test labels
### How to trigger a `detached` pipeline without pushing new commits?
Go to MR web page -> `Pipelines` tab -> click `Run pipeline` button.
In very rare case, this tab will not show up because no merge_request pipeline is created before. Please use web API then.
```shell
curl -X POST --header "PRIVATE-TOKEN: [YOUR PERSONAL ACCESS TOKEN]"[GITLAB_SERVER]/api/v4/projects/103/merge_requests/[MERGE_REQUEST_IID]/pipelines
```
## How to Develop With `rules.yml`?
### General Concepts
-`pattern`: Defined in an array. A GitLab job will be created if the changed files in this MR matched one of the patterns. For example:
```yaml
.patterns-python-files: &patterns-python-files
- "**/*.py"
```
- `label`: Defined in an if clause, similar as the previous bot command. A GitLab job will be created if the pipeline variables contains variables in `BOT_LABEL_xxx` format (DEPRECATED) or included in the MR labels. For example:
- `rule`: A combination of various patterns, and labels. It will be used by GitLab YAML `extends` keyword to tell GitLab in what conditions will this job be created. For example:
```yaml
.rules:build:docs:
rules:
- <<: *if-protected
- <<: *if-label-build_docs
- <<: *if-label-docs
- <<: *if-dev-push
changes: *patterns-docs
```
An example for GitLab job on how to use extends:
```yaml
check_docs_lang_sync:
extends:
- .pre_check_template
- .rules:build:docs
script:
- cd docs
- ./check_lang_folder_sync.sh
```
### How to Add a New `Job`?
check if there's a suitable `.rules:<rules-you-need>` template
1. if there is, put this in the job `extends`. All done, now you can close this window. (`extends` could be array or string)
2. if there isn't
1. check [How to Add a New `Rules` Template?](#how-to-add-a-new-rules-template), create a suitable one
2. follow step 1
### How to Add a New `Rules` Template?
check if this rule is related to `labels`, `patterns`
1. if it is, please refer to [dependencies/README.md](./dependencies/README.md) and add new rules by auto-generating
2. if it isn't, please continue reading
check if there's a suitable `.if-<if-anchor-you-need>` anchor
1. if there is, create a rule following [`rules` Template Naming Rules](#rules-template-naming-rules).For detail information, please refer to [GitLab Documentation `rules-if`](https://docs.gitlab.com/ee/ci/yaml/README.html#rulesif). Here's an example.
```yaml
.rules:patterns:python-files:
rules:
- <<: *if-protected
- <<: *if-dev-push
changes: *patterns-python-files
```
2. if there isn't
1. check [How to Add a New `if` Anchor?](#how-to-add-a-new-if-anchor), create a suitable one
2. follow step 1
### How to Add a New `if` Anchor?
Create an `if` anchor following [`if` Anchors Naming Rules](#if-anchors-naming-rules). For detailed information about how to write the condition clause, please refer to [GitLab Documentation `only/except (advanced)](https://docs.gitlab.com/ee/ci/yaml/README.html#onlyexcept-advanced). Here's an example.
```yaml
.if-schedule:&if-schedule:
if:'$CI_PIPELINE_SOURCE == "schedule"'
```
### Naming Rules
#### Common Naming Rules
if a phrase has multi words, use `_` to concatenate them.
> e.g. `regular_test`
if a name has multi phrases, use `-` to concatenate them.
> e.g. `regular_test-example_test`
#### `if` Anchors Naming Rules
- if it's a label: `.if-label-<label_name>`
- if it's a ref: `.if-ref-<ref_name>`
- if it's a branch: `.if-branch-<branch_name>`
- if it's a tag: `.if-tag-<tag_name>`
- if it's multi-type combination: `.if-ref-<release_name>-branch-<branch_name>`
a combination of `example_test`, `custom_test`, `unit_test`, `component_ut`, `integration_test` and all targets
#### `rules` Template Naming Rules
- if it's tag related: `.rules:tag:<tag_1>-<tag_2>`
- if it's label related: `.rules:labels:<label_1>-<label_2>`
- if it's test related: `.rules:test:<test_type>`
- if it's build related: `.rules:build:<build_type>`
- if it's pattern related: `.rules:patterns:<patterns>`
## Reusable Shell Script `tools/ci/utils.sh`
It is used to put all the reusable shell scripts as small functions. If you want to set `before_script: []` for you job, now you can set `extends: .before_script_slim` instead. it will only run `source tools/ci/utils.sh`
If you're developing CI shell scripts, you can use these functions without `source` them. They're already included in all `before_script`
To run these commands in shell script locally, place `source tools/ci/utils.sh` at the very beginning.
### Functions
#### CI Job Related
-`add_gitlab_ssh_keys`
-`add_github_ssh_keys`
-`add_doc_server_ssh_keys`
-`fetch_submodules`
-`get_all_submodules`
#### Shell Script Related
-`error`: log in red color
-`warning`: log in orange color
-`info`: log in green color
-`run_cmd`: run the command with duration seconds info
-`retry_failed`: run the command with duration seconds info, retry when failed
## Manifest File to Control the Build/Test apps
`.build-test-rules.yml` file is a manifest file to control if the CI is running the build and test job or not. The Supported Targets table in `README.md` for apps would be auto-generated by `pre-commit` from the app's `.build-test-rules.yml`.
