linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Enze Li <lienze@kylinos.cn>
To: Huacai Chen <chenhuacai@kernel.org>
Cc: kernel@xen0n.name,  loongarch@lists.linux.dev,
	 glider@google.com, elver@google.com,  akpm@linux-foundation.org,
	kasan-dev@googlegroups.com,  linux-mm@kvack.org,
	 zhangqing@loongson.cn, yangtiezhu@loongson.cn,
	 dvyukov@google.com
Subject: Re: [PATCH 1/4] LoongArch: mm: Add page table mapped mode support
Date: Fri, 21 Jul 2023 10:12:06 +0800	[thread overview]
Message-ID: <87pm4mf1xl.fsf@kylinos.cn> (raw)
In-Reply-To: <CAAhV-H5pWmd2owMgH9hiqxoWpeAOKGv_=j2V-urA+D87_uCMyg@mail.gmail.com> (Huacai Chen's message of "Wed, 19 Jul 2023 23:29:37 +0800")

On Wed, Jul 19 2023 at 11:29:37 PM +0800, Huacai Chen wrote:

> Hi, Enze,
>
> On Wed, Jul 19, 2023 at 4:34 PM Enze Li <lienze@kylinos.cn> wrote:
>>
>> According to LoongArch documentation online, there are two types of address
>> translation modes: direct mapped address translation mode (direct mapped mode)
>> and page table mapped address translation mode (page table mapped mode).
>>
>> Currently, the upstream code only supports DMM (Direct Mapped Mode).
>> This patch adds a function that determines whether PTMM (Page Table
>> Mapped Mode) should be used, and also adds the corresponding handler
>> funcitons for both modes.
>>
>> For more details on the two modes, see [1].
>>
>> [1]
>> https://loongson.github.io/LoongArch-Documentation/LoongArch-Vol1-EN.html#virtual-address-space-and-address-translation-mode
>>
>> Signed-off-by: Enze Li <lienze@kylinos.cn>
>> ---
>>  arch/loongarch/include/asm/page.h    | 10 ++++++++++
>>  arch/loongarch/include/asm/pgtable.h |  6 ++++++
>>  arch/loongarch/mm/pgtable.c          | 25 +++++++++++++++++++++++++
>>  3 files changed, 41 insertions(+)
>>
>> diff --git a/arch/loongarch/include/asm/page.h b/arch/loongarch/include/asm/page.h
>> index 26e8dccb6619..05919be15801 100644
>> --- a/arch/loongarch/include/asm/page.h
>> +++ b/arch/loongarch/include/asm/page.h
>> @@ -84,7 +84,17 @@ typedef struct { unsigned long pgprot; } pgprot_t;
>>  #define sym_to_pfn(x)          __phys_to_pfn(__pa_symbol(x))
>>
>>  #define virt_to_pfn(kaddr)     PFN_DOWN(PHYSADDR(kaddr))
>> +
>> +#ifdef CONFIG_64BIT
>> +#define virt_to_page(kaddr)                                            \
>> +({                                                                     \
>> +       is_PTMM_addr((unsigned long)kaddr) ?                            \
>> +       PTMM_virt_to_page((unsigned long)kaddr) :                       \
>> +       DMM_virt_to_page((unsigned long)kaddr);                         \
>> +})
> 1, Rename these helpers to
> is_dmw_addr()/dmw_virt_to_page()/tlb_virt_to_page() will be better.
> 2, These helpers are so simple so can be defined as inline function or
> macros in page.h.

Hi Huacai,

Except for tlb_virt_to_page(), the remaining two modifications are easy.

I've run into a lot of problems when trying to make tlb_virt_to_page()
as a macro or inline function.  That's because we need to export this
symbol in order for it to be used by the module that called the
virt_to_page() function, other wise, we got the following errors,

-----------------------------------------------------------------------
  MODPOST Module.symvers
ERROR: modpost: "tlb_virt_to_page" [fs/hfsplus/hfsplus.ko] undefined!
ERROR: modpost: "tlb_virt_to_page" [fs/smb/client/cifs.ko] undefined!
ERROR: modpost: "tlb_virt_to_page" [crypto/gcm.ko] undefined!
ERROR: modpost: "tlb_virt_to_page" [crypto/ccm.ko] undefined!
ERROR: modpost: "tlb_virt_to_page" [crypto/essiv.ko] undefined!
ERROR: modpost: "tlb_virt_to_page" [lib/crypto/libchacha20poly1305.ko] undefined!
ERROR: modpost: "tlb_virt_to_page" [drivers/gpu/drm/ttm/ttm.ko] undefined!
ERROR: modpost: "tlb_virt_to_page" [drivers/gpu/drm/amd/amdgpu/amdgpu.ko] undefined!
ERROR: modpost: "tlb_virt_to_page" [drivers/scsi/iscsi_tcp.ko] undefined!
ERROR: modpost: "tlb_virt_to_page" [drivers/scsi/qla2xxx/qla2xxx.ko] undefined!
WARNING: modpost: suppressed 44 unresolved symbol warnings because there were too many)
-----------------------------------------------------------------------

It seems to me that wrapping it into a common function might be the only
way to successfully compile or link with this modification.

=========================================================
--- a/arch/loongarch/include/asm/pgtable.h
+++ b/arch/loongarch/include/asm/pgtable.h
@@ -360,6 +360,8 @@ static inline void pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *pt
 #define PMD_T_LOG2     (__builtin_ffs(sizeof(pmd_t)) - 1)
 #define PTE_T_LOG2     (__builtin_ffs(sizeof(pte_t)) - 1)
 
+inline struct page *tlb_virt_to_page(unsigned long kaddr);
+

--- a/arch/loongarch/mm/pgtable.c
+++ b/arch/loongarch/mm/pgtable.c
@@ -9,6 +9,12 @@
 #include <asm/pgtable.h>
 #include <asm/tlbflush.h>
 
+inline struct page *tlb_virt_to_page(unsigned long kaddr)
+{
+       return pte_page(*virt_to_kpte(kaddr));
+}
+EXPORT_SYMBOL_GPL(tlb_virt_to_page);
=========================================================

WDYT?

