RTEMS / RTOS / RTEMS¶
Go to Issues or Merge Requests
Issues Summary
Merge Requests Summary
Issues¶
5281 - RTEMS 6.1 GR740 Multiprocessing build fails¶
Id |
5281 |
State |
closed |
Type |
ISSUE |
Author |
Gedare Bloom |
Assignee(s) |
Matteo Concas |
Closed by |
Matteo Concas |
Created |
2025-06-24T20:26:51.026Z |
Closed |
2025-07-04T06:49:45.863Z |
Updated |
2025-07-04T06:49:45.876Z |
Milestone |
6.2 |
Link |
|
Merges |
0 |
Summary¶
When building the GR740 BSP with RTEMS_MULTIPROCESSING set to true , the build will fail.
Steps to reproduce¶
The INI file contents are:
[DEFAULT]
RTEMS_POSIX_API = True
[sparc/gr740_mp]
INHERIT=gr740
RTEMS_MULTIPROCESSING=True
I configure the Waf system with:
./waf configure --prefix=$RTEMS_PREFIX --rtems-config=gr740.ini
And then ./waf to build the BSP. I get the following output:
Waf: Entering directory `/scratch/matteo/rtems-build/src/rtems-6.1/build'
Waf: Leaving directory `/scratch/matteo/rtems-build/src/rtems-6.1/build'
'build' finished successfully (0.081s)
Waf: Entering directory `/scratch/matteo/rtems-build/src/rtems-6.1/build/sparc/gr740_mp'
[ 167/1641] Compiling bsps/sparc/leon3/clock/ckinit.c
In file included from ../../../bsps/sparc/leon3/clock/ckinit.c:46:
../../../bsps/sparc/leon3/clock/ckinit.c: In function 'leon3_clock_initialize':
../../../bsps/sparc/leon3/include/bsp/leon3.h:221:5: error: 'leon3_timer_core_index' undeclared (first use in this function); did you mean 'leon3_timecounter'?
221 | ( leon3_timer_core_index != 0 ? 0 : 2 * LEON3_Cpu_Index )
| ^~~~~~~~~~~~~~~~~~~~~~
../../../bsps/sparc/leon3/clock/ckinit.c:181:36: note: in expansion of macro 'LEON3_CLOCK_INDEX'
181 | timer = &LEON3_Timer_Regs->timer[LEON3_CLOCK_INDEX];
| ^~~~~~~~~~~~~~~~~
../../../bsps/sparc/leon3/include/bsp/leon3.h:221:5: note: each undeclared identifier is reported only once for each function it appears in
221 | ( leon3_timer_core_index != 0 ? 0 : 2 * LEON3_Cpu_Index )
| ^~~~~~~~~~~~~~~~~~~~~~
../../../bsps/sparc/leon3/clock/ckinit.c:181:36: note: in expansion of macro 'LEON3_CLOCK_INDEX'
181 | timer = &LEON3_Timer_Regs->timer[LEON3_CLOCK_INDEX];
| ^~~~~~~~~~~~~~~~~
../../../bsps/sparc/leon3/clock/../../../shared/dev/clock/clockimpl.h: In function '_Clock_Initialize':
../../../bsps/sparc/leon3/include/bsp/leon3.h:221:5: error: 'leon3_timer_core_index' undeclared (first use in this function); did you mean 'leon3_timecounter'?
221 | ( leon3_timer_core_index != 0 ? 0 : 2 * LEON3_Cpu_Index )
| ^~~~~~~~~~~~~~~~~~~~~~
../../../bsps/sparc/leon3/clock/ckinit.c:137:49: note: in expansion of macro 'LEON3_CLOCK_INDEX'
137 | #define Adjust_clkirq_for_node() do { clkirq += LEON3_CLOCK_INDEX; } while(0)
| ^~~~~~~~~~~~~~~~~
../../../bsps/sparc/leon3/clock/ckinit.c:145:7: note: in expansion of macro 'Adjust_clkirq_for_node'
145 | Adjust_clkirq_for_node(); \\
| ^~~~~~~~~~~~~~~~~~~~~~
../../../bsps/sparc/leon3/clock/../../../shared/dev/clock/clockimpl.h:249:3: note: in expansion of macro 'Clock_driver_support_find_timer'
249 | Clock_driver_support_find_timer();
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Waf: Leaving directory `/scratch/matteo/rtems-build/src/rtems-6.1/build/sparc/gr740_mp'
Build failed
-> task in '/bsps/sparc/leon3/obj' failed with exit status 1 (run with -v to display more information)
Expected behaviour¶
Build should complete successfully.
Possible fix¶
From my own investigation this comes from bsps/sparc/leon3/include/bsp/leon3.h . leon3_timer_core_index is only declared if LEON3_GPTIMER_BASE is not defined but later in the file the LEON3_CLOCK_INDEX macro is defined in the following way:
/**
* @brief This constant defines the index of the GPTIMER timer used by the
* clock driver.
*/
#if defined(RTEMS_MULTIPROCESSING)
#define LEON3_CLOCK_INDEX \\
( leon3_timer_core_index != 0 ? 0 : 2 * LEON3_Cpu_Index )
#else
#define LEON3_CLOCK_INDEX 0
#endif
So when RTEMS_MULTIPROCESSING is used the LEON3_CLOCK_INDEX will expand to use the leon3_timer_core_index variable which won’t be declared as the GPTIMER base is defined for the GR740 BSP.
I don’t fully understand why the LEON3_CLOCK_INDEX macro is defined this way but changing it’s define conditional to:
#if defined(RTEMS_MULTIPROCESSING) && !defined(LEON3_GPTIMER_BASE)
Will make the BSP build successfully, I haven’t had the time to test this fix, it might break something else at runtime.
Author: Gedare Bloom
2025-06-24T20:26:51.204Z
cloned from #5258
Author: Gedare Bloom
2025-06-24T20:28:54.720Z
Author: Gedare Bloom
2025-06-24T20:30:26.097Z
@matteo.concas
Author: Gedare Bloom
2025-06-24T20:30:46.754Z
assigned to @matteo.concas
Author: Gedare Bloom
2025-06-24T20:31:37.061Z
mentioned in merge request !532
5288 - Backport RISC-V Double Trap extension fix to 6¶
Id |
5288 |
State |
closed |
Type |
ISSUE |
Author |
Matteo Concas |
Closed by |
Matteo Concas |
Created |
2025-07-03T08:07:26.897Z |
Closed |
2025-07-04T06:48:43.417Z |
Updated |
2025-07-04T06:48:43.431Z |
Milestone |
6.2 |
Labels |
arch:riscv |
Link |
|
Merges |
0 |
If the Double Trap Extension is implemented, the MDT bit of the mstatus (or mstatush in RV32) register will be set when a trap is to be taken.
The MIE (Machine Interrupt Enable) bit can only be set to 1 if the MDT bit is zero.
Thus, we need to clear MDT first if we want to enable interrupts when dispatching a thread.
MDT is also cleared in register a1 before restoring the interrupt frame as writing 1 to MDT will cause MIE to be set to 0. In RV64 this happens regardless of the value written to MIE in the same write.
In RV32, MDT is in mstatush so we do not need to clear during restore as this register is not restored.
With this change all 60 SMP tests pass (compared to 20/60 before the fix). The tests have been run on hardware using two RV64 CPUs that implement the double trap extension.
Author: Matteo Concas
2025-07-03T08:10:45.328Z
mentioned in merge request !548
Author: Kinsey Moore
2025-07-03T13:10:32.134Z
added rtems#10 as parent epic
5283 - Backport 6.2: bsps/noelv: Fix using console in polled mode¶
Id |
5283 |
State |
closed |
Type |
ISSUE |
Author |
Matteo Concas |
Closed by |
Gedare Bloom |
Created |
2025-06-27T07:40:58.098Z |
Closed |
2025-06-28T15:36:17.903Z |
Updated |
2025-06-28T15:36:17.916Z |
Milestone |
6.2 |
Link |
|
Merges |
0 |
Summary¶
bsps/noelv: Fix using console in polled mode
The console driver needs BSP_CONSOLE_USE_INTERRUPTS to be defined or it does not build. The intent was to use polled mode if the macro was equal to zero but not using interrupt means the macro won’t be defined instead.
This is the faulty code:
if (BSP_CONSOLE_USE_INTERRUPTS) {
handler = &apbuart_handler_interrupt;
}
In bsps/riscv/noel/console/console-config.c .
Steps to reproduce¶
Create a INI with the following contents:
[riscv/noel64imafd_polling]
INHERIT = noel64imafd
RTEMS_SMP = True
BSP_CONSOLE_USE_INTERRUPTS = 0
Build the RTEMS kernel, it will fail to build due to BSP_CONSOLE_USE_INTERRUPTS being undefined.
Expected behaviour¶
BSP should build successfully.