### Grammar
#### Operands
- Variables start with `SOC_`. The value would be parsed from components/soc/[TARGET]/include/soc/*_caps.h
-`IDF_TARGET`
-`INCLUDE_DEFAULT` (The default value of officially supported targets is 1, otherwise is 0)
- String, must be double-quoted. e.g. `"esp32"`, `"12345"`
- Integer, support decimal and hex. e.g. `1`, `0xAB`
- List with String and Integer inside, the type could be mixed. e.g. `["esp32", 1]`
#### Operators
-`==`, `!=`, `>`, `>=`, `<`, `<=`
-`and`, `or`
-`in`, `not in` with list
- parentheses
#### Limitation:
- all operators are binary operator. For more than two operands, you may use nested parentheses trick. For example,
-`A == 1 or (B == 2 and C in [1,2,3])`
-`(A == 1 and B == 2) or (C not in ["3", "4", 5])`
### How does it work?
By default, we enable build and test jobs for supported targets. In other words, if an app supports all supported targets, it does not need to be added in a manifest file. The manifest files are files that set the violation rules for apps.
three rules (disable rules are calculated after the `enable` rule):
- enable: run CI build/test jobs for targets that match any of the specified conditions only
- disable: will not run CI build/test jobs for targets that match any of the specified conditions
- disable_test: will not run CI test jobs for targets that match any of the specified conditions
Each key is a test folder. Will apply to all folders inside.
If one sub folder is in a special case, you can overwrite the rules for this folder by adding another entry for this folder itself. Each folder's rules are standalone, and will not inherit its parent's rules. (YAML inheritance is too complicated for reading)
For example in the following codeblock, only `disable` rule exists in `examples/foo/bar`. It's unaware of its parent's `enable` rule.
```yaml
examples/foo:
enable:
- if:IDF_TARGET == "esp32"
examples/foo/bar:
disable:
- if:IDF_TARGET == "esp32s2"
```
### Example
```yaml
examples/foo:
enable:
- if IDF_TARGET in ["esp32", 1, 2, 3]
- if IDF_TARGET not in ["4", "5", 6]
# should be run under all targets!
examples/bluetooth:
disable:# disable both build and tests jobs
- if:SOC_BT_SUPPORTED != 1
# reason is optional if there's no `temporary: true`
disable_test:
- if:IDF_TARGET == "esp32"
temporary:true
reason: lack of ci runners # required when `temporary:true`
examples/bluetooth/test_foo:
# each folder's settings are standalone
disable:
- if:IDF_TARGET == "esp32s2"
temporary:true
reason:noidea
# unlike examples/bluetooth, the apps under this folder would not be build nor test for "no idea" under target esp32s2
examples/get-started/hello_world:
enable:
- if:IDF_TARGET == "linux"
reason:this one only supports linux!
examples/get-started/blink:
enable:
- if:INCLUDE_DEFAULT == 1 or IDF_TARGET == "linux"
reason:This one supports all supported targets and linux
- if [ -n "$LOCAL_GITLAB_HTTPS_HOST" ]; then export LOCAL_CI_REPOSITORY_URL="https://gitlab-ci-token:${CI_JOB_TOKEN}@${LOCAL_GITLAB_HTTPS_HOST}/${CI_PROJECT_PATH}"; fi
- if [ -n "$LOCAL_GIT_MIRROR" ]; then export LOCAL_CI_REPOSITORY_URL="${LOCAL_GIT_MIRROR}/${CI_PROJECT_PATH}"; fi
- echo "Using repository at $LOCAL_CI_REPOSITORY_URL"
1. Auto-generate some labels/rules we need and update them in `rules.yml`
2. Generate a dependency tree graph
## Schema
This file only used basic YAML grammar and has nothing to do with the GitLab version YAML file.
It has five custom keywords:
-`matrix`: An array of sub-arrays, used to replicate rules by formatting strings. You can use the format string everywhere, it will be formatted recursively
-`labels`: An array of `labels`.
-`patterns`: An array of `patterns`. Patterns that not included
-`included_in`: An array of other `rule` names. It indicates the `labels` and `patterns` will be included in all specified `rules` as well
-`deploy`: An array of strings, used to replicate rules by adding postfix `-<item in deploy array>`. It indicates the extra `label` used in `rules`, which will explain later.
## How to use this file to generate `rules.yml`
Let's take a complicated example to help understand the process
```yaml
"test-{0}-{1}":
matrix:
- [a, b]
- [c, d]
labels:
- "{0}-{1}"
patterns:
- "{0}"
- pattern-not-exist
included_in:
- build-{0}
```
1. expand the mapping dicts defined by `matrix`
After this step, it will turn into 4 dicts:
| key | labels | patterns | included_in |
| -------- | ------ | -------- | ----------- |
| test-a-c | a-c | a | build-a |
| test-a-d | a-d | a | build-a |
| test-b-c | b-c | b | build-b |
| test-b-d | b-d | b | build-b |
**Advanced Usage: You can overwrite a mapping by declaring it again later**, For example:
If we concatenate this part to the previous example,
```yaml
# ... The same as the previous example
test-a-c:
labels:
- overwrite
```
`rule` `test-a-c` will be turned into:
| key | labels |
| -------- | --------- |
| test-a-c | overwrite |
**Mappings with the keyword `deploy` will also replicate by adding a postfix `-<item in deploy array>` to the mapping key**
2. create rules by `included_in`
After this step, it will turn into 6 mapping dicts:
| key | labels | patterns |
| -------- | -------- | -------- |
| test-a-c | a-c | a |
| test-a-d | a-d | a |
| test-b-c | b-c | b |
| test-b-d | b-d | b |
| build-a | a-c, a-d | a |
| build-b | b-c, b-d | b |
3. replace the auto-generated region in `rules.yml` with `labels`, and `rules`. Each mapping will generate a `rule` and all the required labels. `patterns` are pre-defined in `rules.yml` and could not be generated automatically. If a mapping is using a `pattern` undefined, the `pattern` will be ignored.
- If a mapping key has postfix `-preview`, no `if-protected-xxx` clause will be added
- else if a mapping key has postfix `-production`, an `if-protected-no_label` clause will be added
- else, an `if-protected` clause will be added
## Graph
All `label` nodes are in green, `pattern` nodes are in cyan, `rule` nodes are in blue
### Requirements
There are a few extra dependencies while generating the dependency tree graph, please refer to [pygraphviz](https://github.com/pygraphviz/pygraphviz/blob/master/INSTALL.txt) documentation to install both `graphviz` and `pygraphviz`
- git diff --exit-code -- ${IDF_TARGET}/esp_efuse_table.c || { echo 'Differences found for ${IDF_TARGET} target. Please run make efuse_common_table or idf.py efuse-common-table and commit the changes.'; exit 1; }
- cd ${IDF_PATH}/components/efuse/test_efuse_host
- ./efuse_tests.py
test_efuse_table_on_host_esp32:
extends:.test_efuse_table_on_host_template
test_efuse_table_on_host_esp32s2:
extends:.test_efuse_table_on_host_template
variables:
IDF_TARGET:esp32s2
test_efuse_table_on_host_esp32s3:
extends:.test_efuse_table_on_host_template
variables:
IDF_TARGET:esp32s3
test_efuse_table_on_host_esp32c3:
extends:.test_efuse_table_on_host_template
variables:
IDF_TARGET:esp32c3
test_efuse_table_on_host_esp32h2:
extends:.test_efuse_table_on_host_template
variables:
IDF_TARGET:esp32h2
test_logtrace_proc:
extends:.host_test_template
artifacts:
when:on_failure
paths:
- tools/esp_app_trace/test/logtrace/output
- tools/esp_app_trace/test/logtrace/.coverage
expire_in:1week
script:
- cd ${IDF_PATH}/tools/esp_app_trace/test/logtrace
- ./test.sh
test_sysviewtrace_proc:
extends:.host_test_template
artifacts:
when:on_failure
paths:
- tools/esp_app_trace/test/sysview/output
- tools/esp_app_trace/test/sysview/.coverage
expire_in:1week
script:
- cd ${IDF_PATH}/tools/esp_app_trace/test/sysview
- ./test.sh
test_mkdfu:
extends:.host_test_template
variables:
LC_ALL:C.UTF-8
script:
- cd ${IDF_PATH}/tools/test_mkdfu
- ./test_mkdfu.py
test_mkuf2:
extends:.host_test_template
script:
- cd ${IDF_PATH}/tools/test_mkuf2
- ./test_mkuf2.py
test_autocomplete:
extends:
- .host_test_template
artifacts:
when:on_failure
paths:
- ${IDF_PATH}/*.out
expire_in:1week
script:
- ${IDF_PATH}/tools/ci/test_autocomplete.py
test_detect_python:
extends:
- .host_test_template
script:
- cd ${IDF_PATH}
- shellcheck -s sh tools/detect_python.sh
- shellcheck -s bash tools/detect_python.sh
- shellcheck -s dash tools/detect_python.sh
- "bash -c '. tools/detect_python.sh && echo Our Python: ${ESP_PYTHON?Python is not set}'"
- "dash -c '. tools/detect_python.sh && echo Our Python: ${ESP_PYTHON?Python is not set}'"
- "zsh -c '. tools/detect_python.sh && echo Our Python: ${ESP_PYTHON?Python is not set}'"
_All texts in italics are instructional and should be replaced by contents or removed._
## Checklist
_This entire section can be deleted if all items are checked._
* [ ] Enough information to help reviewers understand the issue, its root cause, impact, and the proposed solution
* [ ] Enough information to help reviewers understand the feature, its functional description, example, documentation, test cases, test results, feature TODO list
* [ ] The MR Title describes the change, including the component name, ie "lwip: Add support for IP over Pigeon"
* [ ] All related links, including JIRA, backport, submodule MR, are mentioned in the `Related` subsection.
* [ ] Any GitHub issues are linked inside the git commit message and corresponding release notes
* [ ] Add label for the area this MR is part of
* [ ] For documentation updates, check if label `Docs` and `needs translation:CN` or `needs translation:EN` have been added when the other language version still needs the update. Skip adding the label if the document is not yet translated.
* [ ] Check if documents requiring translation fall under get-started section. If yes, add the labels mentioned above. Then the documentation team will assign a translator for you. Please inform the translator to prepare translation once your MR is ready to merge. The translation should be included in your MR to get it merged. For more information, see documentation workflow in Wiki.
* [ ] Any necessary "needs backport" labels are added
* [ ] Check if this is a breaking change. If it is, add notes to the `Breaking change notes` subsection below
* [ ] Release note entry if this is a new public feature, or a fix for an issue introduced in the previous release.
* [ ] The commit log is clean and ready to merge.
* [ ] All relevant CI jobs have been run, i.e. jobs which cover the code changed by the MR.
---
_For issues, put enough information here to help reviewers understand the issue, its root cause, impact, and the proposed solution._
_If this issue is a regression, specify in the `Related` subsection below, in which commit or MR it was introduced. This helps reviewers to check if the backport labels are set correctly._
---
_For features, put enough information here to help reviewers understand the feature, its functional description, example, documentation, test cases, test results, feature TODO list._
---
_For other small/non-public changes, which are not expected to be in the release notes, can be mentioned here, include:_
* changes that don't apply to customers, e.g. some CI fixes, test only MR and etc
## Related
* Mention or related JIRA tasks (e.g. IDF-0000), to make sure they get updated.
* Mention submodule MR, if there is
* Mention backport(ed) MR, if there is
_Don't touch the subsection titles below, they will be parsed by scripts._
## Release notes (Mandatory)
_Changes made in this MR that should go into the **Release Notes** should be listed here. Please use **past tense** and *specify the area (see maintainers page of IDF internal wiki)*. If there is a subscope, include it and separate with slash (`/`). Minor changes can go to the descriptions above without a release notes entry._
_Write all the changes in a **list** (Start at the beginning of the line with `-` or `*`). If multiple changes are made, each of them should take a single line. If there is only one change to list, it should still be the only line of a list. If this MR does not need any release notes, write "No release notes" here without the `-` or `*`. e.g._
_If there are any breaking changes, please mention it here. Talking about (1) what is not accepted any more, (2) the alternative solution and (3) the benefits/reason. e.g._
_Please strictly follow the breaking change restriction, which means, if there is a breaking change but you are merging to non-major versions, you have to separate the breaking part out to another MR for a major version. The breaking change subsection is only accepted in MRs merging to major versions._
* [VFS/UART] Now vfs_uart_set_rts_cts accept one more instance argument, to support configuration to different ports.
Contributions to esp-idf - fixing bugs, adding features, adding documentation - are welcome. We accept contributions via `Github Pull Requests <https://help.github.com/en/github/collaborating-with-issues-and-pull-requests/about-pull-requests>`_.
Before Contributing
-------------------
Before sending us a Pull Request, please consider this list of points:
* Is the contribution entirely your own work, or already licensed under an Apache License 2.0 compatible Open Source License? If not then we unfortunately cannot accept it.
* Does any new code conform to the esp-idf :doc:`Style Guide <style-guide>`?
* Does the code documentation follow requirements in :doc:`documenting-code`?
* Is the code adequately commented for people to understand how it is structured?
* Is there documentation or examples that go with code contributions? There are additional suggestions for writing good examples in :idf:`examples` readme.
* Are comments and documentation written in clear English, with no spelling or grammar errors?
* Example contributions are also welcome. Please check the :doc:`creating-examples` guide for these.
* If the contribution contains multiple commits, are they grouped together into logical changes (one major change per pull request)? Are any commits with names like "fixed typo" `squashed into previous commits <https://eli.thegreenplace.net/2014/02/19/squashing-github-pull-requests-into-a-single-commit/>`_?
* If you're unsure about any of these points, please open the Pull Request anyhow and then ask us for feedback.
Pull Request Process
--------------------
After you open the Pull Request, there will probably be some discussion in the comments field of the request itself.
Once the Pull Request is ready to merge, it will first be merged into our internal git system for in-house automated testing.
If this process passes, it will be merged onto the public github repository.
Legal Part
----------
Before a contribution can be accepted, you will need to sign our :doc:`contributor-agreement`. You will be prompted for this automatically as part of the Pull Request process.
- Please read [the support policy](SUPPORT_POLICY.md) and [the documentation](https://docs.espressif.com/projects/esp-idf/en/latest/esp32/versions.html) for more information about ESP-IDF versions.
- Please see the [End-of-Life Advisories](https://www.espressif.com/en/support/documents/advisories?keys=&field_type_of_advisory_tid%5B%5D=817) for information about ESP-IDF releases with discontinued support.
# ESP-IDF Release and SoC Compatibility
The following table shows ESP-IDF support of Espressif SoCs where ![alt text][preview] and ![alt text][supported] denote preview status and support, respectively. The preview support is usually limited in time and intended for beta versions of chips. Please use an ESP-IDF release where the desired SoC is already supported.
Espressif SoCs released before 2016 (ESP8266 and ESP8285) are supported by [RTOS SDK](https://github.com/espressif/ESP8266_RTOS_SDK) instead.
ESP-IDF is the development framework for Espressif SoCs (released after 2016<sup>[1](#fn1)</sup>) provided for Windows, Linux and macOS.
# Developing With ESP-IDF
@@ -39,9 +14,11 @@ See https://idf.espressif.com/ for links to detailed instructions on how to set
### Non-GitHub forks
ESP-IDF uses relative locations as its submodules URLs ([.gitmodules](.gitmodules)). So they link to GitHub. If ESP-IDF is forked to a Git repository which is not on GitHub, you will need to run the script [tools/set-submodules-to-github.sh](tools/set-submodules-to-github.sh) after git clone.
The script sets absolute URLs for all submodules, allowing `git submodule update --init --recursive` to complete. If cloning ESP-IDF from GitHub, this step is not needed.
ESP-IDF uses relative locations as its submodules URLs ([.gitmodules](.gitmodules)). So they link to GitHub.
If ESP-IDF is forked to a Git repository which is not on GitHub, you will need to run the script
[tools/set-submodules-to-github.sh](tools/set-submodules-to-github.sh) after git clone.
The script sets absolute URLs for all submodules, allowing `git submodule update --init --recursive` to complete.
If cloning ESP-IDF from GitHub, this step is not needed.
## Finding a Project
@@ -109,9 +86,9 @@ After the initial flash, you may just want to build and flash just your app, not
## Erasing Flash
The `idf.py flash` target does not erase the entire flash contents. However it is sometimes useful to set the device back to a totally erased state, particularly when making partition table changes or OTA app updates. To erase the entire flash, run `idf.py erase-flash`.
