On 2021/12/16 19:17, Borislav Petkov wrote: > On Fri, Dec 10, 2021 at 02:55:28PM +0800, Zhen Lei wrote: >> + * reserve_crashkernel() - reserves memory for crash kernel >> + * >> + * This function reserves memory area given in "crashkernel=" kernel command >> + * line parameter. The memory reserved is used by dump capture kernel when >> + * primary kernel is crashing. >> + */ >> +void __init reserve_crashkernel(void) > > As I've already alluded to in another mail, ontop of this there should > be a patch or multiple patches which clean this up more and perhaps even > split it into separate functions doing stuff in this order: > > 1. Parse all crashkernel= cmdline options > > 2. Do all crash_base, crash_size etc checks > > 3. Do the memory reservations > > And all that supplied with comments explaining why stuff is being done. I agree with you. This makes the code look clear. I will do it, try to post v18 next Monday. > > This set of functions is a mess and there's no better time for cleaning > it up and documenting it properly than when you move it to generic code. > > Thx. >
