Skip to contentSkip to conversation

tomoyo: Convert get_user_pages*() to pin_user_pages*()

PATCH In review Souptick Joarder 4 participants 15 messages Nov 7, 2020 – Nov 10, 2020

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/2 tomoyo: Convert get_user_pages*() to pin_user_pages*()

In 2019, we introduced pin_user_pages*() and now we are converting get_user_pages*() to the new API as appropriate. [1] & [2] could be referred for more information. This is case 5 as per document [1].

[1] Documentation/core-api/pin_user_pages.rst

[2] "Explicit pinning of user-space pages":

        https://lwn.net/Articles/807108/
---
 security/tomoyo/domain.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
security/tomoyo/domain.c +2 -2
index dc4ecc0..bd748be 100644
@@ -914,7 +914,7 @@ bool tomoyo_dump_page(struct linux_binprm *bprm, unsigned long pos,
 	 * (represented by bprm).  'current' is the process doing
 	 * the execve().
 	 */
-	if (get_user_pages_remote(bprm->mm, pos, 1,
+	if (pin_user_pages_remote(bprm->mm, pos, 1,
 				FOLL_FORCE, &page, NULL, NULL) <= 0)
 		return false;
 #else
@@ -936,7 +936,7 @@ bool tomoyo_dump_page(struct linux_binprm *bprm, unsigned long pos,
 	}
 	/* Same with put_arg_page(page) in fs/exec.c */
 #ifdef CONFIG_MMU
-	put_page(page);
+	unpin_user_page(page);
 #endif
 	return true;
 }
Signature · 1 line
1.9.1
#1 lore raw

PATCH 2/2 tomoyo: Fixed typo in documentation

Fixed typo s/Poiner/Pointer

Fixes 5b636857fee6 ("TOMOYO: Allow using argv[]/envp[] of execve() as conditions.")
Cc John Hubbard <[email protected]>
---
 security/tomoyo/domain.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
security/tomoyo/domain.c +1 -1
index bd748be..7b2babe 100644
@@ -891,7 +891,7 @@ int tomoyo_find_next_domain(struct linux_binprm *bprm)
  *
  * @bprm: Pointer to "struct linux_binprm".
  * @pos:  Location to dump.
- * @dump: Poiner to "struct tomoyo_page_dump".
+ * @dump: Pointer to "struct tomoyo_page_dump".
  *
  * Returns true on success, false otherwise.
  */
Signature · 1 line
1.9.1
#2 lore raw

Re: [PATCH 2/2] tomoyo: Fixed typo in documentation

Quoted 14 lines · Souptick Joarder

Fixed typo s/Poiner/Pointer

Fixes 5b636857fee6 ("TOMOYO: Allow using argv[]/envp[] of execve() as conditions.")
Cc John Hubbard <[email protected]>
---
  security/tomoyo/domain.c | 2 +-
  1 file changed, 1 insertion(+), 1 deletion(-)
security/tomoyo/domain.c +0 -0
index bd748be..7b2babe 100644

security/tomoyo/domain.c int tomoyo_find_next_domain(struct linux_binprm *bprm) · line 894

Quoted 5 lines
security/tomoyo/domain.c +1 -1
@@ -891,7 +891,7 @@ int tomoyo_find_next_domain(struct linux_binprm *bprm)
   *
   * @bprm: Pointer to "struct linux_binprm".
   * @pos:  Location to dump.
- * @dump: Poiner to "struct tomoyo_page_dump".
+ * @dump: Pointer to "struct tomoyo_page_dump".

Not worth a separate patch, especially since the original comment is merely copying the C sources, and as such, does not add any value.

I'd either a) craft a new documentation line that adds some value, or b) just merge this patch into the previous one, and make a note in the commit description to the effect that you've included a trivial typo fix as long as you're there.

thanks,

Signature · 2 lines
John Hubbard
NVIDIA
Quoted 4 lines
security/tomoyo/domain.c +0 -0
   *
   * Returns true on success, false otherwise.
   */
#7 lore raw
In reply to Souptick Joarder
Quoted 3 lines · Souptick Joarder

In 2019, we introduced pin_user_pages*() and now we are converting get_user_pages*() to the new API as appropriate. [1] & [2] could be referred for more information. This is case 5 as per document [1].

