Update Renesas GCC compiler #89

VanNamDinh · 2020-07-16T07:51:59Z

Description

Port for GCC RX Family (RX100, RX200, RX600 and RX600v2). This work with new FIT version from Renesas as well. The RX GCC of Amazon FreeRTOS has not been updated for a while so that GCC projects can't be built and run.

By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.

Signed-off-by: Dinh Van Nam <[email protected]>

* RX600v2 * RX600 * RX100 Signed-off-by: Dinh Van Nam <[email protected]>

NoMaY-jp · 2020-07-17T14:58:00Z

Hello Dinh Van Nam,

The configINCLUDE_PLATFORM_H_INSTEAD_OF_IODEFINE_H and R_BSP_SECNAME_INTVECTTBL were introduced by me when I ported Amazon FreeRTOS to RX MCU and CC-RX/GNURX. Now I think that the following is better for GNURX.

Your proposal (i.e. Renesas RX FreeRTOS Package v10.0.3):

port.c

/* Hardware specifics. */
#if defined(configINCLUDE_PLATFORM_H_INSTEAD_OF_IODEFINE_H) && (configINCLUDE_PLATFORM_H_INSTEAD_OF_IODEFINE_H == 1)
#include "platform.h"
#else
#include "iodefine.h"
#endif
...omit...

/*
 * Software interrupt handler.  Performs the actual context switch (saving and
 * restoring of registers).  Written in asm code as direct register access is
 * required.
 */
void vPortSoftwareInterruptISR( void ) __attribute__( ( naked ) );
#if defined(configTICK_VECTOR)
void vPortSoftwareInterruptISR( void ) __attribute__((naked, vector( R_BSP_SECNAME_INTVECTTBL, VECT_ICU_SWINT )));
#else
void vPortSoftwareInterruptISR( void ) __attribute__((naked));
#endif

/*
 * The tick interrupt handler.
 */
void vPortTickISR( void ) __attribute__( ( interrupt ) );

#if defined(configTICK_VECTOR)
void vPortTickISR( void ) __attribute__((interrupt( R_BSP_SECNAME_INTVECTTBL, _VECT( configTICK_VECTOR ) )));
#else
void vPortTickISR( void ) __attribute__((interrupt));
#endif

My proposal:

portmacro.h (The following is picked up from [my RX700v3 DPFPU's portmacro.h which is being developed](url https://github.com/NoMaY-jp/FreeRTOS-Kernel/blob/b6bb3a60c1a2138193d2bc9f996c53f51b5d4be4/portable/GCC/RX700v3_DPFPU/portmacro.h).)

/* When the FIT configurator or the Smart Configurator is used, platform.h has to be 
 * used. */
    #ifndef configINCLUDE_PLATFORM_H_INSTEAD_OF_IODEFINE_H
        #define configINCLUDE_PLATFORM_H_INSTEAD_OF_IODEFINE_H 0
    #endif

port.c (The following is picked up from [my RX700v3 DPFPU's port.c which is being developed](url https://github.com/NoMaY-jp/FreeRTOS-Kernel/blob/b6bb3a60c1a2138193d2bc9f996c53f51b5d4be4/portable/GCC/RX700v3_DPFPU/port.c).)

/* Hardware specifics. */
#if ( configINCLUDE_PLATFORM_H_INSTEAD_OF_IODEFINE_H == 1 )

    #include "platform.h"

#else /* configINCLUDE_PLATFORM_H_INSTEAD_OF_IODEFINE_H */

    #include "iodefine.h"

#endif /* configINCLUDE_PLATFORM_H_INSTEAD_OF_IODEFINE_H */

...omit...

/*
 * Software interrupt handler.  Performs the actual context switch (saving and
 * restoring of registers).  Written in asm code as direct register access is
 * required.
 */
#if ( configINCLUDE_PLATFORM_H_INSTEAD_OF_IODEFINE_H == 1 )

    R_BSP_PRAGMA_INTERRUPT( vSoftwareInterruptISR, VECT( ICU, SWINT ) )
    R_BSP_ATTRIB_INTERRUPT void vSoftwareInterruptISR( void ) __attribute__( ( naked ) );

#else /* configINCLUDE_PLATFORM_H_INSTEAD_OF_IODEFINE_H */

    void vSoftwareInterruptISR( void ) __attribute__( ( naked ) );

#endif /* configINCLUDE_PLATFORM_H_INSTEAD_OF_IODEFINE_H  */

/*
 * The tick ISR handler.  The peripheral used is configured by the application
 * via a hook/callback function.
 */
#if ( configINCLUDE_PLATFORM_H_INSTEAD_OF_IODEFINE_H == 1 )

    R_BSP_PRAGMA_INTERRUPT( vTickISR, _VECT( configTICK_VECTOR ) )
    R_BSP_ATTRIB_INTERRUPT void vTickISR( void ); /* Do not add __attribute__( ( interrupt ) ). */

#else /* configINCLUDE_PLATFORM_H_INSTEAD_OF_IODEFINE_H */

    void vTickISR( void ) __attribute__( ( interrupt ) );

#endif /* configINCLUDE_PLATFORM_H_INSTEAD_OF_IODEFINE_H */

Best regards,
NoMaY

NoMaY-jp · 2020-07-17T15:24:49Z

Hello Dinh Van Nam,

The following topic (both threads are the same topic) is not subject of your pull request, but I think that it is better for you to keep this topic in your mind if you don't know it.