Best Regards,
Enze

> 3, CONFIG_64BIT can be removed here.
>
> Huacai
>
>> +#else
>>  #define virt_to_page(kaddr)    pfn_to_page(virt_to_pfn(kaddr))
>> +#endif
>>
>>  extern int __virt_addr_valid(volatile void *kaddr);
>>  #define virt_addr_valid(kaddr) __virt_addr_valid((volatile void *)(kaddr))
>> diff --git a/arch/loongarch/include/asm/pgtable.h b/arch/loongarch/include/asm/pgtable.h
>> index ed6a37bb55b5..0fc074b8bd48 100644
>> --- a/arch/loongarch/include/asm/pgtable.h
>> +++ b/arch/loongarch/include/asm/pgtable.h
>> @@ -360,6 +360,12 @@ static inline void pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *pt
>>  #define PMD_T_LOG2     (__builtin_ffs(sizeof(pmd_t)) - 1)
>>  #define PTE_T_LOG2     (__builtin_ffs(sizeof(pte_t)) - 1)
>>
>> +#ifdef CONFIG_64BIT
>> +struct page *DMM_virt_to_page(unsigned long kaddr);
>> +struct page *PTMM_virt_to_page(unsigned long kaddr);
>> +bool is_PTMM_addr(unsigned long kaddr);
>> +#endif
>> +
>>  extern pgd_t swapper_pg_dir[];
>>  extern pgd_t invalid_pg_dir[];
>>
>> diff --git a/arch/loongarch/mm/pgtable.c b/arch/loongarch/mm/pgtable.c
>> index 36a6dc0148ae..4c6448f996b6 100644
>> --- a/arch/loongarch/mm/pgtable.c
>> +++ b/arch/loongarch/mm/pgtable.c
>> @@ -9,6 +9,31 @@
>>  #include <asm/pgtable.h>
>>  #include <asm/tlbflush.h>
>>
>> +#ifdef CONFIG_64BIT
>> +/* DMM stands for Direct Mapped Mode. */
>> +struct page *DMM_virt_to_page(unsigned long kaddr)
>> +{
>> +       return pfn_to_page(virt_to_pfn(kaddr));
>> +}
>> +EXPORT_SYMBOL_GPL(DMM_virt_to_page);
>> +
>> +/* PTMM stands for Page Table Mapped Mode. */
>> +struct page *PTMM_virt_to_page(unsigned long kaddr)
>> +{
>> +       return pte_page(*virt_to_kpte(kaddr));
>> +}
>> +EXPORT_SYMBOL_GPL(PTMM_virt_to_page);
>> +
>> +bool is_PTMM_addr(unsigned long kaddr)
>> +{
>> +       if (unlikely((kaddr & GENMASK(BITS_PER_LONG - 1, cpu_vabits)) ==
>> +                    GENMASK(BITS_PER_LONG - 1, cpu_vabits)))
>> +               return true;
>> +       return false;
>> +}
>> +EXPORT_SYMBOL_GPL(is_PTMM_addr);
>> +#endif
>> +
>>  pgd_t *pgd_alloc(struct mm_struct *mm)
>>  {
>>         pgd_t *ret, *init;
>> --
>> 2.34.1
>>
>>


  reply	other threads:[~2023-07-21  2:12 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-07-19  8:27 [PATCH 0/4] Add KFENCE support for LoongArch Enze Li
2023-07-19  8:27 ` [PATCH 1/4] LoongArch: mm: Add page table mapped mode support Enze Li
2023-07-19 15:29   ` Huacai Chen
2023-07-21  2:12     ` Enze Li [this message]
2023-07-21  2:21       ` Huacai Chen
2023-07-23  7:17         ` Enze Li
2023-07-25  2:06           ` Huacai Chen
2023-07-25  6:07             ` Enze Li
2023-07-19  8:27 ` [PATCH 2/4] LoongArch: Get stack without NMI when providing regs parameter Enze Li
2023-07-19 15:17   ` Huacai Chen
2023-07-21  1:49     ` Enze Li
2023-07-21  2:17       ` Huacai Chen
2023-07-19  8:27 ` [PATCH 3/4] KFENCE: Deferring the assignment of the local variable addr Enze Li
2023-07-19 10:54   ` Marco Elver
2023-07-19 15:06     ` Huacai Chen
2023-07-19 15:08       ` Marco Elver
2023-07-19  8:27 ` [PATCH 4/4] LoongArch: Add KFENCE support Enze Li
2023-07-19 15:27   ` Huacai Chen
2023-07-21  3:13     ` Enze Li
2023-07-21  3:19       ` Huacai Chen
2023-07-23  7:34         ` Enze Li

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=87pm4mf1xl.fsf@kylinos.cn \
    --to=lienze@kylinos.cn \
    --cc=akpm@linux-foundation.org \
    --cc=chenhuacai@kernel.org \
    --cc=dvyukov@google.com \
    --cc=elver@google.com \
    --cc=glider@google.com \
    --cc=kasan-dev@googlegroups.com \
    --cc=kernel@xen0n.name \
    --cc=linux-mm@kvack.org \
    --cc=loongarch@lists.linux.dev \
    --cc=yangtiezhu@loongson.cn \
    --cc=zhangqing@loongson.cn \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox