You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In linux kernel upstream commit 90a545e98 a new CONFIG was added to the kernel that prevents userspace programs (even root) from accesses MMIO regions that are already claimed by a driver. This breaks certain things in NVMe and was found and discussed in another issue.
The issue tracks an effort to document the issue and provide a better user experience when it happens. I plan to try and add a trap for this and a section to the README.md to educate users.
In time we might fix all the broken commands and then this issue becomes moot.
Cheers
Stephen
The text was updated successfully, but these errors were encountered:
config IO_STRICT_DEVMEM
bool "Filter I/O access to /dev/mem"
depends on STRICT_DEVMEM
---help---
If this option is disabled, you allow userspace (root) access to all
io-memory regardless of whether a driver is actively using that
range. Accidental access to this is obviously disastrous, but
specific access can be used by people debugging kernel drivers.
If this option is switched on, the /dev/mem file only allows
userspace access to *idle* io-memory ranges (see /proc/iomem) This
may break traditional users of /dev/mem (dosemu, legacy X, etc...)
if the driver using a given range cannot be disabled.
If in doubt, say Y.
We've got the one command (show-regs) that relies on this being disabled, but it's not a super critical command. Even with the config enabled, it should be possible to use it with a user space driver like SPDK.
Hi
In linux kernel upstream commit 90a545e98 a new CONFIG was added to the kernel that prevents userspace programs (even root) from accesses MMIO regions that are already claimed by a driver. This breaks certain things in NVMe and was found and discussed in another issue.
The issue tracks an effort to document the issue and provide a better user experience when it happens. I plan to try and add a trap for this and a section to the README.md to educate users.
In time we might fix all the broken commands and then this issue becomes moot.
Cheers
Stephen
The text was updated successfully, but these errors were encountered: