When the architectural support for STM32G4 family was added, the
reference manual (RM0440) was at revision 2. Since then, it has
undergone several revisions. One significant change is in the
table of FLASH wait states: section 3.3.3 table 9. The outcome
of this change is that fewer FLASH wait states are needed for
most CPU clock (HCLK) frequencies. Notably, if running the CPU
clock at the maximum 170 MHz, only 4 FLASH wait states are
needed, rather than the previously programmed 8 wait states.
This gives a noticeable performance boost.
arch/arm/src/stm32/stm32g4xxxx_rcc.c:
* FLASH_ACR_LATENCY_SETTING: Reimplement compile-time logic
that selects the required wait state setting to use the new
updated table.
* Update all comments to indicate that RM0440 Rev 5 is used.
* Update section numbers mentioned in comments in cases where
they have changed due to added sections in the manual.
arch/arm/src/stm32/hardware/stm32_adc_v2g4.h:
* New file.
arch/arm/src/stm32/hardware/stm32_adc.h:
* Distinguish between the normal STM32 ADC IPv2 core and the
modified IPv2 core used in the G4 family, and include either
stm32_adc_v2.h or stm32_adc_v2g4.h as needed.
Summary:
- This commit introduces SP_WFE() and SP_SEV() to be used for spinlock
- Also, use wfe/sev instructions for ARMV7-A to reduce power consumption
Impact:
- ARMV7-a SMP only
Testing:
- sabre-6quad:smp (QEMU, dev board)
- maix-bit:smp, esp32-devkitc:smp, spresense:smp sim:smp (compile only)
Signed-off-by: Masayuki Ishikawa <Masayuki.Ishikawa@jp.sony.com>
Summary:
- Because this_task() returns the current task of the current CPU
Impact:
- SMP only
Testing:
- Tested with sabre-6quad:smp (QEMU)
Signed-off-by: Masayuki Ishikawa <Masayuki.Ishikawa@jp.sony.com>
Summary:
- This commit fixes global IRQ control logic
- In previous implementation, g_cpu_irqset for a remote CPU was
set in sched_add_readytorun(), sched_remove_readytorun() and
up_schedule_sigaction()
- In this implementation, they are removed.
- Instead, in the pause handler, call enter_critical_setion()
which will call up_cpu_paused() then acquire g_cpu_irqlock
- So if a new task with irqcount > 1 restarts on the remote CPU,
the CPU will only hold a critical section. Thus, the issue such as
'POSSIBLE FOR TWO CPUs TO HOLD A CRITICAL SECTION' could be resolved.
- Fix nxsched_resume_scheduler() so that it does not call spin_clrbit()
if a CPU does not hold a g_cpu_irqset
- Fix nxtask_exit() so that it acquires g_cpu_irqlock
- Update TODO
Impact:
- All SMP implementations
Testing:
- Tested with smp, ostest with the following configurations
- Tested with spresense:wifi_smp (NCPUS=2,4)
- Tested with sabre-6quad:smp (QEMU, dev board)
- Tested with maix-bit:smp (QEMU)
- Tested with esp32-core:smp (QEMU)
- Tested with lc823450-xgevk:rndis
Signed-off-by: Masayuki Ishikawa <Masayuki.Ishikawa@jp.sony.com>