Author: Matteo Concas
2025-06-27T07:43:21.023Z
mentioned in merge request !537
Author: Kinsey Moore
2025-06-27T12:59:09.680Z
Author: Matteo Concas
2025-06-28T15:35:42.404Z
mentioned in commit dd490c5c2d02e472c83b7052612b554e8b74358f
Author: Gedare Bloom
2025-06-28T15:36:18.541Z
fixed in https://gitlab.rtems.org/rtems/rtos/rtems/-/commit/dd490c5c2d02e472c83b7052612b554e8b74358f
5254 - i386/pc BSP ATA driver asserts with node errors on FAT disk mount¶
Id |
5254 |
State |
closed |
Type |
ISSUE |
Author |
Chris Johns |
Assignee(s) |
Chris Johns |
Closed by |
Kinsey Moore |
Created |
2025-05-28T02:54:29.731Z |
Closed |
2025-05-30T13:34:36.579Z |
Updated |
2025-05-30T13:34:36.593Z |
Milestone |
6.2 |
Labels |
arch:i386 |
Link |
|
Merges |
0 |
Summary¶
With RTEMS_DEBUG enabled the PC ATA driver asserts with a failure because a node’s link pointer is not set to NULL. The assert is:
assertion "_Chain_Is_node_off_chain( the_node )" failed: file "../../../cpukit/include/rtems/score/chainimpl.h", line 691, function: _Chain_Append_unprotected
A back trace is:
(gdb) bt
#0 0x00162c55 in bsp_reset (source=source@entry=RTEMS_FATAL_SOURCE_ASSERT, code=code@entry=1872624) at ../../../bsps/i386/pc386/start/bspreset.c:39
#1 0x00161b15 in bsp_fatal_extension (source=RTEMS_FATAL_SOURCE_ASSERT, always_set_to_false=false, code=1872624) at ../../../bsps/shared/start/bspfatal-default.c:213
#2 0x001189b3 in _User_extensions_Iterate (arg=arg@entry=0x1c92b8, visitor=0x118900 <_User_extensions_Fatal_visitor>, direction=CHAIN_ITERATOR_FORWARD) at ../../../cpukit/score/src/userextiterate.c:194
#3 0x0011321f in _User_extensions_Fatal (error=1872624, source=RTEMS_FATAL_SOURCE_ASSERT) at ../../../cpukit/include/rtems/score/userextimpl.h:467
#4 _Terminate (the_source=the_source@entry=RTEMS_FATAL_SOURCE_ASSERT, the_error=the_error@entry=1872624) at ../../../cpukit/score/src/interr.c:66
#5 0x001077ec in rtems_fatal (fatal_code=1872624, fatal_source=RTEMS_FATAL_SOURCE_ASSERT) at ../../../cpukit/include/rtems/fatal.h:160
#6 __assert_func (file=file@entry=0x189a64 "../../../cpukit/include/rtems/score/chainimpl.h", line=691, func=0x1a0378 <__func__.0> "_Chain_Append_unprotected", failedexpr=0x189a3c "_Chain_Is_node_off_chain( the_node )") at ../../../cpukit/libcsupport/src/__assert.c:71
#7 0x0016c03d in _Chain_Append_unprotected (the_node=0x1cbbc0, the_chain=0x1aae4c <ata_ide_ctrls+44>) at ../../../cpukit/include/rtems/score/chainimpl.h:691
#8 0x0016c0dc in _Chain_Append_unprotected (the_node=0x1cbbc0, the_chain=0x1aae4c <ata_ide_ctrls+44>) at ../../../cpukit/sapi/src/chainprotected.c:99
#9 rtems_chain_append (chain=chain@entry=0x1aae4c <ata_ide_ctrls+44>, node=0x1cbbc0) at ../../../cpukit/sapi/src/chainprotected.c:97
#10 0x00160cb9 in ata_add_to_controller_queue (ctrl_minor=0, areq=0x1cbbc0) at ../../../bsps/shared/dev/ide/ata.c:540
#11 0x00160ea1 in ata_io_data_request (ata_dev=<optimized out>, req=0x1c9450) at ../../../bsps/shared/dev/ide/ata.c:228
#12 ata_ioctl (dd=0x1bee44, cmd=3222815233, argp=0x1c9450) at ../../../bsps/shared/dev/ide/ata.c:949
#13 0x00102f5e in rtems_bdbuf_execute_transfer_request (dd=dd@entry=0x1bee44, req=req@entry=0x1c9450, cache_locked=cache_locked@entry=true) at ../../../cpukit/libblock/src/bdbuf.c:1878
#14 0x001032c9 in rtems_bdbuf_execute_read_request (dd=dd@entry=0x1bee44, bd=<optimized out>, bd@entry=0x1bbe2c, transfer_count=transfer_count@entry=1) at ../../../cpukit/libblock/src/bdbuf.c:1994
#15 0x001044df in rtems_bdbuf_read (dd=dd@entry=0x1bee44, block=block@entry=0, bd_ptr=bd_ptr@entry=0x1c9558) at ../../../cpukit/libblock/src/bdbuf.c:2094
#16 0x00131f67 in rtems_bdpart_read_record (dd=0x1bee44, index=0, block=0x1c9558) at ../../../cpukit/libblock/src/bdpart-read.c:220
#17 0x00132190 in rtems_bdpart_read (disk_name=disk_name@entry=0x1ca62e "/dev/hda", format=format@entry=0x1c95dc, pt=pt@entry=0x1c9630, count=count@entry=0x1c95d4) at ../../../cpukit/libblock/src/bdpart-read.c:273
#18 0x00119a5e in rtems_bdpart_shell_main (argc=2, argv=0x1c9b30) at ../../../cpukit/libmisc/shell/fdisk.c:236
#19 0x0012dacd in shell_main_loop (shell_env=shell_env@entry=0x1c9ec8, interactive=interactive@entry=true, line_editor_output=<optimized out>) at ../../../cpukit/libmisc/shell/shell.c:1215
#20 0x0012ed18 in rtems_shell_main_loop (shell_env=shell_env@entry=0x1c9ec8) at ../../../cpukit/libmisc/shell/shell.c:1383
#21 0x0012ed38 in rtems_shell_task (task_argument=1875656) at ../../../cpukit/libmisc/shell/shell.c:1038
#22 0x00115adb in _Thread_Handler () at ../../../cpukit/score/src/threadhandler.c:164
#23 0xa5a5a5a5 in ?? ()
Steps to reproduce¶
INI file:
[DEFAULT]
RTEMS_DEBUG = True
RTEMS_POSIX_API = True
[i386/pc686]
BUILD_TESTS = True
Run the fileio.exe test, enter the shell and then enter the following command:
fdisk /dev/hda
The shell session is:
*** BEGIN OF TEST FILE I/O ***
*** TEST VERSION: 7.0.0.34c362fee100659e36a6945204f77e71cb1c2f7e
*** TEST STATE: USER_INPUT
*** TEST BUILD: RTEMS_DEBUG RTEMS_POSIX_API
*** TEST TOOLS: 13.3.0 20240521 (RTEMS 7, RSB 170160cb918e7c68b4b35d8f0a42db6df81168e7, Newlib 1b3dcfd)
Press any key to start file I/O sample (20s remaining)
=========================
RTEMS FILE I/O Test Menu
=========================
p -> part_table_initialize
f -> mount all disks in fs_table
l -> list file
r -> read file
w -> write file
s -> start shell
Enter your selection ==>s
Creating /etc/passwd and group with four useable accounts:
root/pwd
test/pwd
rtems/NO PASSWORD
chroot/NO PASSWORD
Only the root user has access to all available commands.
=========================
starting shell
=========================
Welcome to rtems-7.0.0 (Intel i386/Pentium/pc686)
Copyright (C) 1989, 2021 RTEMS Project and contributors
Login into RTEMS
/dev/foobar login: root
Password:
RTEMS Shell on /dev/foobar. Use 'help' to list commands.
SHLL [/] # fdisk /dev/hda
assertion "_Chain_Is_node_off_chain( the_node )" failed: file "../../../cpukit/include/rtems/score/chainimpl.h", line 691, function: _Chain_Append_unprotected
Pre-set options¶
Author: Chris Johns
2025-05-28T02:54:30.004Z
cloned from #5253
Author: Chris Johns
2025-05-28T02:54:43.589Z
assigned to @chris
Author: Chris Johns
2025-05-28T02:54:53.933Z
Author: Chris Johns
2025-05-28T02:56:00.050Z
changed title from i38{-p-}/pc BSP ATA driver asserts with node errors on FAT disk mount to i38{+6+}/pc BSP ATA driver asserts with node errors on FAT disk mount
Author: Chris Johns
2025-05-28T03:02:54.791Z
mentioned in merge request !492
Author: Chris Johns
2025-05-30T13:34:27.984Z
mentioned in commit cb3fba044770f0bb6a96bd2b93e09df90d1e2e34
5232 - Backport “spec: Install missing header files for GRLIB and SPARC” to 6 branch¶
Id |
5232 |
State |
closed |
Type |
ISSUE |
Author |
Kinsey Moore |
Assignee(s) |
Kinsey Moore |
Closed by |
Kinsey Moore |
Created |
2025-03-26T15:53:40.588Z |
Closed |
2025-03-26T16:50:40.530Z |
Updated |
2025-03-26T16:50:40.542Z |
Milestone |
6.2 |
Labels |
grlib |
Link |
|
Merges |
1 |
The following discussion from !458 should be addressed:
[] @joel started a discussion
Is this also being applied to the 6 branch?
Author: Kinsey Moore
2025-03-26T15:53:40.797Z
assigned to @opticron
Author: Kinsey Moore
2025-03-26T15:53:41.118Z
Author: Kinsey Moore
2025-03-26T15:54:02.665Z
Author: Kinsey Moore
2025-03-26T16:00:38.727Z
changed the description
Author: Matteo Concas
2025-03-26T16:48:52.616Z
mentioned in merge request !459
5173 - powerpc/qoriq: Avoid spurious interrupt fatal errors¶
Id |
5173 |
State |
closed |
Type |
ISSUE |
Author |
Sebastian Huber |
Assignee(s) |
Sebastian Huber |
Closed by |
Gedare Bloom |
Created |
2024-12-05T23:47:28.576Z |
Closed |
2025-02-07T23:22:10.064Z |
Updated |
2025-02-11T05:38:23.897Z |
Milestone |
6.2 |
Labels |
arch:powerpc |
Link |
|
Merges |
2 |
Summary¶
The interrupt dispatching routine erroneously invokes the fatal error handler if a spurious interrupt is acknowledged.
Pre-set options¶
Author: Sebastian Huber
2024-12-05T23:47:28.686Z
assigned to @sebhub
Author: Sebastian Huber
2024-12-05T23:47:36.803Z
cloned to #5174
Author: Sebastian Huber
2024-12-05T23:47:47.386Z
Author: Sebastian Huber
2024-12-06T10:22:32.171Z
mentioned in merge request !345
Author: Chris Johns
2024-12-22T03:38:40.991Z
Author: Gedare Bloom
2025-02-07T23:22:09.483Z
Fixed in e421c922a87842f0613a269054acd041e46c91cd
Author: Sebastian Huber
2025-02-11T05:38:23.477Z
mentioned in commit e421c922a87842f0613a269054acd041e46c91cd
5205 - Back port “grlib/occan: Fix baud rate calculation” to 6¶
Id |
5205 |
State |
closed |
Type |
ISSUE |
Author |
Joel Sherrill |
Closed by |
Gedare Bloom |
Created |
2025-02-07T15:07:07.522Z |
Closed |
2025-02-07T20:37:37.624Z |
Updated |
2025-02-11T05:38:23.820Z |
Milestone |
6.2 |
Link |
|
Merges |
2 |
The following discussion from !340 should be addressed:
[] @gedare started a discussion
@thesummer Should this be back-ported to 6?
Yes. @thesummer please use this issue to back port the fix to 6.
Author: Joel Sherrill
2025-02-07T15:07:26.994Z
Author: Gedare Bloom
2025-02-07T15:51:54.551Z
mentioned in merge request !404
Author: Jan Sommer
2025-02-11T05:38:23.247Z
mentioned in commit a0e4be53f4e30ee02d6d0c9221d613d18c72f426
5189 - Backport libdl ARM trampoline crash fix¶
Id |
5189 |
State |
closed |
Type |
ISSUE |
Author |
Kinsey Moore |
Assignee(s) |
Kinsey Moore |
Closed by |
Chris Johns |
Created |
2025-01-23T22:24:29.147Z |
Closed |
2025-01-24T00:10:34.134Z |
Updated |
2025-02-11T05:38:23.668Z |
Milestone |
6.2 |
Labels |
arch:arm, cpukit::dl |
Link |
|
Merges |
2 |
Summary¶
When a trampoline is not aligned on a word boundary and a ldr.w pc,[pc] instruction is generated, the address loaded is not pc+4, but instead pc+2.
Author: Kinsey Moore
2025-01-23T22:24:29.300Z
assigned to @opticron
Author: Kinsey Moore
2025-01-23T22:24:29.649Z
Author: Kinsey Moore
2025-01-23T22:26:00.055Z
mentioned in merge request !367
Author: Chris Johns
2025-01-24T00:10:28.724Z
changed title from Backpor{-r-}t libdl ARM trampoline crash fix to Backport libdl ARM trampoline crash fix
Author: Ranulfo Raphael
2025-02-11T05:38:23.161Z
mentioned in commit 84c2cf3da9983ef7eadb11d4a676449d6b3ef875
4958 - NTP broadcast client fails to open sockets on lwIP¶
Id |
4958 |
State |
closed |
Type |
ISSUE |
Author |
Kinsey Moore |
Assignee(s) |
Trac Migrate |
Closed by |
Kinsey Moore |
Created |
2023-10-04T19:42:57.000Z |
Closed |
2025-01-31T18:30:49.585Z |
Updated |
2025-01-31T18:30:49.660Z |
Milestone |
6.2 |
Labels |
network::lwip, priority::normal, tickettype::defect, version::6 |
Link |
|
Merges |
0 |
Reproduction: * Install lwIP on top of a compatible BSP * Modify the existing ntp01 test to use the broadcast client * Set NTP_DEBUG to 1 * Replace the first two lines of etc_ntp_conf with: * broadcastclient * disable auth * Build rtems-net-services against the installed BSP with lwIP network stack * Run the modified ntp01 test
Expected result: NTP opens broadcast listener successfully.
Actual result: NTP fails to open broadcast listener with error message: bind(37) AF_INET 0.0.0.0#123 flags 0x409 failed: Address already in use
Workaround: In open_socket() in bsd/freebsd/contrib/ntp/ntpd/ntp_io.c, there is a test for wildcard addresses inside a OS_NEEDS_REUSEADDR_FOR_IFADDRBIND conditional. If the set_wildcard_reuse() call is made unconditional, the socket creation succeeds.
The root cause is that behavior differs between LibBSD and lwIP. On lwIP, early-bound wildcard and non-wildcard address bindings succeed, but late-bound wildcard address bindings fail. Once any any address is bound, lwIP will fail to bind any further wildcard sockets. Either we can change lwIP’s behavior to be closer to LibBSD or we can alter the NTP client to check for lwIP or blindly apply the address reuse flags.
