Re: [PATCH tip] x86/mm: Correct fixmap header usage on adaptable MODULES_END

From: Thomas Garnier
Date: Sun Mar 19 2017 - 12:54:25 EST


On Sun, Mar 19, 2017 at 9:03 AM, Wei Yang <richard.weiyang@xxxxxxxxx> wrote:
> On Fri, Mar 17, 2017 at 10:50:34AM -0700, Thomas Garnier wrote:
>>This patch remove fixmap header usage on non-x86 code that was
>>introduced by the adaptable MODULE_END change.
>
> Hi, Thomas
>
> In this patch, it looks you are trying to do two things for my understanding:
> 1. To include <asm/fixmap.h> in asm/pagetable_64.h and remove the include in
> some of the x86 files
> 2. Remove <asm/fixmap.h> in mm/vmalloc.c
>
> I think your change log covers the second task in the patch, but not not talk
> about the first task you did in the patch. If you could mention it in commit
> log, it would be good for maintain.

I agree, I am not the best at writing commits (by far). What's the
best way for me to correct that? (the bot seem to have taken it).

>
> BTW, I have little knowledge about MODULE_END. By searching the code
> MODULE_END is not used in arch/x86. If you would like to mention the commit
> which introduce the problem, it would be more helpful to review the code.

It is used in many places in arch/x86, kasan, head64, fault etc..:
http://lxr.free-electrons.com/ident?i=MODULES_END

>
>>
>>Signed-off-by: Thomas Garnier <thgarnie@xxxxxxxxxx>
>>---
>>Based on tip:x86/mm
>>---
>> arch/x86/include/asm/pgtable_64.h | 1 +
>> arch/x86/kernel/module.c | 1 -
>> arch/x86/mm/dump_pagetables.c | 1 -
>> arch/x86/mm/kasan_init_64.c | 1 -
>> mm/vmalloc.c | 4 ----
>> 5 files changed, 1 insertion(+), 7 deletions(-)
>>
>>diff --git a/arch/x86/include/asm/pgtable_64.h b/arch/x86/include/asm/pgtable_64.h
>>index 73c7ccc38912..67608d4abc2c 100644
>>--- a/arch/x86/include/asm/pgtable_64.h
>>+++ b/arch/x86/include/asm/pgtable_64.h
>>@@ -13,6 +13,7 @@
>> #include <asm/processor.h>
>> #include <linux/bitops.h>
>> #include <linux/threads.h>
>>+#include <asm/fixmap.h>
>>
>
> Hmm... I see in both pgtable_32.h and pgtable_64.h will include <asm/fixmap.h>
> after this change. And pgtable_32.h and pgtable_64.h will be included only in
> pgtable.h. So is it possible to include <asm/fixmap.h> in pgtable.h for once
> instead of include it in both files? Any concerns you would have?

I am not sure I understood. Only 64-bit need this header to correctly
get MODULES_END, that's why I added it to pgtable_64.h only. I tried
to add it lower before and ran into multiple header errors.

>
>> extern pud_t level3_kernel_pgt[512];
>> extern pud_t level3_ident_pgt[512];
>>diff --git a/arch/x86/kernel/module.c b/arch/x86/kernel/module.c
>>index fad61caac75e..477ae806c2fa 100644
>>--- a/arch/x86/kernel/module.c
>>+++ b/arch/x86/kernel/module.c
>>@@ -35,7 +35,6 @@
>> #include <asm/page.h>
>> #include <asm/pgtable.h>
>> #include <asm/setup.h>
>>-#include <asm/fixmap.h>
>>
>> #if 0
>> #define DEBUGP(fmt, ...) \
>>diff --git a/arch/x86/mm/dump_pagetables.c b/arch/x86/mm/dump_pagetables.c
>>index 75efeecc85eb..58b5bee7ea27 100644
>>--- a/arch/x86/mm/dump_pagetables.c
>>+++ b/arch/x86/mm/dump_pagetables.c
>>@@ -20,7 +20,6 @@
>>
>> #include <asm/kasan.h>
>> #include <asm/pgtable.h>
>>-#include <asm/fixmap.h>
>>
>> /*
>> * The dumper groups pagetable entries of the same type into one, and for
>>diff --git a/arch/x86/mm/kasan_init_64.c b/arch/x86/mm/kasan_init_64.c
>>index 1bde19ef86bd..8d63d7a104c3 100644
>>--- a/arch/x86/mm/kasan_init_64.c
>>+++ b/arch/x86/mm/kasan_init_64.c
>>@@ -9,7 +9,6 @@
>>
>> #include <asm/tlbflush.h>
>> #include <asm/sections.h>
>>-#include <asm/fixmap.h>
>>
>> extern pgd_t early_level4_pgt[PTRS_PER_PGD];
>> extern struct range pfn_mapped[E820_X_MAX];
>>diff --git a/mm/vmalloc.c b/mm/vmalloc.c
>>index b7d2a23349f4..0dd80222b20b 100644
>>--- a/mm/vmalloc.c
>>+++ b/mm/vmalloc.c
>>@@ -36,10 +36,6 @@
>> #include <asm/tlbflush.h>
>> #include <asm/shmparam.h>
>>
>>-#ifdef CONFIG_X86
>>-# include <asm/fixmap.h>
>>-#endif
>>-
>> #include "internal.h"
>>
>> struct vfree_deferred {
>>--
>>2.12.0.367.g23dc2f6d3c-goog
>
> --
> Wei Yang
> Help you, Help me



--
Thomas