Skip to contentSkip to conversation

Various patches to improve Secure Boot support

PATCH Superseded Frediano Ziglio 2 participants 12 messages Jun 16 – Jun 26

v4 of 8 Superseded Latest is v8 →

lore
Share & tools
lore
#0 lore raw

v4 cover Various patches to improve Secure Boot support

These patches improve support for Secure boot.
UEFI CA memory mitigation requires memory pages to be not executable and writable at the same time. So changing permissions and splitting some section is required.
Remove multiboot pieces from EFI executable.

Changes since v1:
- improved some comments;
- merged 2 pacthes removing multiboot support in x86 PE;
- removed a patch dealing with SBAT;
- other minor changes (see single patches).

Changes since v2:
- improved some comments.

Changes since v3:
- Added Acked-by;
- Improve commit message.

Frediano Ziglio (2):
Align relevant sections to 4KB
x86: Split .init section to satisfy UEFI CA memory mitigation

Roger Pau Monné (2):
x86/efi: discard multiboot support for PE binary x86/efi: avoid a relocation in efi_arch_post_exit_boot()

 docs/hypervisor-guide/x86/how-xen-boots.rst |  6 ------
 xen/arch/x86/boot/head.S                    |  3 ++-
 xen/arch/x86/efi/efi-boot.h                 |  7 +++++--
 xen/arch/x86/xen.lds.S                      | 22 +++++++++++----------
 4 files changed, 19 insertions(+), 19 deletions(-)
Signature · 1 line
2.43.0
#1 lore raw

PATCH v4 1/4 Align relevant sections to 4KB

From: Frediano Ziglio <[email protected]>

Required by UEFI CA memory mitigation.

It is a requirement for NX_COMPAT so the PE can be loaded with W^X perms in the pagetables.

NX_COMPAT is a requirement from shim-review,
https://github.com/rhboot/shim-review#do-you-have-the-nx-bit-set-in-your-shim-if-so-is-your-entire-boot-stack-nx-compatible-and-what-testing-have-you-done-to-ensure-such-compatibility

Sections with different permissions must be in separate pages. In the case of debug sections they are contiguous and have the same permissions, including the immediately preceding .reloc section, so it's not an issue if they are not aligned to the page.
Before the .debug sections you could have the .reloc or the SBAT section, either are permission-compatible.

--
Changes since v1:
- Change subject.

Changes since v2:
- Improved commit message and subject.

Changes since v3:
- Added Acked-by;
- Improved commit message.

---
 xen/arch/x86/xen.lds.S | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
xen/arch/x86/xen.lds.S +3 -2
index b9e888e596..f758940674 100644
@@ -162,8 +162,8 @@ SECTIONS
        __note_gnu_build_id_end = .;
   } PHDR(note) PHDR(text)
 #elif defined(BUILD_ID_EFI)
-  /* Workaround bug in binutils < 2.36 */
-  . = ALIGN(32);
+  /* Align to satisfy UEFI CA memory mitigation. */
+  . = ALIGN(PAGE_SIZE);
   DECL_SECTION(.buildid) {
        __note_gnu_build_id_start = .;
        *(.buildid)
@@ -330,6 +330,7 @@ SECTIONS
   __2M_rwdata_end = ALIGN(SECTION_ALIGN);
 
 #ifdef EFI
+  . = ALIGN(PAGE_SIZE);
   .reloc ALIGN(4) : {
     __base_relocs_start = .;
     *(.reloc)
Signature · 1 line
2.43.0
Acked-by on v4 1/4
#2 lore raw

Re: [PATCH v4 1/4] Align relevant sections to 4KB

Quoted 17 lines · Frediano Ziglio

From: Frediano Ziglio <[email protected]>

Required by UEFI CA memory mitigation.

It is a requirement for NX_COMPAT so the PE can be loaded with W^X perms in the pagetables.

NX_COMPAT is a requirement from shim-review,
https://github.com/rhboot/shim-review#do-you-have-the-nx-bit-set-in-your-shim-if-so-is-your-entire-boot-stack-nx-compatible-and-what-testing-have-you-done-to-ensure-such-compatibility

Sections with different permissions must be in separate pages. In the case of debug sections they are contiguous and have the same permissions, including the immediately preceding .reloc section, so it's not an issue if they are not aligned to the page.
Before the .debug sections you could have the .reloc or the SBAT section, either are permission-compatible.

Quoted 2 lines

As per the reply to patch 4:

with ...

Quoted 14 lines
xen/arch/x86/xen.lds.S +2 -2
@@ -162,8 +162,8 @@ SECTIONS
        __note_gnu_build_id_end = .;
   } PHDR(note) PHDR(text)
 #elif defined(BUILD_ID_EFI)
-  /* Workaround bug in binutils < 2.36 */
-  . = ALIGN(32);
+  /* Align to satisfy UEFI CA memory mitigation. */
+  . = ALIGN(PAGE_SIZE);
   DECL_SECTION(.buildid) {
        __note_gnu_build_id_start = .;
        *(.buildid)

xen/arch/x86/xen.lds.S SECTIONS · line 333

Quoted 4 lines
xen/arch/x86/xen.lds.S +1 -0
@@ -330,6 +330,7 @@ SECTIONS
   __2M_rwdata_end = ALIGN(SECTION_ALIGN);
 
 #ifdef EFI
+  . = ALIGN(PAGE_SIZE);

... the same comment added here as well. Can do while committing, as long as you agree.

Jan

#3 lore raw

PATCH v4 2/4 x86/efi: discard multiboot support for PE binary

In reply to Frediano Ziglio

From: Roger Pau Monné <[email protected]>

Multiboot and PVH booting are not supported for PE, hence discards them in the linker script when doing a PE build.

That removes some relocations that otherwise appear due to the usage of the start and __efi64_mb2_start symbols in the multiboot2 header.

Section discarding is not done updating DISCARD_SECTIONS definition as the change is specific for x86.

No functional change intended.

--
Changes since v1:
- improve commit message;
- change section orders to avoid changing code order in final executable;
- merge 2 commits;
- removed deprecated documentation section.

Changes since v2:
- Update commit message, join 2 sentences together.

Changes since v3:
- Added Acked-by.

---
 docs/hypervisor-guide/x86/how-xen-boots.rst | 6 ------
 xen/arch/x86/boot/head.S                    | 3 ++-
 xen/arch/x86/xen.lds.S                      | 5 +++++
 3 files changed, 7 insertions(+), 7 deletions(-)
docs/hypervisor-guide/x86/how-xen-boots.rst +0 -6
index 8b3229005c..b6d852050a 100644
@@ -82,12 +82,6 @@ When a PEI-capable toolchain is found, the objects are linked together and a
 PE32+ binary is created.  It can be run directly from the EFI shell, and has
 ``efi_start`` as its entry symbol.
 
-.. note::
-
-   xen.efi does contain all MB1/MB2/PVH tags included in the rest of the
-   build.  However, entry via anything other than the EFI64 protocol is
-   unsupported, and won't work.
-
 
 Boot
 ----
xen/arch/x86/boot/head.S +2 -1
index 77bb7a9e21..90faf411b9 100644
@@ -152,7 +152,7 @@ vga_text_buffer:
 efi_platform:
         .byte   0
 
-        .section .init.text, "ax", @progbits
+        .section .init.multiboot, "ax", @progbits
 
 early_error: /* Here to improve the disassembly. */
 
@@ -710,6 +710,7 @@ trampoline_setup:
         /* Jump into the relocated trampoline. */
         lret
 
+        .section .init.text, "ax", @progbits
 ENTRY(trampoline_start)
 #include "trampoline.S"
 ENTRY(trampoline_end)
xen/arch/x86/xen.lds.S +5 -0
index f758940674..749d9719cc 100644
@@ -57,6 +57,10 @@ SECTIONS
   __image_base__ = .;
 #else
   . = __image_base__;
+  /DISCARD/ : {
+    *(.text.header)
+    *(.init.multiboot)
+  }
 #endif
 
 #if 0
@@ -195,6 +199,7 @@ SECTIONS
   DECL_SECTION(.init.text) {
 #endif
        _sinittext = .;
+       *(.init.multiboot)
        *(.init.text)
        *(.text.startup)
        _einittext = .;
Signature · 1 line
2.43.0
#4 lore raw

Re: [PATCH v4 2/4] x86/efi: discard multiboot support for PE binary

Quoted 8 lines · Frediano Ziglio

From: Roger Pau Monné <[email protected]>

Multiboot and PVH booting are not supported for PE, hence discards them in the linker script when doing a PE build.

