On 01/19/26 at 03:49pm, Andrew Morton wrote: > On Fri, 19 Dec 2025 17:31:30 +0800 Qiang Ma <[email protected]> wrote: > > > The commit cbc2fe9d9cb2 ("kexec_file: add kexec_file flag to control > > debug printing") added the kexec_file flag to control debug printing. > > However, for arm64, after commit 6f8c1da071a4 ("kexec_file, arm64: print > > out debugging message if required"), when using the kexec_load interface, > > the kexec_image_info function is called to print debug message, but > > it uses kexec_dprintk, which is only available under kexec_file, > > then when specifying 'kexec -d', kexec_load interface will print nothing. > > > > For riscv, commit eb7622d908a09 ("kexec_file, riscv: print out debugging > > message if required") remove kexec_image_info(), because the content has > > been printed out in generic code in kexec_file_load, but not in kexec_load. > > > > Therefore, in order to solve the two problems mentioned above, for this > > patchset, several things need to be done: > > > > 1. Enabling kexec_dprintk is available for kexec_load. > > 2. Add some debugging message of the deleted kexec_image_info > > to generic code. > > 3. Remove duplicate debugging prints for arm64. > > Some additional review input on this series would be good, if someone > has the time.
As I have said before, I am not fan of this patchset becuase I think it's not necessary, even though it has no much harm except of adding code. Leave this to those who like it.
