Messages in this thread | | | Date | Thu, 19 Jan 2023 14:05:26 -0800 | From | Ashok Raj <> | Subject | Re: [PATCH v1 Part2 1/5] x86/microcode: Move late load warning to the same function that taints kernel |
| |
On Thu, Jan 19, 2023 at 10:48:30PM +0100, Thomas Gleixner wrote: > On Fri, Jan 13 2023 at 09:29, Ashok Raj wrote: > > Currently the warning about late loading and tainting are issued from two > > different functions. > > > > Later patches will re-enable microcode late-loading. > > > > Having both messages in the same function helps issuing warnings only > > when required. > > > > Move the warning from microcode_reload_late() -> reload_store() where the > > kernel tainting also happens. > > > > No functional change. > > I had to read this more than once to make sense of it. Let me try a > translation: > > Late microcode loading issues a warning and taints the > kernel. Tainting the kernel and emitting the warning happens in two > different functions. > > The upcoming support for safe late loading under certain conditions > needs to prevent both the warning and the tainting when the safe > conditions are met. That would require to hand the result of the safe > condition check into the function which emits the warning. > > To avoid this awkward construct, move the warning into reload_store() > next to the taint() invocation as that is also the function which will > later contain the safe condition check. > > No functional change. > > Did my decoder get that right? >
Yes, that is accurate.. inspite of my awkward phrasing :-(
I should copy this commit verbatim before i resend :-)
Cheers, Ashok
| |