Rant: Why do G4 headers still use uint32_t in #define?

[Uwe Bonnes]

Hello,

I have reported several time about the miss-use of uint32_t in #define. With G4, STM coders had the chance to get things right from the start. However they did not use the chance. Must every device group at STM make its own errors and does not learn from other groups?

Please remove uint32_t from #define.

And have a look at trailing whitespace. All headers are sprankled with trailing whitespace...

[Imen.D]

Hello @Uwe Bonnes​ ,

>> And have a look at trailing whitespace. All headers are sprankled with trailing whitespace...

Can you please provide more details, or example regarding this feedback.

Regards,

Imen

[Uwe Bonnes]

There are lot of definitions like

> stm32g484xx.h:#define SYSCFG_SWPR_PAGE31_Msk        (uint32_t)(0x1UL << SYSCFG_SWPR_PAGE31_Pos)

The (uint32_t) cast is superfluous. A defintion like

/stm32g484xx.h:#define SYSCFG_SWPR_PAGE31_Msk        (0x1UL << SYSCFG_SWPR_PAGE31_Pos)

is enough. The superfluous cast also inhibits some constructs like

#If (bla) < SYSCFG_SWPR_PAGE31_Msk

[Uwe Bonnes]

Probably I miss-interpreted the DOS Line-endings (CR + LF) that triggered git log --check warnings. I do not see any real trailing whitesspace any more.

[Imen.D]

Thanks @Uwe Bonnes​  for this clarification.

Your feedback is reported internally to fix in the coming release.

Regards,

Imen

[Uwe Bonnes]

STMicroelectronics ST-LINK GDB server. Version 5.3.2 now understands the -a option

[Uwe Bonnes]

Sorry, please disregard, wrong thread.