That removes some relocations that otherwise appear due to the usage of the start and __efi64_mb2_start symbols in the multiboot2 header.

Quoted 8 lines

Section discarding is not done updating DISCARD_SECTIONS definition as the change is specific for x86.

No functional change intended.

While on the surface this looks okay, there are still concerns:

For one, this also discards the PVH entry point. That's technically fine aiui, yet shouldn't go without mentioning.

Otoh you discard call sites of functions without discarding the functions themselves, violating Misra's "no unreachable code" rule. Eclair may not be able to spot this, but imo we should still adhere to the rule. Proper coverage analysis, for example, would likely turn this up.

Jan

#5 lore raw

Re: [PATCH v4 2/4] x86/efi: discard multiboot support for PE binary

Quoted 15 lines · Jan Beulich
Quoted 13 lines · Frediano Ziglio

From: Roger Pau Monné <[email protected]>

Multiboot and PVH booting are not supported for PE, hence discards them in the linker script when doing a PE build.

That removes some relocations that otherwise appear due to the usage of the start and __efi64_mb2_start symbols in the multiboot2 header.

Section discarding is not done updating DISCARD_SECTIONS definition as the change is specific for x86.

No functional change intended.

Quoted 9 lines
Quoted 3 lines

While on the surface this looks okay, there are still concerns:

For one, this also discards the PVH entry point. That's technically fine aiui, yet shouldn't go without mentioning.

Considering that the code/data is not exported in EFI as

#if defined(CONFIG_PVH_GUEST) && !defined(EFI)
/*
* In principle this should be fine to live in .note (below), but let's keep
* it separate in case anyone decided to find these notes by section name. */

  DECL_SECTION(.note.Xen) {
      *(.note.Xen)

} PHDR(note) PHDR(text)
#endif

yes, technically it's surely fine.

There's a mention in the commit log:

Multiboot and PVH booting are not supported for PE, hence discards them in the linker script when doing a PE build.

But not in the subject:

x86/efi: discard multiboot support for PE binary

What about simply changing the subject to:

x86/efi: discard multiboot and PVH support for PE binary

Quoted 5 lines

Otoh you discard call sites of functions without discarding the functions themselves, violating Misra's "no unreachable code" rule. Eclair may not be able to spot this, but imo we should still adhere to the rule. Proper coverage analysis, for example, would likely turn this up.

That makes sense. Given that most code in head.S is now discarded most data sections are now not used and the only thing left will be the trampoline.
It'll take a bit of time to search for removed symbols.

About the "no unreachable code" I think we are violating that anyway. We package "built-in.o" files and then use them to craft the final executable. I don't think the linker will be able to discard unused functions for that reason. That does not mean that more things can be discarded.

About discarding more (I think a bit out of scope here) for the EFI application I don't think we need BIOS/EDD code. Not sure how easy is to do this without having 2 objects (one for EFI and one for ELF).

Quoted 1 line

Jan

Frediano

#6 lore raw

Re: [PATCH v4 2/4] x86/efi: discard multiboot support for PE binary

Quoted 45 lines · Frediano Ziglio
Quoted 15 lines · Jan Beulich
Quoted 13 lines · Frediano Ziglio

From: Roger Pau Monné <[email protected]>

Multiboot and PVH booting are not supported for PE, hence discards them in the linker script when doing a PE build.

That removes some relocations that otherwise appear due to the usage of the start and __efi64_mb2_start symbols in the multiboot2 header.

Section discarding is not done updating DISCARD_SECTIONS definition as the change is specific for x86.

No functional change intended.

Quoted 9 lines
Quoted 3 lines

While on the surface this looks okay, there are still concerns:

For one, this also discards the PVH entry point. That's technically fine aiui, yet shouldn't go without mentioning.

Considering that the code/data is not exported in EFI as

#if defined(CONFIG_PVH_GUEST) && !defined(EFI)
/*
* In principle this should be fine to live in .note (below), but let's keep
* it separate in case anyone decided to find these notes by section name. */

  DECL_SECTION(.note.Xen) {
      *(.note.Xen)

} PHDR(note) PHDR(text)
#endif

yes, technically it's surely fine.

There's a mention in the commit log:

Multiboot and PVH booting are not supported for PE, hence discards them in the linker script when doing a PE build.

Quoted 7 lines

But not in the subject:

x86/efi: discard multiboot support for PE binary

What about simply changing the subject to:

x86/efi: discard multiboot and PVH support for PE binary

Perhaps.

Quoted 6 lines
Quoted 5 lines

Otoh you discard call sites of functions without discarding the functions themselves, violating Misra's "no unreachable code" rule. Eclair may not be able to spot this, but imo we should still adhere to the rule. Proper coverage analysis, for example, would likely turn this up.

Quoted 6 lines

That makes sense. Given that most code in head.S is now discarded most data sections are now not used and the only thing left will be the trampoline.
It'll take a bit of time to search for removed symbols.

About the "no unreachable code" I think we are violating that anyway.

Perhaps, but we should get the number of such violations down, not up.

Quoted 4 lines

We package "built-in.o" files and then use them to craft the final executable. I don't think the linker will be able to discard unused functions for that reason. That does not mean that more things can be discarded.

At least not until we engage it garbage collection, which as per Jason proves problematic when linking xen.efi (due to linker issues as it looks).

Jan

#7 lore raw

Re: [PATCH v4 2/4] x86/efi: discard multiboot support for PE binary

Quoted 49 lines · Jan Beulich
Quoted 47 lines · Frediano Ziglio
Quoted 15 lines · Jan Beulich
Quoted 13 lines · Frediano Ziglio

From: Roger Pau Monné <[email protected]>

Multiboot and PVH booting are not supported for PE, hence discards them in the linker script when doing a PE build.

That removes some relocations that otherwise appear due to the usage of the start and __efi64_mb2_start symbols in the multiboot2 header.

Section discarding is not done updating DISCARD_SECTIONS definition as the change is specific for x86.

No functional change intended.

Quoted 9 lines
Quoted 3 lines

While on the surface this looks okay, there are still concerns:

For one, this also discards the PVH entry point. That's technically fine aiui, yet shouldn't go without mentioning.

Considering that the code/data is not exported in EFI as

#if defined(CONFIG_PVH_GUEST) && !defined(EFI)
/*
* In principle this should be fine to live in .note (below), but let's keep
* it separate in case anyone decided to find these notes by section name. */

  DECL_SECTION(.note.Xen) {
      *(.note.Xen)

} PHDR(note) PHDR(text)
#endif

yes, technically it's surely fine.

There's a mention in the commit log:

Multiboot and PVH booting are not supported for PE, hence discards them in the linker script when doing a PE build.

But not in the subject:

Quoted 8 lines
Quoted 5 lines

x86/efi: discard multiboot support for PE binary

What about simply changing the subject to:

x86/efi: discard multiboot and PVH support for PE binary

Perhaps.

Updated.

Quoted 6 lines
Quoted 6 lines
Quoted 5 lines

Otoh you discard call sites of functions without discarding the functions themselves, violating Misra's "no unreachable code" rule. Eclair may not be able to spot this, but imo we should still adhere to the rule. Proper coverage analysis, for example, would likely turn this up.

Quoted 9 lines
Quoted 6 lines

That makes sense. Given that most code in head.S is now discarded most data sections are now not used and the only thing left will be the trampoline.
It'll take a bit of time to search for removed symbols.

About the "no unreachable code" I think we are violating that anyway.

Perhaps, but we should get the number of such violations down, not up.

It was not meant to be an excuse, more of a question if the problem is known. The "It'll take a bit of time to search for removed symbols" was a "I will do". I now have the fixup patch for "x86/efi: discard multiboot and PVH support for PE binary" (the commit we are talking about here). About sending an updated series, what is the best way to send a fixup patch? Send the fixup as separate? Merge into the base patch and remove the "acked-by"? Keep the "acked-by"?

Quoted 7 lines
Quoted 4 lines

We package "built-in.o" files and then use them to craft the final executable. I don't think the linker will be able to discard unused functions for that reason. That does not mean that more things can be discarded.

At least not until we engage it garbage collection, which as per Jason proves problematic when linking xen.efi (due to linker issues as it looks).

In my experience linker garbage collector on Linux is a bit problematic, I think LTO is more tested and working. Still not that automatic.

I tried to remove the BIOS code using sections "trick" like what was done in this commit without success.

Quoted 1 line

Jan

Frediano

#8 lore raw

Re: [PATCH v4 2/4] x86/efi: discard multiboot support for PE binary

Quoted 20 lines · Frediano Ziglio
Quoted 9 lines · Jan Beulich
Quoted 8 lines · Frediano Ziglio
Quoted 6 lines · Jan Beulich

On 16.06.2026 19:28, Frediano Ziglio wrote:
Otoh you discard call sites of functions without discarding the functions themselves, violating Misra's "no unreachable code" rule. Eclair may not be able to spot this, but imo we should still adhere to the rule. Proper coverage analysis, for example, would likely turn this up.

Quoted 9 lines
Quoted 6 lines

That makes sense. Given that most code in head.S is now discarded most data sections are now not used and the only thing left will be the trampoline.
It'll take a bit of time to search for removed symbols.

About the "no unreachable code" I think we are violating that anyway.

Perhaps, but we should get the number of such violations down, not up.

Quoted 7 lines

It was not meant to be an excuse, more of a question if the problem is known. The "It'll take a bit of time to search for removed symbols" was a "I will do". I now have the fixup patch for "x86/efi: discard multiboot and PVH support for PE binary" (the commit we are talking about here). About sending an updated series, what is the best way to send a fixup patch? Send the fixup as separate? Merge into the base patch and remove the "acked-by"? Keep the "acked-by"?

R-b would need dropping. A-b is more of a judgement call.

Jan

#9 lore raw

PATCH v4 3/4 x86/efi: avoid a relocation in efi_arch_post_exit_boot()

In reply to Frediano Ziglio

From: Roger Pau Monné <[email protected]>

Instead of using the absolute __start_xen address, calculate it as an offset from the current instruction pointer. The relocation would be problematic if the generated PE binary had .init.text as a standalone section with just read and execute permissions."

Removing this relocation is necessary to make it safe to split .init.

No functional change intended.

--
Changes since v1:
- Improve commit message.

Changes since v3:
- Added Acked-by.

---
 xen/arch/x86/efi/efi-boot.h | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)
xen/arch/x86/efi/efi-boot.h +5 -2
index d738b839ee..b983f054b5 100644
@@ -270,7 +270,9 @@ static void __init noreturn efi_arch_post_exit_boot(void)
 
                    /* Jump to higher mappings. */
                    "mov    stack_start(%%rip), %%rsp\n\t"
-                   "movabs $__start_xen, %[rip]\n\t"
+                   "lea    __start_xen(%%rip), %[rip]\n\t"
+                   "add    %[offset], %[rip]\n\t"
+
                    "push   %[cs]\n\t"
                    "push   %[rip]\n\t"
                    "lretq"
@@ -278,7 +280,8 @@ static void __init noreturn efi_arch_post_exit_boot(void)
                      [cr4] "+&r" (cr4)
                    : [cr3] "r" (idle_pg_table),
                      [cs] "i" (__HYPERVISOR_CS),
-                     [ds] "r" (__HYPERVISOR_DS)
+                     [ds] "r" (__HYPERVISOR_DS),
+                     [offset] "r" (__XEN_VIRT_START - xen_phys_start)
                    : "memory" );
     unreachable();
 }
Signature · 1 line
2.43.0
#10 lore raw

PATCH v4 4/4 x86: Split .init section to satisfy UEFI CA memory mitigation

In reply to Frediano Ziglio

From: Frediano Ziglio <[email protected]>

Currently .init section is both writeable and executable, split data and code to have 2 sections satisfying W^X rule.

