Skip to contentSkip to conversation

x86: strict separation of startup code

PATCH Superseded Ard Biesheuvel 5 participants 32 messages Jul 16, 2025 – Jul 18, 2025

Subsystems (5)
x86 x86 MM EFI Xen hypervisor x86 +1 more subsystems

v5 of 7 Superseded Latest is v7 →

lore
Share & tools
lore
#0 lore raw

v5 cover x86: strict separation of startup code

From: Ard Biesheuvel <[email protected]>

This series implements a strict separation between startup code and ordinary code, where startup code is built in a way that tolerates being invoked from the initial 1:1 mapping of memory.

The existing approach of emitting this code into .head.text and checking for absolute relocations in that section is not 100% safe, and produces diagnostics that are sometimes difficult to interpret. [0]

Instead, rely on symbol prefixes, similar to how this is implemented for the EFI stub and for the startup code in the arm64 port. This ensures that startup code can only call other startup code, unless a special symbol alias is emitted that exposes a non-startup routine to the startup code.

This is somewhat intrusive, as there are many data objects that are referenced both by startup code and by ordinary code, and an alias needs to be emitted for each of those. If startup code references anything that has not been made available to it explicitly, a build time link error will occur.

This ultimately allows the .head.text section to be dropped entirely, as it no longer has a special significance. Instead, code that only executes at boot is emitted into .init.text as it should.

The majority of changes is around early SEV code. The main issue is that its use of GHCB pages and SVSM calling areas in code that may run from both the 1:1 mapping and the kernel virtual mapping is problematic as it relies on __pa() to perform VA to PA translations, which are ambiguous in this context. Also, __pa() pulls in non-trivial instrumented code when CONFIG_DEBUG_VIRTUAL=y and so it is better to avoid VA to PA translations altogether in the startup code.

Changes since v4:
- Incorporate feedback from Tom, and add a couple of RBs
- Drop patch that moved the MSR save/restore out of the early page state change helper - this is less efficient but likely negligible in practice
- Drop patch that unified the SEV-SNP hypervisor feature check, which was identified by Nikunj as the one breaking SEV-SNP boot.

Changes since RFT/v3:
- Rebase onto tip/master
- Incorporate Borislav's feedback on v3
- Switch to objtool to check for absolute references in startup code
- Remap inittext R-X when running on EFI implementations that require strict R-X/RW- separation
- Include a kbuild fix to incorporate arch/x86/boot/startup/ in the right manner
- For now, omit the LA57 changes that remove the problematic early 5-level paging checks. We can revisit this once there is agreement on the approach.

Changes since RFT/v2:
- Rebase onto tip/x86/boot and drop the patches from the previous revision that have been applied in the meantime.
- Omit the pgtable_l5_enabled() changes for now, and just expose PIC aliases for the variables in question - this can be sorted later.
- Don't use the boot SVSM calling area in snp_kexec_finish(), but pass down the correct per-CPU one to the early page state API.
- Rename arch/x86/coco/sev/sev-noinstr.o to arch/x86/coco/sev/noinstr.o
- Further reduce the amount of SEV code that needs to be constructed in a special way.

Change since RFC/v1:
- Include a major disentanglement/refactor of the SEV-SNP startup code, so that only code that really needs to run from the 1:1 mapping is included in the startup/ code

- Incorporate some early notes from Ingo

Cc Borislav Petkov <[email protected]>
Cc Ingo Molnar <[email protected]>
Cc Kevin Loughlin <[email protected]>
Cc Tom Lendacky <[email protected]>
Cc Josh Poimboeuf <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Nikunj A Dadhania <[email protected]>

[0] https://lore.kernel.org/all/CAHk-=wj7k9nvJn6cpa3-5Ciwn2RGyE605BMkjWE4MqnvC9E92A@mail.gmail.com/

Ard Biesheuvel (22):
x86/sev: Separate MSR and GHCB based snp_cpuid() via a callback x86/sev: Use MSR protocol for remapping SVSM calling area x86/sev: Use MSR protocol only for early SVSM PVALIDATE call x86/sev: Run RMPADJUST on SVSM calling area page to test VMPL x86/sev: Move GHCB page based HV communication out of startup code x86/sev: Avoid global variable to store virtual address of SVSM area x86/sev: Share implementation of MSR-based page state change x86/sev: Pass SVSM calling area down to early page state change API x86/sev: Use boot SVSM CA for all startup and init code x86/boot: Drop redundant RMPADJUST in SEV SVSM presence check x86/boot: Provide PIC aliases for 5-level paging related constants x86/sev: Provide PIC aliases for SEV related data objects x86/sev: Move __sev_[get|put]_ghcb() into separate noinstr object x86/sev: Export startup routines for later use
objtool: Add action to check for absence of absolute relocations x86/boot: Check startup code for absence of absolute relocations x86/boot: Revert "Reject absolute references in .head.text" x86/kbuild: Incorporate boot/startup/ via Kbuild makefile x86/boot: Create a confined code area for startup code efistub/x86: Remap inittext read-execute when needed x86/boot: Move startup code out of __head section x86/boot: Get rid of the .head.text section

 arch/x86/Kbuild                            |   2 +
 arch/x86/Makefile                          |   1 -
 arch/x86/boot/compressed/Makefile          |   2 +-
 arch/x86/boot/compressed/misc.c            |   2 +
 arch/x86/boot/compressed/sev-handle-vc.c   |   3 +
 arch/x86/boot/compressed/sev.c             | 108 +------
 arch/x86/boot/startup/Makefile             |  22 ++
 arch/x86/boot/startup/exports.h            |  14 +
 arch/x86/boot/startup/gdt_idt.c            |   4 +-
 arch/x86/boot/startup/map_kernel.c         |   4 +-
 arch/x86/boot/startup/sev-shared.c         | 317 ++++++--------------
 arch/x86/boot/startup/sev-startup.c        | 196 ++----------
 arch/x86/boot/startup/sme.c                |  27 +-
 arch/x86/coco/sev/Makefile                 |   6 +-
 arch/x86/coco/sev/core.c                   | 169 ++++++++---
 arch/x86/coco/sev/{sev-nmi.c => noinstr.c} |  74 +++++
 arch/x86/coco/sev/vc-handle.c              |   2 +
 arch/x86/coco/sev/vc-shared.c              | 143 ++++++++-
 arch/x86/include/asm/boot.h                |   2 +
 arch/x86/include/asm/init.h                |   6 -
 arch/x86/include/asm/setup.h               |   1 +
 arch/x86/include/asm/sev-internal.h        |  27 +-
 arch/x86/include/asm/sev.h                 |  17 +-
 arch/x86/kernel/head64.c                   |   5 +-
 arch/x86/kernel/head_32.S                  |   2 +-
 arch/x86/kernel/head_64.S                  |  10 +-
 arch/x86/kernel/vmlinux.lds.S              |   9 +-
 arch/x86/mm/mem_encrypt_amd.c              |   6 -
 arch/x86/mm/mem_encrypt_boot.S             |   6 +-
 arch/x86/platform/pvh/head.S               |   2 +-
 arch/x86/tools/relocs.c                    |   8 +-
 drivers/firmware/efi/libstub/x86-stub.c    |   4 +-
 tools/objtool/builtin-check.c              |   2 +
 tools/objtool/check.c                      |  39 ++-
 tools/objtool/include/objtool/builtin.h    |   1 +
 35 files changed, 620 insertions(+), 623 deletions(-)
 create mode 100644 arch/x86/boot/startup/exports.h
 rename arch/x86/coco/sev/{sev-nmi.c => noinstr.c} (61%)
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
#1 lore raw

PATCH v5 1/22 x86/sev: Separate MSR and GHCB based snp_cpuid() via a callback

From: Ard Biesheuvel <[email protected]>

There are two distinct callers of snp_cpuid(): one where the MSR protocol is always used, and one where the GHCB page based interface is always used.

The snp_cpuid() logic does not care about the distinction, which only matters at a lower level. But the fact that it supports both interfaces means that the GHCB page based logic is pulled into the early startup code where PA to VA conversions are problematic, given that it runs from the 1:1 mapping of memory.

So keep snp_cpuid() itself in the startup code, but factor out the hypervisor calls via a callback, so that the GHCB page handling can be moved out.

Code refactoring only - no functional change intended.

---
 arch/x86/boot/startup/sev-shared.c | 60 ++++----------------
 arch/x86/coco/sev/vc-shared.c      | 49 +++++++++++++++-
 arch/x86/include/asm/sev.h         |  3 +-
 3 files changed, 61 insertions(+), 51 deletions(-)
arch/x86/boot/startup/sev-shared.c +11 -49
index 7a706db87b93..c401d0391537 100644
@@ -342,44 +342,7 @@ static int __sev_cpuid_hv_msr(struct cpuid_leaf *leaf)
 	return ret;
 }
 
-static int __sev_cpuid_hv_ghcb(struct ghcb *ghcb, struct es_em_ctxt *ctxt, struct cpuid_leaf *leaf)
-{
-	u32 cr4 = native_read_cr4();
-	int ret;
-
-	ghcb_set_rax(ghcb, leaf->fn);
-	ghcb_set_rcx(ghcb, leaf->subfn);
-
-	if (cr4 & X86_CR4_OSXSAVE)
-		/* Safe to read xcr0 */
-		ghcb_set_xcr0(ghcb, xgetbv(XCR_XFEATURE_ENABLED_MASK));
-	else
-		/* xgetbv will cause #UD - use reset value for xcr0 */
-		ghcb_set_xcr0(ghcb, 1);
-
-	ret = sev_es_ghcb_hv_call(ghcb, ctxt, SVM_EXIT_CPUID, 0, 0);
-	if (ret != ES_OK)
-		return ret;
-
-	if (!(ghcb_rax_is_valid(ghcb) &&
-	      ghcb_rbx_is_valid(ghcb) &&
-	      ghcb_rcx_is_valid(ghcb) &&
-	      ghcb_rdx_is_valid(ghcb)))
-		return ES_VMM_ERROR;
 