Author: Kinsey Moore
2023-10-04T19:44:27.000Z
Description changed
Reproduction: * Install lwIP on top of a compatible BSP * Modify the existing ntp01 test to use the broadcast client - ** Set NTP_DEBUG to 1 ? - + * Set NTP_DEBUG to 1 ? + - ** Replace the first two lines of etc_ntp_conf with: ? - + * Replace the first two lines of etc_ntp_conf with: ? + - *** broadcastclient ? -- + * broadcastclient ? ++ - *** disable auth ? -- + * disable auth ? ++ * Build rtems-net-services against the installed BSP with lwIP network stack * Run the modified ntp01 test Expected result: NTP opens broadcast listener successfully. Actual result: NTP fails to open broadcast listener with error message: bind(37) AF_INET 0.0.0.0#123 flags 0x409 failed: Address already in use Workaround: In open_socket() in bsd/freebsd/contrib/ntp/ntpd/ntp_io.c, there is a test for wildcard addresses inside a OS_NEEDS_REUSEADDR_FOR_IFADDRBIND conditional. If the set_wildcard_reuse() call is made unconditional, the socket creation succeeds. The root cause is that behavior differs between LibBSD and lwIP. On lwIP, early-bound wildcard and non-wildcard address bindings succeed, but late-bound wildcard address bindings fail. Once any any address is bound, lwIP will fail to bind any further wildcard sockets. Either we can change lwIP's behavior to be closer to LibBSD or we can alter the NTP client to check for lwIP or blindly apply the address reuse flags.
Author: Chris Johns
2023-10-04T21:30:43.000Z
I think NTP is using _normal_ behavior so lwIP is not compliant in this case. Adding a work around to NTP solves this problem but leaves the issue present other packages may trip over.
Have you asked lwIP about this?
Author: Kinsey Moore
2023-10-05T17:08:40.000Z
I haven’t. This was an issue I noticed and wanted to document that it existed with current thoughts, but I hadn’t moved beyond that yet since the broadcast client isn’t on my priority list at the moment. I’ll agree with you that fixing lwIP to be more “normal” is probably the right solution. I’ll get more context from lwIP.
Author: Gedare Bloom
2024-12-04T21:40:36.490Z
changed the description
Author: Gedare Bloom
2024-12-04T21:40:37.588Z
Author: Kinsey Moore
2025-01-31T18:30:49.722Z
moved to rtems/pkg/rtems-net-services#10
5146 - Follow-up from “Draft: timespec.h, etc.: Eliminate use of 32-bit for time_t”¶
Id |
5146 |
State |
closed |
Type |
ISSUE |
Author |
Chris Johns |
Assignee(s) |
Joel Sherrill |
Closed by |
Kinsey Moore |
Created |
2024-11-08T00:37:24.402Z |
Closed |
2025-01-24T01:08:51.602Z |
Updated |
2025-01-24T01:08:52.058Z |
Milestone |
6.2 |
Labels |
rtems::score |
Link |
|
Merges |
1 |
The following discussion from !175 should be addressed:
[] @sebhub started a discussion
Add test case to sptchimespec01 which uses a > 32-bit time?
Author: Chris Johns
2024-11-08T00:37:24.525Z
assigned to @joel
Author: zack liang
2024-11-09T22:19:35.699Z
hey @joel would like to work on this just tell me where the tests file is
Author: zack liang
2024-11-15T01:43:13.170Z
mentioned in merge request !314
Author: Joel Sherrill
2024-12-04T00:56:31.646Z
Author: Gedare Bloom
2024-12-04T21:40:37.917Z
5107 - timespecaddto.c: Use of 32-bit time_t¶
Id |
5107 |
State |
closed |
Type |
ISSUE |
Author |
Joel Sherrill |
Assignee(s) |
Joel Sherrill |
Closed by |
Chris Johns |
Created |
2024-08-07T21:57:13.967Z |
Closed |
2024-11-08T00:40:35.581Z |
Updated |
2024-11-08T00:40:35.686Z |
Milestone |
6.2 |
Labels |
rtems::score, scan::coverity |
Link |
|
Merges |
1 |
Summary¶
Coverity CID 1512527
Coverity spotted this where time_t variables/structure elements are assigned to uint32_t. Notice that the return type is uint32_t which propagates to callers of this such as rtems_timespec_add_to().
46 uint32_t _Timespec_Add_to(
47 struct timespec *time,
48 const struct timespec *add
49)
50{
CID 1512527: (#1 of 1): Use of 32-bit time_t (Y2K38_SAFETY)
1. store_truncates_time_t: A time_t value is stored in an integer with too few bits to accommodate it. The expression add->tv_sec is cast to uint32_t.
51 uint32_t seconds = add->tv_sec;
Steps to reproduce¶
Author: Joel Sherrill
2024-08-07T21:57:14.149Z
assigned to @joel
Author: Joel Sherrill
2024-08-07T22:01:06.155Z
mentioned in merge request !175
4964 - rtems-test can hang indefinitely¶
Id |
4964 |
State |
closed |
Type |
ISSUE |
Author |
Kinsey Moore |
Assignee(s) |
Trac Migrate |
Created |
2023-11-07T15:20:21.000Z |
Updated |
2023-11-07T21:06:51.000Z |
Milestone |
6.2 |
Labels |
priority::low, resolution::duplicate, tickettype::defect, tool::test, version::6 |
Link |
|
Merges |
0 |
When using the RTEMS automated tester configured with TFTP, there is a state that can occur in which the tester never times out a test. This was discovered when testing with the AArch64 Xilinx ZynqMP BSP where the board did not actually power on even though it was instructed to do so. This left the tester running the TFTP server waiting more than 4 full days (left to run over a long weekend) for the target to start without progressing to the next test.
The failure to start on the target side was due to a missing minimum settling time between off and on states that is apparently influenced by the environment.
Author: Chris Johns
2023-11-07T20:52:08.000Z
Can you please try https://lists.rtems.org/pipermail/devel/2023-October/076519.html ?
Looks like I forgot to push it.
Author: Chris Johns
2023-11-07T21:06:51.000Z
Resolution set to ~”duplicate”
Status changed from new to closed
This is a duplicate of #4959
5284 - Backport mvme3100 fixes to 6 (!541) (opened)¶
Id |
5284 |
State |
opened |
Type |
ISSUE |
Author |
Jeremy Lorelli |
Created |
2025-06-30T01:03:53.061Z |
Updated |
2025-06-30T14:56:00.427Z |
Milestone |
6.2 |
Labels |
arch:powerpc, backport |
Link |
|
Merges |
0 |
Summary¶
Backport !541 to RTEMS 6, once it’s merged.
Pre-set options¶
Author: Jeremy Lorelli
2025-06-30T01:04:28.988Z
mentioned in merge request !541
Author: Gedare Bloom
2025-06-30T14:54:49.549Z
changed type from task to issue
Author: Gedare Bloom
2025-06-30T14:56:00.396Z
added rtems#10 as parent epic
5273 - Backport libdebugger setjmp fix to 6 branch (opened)¶
Id |
5273 |
State |
opened |
Type |
ISSUE |
Author |
Kinsey Moore |
Assignee(s) |
Kinsey Moore |
Created |
2025-06-18T18:52:51.524Z |
Updated |
2025-06-19T22:52:53.491Z |
Milestone |
6.2 |
Labels |
arch:aarch64, cpukit::debugger |
Link |
|
Merges |
0 |
The following discussion from !523 should be addressed:
[] @opticron started a discussion:
This needs to be backported to 6.
Author: Kinsey Moore
2025-06-18T18:52:52.066Z
assigned to @opticron
Author: Kinsey Moore
2025-06-18T18:52:53.254Z
Author: Kinsey Moore
2025-06-18T18:55:47.388Z
mentioned in merge request !524
Author: Kinsey Moore
2025-06-19T22:52:52.933Z
mentioned in commit 3d8b56f10c9cabe8cc731ada636af59f71e8e10b
5271 - Backport libdebugger TLS support to the RTEMS 6 branch (opened)¶
Id |
5271 |
State |
opened |
Type |
ISSUE |
Author |
Kinsey Moore |
Assignee(s) |
Kinsey Moore |
Created |
2025-06-17T19:15:41.599Z |
Updated |
2025-06-17T19:19:46.332Z |
Milestone |
6.2 |
Labels |
arch:aarch64, cpukit::debugger |
Link |
|
Merges |
0 |
The following discussion from !517 should be addressed:
[] @opticron started a discussion:
This needs to be backported to the 6 branch.
Author: Kinsey Moore
2025-06-17T19:15:42.193Z
assigned to @opticron
Author: Kinsey Moore
2025-06-17T19:15:42.692Z
Author: Kinsey Moore
2025-06-17T19:19:46.270Z
mentioned in merge request !518
5216 - Deprecate powerpc/mvme5500 in favor of powerpc/beatnik (opened)¶
Id |
5216 |
State |
opened |
Type |
ISSUE |
Author |
Gedare Bloom |
Assignee(s) |
Gedare Bloom |
Created |
2025-02-14T22:36:47.150Z |
Updated |
2025-05-08T23:19:02.124Z |
Milestone |
6.2 |
Labels |
arch:powerpc, bsp |
Link |
|
Merges |
0 |
The beatnik primarily targets the mvme6100 but it also supports the mvme5500. This issue notices that the powerpc/mvme5500 is deprecated in the 6 release series and will be removed in 7.
Author: Gedare Bloom
2025-02-14T22:36:47.351Z
cloned from #4160
Author: Gedare Bloom
2025-02-14T22:36:47.239Z
assigned to @gedare
Author: Gedare Bloom
2025-02-14T22:38:04.310Z
changed title from {-Obsolete -}mvme5500 in favor of {-B-}eatnik to {+Deprecate powerpc/+}mvme5500 in favor of {+powerpc/b+}eatnik
Author: Gedare Bloom
2025-02-14T22:38:04.339Z
changed the description
Author: Gedare Bloom
2025-02-14T22:38:13.212Z
Author: Gedare Bloom
2025-02-14T22:57:48.491Z
[x] add this to the release notes
[x] add a note in Beatnik documentation
Author: Gedare Bloom
2025-02-14T22:57:48.491Z
https://gitlab.rtems.org/rtems/rtos/rtems-release/-/merge_requests/25
Author: Gedare Bloom
2025-02-14T22:57:48.491Z
https://gitlab.rtems.org/rtems/docs/rtems-docs/-/merge_requests/153
Author: Gedare Bloom
2025-02-14T22:53:43.694Z
mentioned in commit gedare/rtems-docs@72286dba503c0879486b8877c2e88522b146bad0
Author: Gedare Bloom
2025-02-14T22:54:05.231Z
mentioned in merge request rtems/docs/rtems-docs!153
Author: Gedare Bloom
2025-02-20T23:18:23.121Z
mentioned in commit gedare/rtems-docs@3ec35416774b6a7baf9106ebc0ac4a935e84cb48
Author: Amar Takhar
2025-05-08T23:16:06.202Z
mentioned in commit amar/rtems-docs@38a25aad78332ef1f507620040c3e8541a1ab048
Author: Amar Takhar
2025-05-08T23:19:01.833Z
mentioned in commit gedare/rtems-docs@701f0230e962b7d559dd7bb55b4443c37b82d67f
5236 - Runtime symbols loading is broken (opened)¶
Id |
5236 |
State |
opened |
Type |
ISSUE |
Author |
Chris Johns |
Assignee(s) |
Chris Johns |
Created |
2025-04-01T00:58:33.568Z |
Updated |
2025-04-01T13:51:00.306Z |
Milestone |
6.2 |
Labels |
cpukit::dl, tool::rtems-tools |
Link |
|
Merges |
0 |
Summary¶
Run time symbol loading from an object file is broken. The change to track dependencies added support to check the origin of a symbol to track dependencies. Runtime loading of symbols requires the symbol rtems_rtl_base_sym_global_add be in the symbol table. Loading an symbol object file causes a constrictor attribute function to be run and that calls rtems_rtl_base_sym_global_add adding the base image global symbols. Currently attempting to load a symbol object file results in an unresolved external to rtems_rtl_base_sym_global_add because the symbols is inserted into the global symbol table and not added to the base image’s global symbols.