It turns out that Case 5 can be implemented via a better pattern, as long as we're just dealing with a page at a time, briefly:

lock_page()
write to page's data
unlock_page()

...which neatly synchronizes with writeback and other fs activities.

I was going to track down the Case 5's and do that [1].

+CC Jan and Matthew, to keep us on the straight and narrow, just in case I'm misunderstanding something.

[1] https://lore.kernel.org/r/[email protected]

thanks,

Signature · 2 lines
John Hubbard
NVIDIA
Quoted 35 lines

[1] Documentation/core-api/pin_user_pages.rst

[2] "Explicit pinning of user-space pages":

         https://lwn.net/Articles/807108/
---
  security/tomoyo/domain.c | 4 ++--
  1 file changed, 2 insertions(+), 2 deletions(-)
security/tomoyo/domain.c +2 -2
index dc4ecc0..bd748be 100644
@@ -914,7 +914,7 @@ bool tomoyo_dump_page(struct linux_binprm *bprm, unsigned long pos,
  	 * (represented by bprm).  'current' is the process doing
  	 * the execve().
  	 */
-	if (get_user_pages_remote(bprm->mm, pos, 1,
+	if (pin_user_pages_remote(bprm->mm, pos, 1,
  				FOLL_FORCE, &page, NULL, NULL) <= 0)
  		return false;
  #else
@@ -936,7 +936,7 @@ bool tomoyo_dump_page(struct linux_binprm *bprm, unsigned long pos,
  	}
  	/* Same with put_arg_page(page) in fs/exec.c */
  #ifdef CONFIG_MMU
-	put_page(page);
+	unpin_user_page(page);
  #endif
  	return true;
  }
#8 lore raw
Quoted 5 lines · John Hubbard
Quoted 3 lines · Souptick Joarder

In 2019, we introduced pin_user_pages*() and now we are converting get_user_pages*() to the new API as appropriate. [1] & [2] could be referred for more information. This is case 5 as per document [1].

Quoted 8 lines

It turns out that Case 5 can be implemented via a better pattern, as long as we're just dealing with a page at a time, briefly:

lock_page()
write to page's data
unlock_page()

...which neatly synchronizes with writeback and other fs activities.

Ahem, I left out a key step: set_page_dirty()!

lock_page()
write to page's data
set_page_dirty()
unlock_page()

thanks,

Signature · 2 lines
John Hubbard
NVIDIA
maintainer
#9 lore raw
Quoted 13 lines · John Hubbard
Quoted 12 lines · John Hubbard
Quoted 3 lines · Souptick Joarder

In 2019, we introduced pin_user_pages*() and now we are converting get_user_pages*() to the new API as appropriate. [1] & [2] could be referred for more information. This is case 5 as per document [1].

It turns out that Case 5 can be implemented via a better pattern, as long as we're just dealing with a page at a time, briefly:

lock_page()
write to page's data
unlock_page()

Quoted 9 lines
Quoted 1 line

...which neatly synchronizes with writeback and other fs activities.

Ahem, I left out a key step: set_page_dirty()!

lock_page()
write to page's data
set_page_dirty()
unlock_page()

Excuse me, but Documentation/core-api/pin_user_pages.rst says "CASE 5: Pinning in order to _write_ to the data within the page" while tomoyo_dump_page() is for "_read_ the data within the page". Do we want to convert to pin_user_pages_remote() or lock_page() ?

#10 lore raw
Quoted 24 lines · Tetsuo Handa
Quoted 13 lines · John Hubbard
Quoted 12 lines · John Hubbard
Quoted 3 lines · Souptick Joarder

In 2019, we introduced pin_user_pages*() and now we are converting get_user_pages*() to the new API as appropriate. [1] & [2] could be referred for more information. This is case 5 as per document [1].

It turns out that Case 5 can be implemented via a better pattern, as long as we're just dealing with a page at a time, briefly:

lock_page()
write to page's data
unlock_page()

Quoted 9 lines
Quoted 1 line

...which neatly synchronizes with writeback and other fs activities.

Ahem, I left out a key step: set_page_dirty()!

lock_page()
write to page's data
set_page_dirty()
unlock_page()

Quoted 5 lines

Excuse me, but Documentation/core-api/pin_user_pages.rst says "CASE 5: Pinning in order to _write_ to the data within the page" while tomoyo_dump_page() is for "_read_ the data within the page". Do we want to convert to pin_user_pages_remote() or lock_page() ?

Sorry, I missed the direction here, was too focused on the Case 5 aspect. Yes. Case 5 (which, again, I think we're about to re-document) is only about *writing* to data within the page.

So in this case, where it is just reading from the page, I think it's already from a gup vs pup point of view.

btw, it's not clear to me whether the current code is susceptible to any sort of problem involving something writing to the page while it is being dumped (I am curious). But changing from gup to pup wouldn't fix that, if it were a problem. It a separate question from this patch.

(Souptick, if you're interested, the Case 5 documentation change and callsite retrofit is all yours if you want it. Otherwise it's on my list.)

thanks,

Signature · 2 lines
John Hubbard
NVIDIA
maintainer
#11 lore raw
Quoted 10 lines · John Hubbard
Quoted 5 lines

Excuse me, but Documentation/core-api/pin_user_pages.rst says "CASE 5: Pinning in order to _write_ to the data within the page" while tomoyo_dump_page() is for "_read_ the data within the page". Do we want to convert to pin_user_pages_remote() or lock_page() ?

Sorry, I missed the direction here, was too focused on the Case 5 aspect. Yes. Case 5 (which, again, I think we're about to re-document) is only about *writing* to data within the page.

Quoted 7 lines

So in this case, where it is just reading from the page, I think it's already from a gup vs pup point of view.

btw, it's not clear to me whether the current code is susceptible to any sort of problem involving something writing to the page while it is being dumped (I am curious). But changing from gup to pup wouldn't fix that, if it were a problem. It a separate question from this patch.

The "struct page" tomoyo_dump_page() accesses is argv/envp arguments passed to execve() syscall. Therefore, these pages are not visible from threads except current thread, and thus there is no possibility that these pages are modified by other threads while current thread is reading.

#12 lore raw
Quoted 19 lines · Tetsuo Handa
Quoted 10 lines · John Hubbard
Quoted 5 lines

Excuse me, but Documentation/core-api/pin_user_pages.rst says "CASE 5: Pinning in order to _write_ to the data within the page" while tomoyo_dump_page() is for "_read_ the data within the page". Do we want to convert to pin_user_pages_remote() or lock_page() ?

Sorry, I missed the direction here, was too focused on the Case 5 aspect. Yes. Case 5 (which, again, I think we're about to re-document) is only about *writing* to data within the page.

Quoted 7 lines

So in this case, where it is just reading from the page, I think it's already from a gup vs pup point of view.

btw, it's not clear to me whether the current code is susceptible to any sort of problem involving something writing to the page while it is being dumped (I am curious). But changing from gup to pup wouldn't fix that, if it were a problem. It a separate question from this patch.

Quoted 5 lines

The "struct page" tomoyo_dump_page() accesses is argv/envp arguments passed to execve() syscall. Therefore, these pages are not visible from threads except current thread, and thus there is no possibility that these pages are modified by other threads while current thread is reading.