-	leaf->eax = ghcb->save.rax;
-	leaf->ebx = ghcb->save.rbx;
-	leaf->ecx = ghcb->save.rcx;
-	leaf->edx = ghcb->save.rdx;
-
-	return ES_OK;
-}
-
-static int sev_cpuid_hv(struct ghcb *ghcb, struct es_em_ctxt *ctxt, struct cpuid_leaf *leaf)
-{
-	return ghcb ? __sev_cpuid_hv_ghcb(ghcb, ctxt, leaf)
-		    : __sev_cpuid_hv_msr(leaf);
-}
 
 /*
  * This may be called early while still running on the initial identity
@@ -484,21 +447,20 @@ snp_cpuid_get_validated_func(struct cpuid_leaf *leaf)
 	return false;
 }
 
-static void snp_cpuid_hv(struct ghcb *ghcb, struct es_em_ctxt *ctxt, struct cpuid_leaf *leaf)
+static void snp_cpuid_hv_msr(void *ctx, struct cpuid_leaf *leaf)
 {
-	if (sev_cpuid_hv(ghcb, ctxt, leaf))
+	if (__sev_cpuid_hv_msr(leaf))
 		sev_es_terminate(SEV_TERM_SET_LINUX, GHCB_TERM_CPUID_HV);
 }
 
-static int __head
-snp_cpuid_postprocess(struct ghcb *ghcb, struct es_em_ctxt *ctxt,
-		      struct cpuid_leaf *leaf)
+static int __head snp_cpuid_postprocess(void (*cpuid_fn)(void *ctx, struct cpuid_leaf *),
+					void *ctx, struct cpuid_leaf *leaf)
 {
 	struct cpuid_leaf leaf_hv = *leaf;
 
 	switch (leaf->fn) {
 	case 0x1:
-		snp_cpuid_hv(ghcb, ctxt, &leaf_hv);
+		cpuid_fn(ctx, &leaf_hv);
 
 		/* initial APIC ID */
 		leaf->ebx = (leaf_hv.ebx & GENMASK(31, 24)) | (leaf->ebx & GENMASK(23, 0));
@@ -517,7 +479,7 @@ snp_cpuid_postprocess(struct ghcb *ghcb, struct es_em_ctxt *ctxt,
 		break;
 	case 0xB:
 		leaf_hv.subfn = 0;
-		snp_cpuid_hv(ghcb, ctxt, &leaf_hv);
+		cpuid_fn(ctx, &leaf_hv);
 
 		/* extended APIC ID */
 		leaf->edx = leaf_hv.edx;
@@ -565,7 +527,7 @@ snp_cpuid_postprocess(struct ghcb *ghcb, struct es_em_ctxt *ctxt,
 		}
 		break;
 	case 0x8000001E:
-		snp_cpuid_hv(ghcb, ctxt, &leaf_hv);
+		cpuid_fn(ctx, &leaf_hv);
 
 		/* extended APIC ID */
 		leaf->eax = leaf_hv.eax;
@@ -586,8 +548,8 @@ snp_cpuid_postprocess(struct ghcb *ghcb, struct es_em_ctxt *ctxt,
  * Returns -EOPNOTSUPP if feature not enabled. Any other non-zero return value
  * should be treated as fatal by caller.
  */
-int __head
-snp_cpuid(struct ghcb *ghcb, struct es_em_ctxt *ctxt, struct cpuid_leaf *leaf)
+int __head snp_cpuid(void (*cpuid_fn)(void *ctx, struct cpuid_leaf *), void *ctx,
+		     struct cpuid_leaf *leaf)
 {
 	const struct snp_cpuid_table *cpuid_table = snp_cpuid_get_table();
 
@@ -621,7 +583,7 @@ snp_cpuid(struct ghcb *ghcb, struct es_em_ctxt *ctxt, struct cpuid_leaf *leaf)
 			return 0;
 	}
 
-	return snp_cpuid_postprocess(ghcb, ctxt, leaf);
+	return snp_cpuid_postprocess(cpuid_fn, ctx, leaf);
 }
 
 /*
@@ -648,7 +610,7 @@ void __head do_vc_no_ghcb(struct pt_regs *regs, unsigned long exit_code)
 	leaf.fn = fn;
 	leaf.subfn = subfn;
 
-	ret = snp_cpuid(NULL, NULL, &leaf);
+	ret = snp_cpuid(snp_cpuid_hv_msr, NULL, &leaf);
 	if (!ret)
 		goto cpuid_done;
 
arch/x86/coco/sev/vc-shared.c +48 -1

63 lines not on this page · raw message

arch/x86/include/asm/sev.h +2 -1
index 89075ff19afa..2cabf617de3c 100644
@@ -552,7 +552,8 @@ struct cpuid_leaf {
 	u32 edx;
 };
 
-int snp_cpuid(struct ghcb *ghcb, struct es_em_ctxt *ctxt, struct cpuid_leaf *leaf);
+int snp_cpuid(void (*cpuid_hv)(void *ctx, struct cpuid_leaf *),
+	      void *ctx, struct cpuid_leaf *leaf);
 
 void __noreturn sev_es_terminate(unsigned int set, unsigned int reason);
 enum es_result sev_es_ghcb_hv_call(struct ghcb *ghcb,
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
Reviewed-by on v5 1/22
#2 lore raw

Re: [PATCH v5 01/22] x86/sev: Separate MSR and GHCB based snp_cpuid() via a callback

Quoted 12 lines · Ard Biesheuvel

From: Ard Biesheuvel <[email protected]>

There are two distinct callers of snp_cpuid(): one where the MSR protocol is always used, and one where the GHCB page based interface is always used.

The snp_cpuid() logic does not care about the distinction, which only matters at a lower level. But the fact that it supports both interfaces means that the GHCB page based logic is pulled into the early startup code where PA to VA conversions are problematic, given that it runs from the 1:1 mapping of memory.

Quoted 7 lines

So keep snp_cpuid() itself in the startup code, but factor out the hypervisor calls via a callback, so that the GHCB page handling can be moved out.

Code refactoring only - no functional change intended.

Just a minor comment below...

Quoted 13 lines
---
 arch/x86/boot/startup/sev-shared.c | 60 ++++----------------
 arch/x86/coco/sev/vc-shared.c      | 49 +++++++++++++++-
 arch/x86/include/asm/sev.h         |  3 +-
 3 files changed, 61 insertions(+), 51 deletions(-)
arch/x86/include/asm/sev.h +0 -0
index 89075ff19afa..2cabf617de3c 100644

arch/x86/include/asm/sev.h struct cpuid_leaf { · line 555

Quoted 5 lines
arch/x86/include/asm/sev.h +1 -1
@@ -552,7 +552,8 @@ struct cpuid_leaf {
 	u32 edx;
 };
 
-int snp_cpuid(struct ghcb *ghcb, struct es_em_ctxt *ctxt, struct cpuid_leaf *leaf);
+int snp_cpuid(void (*cpuid_hv)(void *ctx, struct cpuid_leaf *),

You use cpuid_fn elsewhere in the patch. I'm ok with either one, but they should match name-wise.

Also, you should provide an argument name for the leaf pointer or no argument name for the void pointer (throughout).

Thanks,
Tom

Quoted 4 lines
arch/x86/include/asm/sev.h +1 -0
+	      void *ctx, struct cpuid_leaf *leaf);
 
 void __noreturn sev_es_terminate(unsigned int set, unsigned int reason);
 enum es_result sev_es_ghcb_hv_call(struct ghcb *ghcb,
#3 lore raw

PATCH v5 2/22 x86/sev: Use MSR protocol for remapping SVSM calling area

In reply to Ard Biesheuvel

From: Ard Biesheuvel <[email protected]>

As the preceding code comment already indicates, remapping the SVSM calling area occurs long before the GHCB page is configured, and so calling svsm_perform_call_protocol() is guaranteed to result in a call to svsm_perform_msr_protocol().

So just call the latter directly. This allows most of the GHCB based API infrastructure to be moved out of the startup code in a subsequent patch.

---
 arch/x86/boot/startup/sev-shared.c  | 11 +++++++++++
 arch/x86/boot/startup/sev-startup.c |  5 ++---
 2 files changed, 13 insertions(+), 3 deletions(-)
arch/x86/boot/startup/sev-shared.c +11 -0
index c401d0391537..60ab09b3149d 100644
@@ -723,6 +723,17 @@ static void __head setup_cpuid_table(const struct cc_blob_sev_info *cc_info)
 	}
 }
 
+static int __head svsm_call_msr_protocol(struct svsm_call *call)
+{
+	int ret;
+
+	do {
+		ret = svsm_perform_msr_protocol(call);
+	} while (ret == -EAGAIN);
+
+	return ret;
+}
+
 static void __head svsm_pval_4k_page(unsigned long paddr, bool validate)
 {
 	struct svsm_pvalidate_call *pc;
arch/x86/boot/startup/sev-startup.c +2 -3
index 0b7e3b950183..c30e0eed0131 100644
@@ -295,7 +295,6 @@ static __head struct cc_blob_sev_info *find_cc_blob(struct boot_params *bp)
 static __head void svsm_setup(struct cc_blob_sev_info *cc_info)
 {
 	struct svsm_call call = {};
-	int ret;
 	u64 pa;
 
 	/*
@@ -325,8 +324,8 @@ static __head void svsm_setup(struct cc_blob_sev_info *cc_info)
 	call.caa = svsm_get_caa();
 	call.rax = SVSM_CORE_CALL(SVSM_CORE_REMAP_CA);
 	call.rcx = pa;
-	ret = svsm_perform_call_protocol(&call);
-	if (ret)
+
+	if (svsm_perform_msr_protocol(&call))
 		sev_es_terminate(SEV_TERM_SET_LINUX, GHCB_TERM_SVSM_CA_REMAP_FAIL);
 
 	boot_svsm_caa = (struct svsm_ca *)pa;
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
#4 lore raw

Re: [PATCH v5 02/22] x86/sev: Use MSR protocol for remapping SVSM calling area

Quoted 54 lines · Ard Biesheuvel

From: Ard Biesheuvel <[email protected]>

As the preceding code comment already indicates, remapping the SVSM calling area occurs long before the GHCB page is configured, and so calling svsm_perform_call_protocol() is guaranteed to result in a call to svsm_perform_msr_protocol().

