Thanks, Samuel Would it make sense to rename the shared model to stm32l4_soc and make L4x5 and L496 sibling variants under it?
Regards, Mohamad ________________________________ From: Samuel Tardieu <[email protected]> Sent: Saturday, November 14, 2026 12:40 PM To: Mohammad Odeh <[email protected]> Cc: [email protected] <[email protected]>; Alexandre Iooss <[email protected]>; Pierrick Bouvier <[email protected]>; Peter Maydell <[email protected]>; Paolo Bonzini <[email protected]>; Alistair Francis <[email protected]>; Kevin Wolf <[email protected]>; Hanna Reitz <[email protected]>; Fabiano Rosas <[email protected]>; Laurent Vivier <[email protected]>; Philippe Mathieu-Daudé <[email protected]>; Markus Armbruster <[email protected]>; Cédric Le Goater <[email protected]>; Bernhard Beschow <[email protected]>; Gaurav Sharma <[email protected]>; Jan Kiszka <[email protected]>; Bin Meng <[email protected]>; Shameer Kolothum <[email protected]>; Jamin Lin <[email protected]>; Marc-André Lureau <[email protected]>; [email protected] <[email protected]>; [email protected] <[email protected]> Subject: Re: [PATCH qemu 3/3] hw/arm: Add STM32L496G-DISCO machine ~mohammadodeh <[email protected]> writes: > Introduce the STM32L496xG SoC variant with its flash and SRAM sizes. It is a bit strange to include it inside `stm32l4x5_soc.c` when it doesn't match the description. Maybe the name of the file should be changed to reflect that it also cover the 4x6 SOCs? Also, the STM32L496xx SOC has an alias for the SRAM2 at address 0x2004_0000 (in order to get a contiguously addressable SRAM1+SRAM2 memory area). This alias seems to be absent from the code, as it was not needed for the STM32L4x5 SOCs which have no such feature. Best. Sam -- Samuel Tardieu
