diff mbox series

[v4] Documentation/mm: Initial page table documentation

Message ID 20230613083906.757878-1-linus.walleij@linaro.org (mailing list archive)
State New
Headers show
Series [v4] Documentation/mm: Initial page table documentation | expand

Commit Message

Linus Walleij June 13, 2023, 8:39 a.m. UTC
This is based on an earlier blog post at people.kernel.org,
it describes the concepts about page tables that were hardest
for me to grasp when dealing with them for the first time,
such as the prevalent three-letter acronyms pfn, pgd, p4d,
pud, pmd and pte.

I don't know if this is what people want, but it's what I would
have wanted.

I discussed at one point with Mike Rapoport to bring this into
the kernel documentation, so here is a small proposal.

Cc: Matthew Wilcox <willy@infradead.org>
Cc: Randy Dunlap <rdunlap@infradead.org>
Cc: Mike Rapoport <rppt@kernel.org>
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Reviewed-by: Bagas Sanjaya <bagasdotme@gmail.com>
Link: https://people.kernel.org/linusw/arm32-page-tables
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
---
ChangeLog v3->v4:
- Singularis to pluralis fix pointed out by Jonathan Cameron
- Reword the origin story about hierarchical page tables a bit
  inspired by the input from Mike Rapoport.
ChangeLog v2->v3:
- Fix the page size example, also have examples for both 4K and
  16K pages since people will confront these.
- Add a section explaining a bit why we have hierarchical
  page tables at all.
ChangeLog v1->v2:
- Fixed speling mistakes
- Copyedit the paragraph on page frame numbers.
- Reverse the arrows in the page table hierarchy illustration.
- Reverse the order of description of the page hierarchy levels.
- Create a new section for folding
- Emphasize that architectures should try to be page hierarchy
  neutral
- Trying to better describe the fact that the lowest page table PTE
  is called like that for historical reasons.
---
 Documentation/mm/page_tables.rst | 149 +++++++++++++++++++++++++++++++
 1 file changed, 149 insertions(+)

Comments

Mike Rapoport June 13, 2023, 1:27 p.m. UTC | #1
On Tue, Jun 13, 2023 at 10:39:06AM +0200, Linus Walleij wrote:
> This is based on an earlier blog post at people.kernel.org,
> it describes the concepts about page tables that were hardest
> for me to grasp when dealing with them for the first time,
> such as the prevalent three-letter acronyms pfn, pgd, p4d,
> pud, pmd and pte.
> 
> I don't know if this is what people want, but it's what I would
> have wanted.
> 
> I discussed at one point with Mike Rapoport to bring this into
> the kernel documentation, so here is a small proposal.
> 
> Cc: Matthew Wilcox <willy@infradead.org>
> Cc: Randy Dunlap <rdunlap@infradead.org>
> Cc: Mike Rapoport <rppt@kernel.org>
> Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
> Reviewed-by: Bagas Sanjaya <bagasdotme@gmail.com>
> Link: https://people.kernel.org/linusw/arm32-page-tables
> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>