So just call the latter directly. This allows most of the GHCB based API infrastructure to be moved out of the startup code in a subsequent patch.

---
 arch/x86/boot/startup/sev-shared.c  | 11 +++++++++++
 arch/x86/boot/startup/sev-startup.c |  5 ++---
 2 files changed, 13 insertions(+), 3 deletions(-)
arch/x86/boot/startup/sev-shared.c +11 -0
index c401d0391537..60ab09b3149d 100644
@@ -723,6 +723,17 @@ static void __head setup_cpuid_table(const struct cc_blob_sev_info *cc_info)
 	}
 }
 
+static int __head svsm_call_msr_protocol(struct svsm_call *call)
+{
+	int ret;
+
+	do {
+		ret = svsm_perform_msr_protocol(call);
+	} while (ret == -EAGAIN);
+
+	return ret;
+}
+
 static void __head svsm_pval_4k_page(unsigned long paddr, bool validate)
 {
 	struct svsm_pvalidate_call *pc;
arch/x86/boot/startup/sev-startup.c +0 -1
index 0b7e3b950183..c30e0eed0131 100644
@@ -295,7 +295,6 @@ static __head struct cc_blob_sev_info *find_cc_blob(struct boot_params *bp)
 static __head void svsm_setup(struct cc_blob_sev_info *cc_info)
 {
 	struct svsm_call call = {};
-	int ret;
 	u64 pa;
 
 	/*

arch/x86/boot/startup/sev-startup.c static __head void svsm_setup(struct cc_blob_sev_info *cc_info) · line 328

Quoted 7 lines
arch/x86/boot/startup/sev-startup.c +2 -2
@@ -325,8 +324,8 @@ static __head void svsm_setup(struct cc_blob_sev_info *cc_info)
 	call.caa = svsm_get_caa();
 	call.rax = SVSM_CORE_CALL(SVSM_CORE_REMAP_CA);
 	call.rcx = pa;
-	ret = svsm_perform_call_protocol(&call);
-	if (ret)
+
+	if (svsm_perform_msr_protocol(&call))

This should be svsm_call_msr_protocol().

Thanks,
Tom

Quoted 3 lines
arch/x86/boot/startup/sev-startup.c +0 -0
 		sev_es_terminate(SEV_TERM_SET_LINUX, GHCB_TERM_SVSM_CA_REMAP_FAIL);
 
 	boot_svsm_caa = (struct svsm_ca *)pa;
maintainer
#5 lore raw

Re: [PATCH v5 02/22] x86/sev: Use MSR protocol for remapping SVSM calling area

Quoted 54 lines · Tom Lendacky
Quoted 52 lines · Ard Biesheuvel

From: Ard Biesheuvel <[email protected]>

As the preceding code comment already indicates, remapping the SVSM calling area occurs long before the GHCB page is configured, and so calling svsm_perform_call_protocol() is guaranteed to result in a call to svsm_perform_msr_protocol().

So just call the latter directly. This allows most of the GHCB based API infrastructure to be moved out of the startup code in a subsequent patch.

---
 arch/x86/boot/startup/sev-shared.c  | 11 +++++++++++
 arch/x86/boot/startup/sev-startup.c |  5 ++---
 2 files changed, 13 insertions(+), 3 deletions(-)
arch/x86/boot/startup/sev-shared.c +11 -0
index c401d0391537..60ab09b3149d 100644
@@ -723,6 +723,17 @@ static void __head setup_cpuid_table(const struct cc_blob_sev_info *cc_info)
      }
 }
 
+static int __head svsm_call_msr_protocol(struct svsm_call *call)
+{
+     int ret;
+
+     do {
+             ret = svsm_perform_msr_protocol(call);
+     } while (ret == -EAGAIN);
+
+     return ret;
+}
+
 static void __head svsm_pval_4k_page(unsigned long paddr, bool validate)
 {
      struct svsm_pvalidate_call *pc;
arch/x86/boot/startup/sev-startup.c +0 -1
index 0b7e3b950183..c30e0eed0131 100644
@@ -295,7 +295,6 @@ static __head struct cc_blob_sev_info *find_cc_blob(struct boot_params *bp)
 static __head void svsm_setup(struct cc_blob_sev_info *cc_info)
 {
      struct svsm_call call = {};
-     int ret;
      u64 pa;
 
Quoted 12 lines

arch/x86/boot/startup/sev-startup.c static __head void svsm_setup(struct cc_blob_sev_info *cc_info) · line 328

Quoted 9 lines
arch/x86/boot/startup/sev-startup.c +2 -2
      /*
@@ -325,8 +324,8 @@ static __head void svsm_setup(struct cc_blob_sev_info *cc_info)
      call.caa = svsm_get_caa();
      call.rax = SVSM_CORE_CALL(SVSM_CORE_REMAP_CA);
      call.rcx = pa;
-     ret = svsm_perform_call_protocol(&call);
-     if (ret)
+
+     if (svsm_perform_msr_protocol(&call))

This should be svsm_call_msr_protocol().

OK will fix

#6 lore raw

PATCH v5 3/22 x86/sev: Use MSR protocol only for early SVSM PVALIDATE call

In reply to Ard Biesheuvel

From: Ard Biesheuvel <[email protected]>

The early page state change API performs an SVSM call to PVALIDATE each page when running under a SVSM, and this involves either a GHCB page based call or a call based on the MSR protocol.

The GHCB page based variant involves VA to PA translation of the GHCB address, and this is best avoided in the startup code, where virtual addresses are ambiguous (1:1 or kernel virtual).

As this is the last remaining occurrence of svsm_perform_call_protocol() in the startup code, switch to the MSR protocol exclusively in this particular case, so that the GHCB based plumbing can be moved out of the startup code entirely in a subsequent patch.

---
 arch/x86/boot/compressed/sev.c     | 20 --------------------
 arch/x86/boot/startup/sev-shared.c |  9 ++++++---
 2 files changed, 6 insertions(+), 23 deletions(-)
arch/x86/boot/compressed/sev.c +0 -20
index fd1b67dfea22..b71c1ab6a282 100644
@@ -50,31 +50,11 @@ u64 svsm_get_caa_pa(void)
 	return boot_svsm_caa_pa;
 }
 
-int svsm_perform_call_protocol(struct svsm_call *call);
-
 u8 snp_vmpl;
 
 /* Include code for early handlers */
 #include "../../boot/startup/sev-shared.c"
 
-int svsm_perform_call_protocol(struct svsm_call *call)
-{
-	struct ghcb *ghcb;
-	int ret;
-
-	if (boot_ghcb)
-		ghcb = boot_ghcb;
-	else
-		ghcb = NULL;
-
-	do {
-		ret = ghcb ? svsm_perform_ghcb_protocol(ghcb, call)
-			   : svsm_perform_msr_protocol(call);
-	} while (ret == -EAGAIN);
-
-	return ret;
-}
-
 static bool sev_snp_enabled(void)
 {
 	return sev_status & MSR_AMD64_SEV_SNP_ENABLED;
arch/x86/boot/startup/sev-shared.c +6 -3
index 60ab09b3149d..d9c0c64d80fe 100644
@@ -740,7 +740,6 @@ static void __head svsm_pval_4k_page(unsigned long paddr, bool validate)
 	struct svsm_call call = {};
 	unsigned long flags;
 	u64 pc_pa;
-	int ret;
 
 	/*
 	 * This can be called very early in the boot, use native functions in
@@ -764,8 +763,12 @@ static void __head svsm_pval_4k_page(unsigned long paddr, bool validate)
 	call.rax = SVSM_CORE_CALL(SVSM_CORE_PVALIDATE);
 	call.rcx = pc_pa;
 
-	ret = svsm_perform_call_protocol(&call);
-	if (ret)
+	/*
+	 * Use the MSR protocol exclusively, so that this code is usable in
+	 * startup code where VA/PA translations of the GHCB page's address may
+	 * be problematic.
+	 */
+	if (svsm_call_msr_protocol(&call))
 		sev_es_terminate(SEV_TERM_SET_LINUX, GHCB_TERM_PVALIDATE);
 
 	native_local_irq_restore(flags);
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
#7 lore raw

PATCH v5 4/22 x86/sev: Run RMPADJUST on SVSM calling area page to test VMPL

In reply to Ard Biesheuvel

From: Ard Biesheuvel <[email protected]>

Determining the VMPL at which the kernel runs involves performing a RMPADJUST operation on an arbitrary page of memory, and observing whether it succeeds.

The use of boot_ghcb_page in the core kernel in this case is completely arbitrary, but results in the need to provide a PIC alias for it. So use boot_svsm_ca_page instead, which already needs this alias for other reasons.

---
 arch/x86/boot/compressed/sev.c      | 2 +-
 arch/x86/boot/startup/sev-shared.c  | 5 +++--
 arch/x86/boot/startup/sev-startup.c | 2 +-
 3 files changed, 5 insertions(+), 4 deletions(-)
arch/x86/boot/compressed/sev.c +1 -1
index b71c1ab6a282..3628e9bddc6a 100644
@@ -327,7 +327,7 @@ static bool early_snp_init(struct boot_params *bp)
 	 * running at VMPL0. The CA will be used to communicate with the
 	 * SVSM and request its services.
 	 */
-	svsm_setup_ca(cc_info);
+	svsm_setup_ca(cc_info, rip_rel_ptr(&boot_ghcb_page));
 
 	/*
 	 * Pass run-time kernel a pointer to CC info via boot_params so EFI
arch/x86/boot/startup/sev-shared.c +3 -2
index d9c0c64d80fe..cbf26466e0da 100644
@@ -792,7 +792,8 @@ static void __head pvalidate_4k_page(unsigned long vaddr, unsigned long paddr,
  * Maintain the GPA of the SVSM Calling Area (CA) in order to utilize the SVSM
  * services needed when not running in VMPL0.
  */
-static bool __head svsm_setup_ca(const struct cc_blob_sev_info *cc_info)
+static bool __head svsm_setup_ca(const struct cc_blob_sev_info *cc_info,
+				 void *page)
 {
 	struct snp_secrets_page *secrets_page;
 	struct snp_cpuid_table *cpuid_table;
@@ -815,7 +816,7 @@ static bool __head svsm_setup_ca(const struct cc_blob_sev_info *cc_info)
 	 * routine is running identity mapped when called, both by the decompressor
 	 * code and the early kernel code.
 	 */
-	if (!rmpadjust((unsigned long)rip_rel_ptr(&boot_ghcb_page), RMP_PG_SIZE_4K, 1))
+	if (!rmpadjust((unsigned long)page, RMP_PG_SIZE_4K, 1))
 		return false;
 
 	/*
arch/x86/boot/startup/sev-startup.c +1 -1
index c30e0eed0131..4b9e8ccc0e91 100644
@@ -302,7 +302,7 @@ static __head void svsm_setup(struct cc_blob_sev_info *cc_info)
 	 * running at VMPL0. The CA will be used to communicate with the
 	 * SVSM to perform the SVSM services.
 	 */
-	if (!svsm_setup_ca(cc_info))
+	if (!svsm_setup_ca(cc_info, rip_rel_ptr(&boot_svsm_ca_page)))
 		return;
 
 	/*
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
#8 lore raw

PATCH v5 5/22 x86/sev: Move GHCB page based HV communication out of startup code

In reply to Ard Biesheuvel

From: Ard Biesheuvel <[email protected]>

Both the decompressor and the core kernel implement an early #VC handler, which only deals with CPUID instructions, and full featured one, which can handle any #VC exception.

The former communicates with the hypervisor using the MSR based protocol, whereas the latter uses a shared GHCB page, which is configured a bit later during the boot, when the kernel runs from its ordinary virtual mapping, rather than the 1:1 mapping that the startup code uses.

Accessing this shared GHCB page from the core kernel's startup code is problematic, because it involves converting the GHCB address provided by the caller to a physical address. In the startup code, virtual to physical address translations are problematic, given that the virtual address might be a 1:1 mapped address, and such translations should therefore be avoided.

This means that exposing startup code dealing with the GHCB to callers that execute from the ordinary kernel virtual mapping should be avoided too. So move all GHCB page based communication out of the startup code, now that all communication occurring before the kernel virtual mapping is up relies on the MSR protocol only.

---
 arch/x86/boot/compressed/sev-handle-vc.c |   3 +
 arch/x86/boot/startup/sev-shared.c       | 143 +-------------------
 arch/x86/boot/startup/sev-startup.c      |  42 ------
 arch/x86/coco/sev/core.c                 |  76 +++++++++++
 arch/x86/coco/sev/vc-handle.c            |   2 +
 arch/x86/coco/sev/vc-shared.c            |  94 +++++++++++++
 arch/x86/include/asm/sev-internal.h      |   7 +-
 arch/x86/include/asm/sev.h               |  11 +-
 8 files changed, 190 insertions(+), 188 deletions(-)
arch/x86/boot/compressed/sev-handle-vc.c +3 -0
index 89dd02de2a0f..7530ad8b768b 100644
@@ -1,6 +1,7 @@
 // SPDX-License-Identifier: GPL-2.0
 
 #include "misc.h"
+#include "error.h"
 #include "sev.h"
 
 #include <linux/kernel.h>
@@ -14,6 +15,8 @@
 #include <asm/fpu/xcr.h>
 
 #define __BOOT_COMPRESSED
+#undef __init
+#define __init
 
 /* Basic instruction decoding support needed */
 #include "../../lib/inat.c"
arch/x86/boot/startup/sev-shared.c +4 -139

188 lines not on this page · raw message

arch/x86/boot/startup/sev-startup.c +0 -42

54 lines not on this page · raw message

arch/x86/coco/sev/core.c +76 -0

88 lines not on this page · raw message

arch/x86/coco/sev/vc-handle.c +2 -0
index faf1fce89ed4..9a5e16f70e83 100644
@@ -351,6 +351,8 @@ static enum es_result vc_read_mem(struct es_em_ctxt *ctxt,
 }
 
 #define sev_printk(fmt, ...)		printk(fmt, ##__VA_ARGS__)
+#define error(v)
+#define has_cpuflag(f)			boot_cpu_has(f)
 
 #include "vc-shared.c"
 
arch/x86/coco/sev/vc-shared.c +94 -0

103 lines not on this page · raw message

arch/x86/include/asm/sev-internal.h +2 -5
index 3dfd306d1c9e..6199b35a82e4 100644
@@ -2,7 +2,6 @@
 
 #define DR7_RESET_VALUE        0x400
 
-extern struct ghcb boot_ghcb_page;
 extern u64 sev_hv_features;
 extern u64 sev_secrets_pa;
 
@@ -80,7 +79,8 @@ static __always_inline u64 svsm_get_caa_pa(void)
 		return boot_svsm_caa_pa;
 }
 
-int svsm_perform_call_protocol(struct svsm_call *call);
+enum es_result verify_exception_info(struct ghcb *ghcb, struct es_em_ctxt *ctxt);
+void vc_forward_exception(struct es_em_ctxt *ctxt);
 
 static inline u64 sev_es_rd_ghcb_msr(void)
 {
@@ -97,9 +97,6 @@ static __always_inline void sev_es_wr_ghcb_msr(u64 val)
 	native_wrmsr(MSR_AMD64_SEV_ES_GHCB, low, high);
 }
 
-void snp_register_ghcb_early(unsigned long paddr);
-bool sev_es_negotiate_protocol(void);
-bool sev_es_check_cpu_features(void);
 u64 get_hv_features(void);
 
 const struct snp_cpuid_table *snp_cpuid_get_table(void);
arch/x86/include/asm/sev.h +9 -2
index 2cabf617de3c..135e91a17d04 100644
@@ -503,6 +503,7 @@ static inline int pvalidate(unsigned long vaddr, bool rmp_psize, bool validate)
 }
 
 void setup_ghcb(void);
+void snp_register_ghcb_early(unsigned long paddr);
 void early_snp_set_memory_private(unsigned long vaddr, unsigned long paddr,
 				  unsigned long npages);
 void early_snp_set_memory_shared(unsigned long vaddr, unsigned long paddr,
@@ -540,8 +541,6 @@ static __always_inline void vc_ghcb_invalidate(struct ghcb *ghcb)
 	__builtin_memset(ghcb->save.valid_bitmap, 0, sizeof(ghcb->save.valid_bitmap));
 }
 
-void vc_forward_exception(struct es_em_ctxt *ctxt);
-
 /* I/O parameters for CPUID-related helpers */
 struct cpuid_leaf {
 	u32 fn;
@@ -552,15 +551,23 @@ struct cpuid_leaf {
 	u32 edx;
 };
 
+int svsm_perform_msr_protocol(struct svsm_call *call);
 int snp_cpuid(void (*cpuid_hv)(void *ctx, struct cpuid_leaf *),
 	      void *ctx, struct cpuid_leaf *leaf);
 
+void svsm_issue_call(struct svsm_call *call, u8 *pending);
+int svsm_process_result_codes(struct svsm_call *call);
+
 void __noreturn sev_es_terminate(unsigned int set, unsigned int reason);
 enum es_result sev_es_ghcb_hv_call(struct ghcb *ghcb,
 				   struct es_em_ctxt *ctxt,
 				   u64 exit_code, u64 exit_info_1,
 				   u64 exit_info_2);
 
+bool sev_es_negotiate_protocol(void);
+bool sev_es_check_cpu_features(void);
+
+extern u16 ghcb_version;
 extern struct ghcb *boot_ghcb;
 
 #else	/* !CONFIG_AMD_MEM_ENCRYPT */
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
#9 lore raw

PATCH v5 6/22 x86/sev: Avoid global variable to store virtual address of SVSM area

In reply to Ard Biesheuvel

From: Ard Biesheuvel <[email protected]>

The boottime SVSM calling area is used both by the startup code running from a 1:1 mapping, and potentially later on running from the ordinary kernel mapping.

This SVSM calling area is statically allocated, and so its physical address doesn't change. However, its virtual address depends on the calling context (1:1 mapping or kernel virtual mapping), and even though the variable that holds the virtual address of this calling area gets updated from 1:1 address to kernel address during the boot, it is hard to reason about why this is guaranteed to be safe.

So instead, take the RIP-relative address of the boottime SVSM calling area whenever its virtual address is required, and only use a global variable for the physical address.

