mseal integration in ld.so is going to break relrhack
Categories
(Firefox Build System :: General, defect, P3)
Tracking
(firefox127 affected, firefox128 affected, firefox129 affected)
People
(Reporter: glandium, Assigned: glandium)
Details
Context: https://lwn.net/Articles/978010/
Specifically this change is going to break relrhack: https://lwn.net/ml/all/20240611153220.165430-4-adhemerval.zanella@linaro.org/
|
Assignee | |
Comment 1•1 year ago
|
||
(In reply to Mike Hommey [:glandium] from comment #0)
Specifically this change is going to break relrhack
elfhack too, for that matter, but I don't think it's worth worrying about elfhack.
|
||
Comment 2•1 year ago
|
||
The severity field is not set for this bug.
:ahochheiden, could you have a look please?
For more information, please visit BugBot documentation.
|
||
Updated•1 year ago
|
|
|
Assignee | |
Comment 3•1 year ago
|
||
Good news. The last iteration of the patchset for memory sealing support in glibc is opt-in rather than opt-out, so older versions of Firefox we won't be affected by it, and we won't need adjustments to keep things working for newer versions. We may want to apply memory sealing ourselves, but that's independent of such upstream support (although it would be good to at least wait for the binutils support to land, so that we can use the same flag). As a side note, memory sealing is new to kernel 6.10, so it's far from being common, so it will probably take some time before implementing it in elfhack becomes really useful.
Description
•