It is a requirement for NX_COMPAT so the PE can be loaded with W^X perms in the pagetables.

NX_COMPAT is a requirement from shim-review,
https://github.com/rhboot/shim-review#do-you-have-the-nx-bit-set-in-your-shim-if-so-is-your-entire-boot-stack-nx-compatible-and-what-testing-have-you-done-to-ensure-such-compatibility

--
Change since v1:
- update comment style.

Changes since v3:
- Added Acked-by.

---
 xen/arch/x86/xen.lds.S | 12 ++++--------
 1 file changed, 4 insertions(+), 8 deletions(-)
xen/arch/x86/xen.lds.S +4 -8
index 749d9719cc..8fefda1816 100644
@@ -193,11 +193,7 @@ SECTIONS
   __2M_init_start = .;         /* Start of 2M superpages, mapped RWX (boot only). */
   . = ALIGN(PAGE_SIZE);             /* Init code and data */
   __init_begin = .;
-#ifdef EFI /* EFI wants to merge all of .init.*  ELF doesn't. */
-  DECL_SECTION(.init) {
-#else
   DECL_SECTION(.init.text) {
-#endif
        _sinittext = .;
        *(.init.multiboot)
        *(.init.text)
@@ -210,12 +206,12 @@ SECTIONS
         */
        *(.altinstr_replacement)
 
-#ifdef EFI /* EFI wants to merge all of .init.*  ELF doesn't. */
-       . = ALIGN(SMP_CACHE_BYTES);
-#else
   } PHDR(text)
-  DECL_SECTION(.init.data) {
+#ifdef EFI
+  /* Align to satisfy UEFI CA memory mitigation. */
+  . = ALIGN(SECTION_ALIGN);
 #endif
+  DECL_SECTION(.init.data) {
        *(.init.bss.stack_aligned)
        *(.init.data.page_aligned)
 
Signature · 1 line
2.43.0
Acked-by on v4 4/4
#11 lore raw

Re: [PATCH v4 4/4] x86: Split .init section to satisfy UEFI CA memory mitigation

Quoted 5 lines · Frediano Ziglio

From: Frediano Ziglio <[email protected]>

Currently .init section is both writeable and executable, split data and code to have 2 sections satisfying W^X rule.

Quoted 8 lines

It is a requirement for NX_COMPAT so the PE can be loaded with W^X perms in the pagetables.

NX_COMPAT is a requirement from shim-review,
https://github.com/rhboot/shim-review#do-you-have-the-nx-bit-set-in-your-shim-if-so-is-your-entire-boot-stack-nx-compatible-and-what-testing-have-you-done-to-ensure-such-compatibility

This looks okay in principle, yet I have a nit-like question:

xen/arch/x86/xen.lds.S SECTIONS · line 212

Quoted 12 lines
xen/arch/x86/xen.lds.S +3 -4
@@ -210,12 +206,12 @@ SECTIONS
         */
        *(.altinstr_replacement)
 
-#ifdef EFI /* EFI wants to merge all of .init.*  ELF doesn't. */
-       . = ALIGN(SMP_CACHE_BYTES);
-#else
   } PHDR(text)
-  DECL_SECTION(.init.data) {
+#ifdef EFI
+  /* Align to satisfy UEFI CA memory mitigation. */
+  . = ALIGN(SECTION_ALIGN);

In patch 1 you add two such ALIGN()s, one with comment and one without. Imo we want to be consistent - all otherwise unnecessary ALIGN()s with comment, or (less desirable) just the 1st one. If we go with comments on all three of them:

Jan