---
 arch/x86/boot/compressed/sev.c      | 5 ++---
 arch/x86/boot/startup/sev-shared.c  | 6 ------
 arch/x86/boot/startup/sev-startup.c | 9 +++++----
 arch/x86/coco/sev/core.c            | 9 ---------
 arch/x86/include/asm/sev-internal.h | 3 +--
 arch/x86/include/asm/sev.h          | 2 --
 arch/x86/mm/mem_encrypt_amd.c       | 6 ------
 7 files changed, 8 insertions(+), 32 deletions(-)
arch/x86/boot/compressed/sev.c +2 -3
index 3628e9bddc6a..6c0f91d38595 100644
@@ -37,12 +37,12 @@ struct ghcb *boot_ghcb;
 
 #define __BOOT_COMPRESSED
 
-extern struct svsm_ca *boot_svsm_caa;
 extern u64 boot_svsm_caa_pa;
 
 struct svsm_ca *svsm_get_caa(void)
 {
-	return boot_svsm_caa;
+	/* The decompressor is mapped 1:1 so VA == PA */
+	return (struct svsm_ca *)boot_svsm_caa_pa;
 }
 
 u64 svsm_get_caa_pa(void)
@@ -530,7 +530,6 @@ bool early_is_sevsnp_guest(void)
 
 			/* Obtain the address of the calling area to use */
 			boot_rdmsr(MSR_SVSM_CAA, &m);
-			boot_svsm_caa = (void *)m.q;
 			boot_svsm_caa_pa = m.q;
 
 			/*
arch/x86/boot/startup/sev-shared.c +0 -6
index f9de8b33de6c..51f2110bc509 100644
@@ -26,7 +26,6 @@
  *   early boot, both with identity mapped virtual addresses and proper kernel
  *   virtual addresses.
  */
-struct svsm_ca *boot_svsm_caa __ro_after_init;
 u64 boot_svsm_caa_pa __ro_after_init;
 
 /*
@@ -709,11 +708,6 @@ static bool __head svsm_setup_ca(const struct cc_blob_sev_info *cc_info,
 	if (caa & (PAGE_SIZE - 1))
 		sev_es_terminate(SEV_TERM_SET_LINUX, GHCB_TERM_SVSM_CAA);
 
-	/*
-	 * The CA is identity mapped when this routine is called, both by the
-	 * decompressor code and the early kernel code.
-	 */
-	boot_svsm_caa = (struct svsm_ca *)caa;
 	boot_svsm_caa_pa = caa;
 
 	/* Advertise the SVSM presence via CPUID. */
arch/x86/boot/startup/sev-startup.c +5 -4
index 7b38085c7218..f3e247d205b7 100644
@@ -252,6 +252,7 @@ static __head struct cc_blob_sev_info *find_cc_blob(struct boot_params *bp)
 
 static __head void svsm_setup(struct cc_blob_sev_info *cc_info)
 {
+	struct snp_secrets_page *secrets = (void *)cc_info->secrets_phys;
 	struct svsm_call call = {};
 	u64 pa;
 
@@ -272,21 +273,21 @@ static __head void svsm_setup(struct cc_blob_sev_info *cc_info)
 	pa = (u64)rip_rel_ptr(&boot_svsm_ca_page);
 
 	/*
-	 * Switch over to the boot SVSM CA while the current CA is still
-	 * addressable. There is no GHCB at this point so use the MSR protocol.
+	 * Switch over to the boot SVSM CA while the current CA is still 1:1
+	 * mapped and thus addressable with VA == PA. There is no GHCB at this
+	 * point so use the MSR protocol.
 	 *
 	 * SVSM_CORE_REMAP_CA call:
 	 *   RAX = 0 (Protocol=0, CallID=0)
 	 *   RCX = New CA GPA
 	 */
-	call.caa = svsm_get_caa();
+	call.caa = (struct svsm_ca *)secrets->svsm_caa;
 	call.rax = SVSM_CORE_CALL(SVSM_CORE_REMAP_CA);
 	call.rcx = pa;
 
 	if (svsm_perform_msr_protocol(&call))
 		sev_es_terminate(SEV_TERM_SET_LINUX, GHCB_TERM_SVSM_CA_REMAP_FAIL);
 
-	boot_svsm_caa = (struct svsm_ca *)pa;
 	boot_svsm_caa_pa = pa;
 }
 
arch/x86/coco/sev/core.c +0 -9
index 15be9e52848d..bea67d017bf0 100644
@@ -1643,15 +1643,6 @@ void sev_show_status(void)
 	pr_cont("\n");
 }
 
-void __init snp_update_svsm_ca(void)
-{
-	if (!snp_vmpl)
-		return;
-
-	/* Update the CAA to a proper kernel address */
-	boot_svsm_caa = &boot_svsm_ca_page;
-}
-
 #ifdef CONFIG_SYSFS
 static ssize_t vmpl_show(struct kobject *kobj,
 			 struct kobj_attribute *attr, char *buf)
arch/x86/include/asm/sev-internal.h +1 -2
index 6199b35a82e4..ffe4755962fe 100644
@@ -60,7 +60,6 @@ void early_set_pages_state(unsigned long vaddr, unsigned long paddr,
 DECLARE_PER_CPU(struct svsm_ca *, svsm_caa);
 DECLARE_PER_CPU(u64, svsm_caa_pa);
 
-extern struct svsm_ca *boot_svsm_caa;
 extern u64 boot_svsm_caa_pa;
 
 static __always_inline struct svsm_ca *svsm_get_caa(void)
@@ -68,7 +67,7 @@ static __always_inline struct svsm_ca *svsm_get_caa(void)
 	if (sev_cfg.use_cas)
 		return this_cpu_read(svsm_caa);
 	else
-		return boot_svsm_caa;
+		return rip_rel_ptr(&boot_svsm_ca_page);
 }
 
 static __always_inline u64 svsm_get_caa_pa(void)
arch/x86/include/asm/sev.h +0 -2
index 135e91a17d04..f3acbfcdca9a 100644
@@ -519,7 +519,6 @@ void snp_accept_memory(phys_addr_t start, phys_addr_t end);
 u64 snp_get_unsupported_features(u64 status);
 u64 sev_get_status(void);
 void sev_show_status(void);
-void snp_update_svsm_ca(void);
 int prepare_pte_enc(struct pte_enc_desc *d);
 void set_pte_enc_mask(pte_t *kpte, unsigned long pfn, pgprot_t new_prot);
 void snp_kexec_finish(void);
@@ -600,7 +599,6 @@ static inline void snp_accept_memory(phys_addr_t start, phys_addr_t end) { }
 static inline u64 snp_get_unsupported_features(u64 status) { return 0; }
 static inline u64 sev_get_status(void) { return 0; }
 static inline void sev_show_status(void) { }
-static inline void snp_update_svsm_ca(void) { }
 static inline int prepare_pte_enc(struct pte_enc_desc *d) { return 0; }
 static inline void set_pte_enc_mask(pte_t *kpte, unsigned long pfn, pgprot_t new_prot) { }
 static inline void snp_kexec_finish(void) { }
arch/x86/mm/mem_encrypt_amd.c +0 -6
index faf3a13fb6ba..2f8c32173972 100644
@@ -536,12 +536,6 @@ void __init sme_early_init(void)
 		x86_init.resources.dmi_setup = snp_dmi_setup;
 	}
 
-	/*
-	 * Switch the SVSM CA mapping (if active) from identity mapped to
-	 * kernel mapped.
-	 */
-	snp_update_svsm_ca();
-
 	if (sev_status & MSR_AMD64_SNP_SECURE_TSC)
 		setup_force_cpu_cap(X86_FEATURE_TSC_RELIABLE);
 }
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
#10 lore raw

PATCH v5 7/22 x86/sev: Share implementation of MSR-based page state change

In reply to Ard Biesheuvel

From: Ard Biesheuvel <[email protected]>

Both the decompressor and the SEV startup code implement the exact same sequence for invoking the MSR based communication protocol to effectuate a page state change.

Before tweaking the internal APIs used in both versions, merge them and share them so those tweaks are only needed in a single place.

---
 arch/x86/boot/compressed/sev.c      | 40 ++------------------
 arch/x86/boot/startup/sev-shared.c  | 35 +++++++++++++++++
 arch/x86/boot/startup/sev-startup.c | 29 +-------------
 3 files changed, 39 insertions(+), 65 deletions(-)
arch/x86/boot/compressed/sev.c +3 -37
index 6c0f91d38595..f714235d3222 100644
@@ -60,46 +60,12 @@ static bool sev_snp_enabled(void)
 	return sev_status & MSR_AMD64_SEV_SNP_ENABLED;
 }
 
-static void __page_state_change(unsigned long paddr, enum psc_op op)
-{
-	u64 val, msr;
-
-	/*
-	 * If private -> shared then invalidate the page before requesting the
-	 * state change in the RMP table.
-	 */
-	if (op == SNP_PAGE_STATE_SHARED)
-		pvalidate_4k_page(paddr, paddr, false);
-
-	/* Save the current GHCB MSR value */
-	msr = sev_es_rd_ghcb_msr();
-
-	/* Issue VMGEXIT to change the page state in RMP table. */
-	sev_es_wr_ghcb_msr(GHCB_MSR_PSC_REQ_GFN(paddr >> PAGE_SHIFT, op));
-	VMGEXIT();
-
-	/* Read the response of the VMGEXIT. */
-	val = sev_es_rd_ghcb_msr();
-	if ((GHCB_RESP_CODE(val) != GHCB_MSR_PSC_RESP) || GHCB_MSR_PSC_RESP_VAL(val))
-		sev_es_terminate(SEV_TERM_SET_LINUX, GHCB_TERM_PSC);
-
-	/* Restore the GHCB MSR value */
-	sev_es_wr_ghcb_msr(msr);
-
-	/*
-	 * Now that page state is changed in the RMP table, validate it so that it is
-	 * consistent with the RMP entry.
-	 */
-	if (op == SNP_PAGE_STATE_PRIVATE)
-		pvalidate_4k_page(paddr, paddr, true);
-}
-
 void snp_set_page_private(unsigned long paddr)
 {
 	if (!sev_snp_enabled())
 		return;
 
-	__page_state_change(paddr, SNP_PAGE_STATE_PRIVATE);
+	__page_state_change(paddr, paddr, SNP_PAGE_STATE_PRIVATE);
 }
 
 void snp_set_page_shared(unsigned long paddr)
@@ -107,7 +73,7 @@ void snp_set_page_shared(unsigned long paddr)
 	if (!sev_snp_enabled())
 		return;
 
-	__page_state_change(paddr, SNP_PAGE_STATE_SHARED);
+	__page_state_change(paddr, paddr, SNP_PAGE_STATE_SHARED);
 }
 
 bool early_setup_ghcb(void)
@@ -133,7 +99,7 @@ bool early_setup_ghcb(void)
 void snp_accept_memory(phys_addr_t start, phys_addr_t end)
 {
 	for (phys_addr_t pa = start; pa < end; pa += PAGE_SIZE)
-		__page_state_change(pa, SNP_PAGE_STATE_PRIVATE);
+		__page_state_change(pa, pa, SNP_PAGE_STATE_PRIVATE);
 }
 
 void sev_es_shutdown_ghcb(void)
arch/x86/boot/startup/sev-shared.c +35 -0
index 51f2110bc509..eb241ff1156d 100644
@@ -652,6 +652,41 @@ static void __head pvalidate_4k_page(unsigned long vaddr, unsigned long paddr,
 	}
 }
 
+static void __head __page_state_change(unsigned long vaddr, unsigned long paddr,
+				       enum psc_op op)
+{
+	u64 val, msr;
+
+	/*
+	 * If private -> shared then invalidate the page before requesting the
+	 * state change in the RMP table.
+	 */
+	if (op == SNP_PAGE_STATE_SHARED)
+		pvalidate_4k_page(vaddr, paddr, false);
+
+	/* Save the current GHCB MSR value */
+	msr = sev_es_rd_ghcb_msr();
+
+	/* Issue VMGEXIT to change the page state in RMP table. */
+	sev_es_wr_ghcb_msr(GHCB_MSR_PSC_REQ_GFN(paddr >> PAGE_SHIFT, op));
+	VMGEXIT();
+
+	/* Read the response of the VMGEXIT. */
+	val = sev_es_rd_ghcb_msr();
+	if ((GHCB_RESP_CODE(val) != GHCB_MSR_PSC_RESP) || GHCB_MSR_PSC_RESP_VAL(val))
+		sev_es_terminate(SEV_TERM_SET_LINUX, GHCB_TERM_PSC);
+
+	/* Restore the GHCB MSR value */
+	sev_es_wr_ghcb_msr(msr);
+
+	/*
+	 * Now that page state is changed in the RMP table, validate it so that it is
+	 * consistent with the RMP entry.
+	 */
+	if (op == SNP_PAGE_STATE_PRIVATE)
+		pvalidate_4k_page(vaddr, paddr, true);
+}
+
 /*
  * Maintain the GPA of the SVSM Calling Area (CA) in order to utilize the SVSM
  * services needed when not running in VMPL0.
arch/x86/boot/startup/sev-startup.c +1 -28
index f3e247d205b7..b4e2cb7bc44a 100644
@@ -135,7 +135,6 @@ early_set_pages_state(unsigned long vaddr, unsigned long paddr,
 		      unsigned long npages, enum psc_op op)
 {
 	unsigned long paddr_end;
-	u64 val;
 
 	vaddr = vaddr & PAGE_MASK;
 
@@ -143,37 +142,11 @@ early_set_pages_state(unsigned long vaddr, unsigned long paddr,
 	paddr_end = paddr + (npages << PAGE_SHIFT);
 
 	while (paddr < paddr_end) {
-		/* Page validation must be rescinded before changing to shared */
-		if (op == SNP_PAGE_STATE_SHARED)
-			pvalidate_4k_page(vaddr, paddr, false);
-
-		/*
-		 * Use the MSR protocol because this function can be called before
-		 * the GHCB is established.
-		 */
-		sev_es_wr_ghcb_msr(GHCB_MSR_PSC_REQ_GFN(paddr >> PAGE_SHIFT, op));
-		VMGEXIT();
-
-		val = sev_es_rd_ghcb_msr();
-
-		if (GHCB_RESP_CODE(val) != GHCB_MSR_PSC_RESP)
-			goto e_term;
-
-		if (GHCB_MSR_PSC_RESP_VAL(val))
-			goto e_term;
-
-		/* Page validation must be performed after changing to private */
-		if (op == SNP_PAGE_STATE_PRIVATE)
-			pvalidate_4k_page(vaddr, paddr, true);
+		__page_state_change(vaddr, paddr, op);
 
 		vaddr += PAGE_SIZE;
 		paddr += PAGE_SIZE;
 	}
-
-	return;
-
-e_term:
-	sev_es_terminate(SEV_TERM_SET_LINUX, GHCB_TERM_PSC);
 }
 
 void __head early_snp_set_memory_private(unsigned long vaddr, unsigned long paddr,
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
#11 lore raw

PATCH v5 8/22 x86/sev: Pass SVSM calling area down to early page state change API

In reply to Ard Biesheuvel

From: Ard Biesheuvel <[email protected]>

The early page state change API is mostly only used very early, when only the boot time SVSM calling area is in use. However, this API is also called by the kexec finishing code, which runs very late, and potentially from a different CPU (which uses a different calling area).

To avoid pulling the per-CPU SVSM calling area pointers and related SEV state into the startup code, refactor the page state change API so the SVSM calling area virtual and physical addresses can be provided by the caller.

No functional change intended.

---
 arch/x86/boot/compressed/sev.c      | 12 +++++++++---
 arch/x86/boot/startup/sev-shared.c  | 17 +++++++++--------
 arch/x86/boot/startup/sev-startup.c | 11 +++++++----
 arch/x86/coco/sev/core.c            |  3 ++-
 arch/x86/include/asm/sev-internal.h |  3 ++-
 5 files changed, 29 insertions(+), 17 deletions(-)
arch/x86/boot/compressed/sev.c +9 -3
index f714235d3222..18b0ccf517eb 100644
@@ -65,7 +65,9 @@ void snp_set_page_private(unsigned long paddr)
 	if (!sev_snp_enabled())
 		return;
 
-	__page_state_change(paddr, paddr, SNP_PAGE_STATE_PRIVATE);
+	__page_state_change(paddr, paddr, SNP_PAGE_STATE_PRIVATE,
+			    (struct svsm_ca *)boot_svsm_caa_pa,
+			    boot_svsm_caa_pa);
 }
 
 void snp_set_page_shared(unsigned long paddr)
@@ -73,7 +75,9 @@ void snp_set_page_shared(unsigned long paddr)
 	if (!sev_snp_enabled())
 		return;
 
-	__page_state_change(paddr, paddr, SNP_PAGE_STATE_SHARED);
+	__page_state_change(paddr, paddr, SNP_PAGE_STATE_SHARED,
+			    (struct svsm_ca *)boot_svsm_caa_pa,
+			    boot_svsm_caa_pa);
 }
 
 bool early_setup_ghcb(void)
@@ -99,7 +103,9 @@ bool early_setup_ghcb(void)
 void snp_accept_memory(phys_addr_t start, phys_addr_t end)
 {
 	for (phys_addr_t pa = start; pa < end; pa += PAGE_SIZE)
-		__page_state_change(pa, pa, SNP_PAGE_STATE_PRIVATE);
+		__page_state_change(pa, pa, SNP_PAGE_STATE_PRIVATE,
+				    (struct svsm_ca *)boot_svsm_caa_pa,
+				    boot_svsm_caa_pa);
 }
 
 void sev_es_shutdown_ghcb(void)
arch/x86/boot/startup/sev-shared.c +9 -8
index eb241ff1156d..83ca97df0808 100644
@@ -598,7 +598,8 @@ static int __head svsm_call_msr_protocol(struct svsm_call *call)
 	return ret;
 }
 
-static void __head svsm_pval_4k_page(unsigned long paddr, bool validate)
+static void __head svsm_pval_4k_page(unsigned long paddr, bool validate,
+				     struct svsm_ca *caa, u64 caa_pa)
 {
 	struct svsm_pvalidate_call *pc;
 	struct svsm_call call = {};
@@ -611,10 +612,10 @@ static void __head svsm_pval_4k_page(unsigned long paddr, bool validate)
 	 */
 	flags = native_local_irq_save();
 
-	call.caa = svsm_get_caa();
+	call.caa = caa;
 
 	pc = (struct svsm_pvalidate_call *)call.caa->svsm_buffer;
-	pc_pa = svsm_get_caa_pa() + offsetof(struct svsm_ca, svsm_buffer);
+	pc_pa = caa_pa + offsetof(struct svsm_ca, svsm_buffer);
 
 	pc->num_entries = 1;
 	pc->cur_index   = 0;
@@ -639,12 +640,12 @@ static void __head svsm_pval_4k_page(unsigned long paddr, bool validate)
 }
 
 static void __head pvalidate_4k_page(unsigned long vaddr, unsigned long paddr,
-				     bool validate)
+				     bool validate, struct svsm_ca *caa, u64 caa_pa)
 {
 	int ret;
 
 	if (snp_vmpl) {
-		svsm_pval_4k_page(paddr, validate);
+		svsm_pval_4k_page(paddr, validate, caa, caa_pa);
 	} else {
 		ret = pvalidate(vaddr, RMP_PG_SIZE_4K, validate);
 		if (ret)
@@ -653,7 +654,7 @@ static void __head pvalidate_4k_page(unsigned long vaddr, unsigned long paddr,
 }
 
 static void __head __page_state_change(unsigned long vaddr, unsigned long paddr,
-				       enum psc_op op)
+				       enum psc_op op, struct svsm_ca *caa, u64 caa_pa)
 {
 	u64 val, msr;
 
@@ -662,7 +663,7 @@ static void __head __page_state_change(unsigned long vaddr, unsigned long paddr,
 	 * state change in the RMP table.
 	 */
 	if (op == SNP_PAGE_STATE_SHARED)
-		pvalidate_4k_page(vaddr, paddr, false);
+		pvalidate_4k_page(vaddr, paddr, false, caa, caa_pa);
 
 	/* Save the current GHCB MSR value */
 	msr = sev_es_rd_ghcb_msr();
@@ -684,7 +685,7 @@ static void __head __page_state_change(unsigned long vaddr, unsigned long paddr,
 	 * consistent with the RMP entry.
 	 */
 	if (op == SNP_PAGE_STATE_PRIVATE)
-		pvalidate_4k_page(vaddr, paddr, true);
+		pvalidate_4k_page(vaddr, paddr, true, caa, caa_pa);
 }
 