Also rtems-syms does not correctly export the needed TLS data.
Author: Chris Johns
2025-04-01T00:58:33.714Z
cloned from #5234
Author: Chris Johns
2025-04-01T00:58:33.637Z
assigned to @chris
Author: Chris Johns
2025-04-01T01:34:24.159Z
mentioned in merge request !461
Author: Kinsey Moore
2025-04-01T13:51:00.525Z
5222 - RTEMS Priority range API change (regression) (opened)¶
Id |
5222 |
State |
opened |
Type |
ISSUE |
Author |
Chris Johns |
Assignee(s) |
Chris Johns |
Created |
2025-02-27T21:44:09.559Z |
Updated |
2025-03-05T13:15:56.986Z |
Milestone |
6.2 |
Labels |
api |
Link |
|
Merges |
0 |
The following discussion from !436 should be addressed:
[] @chris started a discussion
Yes I could add
CONFIGURE_MAXIMUM_PRIORITYwith255to resolve the problem in my application. Why should I? The code base that brought this problem to my attention was originally developed on RTEMS 4 on a Zynq and the user wants to use SMP?My investigation to understand the problem has raised a number of concerns and I do not think this MR is the place to discuss them. The MR may in the end be a suitable resolution however we need to first have a wider discussion.
I believe the changing of the priority range is a _regression_ with respect to the historical user interfaces until we decide otherwise. I do not remember a specific discussion about the range changing in regards to this context, the effects, other possible options or how we convey this to users? I believe a number of separate changes over a period of time have come together to bring about my concerns. Each piece on its own makes sense and seems reasonable however together there are issues. We need to respect our users, existing code and the need to make things as simple as possible simple configurations.
I am also using this specific example to bring about a policy change. API changes needs to be announced and a specific discussion happen for the change to accepted. Approval and merging of changes in isolation does not mean an acceptable of change. If this happens we need to discuss the impact, look at solutions I have labelled this MR API however I wonder if we need an
api-changelabel as well?—
A discourse topic discussing this issue can be found at https://users.rtems.org/t/scheduler-maximum-regression-in-rtems-6/78
Author: Chris Johns
2025-02-27T21:44:09.713Z
assigned to @chris
Author: Chris Johns
2025-02-27T21:45:07.643Z
changed the description
Author: Chris Johns
2025-02-28T02:19:24.550Z
changed the description
Author: Sebastian Huber
2025-03-05T13:15:56.927Z
The default SMP scheduler is an EDF scheduler. This has other implications. For example, the use of rate-monotonic objects changes drastically. If you want maximum backward compatibility, the default SMP scheduler could be changed back to deterministic priority.
5212 - machine/_timecounter.h is missing _Timecounter_Getboottime(_tvp) (opened)¶
Id |
5212 |
State |
opened |
Type |
ISSUE |
Author |
Chris Johns |
Assignee(s) |
Chris Johns |
Created |
2025-02-11T22:24:18.786Z |
Updated |
2025-02-11T22:27:23.625Z |
Milestone |
6.2 |
Labels |
rtems::kernel |
Link |
|
Merges |
0 |
The machine/_kernel_time.h header has:
#define getboottime(_tvp) _Timecounter_Getboottime(_tvp)
#define getboottimebin(_bt) _Timecounter_Getboottimebin(_bt)
however machine/_timecounter.h does not have a prototype for _Timecounter_Getboottime and GCC 14 now generates an error.
Author: Chris Johns
2025-02-11T22:24:18.920Z
cloned from #5211
Author: Chris Johns
2025-02-11T22:24:18.850Z
assigned to @chris
Author: Chris Johns
2025-02-11T22:24:31.498Z
Author: Chris Johns
2025-02-11T22:25:03.138Z
GCC 13 does not complain but the issue is still present.
Author: Chris Johns
2025-02-11T22:27:23.402Z
mentioned in merge request !416
5165 - Use of pkg-config –libs is unreliable (opened)¶
Id |
5165 |
State |
opened |
Type |
ISSUE |
Author |
Sebastian Huber |
Assignee(s) |
Sebastian Huber |
Created |
2024-11-28T12:58:42.398Z |
Updated |
2025-02-11T06:26:21.717Z |
Milestone |
6.2 |
Link |
|
Merges |
0 |
Summary¶
I noticed a strange issue with pkg-config for a BSP with special linker flags:
https://github.com/pkgconf/pkgconf/issues/371
I think we should provide the linker flags also through an LDFLAGS variable.
Author: Sebastian Huber
2024-11-28T12:58:42.511Z
assigned to @sebhub
Author: Sebastian Huber
2024-11-28T14:26:59.811Z
mentioned in merge request !335
Author: Sebastian Huber
2024-11-28T14:28:04.148Z
mentioned in merge request !336
Author: Gedare Bloom
2024-12-04T21:39:42.250Z
Author: Sebastian Huber
2024-12-05T03:02:18.776Z
cloned to #5169
Author: Chris Johns
2024-12-22T03:37:31.386Z
Author: Chris Johns
2025-02-11T05:34:34.959Z
This change breaks libbsd and I suspect more. See https://gitlab.rtems.org/rtems/pkg/rtems-libbsd/-/issues/38 for details.
Author: Chris Johns
2025-02-11T05:36:33.803Z
The MR needs to be revert it.
Author: Sebastian Huber
2025-02-11T05:38:22.683Z
mentioned in commit e36ba91110724b5cc8a9e07659d98d0bcb96e4d4
Author: Sebastian Huber
2025-02-11T05:38:56.744Z
I would first like to investigate why this change causes something to break. The LDFLAGS were previously not visible by pkg-config.
Author: Sebastian Huber
2025-02-11T06:26:22.134Z
It seems the removal of the
Ldflags:is the issue here. This part of the patch needs to be reverted. Thertems_wafpackage uses the option--ldflagswhich is not documented bypkg-config --help.
Author: Chris Johns
2025-02-11T06:26:22.134Z
I suggest this change is reverted and then a new change can created, review and merged. Any change has to build
rtems_wafbase packages such aslibbsdto be approved.
Author: Sebastian Huber
2025-02-11T06:26:22.134Z
Why don’t you give people a business day to fix this?
Author: Chris Johns
2025-02-11T06:26:22.134Z
That depends on the fix and how long it takes to be reviewed and merged. A revert works, I have tested it so it lets LibBSD build which is important. I am a little unsure what the problem is with a revert?
Author: Sebastian Huber
2025-02-11T06:26:22.134Z
I am already working on a patch.
Author: Chris Johns
2025-02-11T06:26:22.134Z
Sure and thanks. I am only providing a way to solve this breakage. It would helpful if you could please explain your position of no reverts? Gitlab has a Revert button because it is useful and used so if a change breaks something it is helpful.
Author: Sebastian Huber
2025-02-11T06:26:22.134Z
In the past, patches were not reverted this quickly.
Author: Amar Takhar
2025-02-11T06:27:01.628Z
These should be reverted because it caused the breakage it’s a good way to record the issue in GitLab. New MRs should come with the alternate change.
The reason we moved to GitLab is to make these processes easy we’re not in a race I appreciate you wanting more time but the policy should always be to revert ASAP when we see breakage and have a new change in come via a new MR this is an easier way to track changes. This also means that any future changes can be properly considered without being under pressure due to current breakage.
Author: Sebastian Huber
2025-02-11T06:26:22.134Z
Please go ahead, I will have a look at it in a couple of days.
Author: Chris Johns
2025-02-11T05:50:43.935Z
mentioned in merge request !413
Author: Amar Takhar
2025-02-11T05:58:44.327Z
mentioned in merge request !414
5184 - cpukit/libmisc/capture: rtems_capture_flush incorrectly clears the RTEMS_CAPTURE_INIT flag (opened)¶
Id |
5184 |
State |
opened |
Type |
ISSUE |
Author |
Gedare Bloom |
Assignee(s) |
Gedare Bloom |
Created |
2025-01-20T23:56:21.515Z |
Updated |
2025-02-11T05:38:23.408Z |
Milestone |
6.2 |
Labels |
cpukit::misc |
Link |
|
Merges |
2 |
The following discussion from !337 should be addressed:
[] @gedare started a discussion:
The RTEMS_CAPTURE_OVERFLOW is an overflow flag for each CPU, and its value is the same as RTEMS_CAPTURE_INIT. Executing rtems_capture_flush will set the RTEMS_CAPTURE_INIT flag to 0 in the global flags.
Author: Gedare Bloom
2025-01-20T23:56:21.693Z
assigned to @gedare
Author: Gedare Bloom
2025-01-20T23:56:22.093Z
Author: Gedare Bloom
2025-01-21T00:00:30.526Z
mentioned in merge request !361
Author: Gedare Bloom
2025-01-21T02:25:29.025Z
Author: Zhaoyue Wang
2025-02-11T05:38:22.745Z
mentioned in commit 603c168a0ba828efc949b398d3623ebc9a0aafa3
5206 - Back-port “wscript: Make builds with tar archives reproducible” (opened)¶
Id |
5206 |
State |
opened |
Type |
ISSUE |
Author |
Gedare Bloom |
Assignee(s) |
Gedare Bloom |
Created |
2025-02-07T20:41:02.457Z |
Updated |
2025-02-07T20:41:02.457Z |
Milestone |
6.2 |
Labels |
backport |
Link |
|
Merges |
0 |
The following discussion from !405 should be addressed:
[] @frank_k started a discussion
What do I need to do to get this patch also into RTEMS 6?
Author: Gedare Bloom
2025-02-07T20:41:02.632Z
assigned to @gedare
Author: Gedare Bloom
2025-02-07T20:41:03.799Z
5202 - Termios: cannot set all baud rates with rtems_termios_set_best_baud() (opened)¶
Id |
5202 |
State |
opened |
Type |
ISSUE |
Author |
Gedare Bloom |
Assignee(s) |
Jan Sommer |
Created |
2025-02-03T19:00:52.672Z |
Updated |
2025-02-06T03:35:43.265Z |
Milestone |
6.2 |
Labels |
cpukit |
Link |
|
Merges |
2 |
Summary¶
Steps to reproduce¶
https://gitlab.rtems.org/rtems/rtos/rtems/-/blob/main/cpukit/libcsupport/src/termios_baudtable.c?ref_type=heads https://gitlab.rtems.org/rtems/rtos/rtems/-/blob/main/cpukit/libcsupport/src/termios_setbestbaud.c?ref_type=heads#L43
The algorithm in rtems_termios_set_best_baud() successively iterates over the elements of the rtems_termios_baud_table and searches for a fitting baud rate (checking by “is element lower than the baud parameter, then checks the next element, …).
The table of baud rates is not sorted correctly from a certain point on, therefore the algorithm for choosing the optimal baud rate does not work for every baud rate.
e.g. if the input is 7200 (which is in the table) the search terminates at 9600 as that is larger, the last entry was 4800, so in the check it sets it to either of those, but not the 7200.
We observed it when we entered 38401 (due to measurement errors on the hardware was a bit larger).
We identified the search ending at 57600 and it ended using the last = 28800.
Pre-set options¶
Author: Gedare Bloom
2025-02-03T19:00:52.815Z
cloned from #5162
Author: Gedare Bloom
2025-02-03T19:00:52.746Z
assigned to @thesummer
Author: Gedare Bloom
2025-02-03T19:01:07.472Z
Author: Gedare Bloom
2025-02-03T19:05:34.225Z
mentioned in merge request !381
Author: Jan Sommer
2025-02-06T03:35:43.024Z
mentioned in commit 9641e1e97d8195d567a6218228ef0f0f2d7fdc25
5008 - Create an ALUMNI file for MAINTAINERS in lieu of CODEOWNERS (opened)¶
Id |
5008 |
State |
opened |
Type |
ISSUE |
Author |
Amar Takhar |
Assignee(s) |
Chris Johns, Gedare Bloom, Joel Sherrill |
Created |
2024-05-02T07:17:35.160Z |
Updated |
2025-01-31T18:56:38.667Z |
Milestone |
6.2 |
Labels |
doc |
Link |
|
Merges |
0 |
Summary¶
We have some alumni in the MAINTAINERS file we want to put somewhere in the docs? But it’s confusing to have both let’s make sure that the CODEOWNERS is the canonical version and anything that needs to be carried in docs gets done so.