[Improper access to an on-chip peripheral register in Renesas RX port layer](url https://forums.freertos.org/t/improper-access-to-an-on-chip-peripheral-register-in-renesas-rx-port-layer/9851/) <-- in English

[Question about FreeRTOS code](url https://japan.renesasrulz.com/cafe_rene/f/forum5/5878/freertos/) <-- Please note that original title and all contents are in Japanese

Best regards,
NoMaY

VanNamDinh · 2020-07-17T15:58:56Z

Hi NoMaY,

Thank you so much for your proposal and your contribution on RX MCU.

My team will consider your proposal in detail and resubmit pull request as soon as possible.
Please keep discussing on this channel.

Best regards,
Nam

ravibhagavandas · 2020-08-10T20:42:32Z

Hello @VanNamDinh

We are in the process of reviewing this PR. Are you considering to resubmit this PR with the changes from NoMaY as proposed above ?

ravibhagavandas · 2020-08-12T17:12:25Z

@VanNamDinh
It seems that existing demo RX200_RX231-RSK_GCC_e2studio_IAR build fails with this PR, due to the following error:

FreeRTOS/FreeRTOS/Source/portable/GCC/RX200/port.c:75:66: error: 'R_BSP_SECNAME_INTVECTTBL' undeclared here (not in a function)
 void vSoftwareInterruptISR( void ) __attribute__((naked, vector( R_BSP_SECNAME_INTVECTTBL, VECT_ICU_SWINT )));

Should the declaration above be guarded with configINCLUDE_PLATFORM_H_INSTEAD_OF_IODEFINE_H flag?

NoMaY-jp · 2020-08-12T18:10:57Z

Hello @ravibhagavandas,

Should the declaration above be guarded with configINCLUDE_PLATFORM_H_INSTEAD_OF_IODEFINE_H flag?

Yes. As I wrote before, Dinh Van Nam's proposal is based on my work when I ported Amazon FreeRTOS to RX MCU and CC-RX/GNURX two years ago. At that time, I did not notice that configTICK_VECTOR is defined in FreeRTOSConfig.h of existing GNURX projects in FreeRTOS's zip file and FreeRTOS's repository though configTICK_VECTOR is not used in the GNURX port layers. Your comment makes me noticed that such thing just now. So, I have to say that it is mandatory (not enough 'better') to use configINCLUDE_PLATFORM_H_INSTEAD_OF_IODEFINE_H in the GNURX port layers. I'm sorry for my mistake.

Best regards,
NoMaY

NoMaY-jp · 2020-08-13T01:40:30Z

Hello @ravibhagavandas,

I have reported this problem to Renesas Electronics Corporation via Japanese user forum (in the following post) but Renesas Japan is in company's summer holidays this week. (I have no direct communication method with them and Dinh Van Nam.)

(I'm sorry that all contents are in Japanese.)
https://japan.renesasrulz.com/cafe_rene/f/forum21/5940/e2-studio-v7-5-0-freertos-scfg-rtos-object/36176#36176

Best regards,
NoMaY

ravibhagavandas · 2020-08-24T23:43:49Z

The changes for this PR are merged in PR #135 . Please feel free to open a new PR if you have additional changes.

Co-authored-by: Cobus van Eeden <[email protected]> Co-authored-by: Aniruddha Kanhere <[email protected]>

VanNamDinh added 2 commits July 16, 2020 16:38

Add RX200 GCC compiler

226af68

Signed-off-by: Dinh Van Nam <[email protected]>

Update GCC compiler for:

d7c2805

* RX600v2 * RX600 * RX100 Signed-off-by: Dinh Van Nam <[email protected]>

ravibhagavandas mentioned this pull request Aug 22, 2020

Update Renesas GCC compiler ports #135

Merged

ravibhagavandas closed this Aug 24, 2020

laroche pushed a commit to laroche/FreeRTOS-Kernel that referenced this pull request Apr 18, 2024

fix demo picmz-mplab (FreeRTOS#89)

7c3c168

Co-authored-by: Cobus van Eeden <[email protected]> Co-authored-by: Aniruddha Kanhere <[email protected]>

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Update Renesas GCC compiler #89

Update Renesas GCC compiler #89

Uh oh!

VanNamDinh commented Jul 16, 2020

Uh oh!

NoMaY-jp commented Jul 17, 2020

Uh oh!

NoMaY-jp commented Jul 17, 2020

Uh oh!

VanNamDinh commented Jul 17, 2020

Uh oh!

ravibhagavandas commented Aug 10, 2020

Uh oh!

ravibhagavandas commented Aug 12, 2020 •

edited

Loading

Uh oh!

NoMaY-jp commented Aug 12, 2020 •

edited

Loading

Uh oh!

NoMaY-jp commented Aug 13, 2020

Uh oh!

ravibhagavandas commented Aug 24, 2020

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Update Renesas GCC compiler #89

Update Renesas GCC compiler #89

Uh oh!

Conversation

VanNamDinh commented Jul 16, 2020

Description

Uh oh!

NoMaY-jp commented Jul 17, 2020

Uh oh!

NoMaY-jp commented Jul 17, 2020

Uh oh!

VanNamDinh commented Jul 17, 2020

Uh oh!

ravibhagavandas commented Aug 10, 2020

Uh oh!

ravibhagavandas commented Aug 12, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

NoMaY-jp commented Aug 12, 2020 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

NoMaY-jp commented Aug 13, 2020

Uh oh!

ravibhagavandas commented Aug 24, 2020

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

ravibhagavandas commented Aug 12, 2020 •

edited

Loading

NoMaY-jp commented Aug 12, 2020 •

edited

Loading