Reviewed-by: Mike Rapoport (IBM) <rppt@kernel.org>

> ---
> ChangeLog v3->v4:
> - Singularis to pluralis fix pointed out by Jonathan Cameron
> - Reword the origin story about hierarchical page tables a bit
>   inspired by the input from Mike Rapoport.
> ChangeLog v2->v3:
> - Fix the page size example, also have examples for both 4K and
>   16K pages since people will confront these.
> - Add a section explaining a bit why we have hierarchical
>   page tables at all.
> ChangeLog v1->v2:
> - Fixed speling mistakes
> - Copyedit the paragraph on page frame numbers.
> - Reverse the arrows in the page table hierarchy illustration.
> - Reverse the order of description of the page hierarchy levels.
> - Create a new section for folding
> - Emphasize that architectures should try to be page hierarchy
>   neutral
> - Trying to better describe the fact that the lowest page table PTE
>   is called like that for historical reasons.
> ---
>  Documentation/mm/page_tables.rst | 149 +++++++++++++++++++++++++++++++
>  1 file changed, 149 insertions(+)
> 
> diff --git a/Documentation/mm/page_tables.rst b/Documentation/mm/page_tables.rst
> index 96939571d7bc..a5e202301e7b 100644
> --- a/Documentation/mm/page_tables.rst
> +++ b/Documentation/mm/page_tables.rst
> @@ -3,3 +3,152 @@
>  ===========
>  Page Tables
>  ===========
> +
> +Paged virtual memory was invented along with virtual memory as a concept in
> +1962 on the Ferranti Atlas Computer which was the first computer with paged
> +virtual memory. The feature migrated to newer computers and became a de facto
> +feature of all Unix-like systems as time went by. In 1985 the feature was
> +included in the Intel 80386, which was the CPU Linux 1.0 was developed on.
> +
> +Page tables map virtual addresses as seen by the CPU program counter into
> +physical addresses as seen on the external memory bus.
> +
> +Linux defines page tables as a hierarchy which is currently five levels in
> +height. The target architecture code for each supported architecture will then
> +map this to the restrictions of the target hardware.
> +
> +The physical address corresponding to the virtual address is often referenced
> +by the underlying physical page frame. The **page frame number** or **pfn**
> +is the physical address of the page (as seen on the external memory bus)
> +divided by `PAGE_SIZE`.
> +
> +Physical memory address 0 will be *pfn 0* and the highest pfn will be
> +the last page of physical memory the external address bus of the CPU can
> +address.
> +
> +With a page granularity of 4KB and a address range of 32 bits, pfn 0 is at
> +address 0x00000000, pfn 1 is at address 0x00001000, pfn 2 is at 0x00002000
> +and so on until we reach pfn 0xfffff at 0xfffff000. With 16KB pages pfs are
> +at 0x00004000, 0x00008000 ... 0xffffc000 and pfn goes from 0 to 0x3fffff.
> +
> +As you can see, with 4KB pages the page base address uses bits 12-31 of the
> +address, and this is why `PAGE_SHIFT` in this case is defined as 12 and
> +`PAGE_SIZE` is usually defined in terms of the page shift as `(1 << PAGE_SHIFT)`
> +
> +Over time a deeper hierarchy has been developed in response to increasing memory
> +sizes. When Linux was created, 4KB pages and a single page table called
> +`swapper_pg_dir` with 1024 entries was used, covering 4MB which coincided with
> +the fact that Torvald's first computer had 4MB of physical memory. Entries in
> +this single table were referred to as *PTE*:s - page table entries.
> +
> +The software page table hierarchy reflects the fact that page table hardware has
> +become hierarchical and that in turn is done to save page table memory and
> +speed up mapping.
> +
> +One could of course imagine a single, linear page table with enormous amounts
> +of entries, breaking down the whole memory into single pages. Such a page table
> +would be very sparse, because large portions of the virtual memory usually
> +remains unused. By using hierarchical page tables large holes in the virtual
> +address space does not waste valuable page table memory, because it will suffice
> +to mark large areas as unmapped at a higher level in the page table hierarchy.
> +
> +Additionally, on modern CPUs, a higher level page table entry can point directly
> +to a physical memory range, which allows mapping a contiguous range of several
> +megabytes or even gigabytes in a single high-level page table entry, taking
> +shortcuts in mapping virtual memory to physical memory: there is no need to
> +traverse deeper in the hierarchy when you find a large mapped range like this.
> +
> +The page table hierarchy has now developed into this::
> +
> +  +-----+
> +  | PGD |
> +  +-----+
