diff mbox series

[v2] x86: Put trampoline in separate .init.trampoline section

Message ID 20240911095550.31139-1-frediano.ziglio@cloud.com (mailing list archive)
State Superseded
Headers show
Series [v2] x86: Put trampoline in separate .init.trampoline section | expand

Commit Message

Frediano Ziglio Sept. 11, 2024, 9:55 a.m. UTC
This change put the trampoline in a separate, not executable section.
The trampoline contains a mix of code and data (data which
is modified from C code during early start so must be writable).
This is in preparation for W^X patch in order to satisfy UEFI CA
memory mitigation requirements.
At the moment .init.text and .init.data in EFI mode are put together
so they will be in the same final section as before this patch.
Putting in a separate section (even in final executables) allows
to easily disassembly that section.

Signed-off-by: Frediano Ziglio <frediano.ziglio@cloud.com>
---
Changes since last version:
- use completely separate section even on final executables
  (suggested by Jan Beulich).

Changes since v1:
- remove useless align.
---
 xen/arch/x86/boot/head.S | 5 +++--
 xen/arch/x86/xen.lds.S   | 5 +++++
 2 files changed, 8 insertions(+), 2 deletions(-)

Comments

Jan Beulich Sept. 14, 2024, 6:16 a.m. UTC | #1
On 11.09.2024 11:55, Frediano Ziglio wrote:
> --- a/xen/arch/x86/boot/head.S
> +++ b/xen/arch/x86/boot/head.S
> @@ -882,8 +882,9 @@ cmdline_parse_early:
>  reloc:
>          .incbin "reloc.bin"
>  
> +#include "x86_64.S"
> +
> +        .section .init.trampoline, "aw", @progbits
>  ENTRY(trampoline_start)
>  #include "trampoline.S"
>  ENTRY(trampoline_end)

Hmm, nice - this turns out rather easier than I first thought.

> --- a/xen/arch/x86/xen.lds.S
> +++ b/xen/arch/x86/xen.lds.S
> @@ -269,6 +269,11 @@ SECTIONS
>         __ctors_end = .;
>    } PHDR(text)
>  
> +  . = ALIGN(PAGE_SIZE);

Why? There's no special alignment right now.

> +  DECL_SECTION(.init.trampoline) {
> +       *(.init.trampoline)
> +  } PHDR(text)
> +
>  #ifndef EFI

If this is to be a separate section also for ELF, I think that
wants mentioning explicitly. "Easily disassemble" is too vague
a reason for my taste.

Jan
Frediano Ziglio Sept. 16, 2024, 9:11 a.m. UTC | #2
On Sat, Sep 14, 2024 at 7:16 AM Jan Beulich <jbeulich@suse.com> wrote:
>
> On 11.09.2024 11:55, Frediano Ziglio wrote:
> > --- a/xen/arch/x86/boot/head.S
> > +++ b/xen/arch/x86/boot/head.S
> > @@ -882,8 +882,9 @@ cmdline_parse_early:
> >  reloc:
> >          .incbin "reloc.bin"
> >
> > +#include "x86_64.S"
> > +
> > +        .section .init.trampoline, "aw", @progbits
> >  ENTRY(trampoline_start)
> >  #include "trampoline.S"
> >  ENTRY(trampoline_end)
>
> Hmm, nice - this turns out rather easier than I first thought.
>
> > --- a/xen/arch/x86/xen.lds.S
> > +++ b/xen/arch/x86/xen.lds.S
> > @@ -269,6 +269,11 @@ SECTIONS
> >         __ctors_end = .;
> >    } PHDR(text)
> >
> > +  . = ALIGN(PAGE_SIZE);
>
> Why? There's no special alignment right now.
>

UEFI CA Memory Mitigation requirements, I'll remove from this commit.

> > +  DECL_SECTION(.init.trampoline) {
> > +       *(.init.trampoline)
> > +  } PHDR(text)
> > +
> >  #ifndef EFI
>
> If this is to be a separate section also for ELF, I think that
> wants mentioning explicitly. "Easily disassemble" is too vague
> a reason for my taste.

It's not clear if either you changed your mind on that reason or if
the commit message is not clear. I'm assuming the latter, I'll improve
the commit message.
Not clear why you specify "ELF" in the above sentence, I mean, why
should it matter if it applies to EFI and/or ELF? And why having it
different from ELF to EFI?

>
> Jan

Frediano
Jan Beulich Sept. 16, 2024, 11:49 a.m. UTC | #3
On 16.09.2024 11:11, Frediano Ziglio wrote:
> On Sat, Sep 14, 2024 at 7:16 AM Jan Beulich <jbeulich@suse.com> wrote:
>> On 11.09.2024 11:55, Frediano Ziglio wrote:
>>> +  DECL_SECTION(.init.trampoline) {
>>> +       *(.init.trampoline)
>>> +  } PHDR(text)
>>> +
>>>  #ifndef EFI
>>
>> If this is to be a separate section also for ELF, I think that
>> wants mentioning explicitly. "Easily disassemble" is too vague
>> a reason for my taste.
> 
> It's not clear if either you changed your mind on that reason or if
> the commit message is not clear. I'm assuming the latter, I'll improve
> the commit message.
> Not clear why you specify "ELF" in the above sentence, I mean, why
> should it matter if it applies to EFI and/or ELF? And why having it
> different from ELF to EFI?

Why do we fold sections at all? We could keep them all separate, leading
to a big section table with -f{function,data}-sections. If you want to
change something, you need to clarify why the change is being made, for
every part that is affected. We don't have a need for the separate
section in ELF, so it becoming one without at least saying so (and
ideally also why) it might look like an oversight / mistake to someone
later stumbling across that commit.

Jan
diff mbox series

Patch

diff --git a/xen/arch/x86/boot/head.S b/xen/arch/x86/boot/head.S
index 12bbb97f33..493286a9fb 100644
--- a/xen/arch/x86/boot/head.S
+++ b/xen/arch/x86/boot/head.S
@@ -882,8 +882,9 @@  cmdline_parse_early:
 reloc:
         .incbin "reloc.bin"
 
+#include "x86_64.S"
+
+        .section .init.trampoline, "aw", @progbits
 ENTRY(trampoline_start)
 #include "trampoline.S"
 ENTRY(trampoline_end)
-
-#include "x86_64.S"
diff --git a/xen/arch/x86/xen.lds.S b/xen/arch/x86/xen.lds.S
index d48de67cfd..390870e463 100644
--- a/xen/arch/x86/xen.lds.S
+++ b/xen/arch/x86/xen.lds.S
@@ -269,6 +269,11 @@  SECTIONS
        __ctors_end = .;
   } PHDR(text)
 
+  . = ALIGN(PAGE_SIZE);
+  DECL_SECTION(.init.trampoline) {
+       *(.init.trampoline)
+  } PHDR(text)
+
 #ifndef EFI
   /*
    * With --orphan-sections=warn (or =error) we need to handle certain linker