The `idf.py flash` target does not erase the entire flash contents. However it is sometimes useful to set the device back to a totally erased state, particularly when making partition table changes or OTA app updates. To erase the entire flash, run `idf.py erase_flash`.
This can be combined with other targets, ie `idf.py -p PORT erase-flash flash` will erase everything and then re-flash the new app, bootloader and partition table.
This can be combined with other targets, ie `idf.py -p PORT erase_flash flash` will erase everything and then re-flash the new app, bootloader and partition table.
# Resources
@@ -122,3 +99,8 @@ This can be combined with other targets, ie `idf.py -p PORT erase-flash flash` w
* [Check the Issues section on github](https://github.com/espressif/esp-idf/issues) if you find a bug or have a feature request. Please check existing Issues before opening a new one.
* If you're interested in contributing to ESP-IDF, please check the [Contributions Guide](https://docs.espressif.com/projects/esp-idf/en/latest/contribute/index.html).
________
<a name="fn1">1</a>: ESP8266 and ESP8285 are not supported in ESP-IDF. See [RTOS SDK](https://github.com/espressif/ESP8266_RTOS_SDK) instead.
Please refer to https://docs.espressif.com/projects/esp-idf/en/latest/esp32/versions.html#support-periods for more details on ESP-IDF supported versions and support period policy.
## Reporting a Vulnerability
If you think you have found a security vulnerability in Espressif solutions (including ESP-IDF), then please send an email to our Bug Bounty team at bugbounty@espressif.com. Please do **NOT** create a public GitHub issue.
Each ESP-IDF major and minor release (V4.1, V4.2, etc) is supported for 30 months after the initial stable release date.
Each ESP-IDF major and minor release (V4.0, V4.1, etc) is supported for
18 months after the initial stable release date.
Supported means that the ESP-IDF team will continue to apply bug fixes, security fixes, etc to the release branch on GitHub, and periodically make new bugfix releases as needed.
Supported means that the ESP-IDF team will continue to apply bug fixes,
security fixes, etc to the release branch on GitHub, and periodically
make new bugfix releases as needed.
Support period is divided into "Service" and "Maintenance" period:
Users are encouraged to upgrade to a newer ESP-IDF release before the
support period finishes and the release becomes End of Life (EOL). It is
our policy to not continue fixing bugs in End of Life releases.
| Period | Duration | Recommended for new projects? |
Pre-release versions (betas, previews, `-rc` and `-dev` versions, etc)
are not covered by any support period. Sometimes a particular feature is
marked as \"Preview\" in a release, which means it is also not covered
by the support period.
During the Service period, bugfixes releases are more frequent. In some cases, support for new features may be added during the Service period (this is reserved for features which are needed to meet particular regulatory requirements or standards for new products, and which carry a very low risk of introducing regressions.)
The ESP-IDF Programming Guide has information about the
[different versions of ESP-IDF](https://docs.espressif.com/projects/esp-idf/en/latest/versions.html)
(major, minor, bugfix, etc).
During the Maintenance period, the version is still supported but only bugfixes for high severity issues or security issues will be applied.
Long Term Support releases
--------------------------
Using an “In Service” version is recommended when starting a new project.
Some releases (starting with ESP-IDF V3.3) are designated Long Term
Support (LTS). LTS releases are supported for 30 months (2.5 years)
after the initial stable release date.
Users are encouraged to upgrade all projects to a newer ESP-IDF release before the support period finishes and the release becomes End of Life (EOL). It is our policy to not continue fixing bugs in End of Life releases.
Pre-release versions (betas, previews, `-rc` and `-dev` versions, etc) are not covered by any support period. Sometimes a particular feature is marked as "Preview" in a release, which means it is also not covered by the support period.
ESP-IDF should be used in an up-to-date software environment. The operating system and other third-party tools should be supported by their maintainers. ESP-IDF cannot keep compatibility with unsupported third-party tools.
The ESP-IDF Programming Guide has information about the [different versions of ESP-IDF](https://docs.espressif.com/projects/esp-idf/en/latest/versions.html) (major, minor, bugfix, etc).
A new LTS release will be made at least every 18 months. This means
there will always be a period of at least 12 months to upgrade from the
previous LTS release to the following LTS release.
Example
-------
ESP-IDF V3.3 was released in September 2019. It was supported for 30 months until February 2022.
ESP-IDF V3.3 was released in September 2019 and is a Long Term Support
(LTS) release, meaning it will be supported for 30 months until February
2022.
- The first V3.3 release was `v3.3` in September 2019.
- The ESP-IDF team continues to backport bug fixes, security fixes, etc to the release branch `release/v3.3`.
- Periodically stable bugfix releases are created from the release branch. For example `v3.3.1`, `v3.3.2`, etc. Users are encouraged to always update to the latest bugfix release.
-V3.3 bugfix releases continue until February 2022, when all V3.3.x releases become End of Life.
- The ESP-IDF team continues to backport bug fixes, security fixes,
etc to the release branch `release/v3.3`.
-Periodically stable bugfix releases are created from the release
branch. For example `v3.3.1`, `v3.3.2`, etc. Users are encouraged to
always update to the latest bugfix release.
- V3.3 bugfix releases continue until February 2022, when all V3.3.x
releases become End of Life.
Existing Releases
-----------------
ESP-IDF release V4.1 and all newer releases will follow this support period policy. The support period for each release will be announced when the release is made.