> +     |
> +     |   +-----+
> +     +-->| P4D |
> +         +-----+
> +            |
> +            |   +-----+
> +            +-->| PUD |
> +                +-----+
> +                   |
> +                   |   +-----+
> +                   +-->| PMD |
> +                       +-----+
> +                          |
> +                          |   +-----+
> +                          +-->| PTE |
> +                              +-----+
> +
> +
> +Symbols on the different levels of the page table hierarchy have the following
> +meaning beginning from the bottom:
> +
> +- **pte**, `pte_t`, `pteval_t` = **Page Table Entry** - mentioned earlier.
> +  The *pte* is an array of `PTRS_PER_PTE` elements of the `pteval_t` type, each
> +  mapping a single page of virtual memory to a single page of physical memory.
> +  The architecture defines the size and contents of `pteval_t`.
> +
> +  A typical example is that the `pteval_t` is a 32- or 64-bit value with the
> +  upper bits being a **pfn** (page frame number), and the lower bits being some
> +  architecture-specific bits such as memory protection.
> +
> +  The **entry** part of the name is a bit confusing because while in Linux 1.0
> +  this did refer to a single page table entry in the single top level page
> +  table, it was retrofitted to be an array of mapping elements when two-level
> +  page tables were first introduced, so the *pte* is the lowermost page
> +  *table*, not a page table *entry*.
> +
> +- **pmd**, `pmd_t`, `pmdval_t` = **Page Middle Directory**, the hierarchy right
> +  above the *pte*, with `PTRS_PER_PMD` references to the *pte*:s.
> +
> +- **pud**, `pud_t`, `pudval_t` = **Page Upper Directory** was introduced after
> +  the other levels to handle 4-level page tables. It is potentially unused,
> +  or *folded* as we will discuss later.
> +
> +- **p4d**, `p4d_t`, `p4dval_t` = **Page Level 4 Directory** was introduced to
> +  handle 5-level page tables after the *pud* was introduced. Now it was clear
> +  that we needed to replace *pgd*, *pmd*, *pud* etc with a figure indicating the
> +  directory level and that we cannot go on with ad hoc names any more. This
> +  is only used on systems which actually have 5 levels of page tables, otherwise
> +  it is folded.
> +
> +- **pgd**, `pgd_t`, `pgdval_t` = **Page Global Directory** - the Linux kernel
> +  main page table handling the PGD for the kernel memory is still found in
> +  `swapper_pg_dir`, but each userspace process in the system also has its own
> +  memory context and thus its own *pgd*, found in `struct mm_struct` which
> +  in turn is referenced to in each `struct task_struct`. So tasks have memory
> +  context in the form of a `struct mm_struct` and this in turn has a
> +  `struct pgt_t *pgd` pointer to the corresponding page global directory.
> +
> +To repeat: each level in the page table hierarchy is a *array of pointers*, so
> +the **pgd** contains `PTRS_PER_PGD` pointers to the next level below, **p4d**
> +contains `PTRS_PER_P4D` pointers to **pud** items and so on. The number of
> +pointers on each level is architecture-defined.::
> +
> +        PMD
> +  --> +-----+           PTE
> +      | ptr |-------> +-----+
> +      | ptr |-        | ptr |-------> PAGE
> +      | ptr | \       | ptr |
> +      | ptr |  \        ...
> +      | ... |   \
> +      | ptr |    \         PTE
> +      +-----+     +----> +-----+
> +                         | ptr |-------> PAGE
> +                         | ptr |
> +                           ...
> +
> +
> +Page Table Folding
> +==================
> +
> +If the architecture does not use all the page table levels, they can be *folded*
> +which means skipped, and all operations performed on page tables will be
> +compile-time augmented to just skip a level when accessing the next lower
> +level.
> +
> +Page table handling code that wishes to be architecture-neutral, such as the
> +virtual memory manager, will need to be written so that it traverses all of the
> +currently five levels. This style should also be preferred for
> +architecture-specific code, so as to be robust to future changes.
> -- 
> 2.40.1
>
Matthew Wilcox (Oracle) June 13, 2023, 2:45 p.m. UTC | #2
On Tue, Jun 13, 2023 at 10:39:06AM +0200, Linus Walleij wrote:
> +Paged virtual memory was invented along with virtual memory as a concept in
> +1962 on the Ferranti Atlas Computer which was the first computer with paged
> +virtual memory. The feature migrated to newer computers and became a de facto
> +feature of all Unix-like systems as time went by. In 1985 the feature was
> +included in the Intel 80386, which was the CPU Linux 1.0 was developed on.

I still don't think the origin story is useful.  It's trivia and doesn't
help someone understand what they need to know.

> +Page tables map virtual addresses as seen by the CPU program counter into
> +physical addresses as seen on the external memory bus.

This makes it sound like virtual addresses are only used for
instructions.  I had better wording earlier, but there's no point in
repeating it.  Just: I dissent. 

> +Linux defines page tables as a hierarchy which is currently five levels in
> +height. The target architecture code for each supported architecture will then
> +map this to the restrictions of the target hardware.

The word "target" isn't adding any value in this paragraph.

Honestly, I don't like much about this document.  The writing is
flabby and untargetted.  Much of my last review was ignored.  I'm just
going to stop here since I have low confidence that any suggestions
would be incorporated.
Linus Walleij June 14, 2023, 7:12 a.m. UTC | #3
On Tue, Jun 13, 2023 at 4:45 PM Matthew Wilcox <willy@infradead.org> wrote:
> On Tue, Jun 13, 2023 at 10:39:06AM +0200, Linus Walleij wrote:

> > +Paged virtual memory was invented along with virtual memory as a concept in
> > +1962 on the Ferranti Atlas Computer which was the first computer with paged
> > +virtual memory. The feature migrated to newer computers and became a de facto
> > +feature of all Unix-like systems as time went by. In 1985 the feature was
> > +included in the Intel 80386, which was the CPU Linux 1.0 was developed on.
>
> I still don't think the origin story is useful.  It's trivia and doesn't
> help someone understand what they need to know.

I think history is pretty important for understanding concepts, otherwise
they appear as not invented but emergent. However the question about
what is important and what is trivia is a question of professional technical
writing. The documentation maintainer works with technical writing and
can decide the value of this.

> > +Page tables map virtual addresses as seen by the CPU program counter into
> > +physical addresses as seen on the external memory bus.
>
> This makes it sound like virtual addresses are only used for
> instructions.  I had better wording earlier, but there's no point in
> repeating it.  Just: I dissent.

You are right I should reword this to be about memory accesses, I
might have missed some of your feedback so I will go back and check
what you said.

> > +Linux defines page tables as a hierarchy which is currently five levels in
> > +height. The target architecture code for each supported architecture will then
> > +map this to the restrictions of the target hardware.
>
> The word "target" isn't adding any value in this paragraph.

Thanks dropping it.

I guess the word is a byproduct of doing too much cross compiling...

> Honestly, I don't like much about this document.  The writing is
> flabby and untargetted.

Please refarain from using this kind of unproffessional wording in your
professional communication. Be technical, precise and to the point and do
not use value statements.

> Much of my last review was ignored.  I'm just
> going to stop here since I have low confidence that any suggestions
> would be incorporated.

Your statement is incorrect. Your feedback is seen as valuable
and it is read, reacted on and incorporated.

I did rewrite the document thoroughly in reaction to your comments,
and in order so you feel respected I am going to illustrate.

You wrote:

> This reads backwards to me.  The index point in the hierarchy (what an
> unusual turn of phrase!) is surely the virtual address, since the
> hierarchy is indexed by virtual addresses.  If this paragraph is
> supposed to define what a pfn is, how about simply:
>
> The pfn of a page of memory is the physical address of the page divided
> by PAGE_SIZE

Which I took as a good suggestion, and the paragraph is rewritten
from this (which you criticized):

> +The physical address corresponding to the virtual address is commonly
> +defined by the index point in the hierarchy, and this is called a **page frame
> +number** or **pfn**. The first entry on the top level to the first entry in the
> +second and so on down the hierarchy will point out the virtual address for the
> +physical memory address 0, which will be *pfn 0* and the highest pfn will be
> +the last page of physical memory the external address bus of the CPU can
> +address.

To this (current wording):

> The physical address corresponding to the virtual address is often referenced
> by the underlying physical page frame. The **page frame number** or **pfn**
> is the physical address of the page (as seen on the external memory bus)
> divided by `PAGE_SIZE`.