We can have an ALUMNI file plenty of projects have that and I support this method since it’s carried in the source I don’t think they should need to go seek somewhere else.
Steps to reproduce¶
Author: Amar Takhar
2024-05-02T07:17:35.264Z
assigned to @chris, @gedare, and @joel
Author: Gedare Bloom
2024-12-04T21:40:37.675Z
Author: Gedare Bloom
2025-01-31T18:56:38.719Z
changed title from {-Remove-} MAINTAINERS in lieu of CODEOWNERS to {+Create an ALUMNI file for+} MAINTAINERS in lieu of CODEOWNERS
4956 - Deadlock caused by locking inversion in JFFS2 (opened)¶
Id |
4956 |
State |
opened |
Type |
ISSUE |
Author |
Kinsey Moore |
Assignee(s) |
Kinsey Moore |
Created |
2023-09-20T16:58:43.000Z |
Updated |
2024-12-04T21:40:36.315Z |
Milestone |
6.2 |
Labels |
priority::normal, tickettype::defect |
Link |
|
Merges |
0 |
When the code for delayed writes was added to support JFFS2 on NAND, it introduced a possible locking inversion when managing the delayed write work queue.
This has been partially addressed by fa85760c99a72b8186c70c896b98efe0d7cde2b6, but there are other issues with this code that need to be addressed as well.
First is that when not compiling with RTEMS_DEBUG, nodes are not automatically set as off the chain upon extraction and this behavior is depended upon.
Second is that protected (ISR-locked) chain calls are used when they are not necessary due to already being locked.
Third is that the bug fixed by the referenced commit needs a test that verifies the fix.
Author: Trac Migrate
2023-09-26T00:41:07.000Z
Hey i’m interested working on this ticket would the nodes be set off only if
Author: Kinsey Moore
2023-09-26T13:16:46.000Z
Hey, thanks for the interest. I actually have a patch on the mailing list that just got approved for fixing this, so it should be going into the codebase shortly. I still need to write a test that produces the original issue, though.
Author: Trac Migrate
2023-09-26T13:38:00.000Z
In [changeset:”99c637fbd973a3d2f6e5adf38aac8f3f588cbb7e/rtems” 99c637f/rtems]:
cpukit/jffs2: Avoid use of off-chain semantics This reworks the JFFS2 delayed work queue to avoid use of on-chain/off-chain semantics since they vary in behavior under RTEMS_DEBUG and are not guaranteed to be safe to use in SMP systems. This adds all delayed work structs to the chain on FS init and does not remove them until umount. Updates #4956
Author: Kinsey Moore
2024-02-16T00:35:11.000Z
Owner set to @kinsey
Status changed from new to assigned
Author: Kinsey Moore
2024-05-02T23:19:35.303Z
A baseline set of tests has been added to the RTEMS testsuite to test JFFS2 on a simulated NAND device. This baseline set of tests does not trigger the original condition that necessitated this ticket and I have thus far been unable to trigger the deadlock from a test with rolled-back code.
Author: Kinsey Moore
2024-05-17T02:59:17.993Z
changed the description
Author: Gedare Bloom
2024-12-04T21:40:37.334Z
4895 - Add support for RTEMS cache API to MicroBlaze (opened)¶
Id |
4895 |
State |
opened |
Type |
ISSUE |
Author |
Trac Migrate |
Assignee(s) |
Trac Migrate |
Created |
2023-04-12T03:23:57.000Z |
Updated |
2024-12-04T21:40:36.095Z |
Milestone |
6.2 |
Labels |
bsp, priority::normal, tickettype::enhancement, version::6 |
Link |
|
Merges |
0 |
Original author: Alex
Some cache management functions have already been added to MicroBlaze. The functions are declared in bsps/microblaze/microblaze_fpga/include/bsp.h. These should instead be called by the wrapper functions of the RTEMS cache management API from bsps/shared/cache/cacheimpl.h to match the API used in other BSPs. The cache management functions should be defined in bsps/microblaze/shared/cache/cache.c.
Author: Joel Sherrill
2023-06-22T20:54:27.000Z
Original author: Alex
Alex .. Is this required for 6.1 or can the milestone be moved to 6.2, 7.1, or indefinite? Is it in your work queue?
Author: Trac Migrate
2023-09-20T15:08:40.000Z
Original author: Alex
Replying to Joel Sherrill:
Alex .. Is this required for 6.1 or can the milestone be moved to 6.2, 7.1, or indefinite? Is it in your work queue?
This can be moved to 6.2. It’s still on the backlog.
Author: Kinsey Moore
2023-09-21T13:20:48.000Z
Original author: Alex
Milestone changed from %”6.1” to %”6.2”
Author: Chris Johns
2023-09-21T23:45:16.000Z
Original author: Alex
Replying to Alex White:
Replying to Joel Sherrill:
Alex .. Is this required for 6.1 or can the milestone be moved to 6.2, 7.1, or indefinite? Is it in your work queue?
This can be moved to 6.2. It’s still on the backlog.
Does this change mean an cache API change for the Microblaze?
Author: Gedare Bloom
2024-12-04T21:40:36.178Z
changed the description
Author: Gedare Bloom
2024-12-04T21:40:37.155Z
4675 - sleep should be a thread cancellation point (opened)¶
Id |
4675 |
State |
opened |
Type |
ISSUE |
Author |
Trac Migrate |
Assignee(s) |
Kinsey Moore |
Created |
2022-07-07T10:17:28.000Z |
Updated |
2024-12-04T21:40:35.983Z |
Milestone |
6.2 |
Labels |
api::posix, priority::normal, tickettype::defect, version::6 |
Link |
|
Merges |
0 |
Original author: aoliva
GCC has worked around, in the libstdc++ testsuite, the fact that nanosleep (as well as sleep and clock_nanosleep) should be thread cancellation points, per POSIX, but RTEMS may want to make them so. https://gcc.gnu.org/pipermail/gcc-patches/2022-June/597003.html
(I haven’t checked that other functions that should be thread cancellation points are, I only noticed that this one wasn’t, and decided to file this ticket because for other issues in which I explicitly identified RTEMS I got a request to do so)
Author: Trac Migrate
2022-07-07T12:40:48.000Z
Original author: aoliva
Thanks for reporting the issue. I think we have currently no cancellation points in RTEMS at all. I think this is intentional, so I would mark this bug as WONTFIX.
Author: Joel Sherrill
2023-03-18T17:43:07.000Z
Original author: aoliva
Do you have a test to show something isn’t right? sleep() and nanonsleep() ultimately call clock_nanosleep() and the requested blocking state is “STATES_WAITING_FOR_TIME | STATES_INTERRUPTIBLE_BY_SIGNAL”
https://git.rtems.org/rtems/tree/cpukit/posix/src/clocknanosleep.c#n77
RTEMS state bits consist of the type of object is being blocked on and attributes such as interruptible. It has been this way for a LONG time.
As a random thought, if you don’t build RTEMS with POSIX enabled, then POSIX signals are disabled and then this likely wouldn’t work.
Author: Joel Sherrill
2023-03-18T17:43:24.000Z
Original author: aoliva
Milestone set to %”6.1”
Author: Trac Migrate
2023-04-22T05:32:10.000Z
Original author: aoliva
The unpatched libstdc++ testcase is a test that shows the problem. An explicit pthread_testcancel shouldn’t be needed. Functions that are cancellation points should (as-if) call it implicitly.
Author: Joel Sherrill
2023-06-22T20:52:53.000Z
Original author: aoliva
Owner set to @kinsey
Status changed from new to assigned
Author: Amar Takhar
2024-05-17T00:42:51.649Z
Hi there Kinsey can this be moved to %6.2?
Author: Amar Takhar
2024-05-17T01:08:02.038Z
changed the description
Author: Kinsey Moore
2024-05-17T02:53:23.124Z
@joel Do you have opinions on this? It’s in my backlog of things to get to, but I have many things that come before it.
Author: Amar Takhar
2024-05-17T03:23:52.870Z
@opticron this was already moved to %6.2 :) .. Joel asked to move it.
Author: Kinsey Moore
2024-05-17T03:23:52.870Z
I saw that it got moved, but not the rationale. I was driving at the time.
Author: Amar Takhar
2024-05-17T03:23:52.870Z
Ah, makes sense I should have noted that it was Joel that asked, too.
Author: Gedare Bloom
2024-12-04T21:40:37.120Z
4364 - rtems-test Fails to list failed tests to user (opened)¶
Id |
4364 |
State |
opened |
Type |
ISSUE |
Author |
Joel Sherrill |
Assignee(s) |
Chris Johns |
Created |
2021-03-28T15:51:47.000Z |
Updated |
2024-12-04T21:40:35.871Z |
Milestone |
6.2 |
Labels |
priority::normal, tickettype::defect, tool::rtems-source-builder, version::6 |
Link |
|
Merges |
0 |
rtems-test does not report the set of failed tests to the user at the conclusion of its run. It only gives a summary by category. They ARE in the log file. I assume this is a minor issue now that I have spotted the pattern.
I assume this also applies to 5 but I spotted it on 6 during the recent heavy round of manual testing.
Author: Amar Takhar
2024-05-17T02:02:50.798Z
Joel are you OK with this being moved to %6.2 Since there is already a solution this is more of an ease-of-use feature.
Author: Joel Sherrill
2024-05-17T13:46:01.611Z
It is not an ease of use feature. If interactive, you do not get the list of failures unless you enable logging.
I don’t mind moving the milestone though. This has been sitting here for 3 years already.
Author: Amar Takhar
2024-05-17T15:09:16.482Z
changed the description
Author: Gedare Bloom
2024-12-04T21:40:36.939Z
4292 - rtems-rfs-bitmaps.c: Useless call (CID 1399749) (opened)¶
Id |
4292 |
State |
opened |
Type |
ISSUE |
Author |
Joel Sherrill |
Assignee(s) |
Chris Johns |
Created |
2021-03-03T21:04:09.000Z |
Updated |
2024-12-04T21:40:35.770Z |
Milestone |
6.2 |
Labels |
filesystem::rfs, priority::normal, tickettype::defect, version::6 |
Link |
|
Merges |
0 |
Coverity Scan spotted this in rtems-rfs-bitmaps.c as CID 1399749). On the face of it, Scan appears to be correct but this issue should be reviewed by the original author of this code.
}
623 else
624 {
625 bits = *map;
626 available = rtems_rfs_bitmap_element_bits ();
627 }
628
629 if (rtems_rfs_bitmap_match (bits, RTEMS_RFS_BITMAP_ELEMENT_SET))
CID 1399749 (#1 of 1): Useless call (USELESS_CALL)
side_effect_free: Calling rtems_rfs_bitmap_set(*search_map, bit) is only useful for its return value, which is ignored. [show details]
630 rtems_rfs_bitmap_set (*search_map, bit);
631 else
632 {
Author: Joel Sherrill
2021-03-03T21:04:21.000Z
Owner set to @chrisj
Status changed from new to assigned
Author: Chris Johns
2024-05-16T23:08:01.788Z
changed the description
Author: Chris Johns
2024-05-16T23:08:05.781Z
Please create merge request.