ESP-IDF release V3.3 and all newer releases will follow this support
period policy. The support period for each release will be announced
when the release is made.
For releases made before the current support period policy was announced, the original support periods apply:
For releases made before the support period policy was announced,
the following support periods apply:
* ESP-IDF V4.0.x will be supported until October 2021
* ESP-IDF V3.3.x will be supported until February 2022
* ESP-IDF versions before V3.3 are already End of Life.
Policy History
--------------
* September 2019. This policy splits ESP-IDF releases into Standard and Long Term Support.
* July 2020. All releases from now will have the same support period, which is equal to the previous Long Term Support period. Added “In Service” period, during which versions will receive more updates.
- ESP-IDF V3.1.x and V3.2.x will both be supported until October 2020.
- ESP-IDF V3.0.9 (planned for October 2019) will be the last V3.0
bugfix release. ESP-IDF V3.0.x is End of Life from October 2019.
- ESP-IDF versions before V3.0 are already End of Life.
This document contains details about what the core components are, what they contain, and how they are organized.
## Organization
The core components are organized into two groups.
The first group (referred to as `G0` from now on) contains `hal`, `xtensa` and `riscv` (referred to as `arch` components from now on), `esp_rom`, `esp_common`, and `soc`. This
group contain information about and low-level access to underlying hardware; or in the case of `esp_common`, hardware-agnostic code and utilities.
These components can depend on each other, but as much as possible have no dependencies outside the group. The reason for this is that, due to the
nature of what these components contain, the likelihood is high that a lot of other components will require these. Ideally, then, the dependency
relationship only goes one way. This makes it easier for these components, as a group, to be usable in another project. One can conceivably implement
a competing SDK to ESP-IDF on top of these components.
The second group (referred to as `G1` from now on) sits at a higher level than the first group. This group contains the components `esp_hw_support`, `esp_system`, `newlib`, `spi_flash`,
`freertos`, `log`, and `heap`. Like the first group, circular dependencies within the group are allowed; and being at a higher level, dependency on the first group
is allowed. These components represent software mechanisms essential to building other components.
## Descriptions
The following is a short description of the components mentioned above.
### `G0` Components
#### `hal`
Contains the hardware abstraction layer and low-level operation implementations for the various peripherals. The low-level functions assign meaningful names to register-level manipulations; the hardware abstraction provide operations one level above this, grouping these low-level functions
into routines that achieve a meaningful action or state of the peripheral.
Example:
-`spi_flash_ll_set_address` is a low-level function part of the hardware abstraction `spi_flash_hal_read_block`
#### `arch`
Contains low-level architecture operations and definitions, including those for customizations (can be thought of on the same level as the low-level functions of `hal`).
This can also contain files provided by the architecture vendor.
Example:
-`xt_set_exception_handler`
-`rv_utils_intr_enable`
-`ERI_PERFMON_MAX`
#### `esp_common`
Contains hardware-agnostic definitions, constants, macros, utilities, 'pure' and/or algorithmic functions that is useable by all other components (that is, barring there being a more appropriate component to put them in).
Example:
-`BIT(nr)` and other bit manipulation utilities in the future
-`IDF_DEPRECATED(REASON)`
-`ESP_IDF_VERSION_MAJOR`
#### `soc`
Contains description of the underlying hardware: register structure, addresses, pins, capabilities, etc.
Example:
-`DR_REG_DPORT_BASE`
-`SOC_MCPWM_SUPPORTED`
-`uart_dev_s`
#### `esp_rom`
Contains headers, linker scripts, abstraction layer, patches, and other related files to ROM functions.
Example:
-`esp32.rom.eco3.ld`
-`rom/aes.h`
### `G1` Components
#### `spi_flash`
SPI flash device access implementation.
#### `freertos`
FreeRTOS port to targets supported by ESP-IDF.
#### `log`
Logging library.
#### `heap`
Heap implementation.
#### `newlib`
Some functions n the standard library are implemented here, especially those needing other `G1` components.
Example:
-`malloc` is implemented in terms of the component `heap`'s functions
-`gettimeofday` is implemented in terms of system time in `esp_system`
#### `esp_system`
Contains implementation of system services and controls system behavior. The implementations
here may take hardware resources and/or decide on a hardware state needed for support of a system service/feature/mechanism.
Currently, this encompasses the following, but not limited to:
- Startup and initialization
- Panic and debug
- Reset and reset reason
- Task and interrupt watchdogs
#### `esp_hw_support`
Contains implementations that provide hardware operations, arbitration, or resource sharing, especially those that
is used in the system. Unlike `esp_system`, implementations here do not decide on a hardware state or takes hardware resource, acting
merely as facilitator to hardware access. Currently, this encompasses the following, but not limited to:
// In general tracing goes in the following way. User application requests tracing module to send some data by calling esp_apptrace_buffer_get(),
// module allocates necessary buffer in current input trace block. Then user fills received buffer with data and calls esp_apptrace_buffer_put().
// When current input trace block is filled with app data it is exposed to host and the second block becomes input one and buffer filling restarts.
// While target application fills one TRAX block host reads another one via JTAG.
// This module also allows communication in the opposite direction: from host to target. As it was said ESP32 and host can access different TRAX blocks
// simultaneously, so while target writes trace data to one block host can write its own data (e.g. tracing commands) to another one then when
// blocks are switched host receives trace data and target receives data written by host application. Target user application can read host data
// by calling esp_apptrace_read() API.