Which obviously includes some of your wording (divided by PAGE_SIZE etc).
I also got review comments from other reviewers and that is reflected in the
current wording.

You wrote:

> Your arrows are backwards.  The PTE doesn't point to the PMD; the PMD
> points to PTEs.

And this was changed to what you suggested, and it was a pretty
fundamental and important change.

You wrote:

> You have rather too many forward references in this description for my
> taste.  Start with the PTE, then the PMD, then  PUD, P4D, PGD.

And this was changed to exactly what you suggested.

You wrote:

> array, not list

And this was changed (everywhere) to what you suggested.

You wrote:

> I think a document like this that talks about page tables really needs to
> include a description of how some PMDs / PUDs / ... may not be pointers
> to lower levels, but direct pointers to the actual memory (ie THPs /
> hugetlb pages).

And the document now includes this (Mike Rapoport made the same
comment).

So your claim that "Much of my last review was ignored." is simply
incorrect. It was not ignored, rather it has thoroughly shaped the document.

Yours,
Linus Walleij
diff mbox series

Patch

diff --git a/Documentation/mm/page_tables.rst b/Documentation/mm/page_tables.rst
index 96939571d7bc..a5e202301e7b 100644
--- a/Documentation/mm/page_tables.rst
+++ b/Documentation/mm/page_tables.rst
@@ -3,3 +3,152 @@ 
 ===========
 Page Tables
 ===========
+
+Paged virtual memory was invented along with virtual memory as a concept in
+1962 on the Ferranti Atlas Computer which was the first computer with paged
+virtual memory. The feature migrated to newer computers and became a de facto
+feature of all Unix-like systems as time went by. In 1985 the feature was
+included in the Intel 80386, which was the CPU Linux 1.0 was developed on.
+
+Page tables map virtual addresses as seen by the CPU program counter into
+physical addresses as seen on the external memory bus.
+
+Linux defines page tables as a hierarchy which is currently five levels in
+height. The target architecture code for each supported architecture will then
+map this to the restrictions of the target hardware.
+
+The physical address corresponding to the virtual address is often referenced
+by the underlying physical page frame. The **page frame number** or **pfn**
+is the physical address of the page (as seen on the external memory bus)
+divided by `PAGE_SIZE`.
+
+Physical memory address 0 will be *pfn 0* and the highest pfn will be
+the last page of physical memory the external address bus of the CPU can
+address.
+
+With a page granularity of 4KB and a address range of 32 bits, pfn 0 is at
+address 0x00000000, pfn 1 is at address 0x00001000, pfn 2 is at 0x00002000
+and so on until we reach pfn 0xfffff at 0xfffff000. With 16KB pages pfs are
+at 0x00004000, 0x00008000 ... 0xffffc000 and pfn goes from 0 to 0x3fffff.
+
+As you can see, with 4KB pages the page base address uses bits 12-31 of the
+address, and this is why `PAGE_SHIFT` in this case is defined as 12 and
+`PAGE_SIZE` is usually defined in terms of the page shift as `(1 << PAGE_SHIFT)`
+
+Over time a deeper hierarchy has been developed in response to increasing memory
+sizes. When Linux was created, 4KB pages and a single page table called
+`swapper_pg_dir` with 1024 entries was used, covering 4MB which coincided with
+the fact that Torvald's first computer had 4MB of physical memory. Entries in
+this single table were referred to as *PTE*:s - page table entries.
+
+The software page table hierarchy reflects the fact that page table hardware has
+become hierarchical and that in turn is done to save page table memory and
+speed up mapping.
+
+One could of course imagine a single, linear page table with enormous amounts
+of entries, breaking down the whole memory into single pages. Such a page table
+would be very sparse, because large portions of the virtual memory usually
+remains unused. By using hierarchical page tables large holes in the virtual
+address space does not waste valuable page table memory, because it will suffice
+to mark large areas as unmapped at a higher level in the page table hierarchy.
+
+Additionally, on modern CPUs, a higher level page table entry can point directly
+to a physical memory range, which allows mapping a contiguous range of several
