Le 07/02/2019 à 05:53, Nicholas Piggin a écrit :
Michael Ellerman's on February 5, 2019 9:32 pm:From: Christophe Leroy <christophe.le...@c-s.fr> This patch activates CONFIG_THREAD_INFO_IN_TASK which moves the thread_info into task_struct. Moving thread_info into task_struct has the following advantages: - It protects thread_info from corruption in the case of stack overflows. - Its address is harder to determine if stack addresses are leaked, making a number of attacks more difficult. This has the following consequences: - thread_info is now located at the beginning of task_struct. - The 'cpu' field is now in task_struct, and only exists when CONFIG_SMP is active. - thread_info doesn't have anymore the 'task' field. This patch: - Removes all recopy of thread_info struct when the stack changes. - Changes the CURRENT_THREAD_INFO() macro to point to current. - Selects CONFIG_THREAD_INFO_IN_TASK. - Modifies raw_smp_processor_id() to get ->cpu from current without including linux/sched.h to avoid circular inclusion and without including asm/asm-offsets.h to avoid symbol names duplication between ASM constants and C constants.Come to think of it, can this patch be split out entirely and moved earlier as a 32-bit patch? 64-bit does not require that change or the additional build step AFAIKS?
Euh ... we may do that but the change in smp.h cannot go as is until thread_info is moved into current. So it would mean only having the Makefile change and the GENERATING_ASM_OFFSETS define in asm-offsets.c, and eventually an intermediate version of raw_smp_processor_id() in smp.h, that would have to get modified in the activation patch anyway.
So I'm not sure this is really worth it. Christophe