 /*
arch/x86/boot/startup/sev-startup.c +7 -4
index b4e2cb7bc44a..7aabda0b378e 100644
@@ -132,7 +132,8 @@ noinstr void __sev_put_ghcb(struct ghcb_state *state)
 
 void __head
 early_set_pages_state(unsigned long vaddr, unsigned long paddr,
-		      unsigned long npages, enum psc_op op)
+		      unsigned long npages, enum psc_op op,
+		      struct svsm_ca *caa, u64 caa_pa)
 {
 	unsigned long paddr_end;
 
@@ -142,7 +143,7 @@ early_set_pages_state(unsigned long vaddr, unsigned long paddr,
 	paddr_end = paddr + (npages << PAGE_SHIFT);
 
 	while (paddr < paddr_end) {
-		__page_state_change(vaddr, paddr, op);
+		__page_state_change(vaddr, paddr, op, caa, caa_pa);
 
 		vaddr += PAGE_SIZE;
 		paddr += PAGE_SIZE;
@@ -165,7 +166,8 @@ void __head early_snp_set_memory_private(unsigned long vaddr, unsigned long padd
 	  * Ask the hypervisor to mark the memory pages as private in the RMP
 	  * table.
 	  */
-	early_set_pages_state(vaddr, paddr, npages, SNP_PAGE_STATE_PRIVATE);
+	early_set_pages_state(vaddr, paddr, npages, SNP_PAGE_STATE_PRIVATE,
+			      svsm_get_caa(), svsm_get_caa_pa());
 }
 
 void __head early_snp_set_memory_shared(unsigned long vaddr, unsigned long paddr,
@@ -181,7 +183,8 @@ void __head early_snp_set_memory_shared(unsigned long vaddr, unsigned long paddr
 		return;
 
 	 /* Ask hypervisor to mark the memory pages shared in the RMP table. */
-	early_set_pages_state(vaddr, paddr, npages, SNP_PAGE_STATE_SHARED);
+	early_set_pages_state(vaddr, paddr, npages, SNP_PAGE_STATE_SHARED,
+			      svsm_get_caa(), svsm_get_caa_pa());
 }
 
 /*
arch/x86/coco/sev/core.c +2 -1
index bea67d017bf0..7a86a2fe494d 100644
@@ -585,7 +585,8 @@ static void set_pages_state(unsigned long vaddr, unsigned long npages, int op)
 
 	/* Use the MSR protocol when a GHCB is not available. */
 	if (!boot_ghcb)
-		return early_set_pages_state(vaddr, __pa(vaddr), npages, op);
+		return early_set_pages_state(vaddr, __pa(vaddr), npages, op,
+					     svsm_get_caa(), svsm_get_caa_pa());
 
 	vaddr = vaddr & PAGE_MASK;
 	vaddr_end = vaddr + (npages << PAGE_SHIFT);
arch/x86/include/asm/sev-internal.h +2 -1
index ffe4755962fe..3b72d8217827 100644
@@ -55,7 +55,8 @@ DECLARE_PER_CPU(struct sev_es_runtime_data*, runtime_data);
 DECLARE_PER_CPU(struct sev_es_save_area *, sev_vmsa);
 
 void early_set_pages_state(unsigned long vaddr, unsigned long paddr,
-			   unsigned long npages, enum psc_op op);
+			   unsigned long npages, enum psc_op op,
+			   struct svsm_ca *ca, u64 caa_pa);
 
 DECLARE_PER_CPU(struct svsm_ca *, svsm_caa);
 DECLARE_PER_CPU(u64, svsm_caa_pa);
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
#12 lore raw

PATCH v5 9/22 x86/sev: Use boot SVSM CA for all startup and init code

In reply to Ard Biesheuvel

From: Ard Biesheuvel <[email protected]>

To avoid having to reason about whether or not to use the per-CPU SVSM calling area when running startup and init code on the boot CPU, reuse the boot SVSM calling area as the per-CPU area for CPU #0.

This removes the need to make the per-CPU variables and associated state in sev_cfg accessible to the startup code once confined.

---
 arch/x86/boot/compressed/sev.c      | 13 ------
 arch/x86/boot/startup/sev-startup.c |  7 +--
 arch/x86/coco/sev/core.c            | 47 +++++++++-----------
 arch/x86/include/asm/sev-internal.h | 16 -------
 4 files changed, 24 insertions(+), 59 deletions(-)
arch/x86/boot/compressed/sev.c +0 -13
index 18b0ccf517eb..4bdf5595ed96 100644
@@ -37,19 +37,6 @@ struct ghcb *boot_ghcb;
 
 #define __BOOT_COMPRESSED
 
-extern u64 boot_svsm_caa_pa;
-
-struct svsm_ca *svsm_get_caa(void)
-{
-	/* The decompressor is mapped 1:1 so VA == PA */
-	return (struct svsm_ca *)boot_svsm_caa_pa;
-}
-
-u64 svsm_get_caa_pa(void)
-{
-	return boot_svsm_caa_pa;
-}
-
 u8 snp_vmpl;
 
 /* Include code for early handlers */
arch/x86/boot/startup/sev-startup.c +2 -5
index 7aabda0b378e..8e804369cc60 100644
@@ -50,9 +50,6 @@ u64 sev_secrets_pa __ro_after_init;
 /* For early boot SVSM communication */
 struct svsm_ca boot_svsm_ca_page __aligned(PAGE_SIZE);
 
-DEFINE_PER_CPU(struct svsm_ca *, svsm_caa);
-DEFINE_PER_CPU(u64, svsm_caa_pa);
-
 /*
  * Nothing shall interrupt this code path while holding the per-CPU
  * GHCB. The backup GHCB is only for NMIs interrupting this path.
@@ -167,7 +164,7 @@ void __head early_snp_set_memory_private(unsigned long vaddr, unsigned long padd
 	  * table.
 	  */
 	early_set_pages_state(vaddr, paddr, npages, SNP_PAGE_STATE_PRIVATE,
-			      svsm_get_caa(), svsm_get_caa_pa());
+			      rip_rel_ptr(&boot_svsm_ca_page), boot_svsm_caa_pa);
 }
 
 void __head early_snp_set_memory_shared(unsigned long vaddr, unsigned long paddr,
@@ -184,7 +181,7 @@ void __head early_snp_set_memory_shared(unsigned long vaddr, unsigned long paddr
 
 	 /* Ask hypervisor to mark the memory pages shared in the RMP table. */
 	early_set_pages_state(vaddr, paddr, npages, SNP_PAGE_STATE_SHARED,
-			      svsm_get_caa(), svsm_get_caa_pa());
+			      rip_rel_ptr(&boot_svsm_ca_page), boot_svsm_caa_pa);
 }
 
 /*
arch/x86/coco/sev/core.c +22 -25

67 lines not on this page · raw message

arch/x86/include/asm/sev-internal.h +0 -16
index 3b72d8217827..bdfe008120f3 100644
@@ -63,22 +63,6 @@ DECLARE_PER_CPU(u64, svsm_caa_pa);
 
 extern u64 boot_svsm_caa_pa;
 
-static __always_inline struct svsm_ca *svsm_get_caa(void)
-{
-	if (sev_cfg.use_cas)
-		return this_cpu_read(svsm_caa);
-	else
-		return rip_rel_ptr(&boot_svsm_ca_page);
-}
-
-static __always_inline u64 svsm_get_caa_pa(void)
-{
-	if (sev_cfg.use_cas)
-		return this_cpu_read(svsm_caa_pa);
-	else
-		return boot_svsm_caa_pa;
-}
-
 enum es_result verify_exception_info(struct ghcb *ghcb, struct es_em_ctxt *ctxt);
 void vc_forward_exception(struct es_em_ctxt *ctxt);
 
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
#13 lore raw

PATCH v5 10/22 x86/boot: Drop redundant RMPADJUST in SEV SVSM presence check

In reply to Ard Biesheuvel

From: Ard Biesheuvel <[email protected]>

snp_vmpl will be assigned a non-zero value when executing at a VMPL other than 0, and this is inferred from a call to RMPADJUST, which only works when running at VMPL0.

This means that testing snp_vmpl is sufficient, and there is no need to perform the same check again.

---
 arch/x86/boot/compressed/sev.c | 20 +++-----------------
 1 file changed, 3 insertions(+), 17 deletions(-)
arch/x86/boot/compressed/sev.c +3 -17
index 4bdf5595ed96..d62722dd2de1 100644
@@ -392,30 +392,16 @@ void sev_enable(struct boot_params *bp)
 	 */
 	if (sev_status & MSR_AMD64_SEV_SNP_ENABLED) {
 		u64 hv_features;
-		int ret;
 
 		hv_features = get_hv_features();
 		if (!(hv_features & GHCB_HV_FT_SNP))
 			sev_es_terminate(SEV_TERM_SET_GEN, GHCB_SNP_UNSUPPORTED);
 
 		/*
-		 * Enforce running at VMPL0 or with an SVSM.
-		 *
-		 * Use RMPADJUST (see the rmpadjust() function for a description of
-		 * what the instruction does) to update the VMPL1 permissions of a
-		 * page. If the guest is running at VMPL0, this will succeed. If the
-		 * guest is running at any other VMPL, this will fail. Linux SNP guests
-		 * only ever run at a single VMPL level so permission mask changes of a
-		 * lesser-privileged VMPL are a don't-care.
+		 * Running at VMPL0 is required unless an SVSM is present and
+		 * the hypervisor supports the required SVSM GHCB events.
 		 */
-		ret = rmpadjust((unsigned long)&boot_ghcb_page, RMP_PG_SIZE_4K, 1);
-
-		/*
-		 * Running at VMPL0 is not required if an SVSM is present and the hypervisor
-		 * supports the required SVSM GHCB events.
-		 */
-		if (ret &&
-		    !(snp_vmpl && (hv_features & GHCB_HV_FT_SNP_MULTI_VMPL)))
+		if (snp_vmpl > 0 && !(hv_features & GHCB_HV_FT_SNP_MULTI_VMPL))
 			sev_es_terminate(SEV_TERM_SET_LINUX, GHCB_TERM_NOT_VMPL0);
 	}
 
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
#14 lore raw

PATCH v5 11/22 x86/boot: Provide PIC aliases for 5-level paging related constants

In reply to Ard Biesheuvel

From: Ard Biesheuvel <[email protected]>

Provide PIC aliases for the global variables related to 5-level paging, so that the startup code can access them in order to populate the kernel page tables.

