Posts by eggplant886

Hello
1. When using SES with GDB server for debugging, SES would send packages like '$vCont;s:0#22' '$vCont;c:0#12'. It uses thread id 0 here which means 'an arbitrary thread' according to GDB protocol.
Some GDB server does not accept such an ambiguous identity, could SES team change the id from 0 to 1 which means the first thread or change to -1 to represent all threads?
2. SES does not support GDB RLE, which uses * to represent a bunch of repetive data. Could SES team add support for it?

I've compared the description between J-Link PLUS V11 - SEGGER Wiki and J-Link PLUS V12 - SEGGER Wiki. It seems V12 only differs in USB Web Server feature. Is there any other limitations for V11? I'm currently using two V11 probes, both works just fine.
Also please tell me the software maintanance plan for V11 hardwares, especially the firmware maintanance, such as new device support. In the newest Jlink software release note, only V12 gets Arm Cortex-M52 support.
I want to know if software maintanance for V11 is stopped or it would be stopped soon.

I'm using J-link for a RISC-V application, when the commander connects to the target, it prompts some strange hint 'RISC-V: The connected J-Link (S/N 851005533) uses an old firmware module V1 with known problems / limitations.' but the firmware is up-to-date, please help me know what it means and what limitations are they?
Log when connecting to target:

SEGGER J-Link Commander V7.94f (Compiled Jan 17 2024 15:36:28)
DLL version V7.94f, compiled Jan 17 2024 15:35:18

Connecting to J-Link via USB...O.K.
Firmware: J-Link V11 compiled Dec 4 2023 10:22:45
Hardware version: V11.00
J-Link uptime (since boot): 0d 00h 00m 11s
S/N: 851005533
License(s): RDI, FlashBP, FlashDL, JFlash, GDB
USB speed mode: High speed (480 MBit/s)
VTref=3.309V

Type "connect" to establish a target connection, '?' for help
J-Link>connect
Please specify device / core. <Default>: HPM5301XEGX
Type '?' for selection dialog
Device>?
Please specify target interface:
J) JTAG (Default)
S) SWD
T) cJTAG
TIF>
Device position in JTAG chain (IRPre,DRPre) <Default>: -1,-1 => Auto-detect
JTAGConf>
Specify target interface speed [kHz]. <Default>: 4000 kHz
Speed>15000
Device "HPM6364XPAX" selected.

Connecting to target via JTAG
ConfigTargetSettings() start
ConfigTargetSettings() end - Took 472us
TotalIRLen = 5, IRPrint = 0x01
JTAG chain detection found 1 devices:
#0 Id: 0x1000563D, IRLen: 05, RV32
Debug architecture:
RISC-V debug: 0.13
AddrBits: 7
DataBits: 32
IdleClks: 7
Memory access:
Via system bus: Yes (8/16/32-bit accesses are supported)
Via ProgBuf: Yes (8 ProgBuf entries)
Via abstract command (AAM): May be tried as last resort
DataBuf: 4 entries
autoexec[0] implemented: Yes
Detected: RV32 core
RISC-V: The connected J-Link (S/N 851005533) uses an old firmware module V1 with known problems / limitations.
Temp. halting CPU for for feature detection...
HW instruction/data BPs: 2
Support set/clr BPs while running: No
HW data BPs trigger before execution of inst
CSR access via abs. commands: Yes
Compressed instruction support: Yes
Feature detection done. Restarting core...
BG memory access support: Via SBA
Memory zones:
Zone: "Default" Description: Default access mode
RISC-V identified.
J-Link>

Hello my dear team.
I want to know if Segger Embedded studio For RISC-V support the P extension for faster DSP processing? I cannot find relavant info about if it's supported. But emVDSP seems to support it, so please tell me if SES support too.

Hello my dear Segger Team. I'd love to use Ozone which provides excellent user experience. But I have a little trouble and I'd like to ask for help.
Many RTOS like ThreadX, RT-Thread support dynamic module. We need to create at least wo projects, one is the module manager, and the other is the module.
For debugging, Ozone needs to load symbols from the two ELF files corresponding to different flash address ranges. I searched the Ozone manual, I didn't find trace of support for this.
Using File.Open or File.Load, previously opened symbol would be deleted. I cannot make Ozone load both ELF file, and I'd like to know if it can be realized.

Thank you, Nino. And I'd like to know whether J-Link Edu would have a compact form as the Base and Plus product line ?

Ray Chang, [2021/12/24 9:17]
Thank you Nino. The J-link is quite an amazing debug probe. I have one but it's not as tiny as a STLINK. I would try it later.
As for the flash programming issue, I still have a feature wish. When I restart a debug session in SES, a flash pogramming would be executed anyway but it's not needed since the code hasn't changed during the whole debug session. Can SES change the logic of restarting a debug session into directing reset the MCU instead of reprogramming the chip and then reset it?

The SES does not adapt to small screen well, the problem lies in the 'Project Explorer' panel, as it does not have a horizontal scroll bar on the bottom, when the width of this panel is small, the toolbar buttons and files shown inside would not display completely, as shown in the attachments. I think it's a good idea to add a scroll bar in the bottom of this panel, and collaps the toolbar buttons into a click-and-shown context menu so all buttons are accessble even when the window size of SES is small, and the width of the configuration comobox can adapts to the contents.

The Segger Embedded Studio is a very good IDE, but I think the debugger could change a little bit better. Now I use the SES for a STM32 based development using STLINK GDB server as the debugging device in SES. When I restart the debugger session. the binary would be downloaded into the Flash inside the MCU but it's already there, so I think SES should not download the flash in this situation, the enduration of flash is limited. I think SES should verify the contents before downloading, if matched the download is bypasses, and the verification fuction is already here in the IDE.