Author: Gedare Bloom
2024-12-04T21:40:37.082Z
4125 - Update wiki pages to reference waf not autoconf based build (opened)¶
Id |
4125 |
State |
opened |
Type |
ISSUE |
Author |
Joel Sherrill |
Assignee(s) |
Joel Sherrill |
Created |
2020-10-06T18:27:32.000Z |
Updated |
2024-12-04T21:40:35.664Z |
Milestone |
6.2 |
Labels |
priority::normal, tickettype::defect, version::6 |
Link |
|
Merges |
0 |
Since configure is a horrible thing to search for, I searched for “enable-rtems” in the wiki. The results are https://devel.rtems.org/search?q=enable-rtems&noquickjump=1&wiki=on and show about 20 pages that need review.
Some may be ok since they are historical on GSoC projects but others will need updating. Some are BSP pages which really should be converted to the Users Guide and updated.
We have an ongoing effort to reduce the wiki and some of these pages may be good candidates. Let’s use this as an excuse to do some clean up.
Author: Joel Sherrill
2020-10-06T18:27:51.000Z
Summary changed from Update wiki to reference waf to Update wiki pages to reference waf not autoconf based build
Author: Trac Migrate
2020-11-02T18:48:43.000Z
I don’t think this is a blocker to remove the old build system. The user manual is the source of reference.
Author: Chris Johns
2022-11-29T23:23:25.000Z
Owner set to @joel
Status changed from new to assigned
Author: Amar Takhar
2024-05-17T00:47:34.904Z
changed the description
Author: Amar Takhar
2024-05-17T00:47:37.779Z
Hi Joel since we have no more wiki what’s the alternate solution for this? Moving it to %6.2 since it’s not really related to %6.1 anymore.
Author: Gedare Bloom
2024-12-04T21:40:36.815Z
4095 - Remove doc from rtems-tools (opened)¶
Id |
4095 |
State |
opened |
Type |
ISSUE |
Author |
Chris Johns |
Assignee(s) |
Kinsey Moore |
Created |
2020-09-24T23:53:13.000Z |
Updated |
2024-12-04T21:40:35.559Z |
Milestone |
6.2 |
Labels |
priority::normal, tickettype::task, tool, version::6 |
Link |
|
Merges |
0 |
This directory is not used and needs to go. I will push the change directly to the repo to avoid the clutter a review would generate as most of the files and imported.
Author: Joel Sherrill
2022-03-03T22:54:55.000Z
Is this done? Easy to close?
Author: Joel Sherrill
2022-12-28T19:51:02.000Z
Please check if this can be closed.
Author: Gedare Bloom
2023-01-23T20:41:52.000Z
https://git.rtems.org/rtems-tools/tree/doc
Still exists.
Author: Joel Sherrill
2023-02-09T16:04:32.000Z
I think that much of the content in rtems-tools/doc is NOT in the Users Guide. The discussion of the RTEMS Tester in the Users Guide takes a different view and appears to never show how to use it which is the focus of much of this documentation. I don’t think this is a simple remove some files.
Author: Joel Sherrill
2023-06-22T21:21:57.000Z
Owner set to @kinsey
Author: Amar Takhar
2024-05-17T02:01:29.459Z
Can this be moved to %6.2?
Author: Kinsey Moore
2024-05-17T02:54:17.851Z
This is fine to move to %”6.2”.
Author: Kinsey Moore
2024-05-17T02:54:22.287Z
changed the description
Author: Gedare Bloom
2024-12-04T21:40:36.469Z
3230 - RSB does not report –rsb-file for patches correctly. (opened)¶
Id |
3230 |
State |
opened |
Type |
ISSUE |
Author |
Chris Johns |
Assignee(s) |
Chris Johns |
Created |
2017-11-12T04:45:41.000Z |
Updated |
2024-12-04T21:40:35.420Z |
Milestone |
6.2 |
Labels |
priority::normal, tickettype::defect, tool::rtems-source-builder, version::5 |
Link |
|
Merges |
0 |
The reports from the RSB do not report the --rsb-file option correctly.
Author: Chris Johns
2017-11-12T04:46:01.000Z
Status changed from assigned to accepted
Author: Chris Johns
2019-12-12T23:32:22.000Z
Milestone changed from %”5.1” to %”6.1”
Author: Amar Takhar
2024-05-17T01:59:37.817Z
changed the description
Author: Amar Takhar
2024-05-17T02:00:00.357Z
Moving to %6.2 since it’s been bumped once already.
Author: Gedare Bloom
2024-12-04T21:40:35.716Z
Merge Requests¶
548 - cpu/riscv: Add Smdbltrp extension compatibility to 6 branch¶
Id |
548 |
State |
merged |
Merged by |
Kinsey Moore |
Merged at |
2025-07-03T13:20:06.894Z |
Created |
2025-07-03T08:10:40.579Z |
Updated |
2025-07-03T13:20:06.437Z |
Milestone |
6.2 |
Labels |
arch:riscv |
Link |
https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/548 |
Commit |
Summary¶
cpu/riscv: Add Smdbltrp extension compatibility
If the Double Trap Extension is implemented, the MDT bit of the mstatus (or mstatush in RV32) register will be set when a trap is to be taken.
The MIE (Machine Interrupt Enable) bit can only be set to 1 if the MDT bit is zero.
Thus, we need to clear MDT first if we want to enable interrupts when dispatching a thread.
MDT is also cleared in register a1 before restoring the interrupt frame as writing 1 to MDT will cause MIE to be set to 0. In RV64 this happens regardless of the value written to MIE in the same write.
In RV32, MDT is in the mstatush so we do not need to clear during restore as this register is not restored.
With this change all 60 SMP tests pass (compared to 20/60 before the fix). The tests have been run on hardware using two RV64 CPUs that implement the double trap extension.
Close #5288
(cherry picked from commit 19f12d2dca02da3072c0c31e2da281f6a447699e)
Author: Kinsey Moore
2025-07-03T13:11:23.395Z
approved this merge request
Author: Kinsey Moore
2025-07-03T13:11:30.083Z
enabled an automatic merge when all merge checks for e7fa90ab0ccd0566dd0b136c62aa729515c3b011 pass
Author: Joel Sherrill
2025-07-03T13:20:01.821Z
approved this merge request
537 - bsps/noelv: Backport polled mode console to 6¶
Id |
537 |
State |
merged |
Merged by |
Gedare Bloom |
Merged at |
2025-06-28T15:35:41.336Z |
Created |
2025-06-26T07:02:00.422Z |
Updated |
2025-06-28T15:36:37.701Z |
Milestone |
6.2 |
Labels |
arch:riscv, backport |
Link |
https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/537 |
Commit |
Summary¶
bsps/noelv: Fix using console in polled mode
Before, the console driver needed BSP_CONSOLE_USE_INTERRUPTS to be defined or it would not build. The intent was to use polled mode if the macro was equal to zero.
This change makes it so interrupt mode is used if the macro is defined and polled mode is used if the macro is not defined.
(cherry picked from commit 5e0a68d3ab6bedb16e465267310c59c25803b783)
Close #5283
Author: Joel Sherrill
2025-06-27T13:32:10.089Z
This needs an issue for the backport and the issue needs to be associated with the 6.2 release epic.
If that needs clarification, this is documented somewhere. Ask on Discord and hopefully someone can give you a pointer.
If you need help, ask on Discord.
Author: Matteo Concas
2025-06-27T13:32:10.089Z
I created issue #5283, but I can’t seem to add it to the 6.2 release epic or add labels to it. I guess only maintainers have these rights.
Author: Matteo Concas
2025-06-27T07:42:39.176Z
added 1 commit
dd490c5c - bsps/noelv: Fix using console in polled mode
Author: Matteo Concas
2025-06-27T07:43:20.752Z
changed the description
Author: Kinsey Moore
2025-06-27T13:32:10.120Z
resolved all threads
Author: Kinsey Moore
2025-06-27T13:32:14.452Z
approved this merge request
Author: Gedare Bloom
2025-06-28T14:53:54.575Z
approved this merge request
532 - bsps/sparc/leon3: Backport GPTIMER timer index logic fix to 6¶
Id |
532 |
State |
merged |
Merged by |
Kinsey Moore |
Merged at |
2025-06-26T02:36:21.764Z |
Assignee(s) |
Matteo Concas |
Created |
2025-06-23T07:28:42.604Z |
Updated |
2025-06-26T02:36:21.748Z |
Milestone |
6.2 |
Labels |
backport |
Link |
https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/532 |
Commit |
Summary¶
bsps/sparc/leon3: Fix GPTIMER timer index logic
The old logic would lead to an error when multiprocessing was enabled and LEON3_GPTIMER_BASE was defined due to leon3_timer_core_index being undefined.
The new logic fixes this and keeps the same intent:
If multiprocessing is not enabled, the timer index is 0
If multiprocessing is enabled and
LEON3_GPTIMER_BASEis defined, the timer index is twice the CPU boot indexIf multiprocessing is enabled and
LEON3_GPTIMER_BASEis not defined, we fallback to the old logic using the GPTIMER core index.
Author: Matteo Concas
2025-06-23T07:29:20.225Z
changed title from bsps/sparc/leon3: {-Fix-} GPTIMER timer index logic to bsps/sparc/leon3: {+Backport+} GPTIMER timer index logic{+ fix to 6+}
Author: Matteo Concas
2025-06-23T07:30:22.977Z
mentioned in merge request !514
Author: Gedare Bloom
2025-06-25T16:24:56.538Z
@matteo.concas please add closes #5281 to the commit message someone will have to manually close it also, after the backport
Author: Matteo Concas
2025-06-25T16:24:56.538Z
I have added “Close #5281” to the commit message.
Author: Gedare Bloom
2025-06-24T20:31:45.106Z
assigned to @matteo.concas
Author: Matteo Concas
2025-06-25T06:23:00.842Z
added 1 commit
d010f6ae - bsps/sparc/leon3: Fix GPTIMER timer index logic
Author: Gedare Bloom
2025-06-25T16:24:56.595Z
resolved all threads
Author: Gedare Bloom
2025-06-25T16:26:16.722Z
approved this merge request
Author: Joel Sherrill
2025-06-25T22:20:48.635Z
approved this merge request
Author: Kinsey Moore
2025-06-26T02:36:07.135Z
enabled an automatic merge when the pipeline for d010f6ae8bc4e2ccfcc162b256e18ed61b71820d succeeds
524 - cpukit/libdebugger: Prevent hang on memory access¶
Id |
524 |
State |
merged |
Merged by |
Kinsey Moore |
Merged at |
2025-06-19T22:52:50.613Z |
Assignee(s) |
Kinsey Moore |
Reviewer(s) |
Chris Johns |
Created |
2025-06-18T18:55:15.610Z |
Updated |
2025-06-19T22:52:53.452Z |
Milestone |
6.2 |
Labels |
arch:aarch64, cpukit::debugger |
Link |
https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/524 |
Commit |
Summary¶
cpukit/libdebugger: Prevent hang on memory access
When memory is accessed by the remote debugging client, the access is sandboxed with setjmp/longjmp and appropriate exception handlers to prevent the attempted access from causing a failure of the debugger or otherwise altering execution. The existing implementation works as expected when the context executing the memory access and the exception context resulting from a failed access do not share a stack.
In the case of AArch64, a failed access when the debugger is already in exception context causes a re-entry into exception context where the machine state is pushed onto the same stack that was in use where the exception occurred. When setjmp is called inside a stack frame and the exception occurs outside that stack frame, the stack frame is unwound before the exception occurs and the exception entry overwrites the area previously occupied by the stack frame housing the setjmp and corrupting the link register that is stored there. After restoration of state using longjmp(), this corrupted link register information is loaded from the stack frame and undesired behavior occurs.
In the instance of this bug that was encountered, the corrupted link register contained an unaligned pointer which caused an unending cascade of prefetch abort exceptions presenting as a hard hang.
Closes #5273
Author: Kinsey Moore
2025-06-18T18:55:16.294Z
requested review from @chris
Author: Kinsey Moore
2025-06-18T18:55:16.326Z
assigned to @opticron
Author: Chris Johns
2025-06-19T22:36:09.770Z
approved this merge request
Author: Joel Sherrill
2025-06-19T22:48:37.839Z
approved this merge request
354 - score: Fix RTEMS_DEBUG build¶
Id |
354 |
State |
merged |
Merged by |
Kinsey Moore |
Merged at |
2025-01-08T13:31:58.965Z |
Assignee(s) |
Sebastian Huber |
Created |
2025-01-08T05:42:37.595Z |
Updated |
2025-05-09T00:10:25.214Z |
Milestone |
6.2 |
Labels |
rtems::score |
Link |
https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/354 |
Commit |
Summary¶
score: Fix RTEMS_DEBUG build
Close #5159.
