> On Nov 19, 2019, at 3:38 PM, Peter Zijlstra wrote: > > On Tue, Nov 19, 2019 at 09:51:49AM -0600, Tianlin Li wrote: >> Right now several architectures allow their set_memory_*() family of >> functions to fail, but callers may not be checking the return values. We >> need to fix the callers and add the __must_check attribute. They also may >> not provide any level of atomicity, in the sense that the memory >> protections may be left incomplete on failure. This issue likely has a few >> steps on effects architectures[1]: >> 1)Have all callers of set_memory_*() helpers check the return value. >> 2)Add __much_check to all set_memory_*() helpers so that new uses do not >> ignore the return value. >> 3)Add atomicity to the calls so that the memory protections aren't left in >> a partial state. >> >> Ideally, the failure of set_memory_*() should be passed up the call stack, >> and callers should examine the failure and deal with it. But currently, >> some callers just have void return type. >> >> We need to fix the callers to handle the return all the way to the top of >> stack, and it will require a large series of patches to finish all the three >> steps mentioned above. I start with kernel/module, and will move onto other >> subsystems. I am not entirely sure about the failure modes for each caller. >> So I would like to get some comments before I move forward. This single >> patch is just for fixing the return value of set_memory_*() function in >> kernel/module, and also the related callers. Any feedback would be greatly >> appreciated. > > Please have a look here: > > https://lkml.kernel.org/r/20191111131252.921588318@infradead.org > > Much of the code you're patching is slated for removal. > > Josh also has patches reworking KLP and there's some ARM64 patches > pending at which point we can also delete module_disable_ro(). Thanks for the information. I will check the code.