fix: re-enable config_fs_size guard in mount_config_filesystem and document dmgpio hard fault

[Copilot]

Two Renode CI issues: the test permanently times out (infinite loop) due to a commented-out size guard, and — once that is fixed — a hard fault fires when dmdevfs initialises dmgpio_port against STM32 GPIO hardware registers.

Fix: infinite loop in mount_config_filesystem

The if(config_fs_size > 0) guard was commented out, so dmffs was always invoked even when no config FS was embedded (size=0 in the emulation build). dmffs reads flash location from FLASH_FS_ADDR/FLASH_FS_SIZE env vars — not mount_opts as the old code assumed — and hung when both were unset and size was 0.

// Before: guard disabled, wrong mount_opts keys, no env vars set
// if(config_fs_size > 0)
{
    char mount_opts[128];
    Dmod_SnPrintf(mount_opts, sizeof(mount_opts), "flash_addr=0x%X;flash_size=%u", ...);
    dmvfs_mount_fs("dmffs", "/configs", mount_opts);
}

// After: guard restored, env vars set, mount_opts dropped
if(config_fs_size > 0)
{
    dmenv_seti(env_ctx, "FLASH_FS_ADDR", (uint32_t)(uintptr_t)config_fs_start);
    dmenv_seti(env_ctx, "FLASH_FS_SIZE", (uint32_t)config_fs_size);
    dmvfs_mount_fs("dmffs", "/configs", NULL);
}

dmenv_ctx is now threaded through mount_embedded_filesystems() → mount_config_filesystem().

Documentation: dmgpio hard fault in Renode

Added docs/dmgpio-renode-hard-fault.md with reproduction steps, root cause analysis, and proposed fixes for the external repository. The fault occurs inside dmvfs_mount_fs("dmdevfs", "/dev", "/configs") when dmgpio_port v0.4 writes to STM32 GPIO/RCC MMIO registers that Renode's stm32f7_discovery-bb.repl does not fully emulate. The fix belongs in dmgpio (guard hardware access under emulation) or dmdevfs (tolerate driver init failure instead of propagating the exception).

Original prompt

---

This section details on the original issue you should resolve

<issue_title>Naprawić problemy na renode</issue_title>
<issue_description>Mamy problemy na CI:
https://productionresultssa2.blob.core.windows.net/actions-results/50fe23c4-0347-4366-ba0e-1d4ac3e7f976/workflow-job-run-bf55618c-f2ea-51fc-b435-fad2a2ca0927/logs/job/job-logs.txt?rsct=text%2Fplain&se=2026-03-09T22%3A05%3A10Z&sig=r%2BmhVpSLPzT%2FT0bvdTq2Bb40NtwcCBlynbBC%2BxZSbTA%3D&ske=2026-03-10T01%3A52%3A13Z&skoid=ca7593d4-ee42-46cd-af88-8b886a2f84eb&sks=b&skt=2026-03-09T21%3A52%3A13Z&sktid=398a6654-997b-47e9-b12b-9515b896b4de&skv=2025-11-05&sp=r&spr=https&sr=b&st=2026-03-09T21%3A55%3A05Z&sv=2025-11-05

Aktualnie jest jakiś problem prawdopodobnie infinit loop, który sprawia, że test failuje. Gorsze jest jednak to, co się dzieje potem (po naprawie tego błędu) - z jakiegoś powodu po zamontowaniu dmgpio w dmdevfs następuje crash (hard fault). Celem tego zadania jest ustalenie jak największej ilości szczegółów na temat tego crasha i przygotowanie opisu zadania dla repozytorium dmdevfs lub dmgpio (chyba, że błąd jest po naszej stronie, to wtedy naprawienie). Nie wprowadzaj proszę żadnych workaroundów - jeśli błąd jest w innym repozytorium to dokładnie go opisz, a nie próbuj workaroundować </issue_description>

Comments on the Issue (you are @copilot in this section)

• Fixes Naprawić problemy na renode #89

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.