Author: Sebastian Huber
2025-01-08T05:42:37.873Z
assigned to @sebhub
Author: Chris Johns
2025-01-08T05:43:50.581Z
approved this merge request
Author: Kinsey Moore
2025-01-08T13:31:38.056Z
approved this merge request
416 - machine/timecounters: Add _Timecounter_Getboottime, _Timecounter_Getboottimebin decls¶
Id |
416 |
State |
merged |
Merged by |
Kinsey Moore |
Merged at |
2025-05-08T22:38:11.262Z |
Assignee(s) |
Chris Johns |
Created |
2025-02-11T22:27:17.904Z |
Updated |
2025-05-08T22:38:15.309Z |
Milestone |
6.2 |
Labels |
rtems::kernel |
Link |
https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/416 |
Commit |
Summary¶
Add _Timecounter_Getboottime, _Timecounter_Getboottimebin decls
Closes #5212
Author: Chris Johns
2025-02-11T22:27:18.157Z
assigned to @chris
Author: Kinsey Moore
2025-05-08T22:36:28.879Z
This fails the commit message line length checks.
Author: Chris Johns
2025-05-08T22:36:01.285Z
added 1 commit
993962be - machine/timecounters: Add missing _Timecounter_ decls
Author: Joel Sherrill
2025-05-08T22:36:28.914Z
resolved all threads
Author: Joel Sherrill
2025-05-08T22:36:37.023Z
approved this merge request
Author: Kinsey Moore
2025-05-08T22:36:54.367Z
approved this merge request
Author: Kinsey Moore
2025-05-08T22:37:04.928Z
added 4 commits
993962be…f769b20c - 3 commits from branch
rtems/rtos:6328e83ec - machine/timecounters: Add missing _Timecounter_ decls
Author: Chris Johns
2025-05-08T22:37:32.553Z
added 1 commit
6335d7e4 - machine/timecounters: Add missing _Timecounter_ decls
461 - cpukit/libdl: Fix loading symbols from an object file at runtime¶
Id |
461 |
State |
merged |
Merged by |
Amar Takhar |
Merged at |
2025-04-04T00:06:34.500Z |
Assignee(s) |
Chris Johns |
Created |
2025-04-01T01:34:14.150Z |
Updated |
2025-04-04T00:06:36.822Z |
Milestone |
6.2 |
Labels |
cpukit::dl, tool::rtems-tools |
Link |
https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/461 |
Commit |
Summary¶
cpukit/libdl: Fix loading symbols from an object file at runtime
Assume a relocation record with a symbol name with a length of 0 is resolved. ARM seems to create a symbol with no name for R_ARM_V4BX relocation records.
Move the addition of the rtems_rtl_base_sym_global_add symbol to the global symbol table to the weak rtems_rtl_base_global_syms_init call. If symbols are embedded the support for runtime loading symbols is over loaded. This change is required so the base object has a valid global symbol table attached to track dependencies.
Fixes #5236
Author: Chris Johns
2025-04-01T01:34:14.397Z
assigned to @chris
Author: Kinsey Moore
2025-04-03T23:01:59.004Z
approved this merge request
Author: Amar Takhar
2025-04-04T00:06:28.081Z
approved this merge request
459 - Backport “spec: Install missing header files for GRLIB and SPARC” to 6 branch¶
Id |
459 |
State |
merged |
Merged by |
Joel Sherrill |
Merged at |
2025-03-26T16:50:31.035Z |
Created |
2025-03-26T16:48:45.767Z |
Updated |
2025-03-26T16:50:33.271Z |
Milestone |
6.2 |
Link |
https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/459 |
Commit |
Backport commit spec: Install missing header files for GRLIB and SPARC on main to 6 branch.
Closes issue #5232
Author: Kinsey Moore
2025-03-26T16:49:31.177Z
approved this merge request
Author: Joel Sherrill
2025-03-26T16:50:20.843Z
approved this merge request
418 - gitlab: Add link to CI¶
Id |
418 |
State |
merged |
Merged by |
Chris Johns |
Merged at |
2025-02-13T03:24:20.100Z |
Assignee(s) |
Amar Takhar |
Created |
2025-02-12T13:40:24.076Z |
Updated |
2025-03-13T22:44:56.007Z |
Milestone |
6.2 |
Labels |
ci |
Link |
https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/418 |
Commit |
Summary¶
gitlab: Add link to CI
This runs only the commit message and merge request checker.
Required so we can have ‘all pipelines must pass’ enabled due to a bug in Gitlab
Author: Amar Takhar
2025-02-12T13:40:24.364Z
assigned to @amar
Author: Amar Takhar
2025-02-12T14:04:32.289Z
Once this is merged I can enable enforcing pipelines must pass.
Author: Kinsey Moore
2025-02-12T14:46:29.753Z
approved this merge request
Author: Chris Johns
2025-02-13T03:18:41.727Z
changed the description
Author: Chris Johns
2025-02-13T03:24:10.796Z
Commit message as “ue” instead of “due”
Author: Amar Takhar
2025-02-13T03:22:35.416Z
added 1 commit
b4149b22 - gitlab: Add link to CI
Author: Chris Johns
2025-02-13T03:24:10.814Z
resolved all threads
Author: Chris Johns
2025-02-13T03:24:15.303Z
approved this merge request
413 - Revert “build: Provide LDFLAGS for pkg-config”¶
Id |
413 |
State |
merged |
Merged by |
Kinsey Moore |
Merged at |
2025-02-11T13:51:35.086Z |
Assignee(s) |
Chris Johns |
Created |
2025-02-11T05:43:37.231Z |
Updated |
2025-02-11T13:51:37.304Z |
Milestone |
6.2 |
Link |
https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/413 |
Commit |
Summary¶
Revert “build: Provide LDFLAGS for pkg-config”
This reverts commit e36ba91110724b5cc8a9e07659d98d0bcb96e4d4
Updates #5165
See https://gitlab.rtems.org/rtems/pkg/rtems-libbsd/-/issues/38
Author: Chris Johns
2025-02-11T05:50:43.847Z
changed the description
Author: Chris Johns
2025-02-11T05:50:43.893Z
assigned to @chris
Author: Chris Johns
2025-02-11T05:52:05.406Z
changed the description
Author: Amar Takhar
2025-02-11T05:55:01.814Z
approved this merge request
Author: Kinsey Moore
2025-02-11T13:51:20.391Z
approved this merge request
345 - bsp/qoriq: Ignore spurious interrupts¶
Id |
345 |
State |
merged |
Merged by |
Kinsey Moore |
Merged at |
2025-02-07T23:20:29.334Z |
Assignee(s) |
Sebastian Huber |
Created |
2024-12-06T10:22:28.402Z |
Updated |
2025-02-07T23:20:33.215Z |
Milestone |
6.2 |
Labels |
arch:powerpc, backport |
Link |
https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/345 |
Commit |
Summary¶
bsp/qoriq: Ignore spurious interrupts
For example, with edge triggered external interrupts we may see spurious interrupts. Ignore them instead of issuing a fatal error.
Use eieio to synchronize access to the IACK and EOI registers.
Use a loop to immediately services the next pending interrupt without having to go through the exception epiloge and prologue.
Close #5173.
Author: Sebastian Huber
2024-12-06T10:22:28.560Z
assigned to @sebhub
Author: Gedare Bloom
2024-12-06T16:10:09.706Z
approved this merge request
Author: Kinsey Moore
2024-12-06T16:19:36.472Z
added 2 commits
a357119b - 1 commit from branch
rtems/rtos:681ed4b2d - bsp/qoriq: Ignore spurious interrupts
Author: Kinsey Moore
2025-02-07T22:57:58.139Z
This needs the same fix as the other MR.
Author: Gedare Bloom
2025-02-07T23:17:29.509Z
changed this line in version 6 of the diff
Author: Kinsey Moore
2024-12-06T16:20:26.526Z
requested changes
Author: Gedare Bloom
2025-02-03T18:59:36.917Z
unapproved this merge request
Author: Kinsey Moore
2025-02-06T23:49:46.426Z
added 9 commits
81ed4b2d…9641e1e9 - 8 commits from branch
rtems/rtos:670cb4a1c - bsp/qoriq: Ignore spurious interrupts
Author: Kinsey Moore
2025-02-07T22:56:47.362Z
added 3 commits
70cb4a1c…e36ba911 - 2 commits from branch
rtems/rtos:6f1e807b2 - bsp/qoriq: Ignore spurious interrupts
Author: Kinsey Moore
2025-02-07T22:57:45.129Z
added 1 commit
18c59283 - bsp/qoriq: Ignore spurious interrupts
Author: Kinsey Moore
2025-02-07T22:57:58.154Z
resolved all threads
Author: Kinsey Moore
2025-02-07T22:58:41.186Z
approved this merge request
Author: Gedare Bloom
2025-02-07T23:17:29.818Z
added 1 commit
be39d7cc - bsp/qoriq: Ignore spurious interrupts
Author: Gedare Bloom
2025-02-07T23:17:45.099Z
reset approvals from @opticron by pushing to the branch
Author: Gedare Bloom
2025-02-07T23:18:36.678Z
added 1 commit
e421c922 - bsp/qoriq: Ignore spurious interrupts
Author: Gedare Bloom
2025-02-07T23:19:38.690Z
approved this merge request
Author: Kinsey Moore
2025-02-07T23:20:19.442Z
approved this merge request
335 - build: Provide LDFLAGS for pkg-config (6.1)¶
Id |
335 |
State |
merged |
Merged by |
Joel Sherrill |
Merged at |
2025-02-07T18:29:24.490Z |
Assignee(s) |
Sebastian Huber |
Created |
2024-11-28T14:26:56.054Z |
Updated |
2025-02-07T18:29:26.896Z |
Milestone |
6.2 |
Link |
https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/335 |
Commit |
Summary¶
build: Provide LDFLAGS for pkg-config
Some pkg-config variants perform transformations on the –libs options. This may lead to completely broken linker options. Provide the LDFLAGS as a variable. Remove “Ldflags:” since this is an unsupported field.
Update #5165.