// To control buffer switching and for other communication purposes this implementation uses some TRAX registers. It is safe since HW TRAX tracing
// can not be used along with application tracing feature so these registers are freely readable/writeable via JTAG from host and via ERI from ESP32 cores.
// Overhead of this implementation on target CPU is produced only by allocating/managing buffers and copying of data.
// On the host side special OpenOCD command must be used to read trace data.
// 2. TRAX Registers layout
// ========================
// This module uses two TRAX HW registers to communicate with host SW (OpenOCD).
// - Control register uses TRAX_DELAYCNT as storage. Only lower 24 bits of TRAX_DELAYCNT are writable. Control register has the following bitfields:
// 14..0 bits - actual length of user data in trace memory block. Target updates it every time it fills memory block and exposes it to host.
// Host writes zero to this field when it finishes reading exposed block;
// 21..15 bits - trace memory block transfer ID. Block counter. It can overflow. Updated by target, host should not modify it. Actually can be 2 bits;
// 22 bit - 'host data present' flag. If set to one there is data from host, otherwise - no host data;
// 23 bit - 'host connected' flag. If zero then host is not connected and tracing module works in post-mortem mode, otherwise in streaming mode;
// - Status register uses TRAX_TRIGGERPC as storage. If this register is not zero then current CPU is changing TRAX registers and
// this register holds address of the instruction which application will execute when it finishes with those registers modifications.
// See 'Targets Connection' setion for details.
// 3. Modes of operation
// =====================
// This module supports two modes of operation:
// - Post-mortem mode. This is the default mode. In this mode application tracing module does not check whether host has read all the data from block
// exposed to it and switches block in any case. The mode does not need host interaction for operation and so can be useful when only the latest
// trace data are necessary, e.g. for analyzing crashes. On panic the latest data from current input block are exposed to host and host can read them.
// It can happen that system panic occurs when there are very small amount of data which are not exposed to host yet (e.g. crash just after the
// TRAX block switch). In this case the previous 16KB of collected data will be dropped and host will see the latest, but very small piece of trace.
// It can be insufficient to diagnose the problem. To avoid such situations there is menuconfig option
// CONFIG_APPTRACE_POSTMORTEM_FLUSH_THRESH
// which controls the threshold for flushing data in case of panic.
// - Streaming mode. Tracing module enters this mode when host connects to target and sets respective bits in control registers (per core).
// In this mode before switching the block tracing module waits for the host to read all the data from the previously exposed block.
// On panic tracing module also waits (timeout is configured via menuconfig via CONFIG_APPTRACE_ONPANIC_HOST_FLUSH_TMO) for the host to read all data.
// 4. Communication Protocol
// =========================
// 4.1 Trace Memory Blocks
// -----------------------
// Communication is controlled via special register. Host periodically polls control register on each core to find out if there are any data available.
// When current input memory block is filled it is exposed to host and 'block_len' and 'block_id' fields are updated in the control register.
// Host reads new register value and according to it's value starts reading data from exposed block. Meanwhile target starts filling another trace block.
// When host finishes reading the block it clears 'block_len' field in control register indicating to the target that it is ready to accept the next one.
// If the host has some data to transfer to the target it writes them to trace memory block before clearing 'block_len' field. Then it sets
// 'host_data_present' bit and clears 'block_len' field in control register. Upon every block switch target checks 'host_data_present' bit and if it is set
// reads them to down buffer before writing any trace data to switched TRAX block.
// 4.2 User Data Chunks Level
// --------------------------
// Since trace memory block is shared between user data chunks and data copying is performed on behalf of the API user (in its normal context) in
// multithreading environment it can happen that task/ISR which copies data is preempted by another high prio task/ISR. So it is possible situation
// that task/ISR will fail to complete filling its data chunk before the whole trace block is exposed to the host. To handle such conditions tracing
// module prepends all user data chunks with header which contains allocated buffer size and actual data length within it. OpenOCD command
// which reads application traces reports error when it reads incomplete user data block.
// Data which are transffered from host to target are also prepended with a header. Down channel data header is simple and consists of one two bytes field
// containing length of host data following the header.
// 4.3 Data Buffering
// ------------------
// It takes some time for the host to read TRAX memory block via JTAG. In streaming mode it can happen that target has filled its TRAX block, but host
// has not completed reading of the previous one yet. So in this case time critical tracing calls (which can not be delayed for too long time due to
// the lack of free memory in TRAX block) can be dropped. To avoid such scenarios tracing module implements data buffering. Buffered data will be sent
// to the host later when TRAX block switch occurs. The maximum size of the buffered data is controlled by menuconfig option
// CONFIG_APPTRACE_PENDING_DATA_SIZE_MAX.
// 4.4 Target Connection/Disconnection
// -----------------------------------
// When host is going to start tracing in streaming mode it needs to put both ESP32 cores into initial state when 'host connected' bit is set
// on both cores. To accomplish this host halts both cores and sets this bit in TRAX registers. But target code can be halted in state when it has read control
// register but has not updated its value. To handle such situations target code indicates to the host that it is updating control register by writing
// non-zero value to status register. Actually it writes address of the instruction which it will execute when it finishes with
// the registers update. When target is halted during control register update host sets breakpoint at the address from status register and resumes CPU.
// After target code finishes with register update it is halted on breakpoint, host detects it and safely sets 'host connected' bit. When both cores
// are set up they are resumed. Tracing starts without further intrusion into CPUs work.
// When host is going to stop tracing in streaming mode it needs to disconnect targets. Disconnection process is done using the same algorithm
// as for connecting, but 'host connected' bits are cleared on ESP32 cores.