+megabytes or even gigabytes in a single high-level page table entry, taking
+shortcuts in mapping virtual memory to physical memory: there is no need to
+traverse deeper in the hierarchy when you find a large mapped range like this.
+
+The page table hierarchy has now developed into this::
+
+  +-----+
+  | PGD |
+  +-----+
+     |
+     |   +-----+
+     +-->| P4D |
+         +-----+
+            |
+            |   +-----+
+            +-->| PUD |
+                +-----+
+                   |
+                   |   +-----+
+                   +-->| PMD |
+                       +-----+
+                          |
+                          |   +-----+
+                          +-->| PTE |
+                              +-----+
+
+
+Symbols on the different levels of the page table hierarchy have the following
+meaning beginning from the bottom:
+
+- **pte**, `pte_t`, `pteval_t` = **Page Table Entry** - mentioned earlier.
+  The *pte* is an array of `PTRS_PER_PTE` elements of the `pteval_t` type, each
+  mapping a single page of virtual memory to a single page of physical memory.
+  The architecture defines the size and contents of `pteval_t`.
+
+  A typical example is that the `pteval_t` is a 32- or 64-bit value with the
+  upper bits being a **pfn** (page frame number), and the lower bits being some
+  architecture-specific bits such as memory protection.
+
+  The **entry** part of the name is a bit confusing because while in Linux 1.0
+  this did refer to a single page table entry in the single top level page
+  table, it was retrofitted to be an array of mapping elements when two-level
+  page tables were first introduced, so the *pte* is the lowermost page
+  *table*, not a page table *entry*.
+
+- **pmd**, `pmd_t`, `pmdval_t` = **Page Middle Directory**, the hierarchy right
+  above the *pte*, with `PTRS_PER_PMD` references to the *pte*:s.
+
+- **pud**, `pud_t`, `pudval_t` = **Page Upper Directory** was introduced after
+  the other levels to handle 4-level page tables. It is potentially unused,
+  or *folded* as we will discuss later.
+
+- **p4d**, `p4d_t`, `p4dval_t` = **Page Level 4 Directory** was introduced to
+  handle 5-level page tables after the *pud* was introduced. Now it was clear
+  that we needed to replace *pgd*, *pmd*, *pud* etc with a figure indicating the
+  directory level and that we cannot go on with ad hoc names any more. This
+  is only used on systems which actually have 5 levels of page tables, otherwise
+  it is folded.
+
+- **pgd**, `pgd_t`, `pgdval_t` = **Page Global Directory** - the Linux kernel
+  main page table handling the PGD for the kernel memory is still found in
+  `swapper_pg_dir`, but each userspace process in the system also has its own
+  memory context and thus its own *pgd*, found in `struct mm_struct` which
+  in turn is referenced to in each `struct task_struct`. So tasks have memory
+  context in the form of a `struct mm_struct` and this in turn has a
+  `struct pgt_t *pgd` pointer to the corresponding page global directory.
+
+To repeat: each level in the page table hierarchy is a *array of pointers*, so
+the **pgd** contains `PTRS_PER_PGD` pointers to the next level below, **p4d**
+contains `PTRS_PER_P4D` pointers to **pud** items and so on. The number of
+pointers on each level is architecture-defined.::
+
+        PMD
+  --> +-----+           PTE
+      | ptr |-------> +-----+
+      | ptr |-        | ptr |-------> PAGE
+      | ptr | \       | ptr |
+      | ptr |  \        ...
+      | ... |   \
+      | ptr |    \         PTE
+      +-----+     +----> +-----+
+                         | ptr |-------> PAGE
+                         | ptr |
+                           ...
+
+
+Page Table Folding
+==================
+
+If the architecture does not use all the page table levels, they can be *folded*
+which means skipped, and all operations performed on page tables will be
+compile-time augmented to just skip a level when accessing the next lower
+level.
+
+Page table handling code that wishes to be architecture-neutral, such as the
+virtual memory manager, will need to be written so that it traverses all of the
+currently five levels. This style should also be preferred for
+architecture-specific code, so as to be robust to future changes.