Perfect. So since I accidentally left out the word "correct" above (I meant to write, "it's already correct"), let me be extra clear: Souptick, we should just drop this patch.

thanks,

Signature · 2 lines
John Hubbard
NVIDIA
#14 lore raw
In reply to John Hubbard
Quoted 39 lines · John Hubbard
Quoted 24 lines · Tetsuo Handa
Quoted 13 lines · John Hubbard
Quoted 12 lines · John Hubbard
Quoted 3 lines · Souptick Joarder

In 2019, we introduced pin_user_pages*() and now we are converting get_user_pages*() to the new API as appropriate. [1] & [2] could be referred for more information. This is case 5 as per document [1].

It turns out that Case 5 can be implemented via a better pattern, as long as we're just dealing with a page at a time, briefly:

lock_page()
write to page's data
unlock_page()

Quoted 9 lines
Quoted 1 line

...which neatly synchronizes with writeback and other fs activities.

Ahem, I left out a key step: set_page_dirty()!

lock_page()
write to page's data
set_page_dirty()
unlock_page()

Quoted 5 lines

Excuse me, but Documentation/core-api/pin_user_pages.rst says "CASE 5: Pinning in order to _write_ to the data within the page" while tomoyo_dump_page() is for "_read_ the data within the page". Do we want to convert to pin_user_pages_remote() or lock_page() ?

Sorry, I missed the direction here, was too focused on the Case 5 aspect. Yes. Case 5 (which, again, I think we're about to re-document) is only about *writing* to data within the page.

So in this case, where it is just reading from the page, I think it's already from a gup vs pup point of view.

Quoted 8 lines

btw, it's not clear to me whether the current code is susceptible to any sort of problem involving something writing to the page while it is being dumped (I am curious). But changing from gup to pup wouldn't fix that, if it were a problem. It a separate question from this patch.

(Souptick, if you're interested, the Case 5 documentation change and callsite retrofit is all yours if you want it. Otherwise it's on my list.)

Sure John, I will take it.

#13 lore raw
In reply to John Hubbard
Quoted 27 lines · John Hubbard
Quoted 19 lines · Tetsuo Handa
Quoted 10 lines · John Hubbard
Quoted 5 lines

Excuse me, but Documentation/core-api/pin_user_pages.rst says "CASE 5: Pinning in order to _write_ to the data within the page" while tomoyo_dump_page() is for "_read_ the data within the page". Do we want to convert to pin_user_pages_remote() or lock_page() ?

Sorry, I missed the direction here, was too focused on the Case 5 aspect. Yes. Case 5 (which, again, I think we're about to re-document) is only about *writing* to data within the page.

Quoted 7 lines

So in this case, where it is just reading from the page, I think it's already from a gup vs pup point of view.

btw, it's not clear to me whether the current code is susceptible to any sort of problem involving something writing to the page while it is being dumped (I am curious). But changing from gup to pup wouldn't fix that, if it were a problem. It a separate question from this patch.

Quoted 5 lines

The "struct page" tomoyo_dump_page() accesses is argv/envp arguments passed to execve() syscall. Therefore, these pages are not visible from threads except current thread, and thus there is no possibility that these pages are modified by other threads while current thread is reading.

Quoted 4 lines

Perfect. So since I accidentally left out the word "correct" above (I meant to write, "it's already correct"), let me be extra clear: Souptick, we should just drop this patch.

Agreed. I will drop this patch.

#3 lore raw

Re: [PATCH 2/2] tomoyo: Fixed typo in documentation

In reply to John Hubbard
Quoted 25 lines · John Hubbard
Quoted 14 lines · Souptick Joarder

Fixed typo s/Poiner/Pointer

Fixes 5b636857fee6 ("TOMOYO: Allow using argv[]/envp[] of execve() as conditions.")
Cc John Hubbard <[email protected]>
---
  security/tomoyo/domain.c | 2 +-
  1 file changed, 1 insertion(+), 1 deletion(-)
security/tomoyo/domain.c +0 -0
index bd748be..7b2babe 100644

security/tomoyo/domain.c int tomoyo_find_next_domain(struct linux_binprm *bprm) · line 894

Quoted 5 lines
security/tomoyo/domain.c +1 -1
@@ -891,7 +891,7 @@ int tomoyo_find_next_domain(struct linux_binprm *bprm)
   *
   * @bprm: Pointer to "struct linux_binprm".
   * @pos:  Location to dump.
- * @dump: Poiner to "struct tomoyo_page_dump".
+ * @dump: Pointer to "struct tomoyo_page_dump".

