On 21.02.2023 12:35, Xenia Ragiadakou wrote: > On 2/21/23 13:26, Jan Beulich wrote: >> On 17.02.2023 19:48, Xenia Ragiadakou wrote: >>> Do not include the headers: >>> asm/hvm/vpic.h >>> asm/hvm/vpt.h >>> asm/io.h >>> asm/mce.h >>> asm/mem_sharing.h >>> asm/regs.h >>> public/arch-x86/cpuid.h >>> public/hvm/save.h >>> because none of the declarations and macro definitions in them is used. >>> Sort alphabetically the rest of the headers. >>> >>> Rearrange the code to replace all forward declarations with the function >>> definitions. >>> >>> Replace double new lines with one. >>> >>> Reduce scope of nvmx_enqueue_n2_exceptions() to static because it is used >>> only in this file. >>> >>> Move vmx_update_debug_state() in vmcs.c because it is used only in this file >>> and limit its scope to this file by declaring it static and removing its >>> declaration from vmx.h. >>> >>> Take the opportunity to remove all trailing spaces in vmx.c. >>> >>> No functional change intended. >>> >>> Signed-off-by: Xenia Ragiadakou <burzalod...@gmail.com> >>> --- >>> xen/arch/x86/hvm/vmx/vmcs.c | 12 + >>> xen/arch/x86/hvm/vmx/vmx.c | 5844 ++++++++++++------------ >>> xen/arch/x86/include/asm/hvm/vmx/vmx.h | 1 - >>> 3 files changed, 2913 insertions(+), 2944 deletions(-) >> >> I'm afraid this is close to unreviewable and hence absolutely needs >> splitting. >> With this massive amount of re-arrangement (it's half of vmx.c, after all) >> I'm >> also concerned of losing "git blame"-ability for fair parts of the code >> there. > > I understand. Let me split the changes apart from the one that > rearranges the code. Do you agree in principle? or do you want me to > except and sth else?
Well, the large amount of code movement wants at least one other party (e.g. Kevin, Andrew, or Roger) agreeing with your approach. As said, I for one don't like this interruption in half-way easy history determination (which can be particularly helpful e.g. when you want to find a commit to put in a Fixes: tag). Jan