Author: Sebastian Huber
2024-11-28T14:26:56.172Z
assigned to @sebhub
Author: Sebastian Huber
2024-11-28T14:28:45.629Z
changed title from build: Provide LDFLAGS for pkg-config to build: Provide LDFLAGS for pkg-config{+ (6.1)+}
Author: Chris Johns
2024-11-28T18:55:33.768Z
approved this merge request
Author: Chris Johns
2024-11-28T18:56:10.244Z
unapproved this merge request
Author: Kinsey Moore
2025-01-17T01:21:22.921Z
approved this merge request
Author: Kinsey Moore
2025-01-17T01:21:50.110Z
added 4 commits
0de8e928…34ba74f4 - 3 commits from branch
rtems/rtos:6b337ac32 - build: Provide LDFLAGS for pkg-config
Author: Kinsey Moore
2025-01-17T03:12:28.532Z
unapproved this merge request
Author: Kinsey Moore
2025-01-17T03:12:38.050Z
approved this merge request
Author: Sebastian Huber
2025-01-20T23:32:47.963Z
added 2 commits
f7de6d54 - 1 commit from branch
rtems/rtos:6c6242efc - build: Provide LDFLAGS for pkg-config
Author: Sebastian Huber
2025-01-20T23:33:00.694Z
reset approvals from @opticron by pushing to the branch
Author: Kinsey Moore
2025-01-24T01:06:37.734Z
added 5 commits
c6242efc…f1c201c5 - 4 commits from branch
rtems/rtos:66c845593 - build: Provide LDFLAGS for pkg-config
Author: Kinsey Moore
2025-01-24T01:06:45.116Z
approved this merge request
Author: Kinsey Moore
2025-02-06T23:48:13.858Z
added 2 commits
9641e1e9 - 1 commit from branch
rtems/rtos:67654de7c - build: Provide LDFLAGS for pkg-config
Author: Kinsey Moore
2025-02-07T18:28:08.013Z
added 2 commits
a0e4be53 - 1 commit from branch
rtems/rtos:6e36ba911 - build: Provide LDFLAGS for pkg-config
Author: Joel Sherrill
2025-02-07T18:29:12.972Z
approved this merge request
404 - grlib/occan: Fix baud rate calculation¶
Id |
404 |
State |
merged |
Merged by |
Joel Sherrill |
Merged at |
2025-02-07T18:03:48.749Z |
Assignee(s) |
Gedare Bloom |
Created |
2025-02-07T15:51:47.539Z |
Updated |
2025-02-07T18:03:50.977Z |
Milestone |
6.2 |
Labels |
backport, grlib |
Link |
https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/404 |
Commit |
Summary¶
grlib/occan: Fix baud rate calculation
Fixes #5205
Closes #5205
Author: Gedare Bloom
2025-02-07T15:51:47.744Z
assigned to @gedare
Author: Kinsey Moore
2025-02-07T18:01:17.302Z
approved this merge request
Author: Joel Sherrill
2025-02-07T18:03:40.618Z
approved this merge request
381 - cpukit/termios: Fix ordering of baud rate table¶
Id |
381 |
State |
merged |
Merged by |
Sebastian Huber |
Merged at |
2025-02-06T03:35:42.316Z |
Assignee(s) |
Gedare Bloom |
Created |
2025-02-03T19:05:27.637Z |
Updated |
2025-02-06T03:35:45.233Z |
Milestone |
6.2 |
Labels |
cpukit |
Link |
https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/381 |
Commit |
Summary¶
cpukit/termios: Fix ordering of baud rate table
rtems_termios_set_best_baud expects a sorted baud rate table.
Fixes #5202
Closes #5202
Author: Gedare Bloom
2025-02-03T19:05:27.858Z
assigned to @gedare
Author: Kinsey Moore
2025-02-03T19:10:11.840Z
approved this merge request
Author: Sebastian Huber
2025-02-06T03:35:22.184Z
approved this merge request
361 - cpukit/libmisc: fix flag in capture.c¶
Id |
361 |
State |
merged |
Merged by |
Kinsey Moore |
Merged at |
2025-01-23T23:49:48.189Z |
Assignee(s) |
Gedare Bloom |
Created |
2025-01-21T00:00:22.382Z |
Updated |
2025-01-23T23:49:50.563Z |
Milestone |
6.2 |
Labels |
cpukit::misc |
Link |
https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/361 |
Commit |
Summary¶
cpukit/libmisc: fix flag in capture.c
The RTEMS_CAPTURE_OVERFLOW is an overflow flag for each CPU, and its value is the same as RTEMS_CAPTURE_INIT. Executing rtems_capture_flush will set the RTEMS_CAPTURE_INIT flag to 0 in the global flags.
Fixes #5184.
Closes #5184
Author: Gedare Bloom
2025-01-21T00:00:24.030Z
assigned to @gedare
Author: Chris Johns
2025-01-23T23:45:11.771Z
added 3 commits
151074b3…84c2cf3d - 2 commits from branch
rtems/rtos:6603c168a - cpukit/libmisc: fix flag in capture.c
Author: Chris Johns
2025-01-23T23:45:36.805Z
approved this merge request
Author: Kinsey Moore
2025-01-23T23:49:43.026Z
approved this merge request
367 - cpukit/libdl/arm: Fix trampoline alignment¶
Id |
367 |
State |
merged |
Merged by |
Chris Johns |
Merged at |
2025-01-23T23:37:30.167Z |
Created |
2025-01-23T22:05:59.402Z |
Updated |
2025-01-23T23:37:32.473Z |
Milestone |
6.2 |
Labels |
arch:arm, cpukit::dl |
Link |
https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/367 |
Commit |
Summary¶
cpukit/libdl/arm: Fix trampoline alignment
This commit aligns trampolines for THUMB and ARM instructions (CALL/JUMP24 and THM_JUMP24/THM_PC22).
According to the ARM technical reference in section “Register-relative and PC-relative expressions”:
In Thumb code: * For B, BL, CBNZ, and CBZ instructions, the value of the PC is the address of the current instruction plus 4 bytes. * For all other instructions that use labels, the value of the PC is the address of the current instruction plus 4 bytes, with bit[1] of the result cleared to 0 to make it word-aligned.
Closes #5189
Author: Kinsey Moore
2025-01-23T22:49:30.326Z
I have created #5189 for this MR to reference with appropriate epic and milestone. Could you please reference it with “Closes #5189” on a new line at the end of the commit message?
Author: Ranulfo Raphael
2025-01-23T22:49:30.326Z
Done. Thanks.
Author: Ranulfo Raphael
2025-01-23T22:41:09.917Z
added 1 commit
84c2cf3d - cpukit/libdl/arm: Fix trampoline alignment
Author: Ranulfo Raphael
2025-01-23T22:47:35.033Z
changed the description
Author: Ranulfo Raphael
2025-01-23T22:49:30.342Z
resolved all threads
Author: Kinsey Moore
2025-01-23T23:03:23.524Z
approved this merge request
Author: Chris Johns
2025-01-23T23:37:24.885Z
approved this merge request
348 - libmisc/stackchk: Add stack smashing protection (opened)¶
Id |
348 |
State |
opened |
Assignee(s) |
Joel Sherrill |
Reviewer(s) |
Gedare Bloom |
Created |
2024-12-19T06:34:14.290Z |
Updated |
2025-07-03T01:13:57.653Z |
Milestone |
6.2 |
Labels |
backport |
Link |
https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/348 |
Commit |
Summary¶
libmisc/stackchk: Add stack smashing protection
This is a backport for merge request 277! to 6 branch.
Note¶
Closes #5180
Author: Mohamed Hassan
2024-12-19T06:41:59.787Z
mentioned in issue #5180
Author: Gedare Bloom
2024-12-19T16:33:40.221Z
assigned to @joel
Author: Gedare Bloom
2024-12-19T16:35:21.370Z
requested review from @gedare
Author: Gedare Bloom
2024-12-19T16:35:37.534Z
Needs the same fixes as !277
Author: Mohamed Hassan
2024-12-20T15:04:15.023Z
Done.
Author: Gedare Bloom
2025-02-07T18:30:50.708Z
Please add Closes #5180
Author: Mohamed Hassan
2025-02-07T18:30:50.708Z
Understood & done.
Author: Mohamed Hassan
2024-12-20T14:49:15.952Z
added 1 commit
066b355c - libmisc/stackchk: Add stack smashing protection
Author: Mohamed Hassan
2024-12-20T15:02:21.630Z
added 1 commit
30d7008a - libmisc/stackchk: Add stack smashing protection
Author: Mohamed Hassan
2024-12-20T15:03:35.573Z
changed the description
Author: Mohamed Hassan
2024-12-23T20:25:19.593Z
added 2 commits
d61a739e - 1 commit from branch
rtems/rtos:6b50f74c9 - libmisc/stackchk: Add stack smashing protection
Author: Kinsey Moore
2025-02-06T23:51:32.758Z
added 8 commits
b50f74c9…9641e1e9 - 7 commits from branch
rtems/rtos:69c5f1ae7 - libmisc/stackchk: Add stack smashing protection
Author: Amar Takhar
2025-07-03T01:14:00.936Z
added 13 commits
9c5f1ae7…dd490c5c - 12 commits from branch
rtems/rtos:66a6881cb - libmisc/stackchk: Add stack smashing protection
424 - build: Provide LDFLAGS for pkg-config (opened)¶
Id |
424 |
State |
opened |
Assignee(s) |
Sebastian Huber |
Created |
2025-02-15T05:42:39.462Z |
Updated |
2025-07-02T18:25:40.016Z |
Milestone |
6.2 |
Link |
https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/424 |
Commit |
Summary¶
Some pkg-config variants perform transformations on the –libs options. This may lead to completely broken linker options:
https://github.com/pkgconf/pkgconf/issues/371
Provide the LDFLAGS as a variable. Do not add the ABI_FLAGS to LDFLAGS since they are already contained in the CFLAGS. The linker command line should include the compiler flags so that link-time optimization works properly.
Update #5165.
Author: Sebastian Huber
2025-02-15T05:42:39.627Z
assigned to @sebhub
Author: Sebastian Huber
2025-02-15T05:44:12.602Z
changed title from {-cpukit/jffs2/rtime: Fix off-by-one error in decompression check-} to {+build: Provide LDFLAGS for pkg-config+}
Author: Sebastian Huber
2025-02-15T05:44:12.631Z
changed the description
Author: Sebastian Huber
2025-02-15T05:44:12.670Z
changed target branch from
mainto6
Author: Sebastian Huber
2025-02-15T05:48:57.400Z
I created the MR with the wrong target branch. Now the pipeline is confused.
Author: Amar Takhar
2025-02-15T05:59:03.534Z
Yes, we actually enacted a new policy and adding to gitlab that branches targeted towards branches need to start with
5-*or6-*gitlab will then automatically set the branch.This just happened no doc updating the old way is still fine but this way this CI issue won’t happen. Let me try and see if restarting the pipeline manually fixes it.
Author: Amar Takhar
2025-02-15T06:00:16.820Z
Nope that didn’t fix it let me see how to properly handle this situation it may be how I have the CI setup
Author: Amar Takhar
2025-02-16T09:50:17.653Z
I think we’re stuck unless you do an update the source or we have to rebase on it. GitLab will snapshot the state when an MR comes in and use that. I have seen complaints about this exact issue but no solutions anywhere so we’ll wait until there is a chance to rebase.
Author: Kinsey Moore
2025-03-04T15:20:39.238Z
approved this merge request
Author: Amar Takhar
2025-07-02T18:25:44.075Z
added 8 commits
a5e5c058…dd490c5c - 7 commits from branch
rtems/rtos:63e90c378 - build: Provide LDFLAGS for pkg-config
Author: Amar Takhar
2025-07-02T18:25:58.823Z
reset approvals from @opticron by pushing to the branch
518 - cpukit/libdebugger: Add support for TLS variables (opened)¶
Id |
518 |
State |
opened |
Assignee(s) |
Kinsey Moore |
Reviewer(s) |
Chris Johns |
Created |
2025-06-17T19:19:24.345Z |
Updated |
2025-06-26T00:48:21.103Z |
Milestone |
6.2 |
Labels |
backport |
Link |
https://gitlab.rtems.org/rtems/rtos/rtems/-/merge_requests/518 |
Commit |
Summary¶
cpukit/libdebugger: Add support for TLS variables
This adds support for the “vGetTLSAddr” GDB query which retrieves the address of a TLS variable by offset from the beginning of the TLS memory space for the given thread. This offset does not include the size of the thread control block which is at the beginning of every TLS area as used by RTEMS. Notably, the returned address is big-endian rather than the little-endian typical with other responses.
This functionality does not include retrieval of addresses for TLS variables hosted in loadable modules, only TLS variables in the host binary.
Author: Kinsey Moore
2025-06-17T19:19:24.986Z
requested review from @chris
Author: Kinsey Moore
2025-06-17T19:19:25.012Z
assigned to @opticron
Author: Kinsey Moore
2025-06-18T18:59:43.036Z
changed the description
Author: Kinsey Moore
2025-06-18T19:01:15.967Z
mentioned in issue #4883
Author: Kinsey Moore
2025-06-25T11:42:20.014Z
added 1 commit
14329bc2 - cpukit/libdebugger: Add support for TLS variables
Author: Gedare Bloom
2025-06-25T16:35:42.812Z
approved this merge request
Author: Kinsey Moore
2025-06-26T00:48:21.060Z
added 1 commit
4f819d02 - cpukit/libdebugger: Add support for TLS variables