Not worth a separate patch, especially since the original comment is merely copying the C sources, and as such, does not add any value.

Quoted 5 lines

I'd either a) craft a new documentation line that adds some value, or b) just merge this patch into the previous one, and make a note in the commit description to the effect that you've included a trivial typo fix as long as you're there.

John, as patch[1/2] is dropped, can we take this patch forward with some more updates in documentations ?

#4 lore raw

Re: [PATCH 2/2] tomoyo: Fixed typo in documentation

Quoted 26 lines · Souptick Joarder
Quoted 25 lines · John Hubbard
Quoted 14 lines · Souptick Joarder

Fixed typo s/Poiner/Pointer

Fixes 5b636857fee6 ("TOMOYO: Allow using argv[]/envp[] of execve() as conditions.")
Cc John Hubbard <[email protected]>
---
   security/tomoyo/domain.c | 2 +-
   1 file changed, 1 insertion(+), 1 deletion(-)
security/tomoyo/domain.c +0 -0
index bd748be..7b2babe 100644

security/tomoyo/domain.c int tomoyo_find_next_domain(struct linux_binprm *bprm) · line 894

Quoted 5 lines
security/tomoyo/domain.c +1 -1
@@ -891,7 +891,7 @@ int tomoyo_find_next_domain(struct linux_binprm *bprm)
    *
    * @bprm: Pointer to "struct linux_binprm".
    * @pos:  Location to dump.
- * @dump: Poiner to "struct tomoyo_page_dump".
+ * @dump: Pointer to "struct tomoyo_page_dump".

Not worth a separate patch, especially since the original comment is merely copying the C sources, and as such, does not add any value.

Quoted 9 lines
Quoted 5 lines

I'd either a) craft a new documentation line that adds some value, or b) just merge this patch into the previous one, and make a note in the commit description to the effect that you've included a trivial typo fix as long as you're there.

John, as patch[1/2] is dropped, can we take this patch forward with some more updates in documentations ?

That's really up to the folks who work on this code. Personally I would rarely post a patch *just* for this, but on the other hand it is a correction. Either way is fine with me of course.

thanks,

Signature · 2 lines
John Hubbard
NVIDIA
maintainer
#5 lore raw

Re: [PATCH 2/2] tomoyo: Fixed typo in documentation

Quoted 8 lines · John Hubbard
Quoted 3 lines · Souptick Joarder

John, as patch[1/2] is dropped, can we take this patch forward with some more updates in documentations ?

That's really up to the folks who work on this code. Personally I would rarely post a patch *just* for this, but on the other hand it is a correction. Either way is fine with me of course.

Xiaoming Ni has posted a patch for fixing spelling errors in security/ directory at https://lkml.kernel.org/r/[email protected] .

James, can we apply that patch? Or, should we split that patch into each subsystem?

maintainer
#6 lore raw

Re: [PATCH 2/2] tomoyo: Fixed typo in documentation

Quoted 6 lines · Tetsuo Handa
Quoted 5 lines · John Hubbard
Quoted 3 lines · Souptick Joarder

John, as patch[1/2] is dropped, can we take this patch forward with some more updates in documentations ?

Quoted 9 lines
Quoted 3 lines

That's really up to the folks who work on this code. Personally I would rarely post a patch *just* for this, but on the other hand it is a correction. Either way is fine with me of course.

Xiaoming Ni has posted a patch for fixing spelling errors in security/ directory at https://lkml.kernel.org/r/[email protected] .

James, can we apply that patch? Or, should we split that patch into each subsystem?

Probably easiest to go via my tree, the author should resend and include me on the To: line.

Signature · 2 lines
James Morris
<[email protected]>