// 5. Module Access Synchronization
// ================================
// Access to internal module's data is synchronized with custom mutex. Mutex is a wrapper for portMUX_TYPE and uses almost the same sync mechanism as in
// vPortCPUAcquireMutex/vPortCPUReleaseMutex. The mechanism uses S32C1I Xtensa instruction to implement exclusive access to module's data from tasks and
// ISRs running on both cores. Also custom mutex allows specifying timeout for locking operation. Locking routine checks underlaying mutex in cycle until
// it gets its ownership or timeout expires. The differences of application tracing module's mutex implementation from vPortCPUAcquireMutex/vPortCPUReleaseMutex are:
// - Support for timeouts.
// - Local IRQs for CPU which owns the mutex are disabled till the call to unlocking routine. This is made to avoid possible task's prio inversion.
// When low prio task takes mutex and enables local IRQs gets preempted by high prio task which in its turn can try to acquire mutex using infinite timeout.
// So no local task switch occurs when mutex is locked. But this does not apply to tasks on another CPU.
// WARNING: Priority inversion can happen when low prio task works on one CPU and medium and high prio tasks work on another.
// WARNING: Care must be taken when selecting timeout values for trace calls from ISRs. Tracing module does not care about watchdogs when waiting
// on internal locks and for host to complete previous block reading, so if timeout value exceeds watchdog's one it can lead to the system reboot.
// 6. Timeouts
// ===========
// Timeout mechanism is based on xthal_get_ccount() routine and supports timeout values in microseconds.
// There are two situations when task/ISR can be delayed by tracing API call. Timeout mechanism takes into account both conditions:
// - Trace data are locked by another task/ISR. When wating on trace data lock.
// - Current TRAX memory input block is full when working in streaming mode (host is connected). When waiting for host to complete previous block reading.
// When wating for any of above conditions xthal_get_ccount() is called periodically to calculate time elapsed from trace API routine entry. When elapsed
// time exceeds specified timeout value operation is canceled and ESP_ERR_TIMEOUT code is returned.
* SPDX-FileCopyrightText: 2015-2023 Espressif Systems (Shanghai) CO LTD
*
* SPDX-License-Identifier: Apache-2.0
*/
// Copyright 2015-2016 Espressif Systems (Shanghai) PTE LTD
//
// Licensed under the Apache License, Version 2.0 (the "License");
// you may not use this file except in compliance with the License.
// You may obtain a copy of the License at
// http://www.apache.org/licenses/LICENSE-2.0
//
// Unless required by applicable law or agreed to in writing, software
// distributed under the License is distributed on an "AS IS" BASIS,
// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
// See the License for the specific language governing permissions and
// limitations under the License.
#ifndef _OTA_OPS_H
#define _OTA_OPS_H
@@ -12,9 +20,8 @@
#include<stddef.h>
#include"esp_err.h"
#include"esp_partition.h"
#include"esp_app_desc.h"
#include"esp_image_format.h"
#include"esp_flash_partitions.h"
#include"soc/soc_caps.h"
#ifdef __cplusplus
extern"C"
@@ -22,7 +29,6 @@ extern "C"
#endif
#define OTA_SIZE_UNKNOWN 0xffffffff /*!< Used for esp_ota_begin() if new image size is unknown */
#define OTA_WITH_SEQUENTIAL_WRITES 0xfffffffe /*!< Used for esp_ota_begin() if new image size is unknown and erase can be done in incremental manner (assuming write operation is in continuous sequence) */
#define ESP_ERR_OTA_BASE 0x1500 /*!< Base error code for ota_ops api */
#define ESP_ERR_OTA_PARTITION_CONFLICT (ESP_ERR_OTA_BASE + 0x01) /*!< Error if request was to write or erase the current running partition */
* @brief Revokes the signature digest denoted by the given index. This should be called in the application only after the rollback logic otherwise the device may end up in unrecoverable state.
* @brief Revokes the old signature digest. To be called in the application after the rollback logic.
*
* Relevant for Secure boot v2 on ESP32-S2, ESP32-S3, ESP32-C3, ESP32-C6, ESP32-H2 where upto 3 key digests can be stored (Key \#N-1, Key \#N, Key \#N+1).
* When a key used to sign an app is invalidated, an OTA update is to be sent with an app signed with at least one of the other two keys which has not been revoked already.
* After successfully booting the OTA app should call this function to revoke Key \#N-1.
* Relevant for Secure boot v2 on ESP32-S2 where upto 3 key digests can be stored (Key N-1, Key N, Key N+1).
* When key N-1 used to sign an app is invalidated, an OTA update is to be sent with an app signed with key N-1 & Key N.
* After successfully booting the OTA app should call this function to revoke Key N-1.
*
* @param index - The index of the signature block to be revoked
*
@@ -346,7 +325,7 @@ typedef enum {
* - ESP_FAIL: If secure boot v2 has not been enabled.
// 2 Stage: run factory -> check it -> copy factory to OTA0 -> reboot --//--
// 3 Stage: run OTA0 -> check it -> erase OTA_DATA for next tests -> PASS
TEST_CASE_MULTIPLE_STAGES("Switching between factory, OTA0 using esp_ota_write_with_offset","[app_update][timeout=90][reset=DEEPSLEEP_RESET, DEEPSLEEP_RESET]",start_test,test_flow6,test_flow6);
//IDF-5145
TEST_CASE("Test bootloader_common_get_sha256_of_partition returns ESP_ERR_IMAGE_INVALID when image is ivalid","[partitions]")
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