---
 arch/x86/kernel/head64.c | 3 +++
 1 file changed, 3 insertions(+)
arch/x86/kernel/head64.c +3 -0
index 533fcf5636fc..1bc40d0785ee 100644
@@ -52,10 +52,13 @@ SYM_PIC_ALIAS(next_early_pgt);
 pmdval_t early_pmd_flags = __PAGE_KERNEL_LARGE & ~(_PAGE_GLOBAL | _PAGE_NX);
 
 unsigned int __pgtable_l5_enabled __ro_after_init;
+SYM_PIC_ALIAS(__pgtable_l5_enabled);
 unsigned int pgdir_shift __ro_after_init = 39;
 EXPORT_SYMBOL(pgdir_shift);
+SYM_PIC_ALIAS(pgdir_shift);
 unsigned int ptrs_per_p4d __ro_after_init = 1;
 EXPORT_SYMBOL(ptrs_per_p4d);
+SYM_PIC_ALIAS(ptrs_per_p4d);
 
 unsigned long page_offset_base __ro_after_init = __PAGE_OFFSET_BASE_L4;
 EXPORT_SYMBOL(page_offset_base);
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
#15 lore raw

PATCH v5 12/22 x86/sev: Provide PIC aliases for SEV related data objects

In reply to Ard Biesheuvel

From: Ard Biesheuvel <[email protected]>

Provide PIC aliases for data objects that are shared between the SEV startup code and the SEV code that executes later. This is needed so that the confined startup code is permitted to access them.

This requires some of these variables to be moved into a source file that is not part of the startup code, as the PIC alias is already implied, and exporting variables in the opposite direction is not supported.

Move ghcb_version as well, but don't provide a PIC alias as it is not actually needed.

---
 arch/x86/boot/compressed/sev.c      |  3 ++
 arch/x86/boot/startup/sev-shared.c  | 19 -----------
 arch/x86/boot/startup/sev-startup.c |  9 ------
 arch/x86/coco/sev/core.c            | 34 ++++++++++++++++++++
 4 files changed, 37 insertions(+), 28 deletions(-)
arch/x86/boot/compressed/sev.c +3 -0
index d62722dd2de1..faa6cc2f9990 100644
@@ -38,6 +38,9 @@ struct ghcb *boot_ghcb;
 #define __BOOT_COMPRESSED
 
 u8 snp_vmpl;
+u16 ghcb_version;
+
+u64 boot_svsm_caa_pa;
 
 /* Include code for early handlers */
 #include "../../boot/startup/sev-shared.c"
arch/x86/boot/startup/sev-shared.c +0 -19
index 83ca97df0808..cc14daf816e8 100644
@@ -18,25 +18,6 @@
 #define WARN(condition, format...) (!!(condition))
 #endif
 
-/*
- * SVSM related information:
- *   During boot, the page tables are set up as identity mapped and later
- *   changed to use kernel virtual addresses. Maintain separate virtual and
- *   physical addresses for the CAA to allow SVSM functions to be used during
- *   early boot, both with identity mapped virtual addresses and proper kernel
- *   virtual addresses.
- */
-u64 boot_svsm_caa_pa __ro_after_init;
-
-/*
- * Since feature negotiation related variables are set early in the boot
- * process they must reside in the .data section so as not to be zeroed
- * out when the .bss section is later cleared.
- *
- * GHCB protocol version negotiated with the hypervisor.
- */
-u16 ghcb_version __ro_after_init;
-
 /* Copy of the SNP firmware's CPUID page. */
 static struct snp_cpuid_table cpuid_table_copy __ro_after_init;
 
arch/x86/boot/startup/sev-startup.c +0 -9
index 8e804369cc60..733491482cbb 100644
@@ -41,15 +41,6 @@
 #include <asm/cpuid/api.h>
 #include <asm/cmdline.h>
 
-/* Bitmap of SEV features supported by the hypervisor */
-u64 sev_hv_features __ro_after_init;
-
-/* Secrets page physical address from the CC blob */
-u64 sev_secrets_pa __ro_after_init;
-
-/* For early boot SVSM communication */
-struct svsm_ca boot_svsm_ca_page __aligned(PAGE_SIZE);
-
 /*
  * Nothing shall interrupt this code path while holding the per-CPU
  * GHCB. The backup GHCB is only for NMIs interrupting this path.
arch/x86/coco/sev/core.c +34 -0
index 4fe0928bc0ad..be89f0a4a28f 100644
@@ -46,6 +46,29 @@
 #include <asm/cmdline.h>
 #include <asm/msr.h>
 
+/* Bitmap of SEV features supported by the hypervisor */
+u64 sev_hv_features __ro_after_init;
+SYM_PIC_ALIAS(sev_hv_features);
+
+/* Secrets page physical address from the CC blob */
+u64 sev_secrets_pa __ro_after_init;
+SYM_PIC_ALIAS(sev_secrets_pa);
+
+/* For early boot SVSM communication */
+struct svsm_ca boot_svsm_ca_page __aligned(PAGE_SIZE);
+SYM_PIC_ALIAS(boot_svsm_ca_page);
+
+/*
+ * SVSM related information:
+ *   During boot, the page tables are set up as identity mapped and later
+ *   changed to use kernel virtual addresses. Maintain separate virtual and
+ *   physical addresses for the CAA to allow SVSM functions to be used during
+ *   early boot, both with identity mapped virtual addresses and proper kernel
+ *   virtual addresses.
+ */
+u64 boot_svsm_caa_pa __ro_after_init;
+SYM_PIC_ALIAS(boot_svsm_caa_pa);
+
 DEFINE_PER_CPU(struct svsm_ca *, svsm_caa);
 DEFINE_PER_CPU(u64, svsm_caa_pa);
 
@@ -119,6 +142,17 @@ DEFINE_PER_CPU(struct sev_es_save_area *, sev_vmsa);
  */
 u8 snp_vmpl __ro_after_init;
 EXPORT_SYMBOL_GPL(snp_vmpl);
+SYM_PIC_ALIAS(snp_vmpl);
+
+/*
+ * Since feature negotiation related variables are set early in the boot
+ * process they must reside in the .data section so as not to be zeroed
+ * out when the .bss section is later cleared.
+ *
+ * GHCB protocol version negotiated with the hypervisor.
+ */
+u16 ghcb_version __ro_after_init;
+SYM_PIC_ALIAS(ghcb_version);
 
 /* For early boot hypervisor communication in SEV-ES enabled guests */
 static struct ghcb boot_ghcb_page __bss_decrypted __aligned(PAGE_SIZE);
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
#16 lore raw

PATCH v5 13/22 x86/sev: Move __sev_[get|put]_ghcb() into separate noinstr object

In reply to Ard Biesheuvel

From: Ard Biesheuvel <[email protected]>

Rename sev-nmi.c to noinstr.c, and move the get/put GHCB routines into it too, which are also annotated as 'noinstr' and suffer from the same problem as the NMI code, i.e., that GCC may ignore the __no_sanitize_address__ function attribute implied by 'noinstr' and insert KASAN instrumentation anyway.

---
 arch/x86/boot/startup/sev-startup.c        | 74 --------------------
 arch/x86/coco/sev/Makefile                 |  6 +-
 arch/x86/coco/sev/{sev-nmi.c => noinstr.c} | 74 ++++++++++++++++++++
 3 files changed, 77 insertions(+), 77 deletions(-)
arch/x86/boot/startup/sev-startup.c +0 -74
index 733491482cbb..e9238149f2cf 100644
@@ -41,83 +41,9 @@
 #include <asm/cpuid/api.h>
 #include <asm/cmdline.h>
 
-/*
- * Nothing shall interrupt this code path while holding the per-CPU
- * GHCB. The backup GHCB is only for NMIs interrupting this path.
- *
- * Callers must disable local interrupts around it.
- */
-noinstr struct ghcb *__sev_get_ghcb(struct ghcb_state *state)
-{
-	struct sev_es_runtime_data *data;
-	struct ghcb *ghcb;
-
-	WARN_ON(!irqs_disabled());
-
-	data = this_cpu_read(runtime_data);
-	ghcb = &data->ghcb_page;
-
-	if (unlikely(data->ghcb_active)) {
-		/* GHCB is already in use - save its contents */
-
-		if (unlikely(data->backup_ghcb_active)) {
-			/*
-			 * Backup-GHCB is also already in use. There is no way
-			 * to continue here so just kill the machine. To make
-			 * panic() work, mark GHCBs inactive so that messages
-			 * can be printed out.
-			 */
-			data->ghcb_active        = false;
-			data->backup_ghcb_active = false;
-
-			instrumentation_begin();
-			panic("Unable to handle #VC exception! GHCB and Backup GHCB are already in use");
-			instrumentation_end();
-		}
-
-		/* Mark backup_ghcb active before writing to it */
-		data->backup_ghcb_active = true;
-
-		state->ghcb = &data->backup_ghcb;
-
-		/* Backup GHCB content */
-		*state->ghcb = *ghcb;
-	} else {
-		state->ghcb = NULL;
-		data->ghcb_active = true;
-	}
-
-	return ghcb;
-}
-
 /* Include code shared with pre-decompression boot stage */
 #include "sev-shared.c"
 
-noinstr void __sev_put_ghcb(struct ghcb_state *state)
-{
-	struct sev_es_runtime_data *data;
-	struct ghcb *ghcb;
-
-	WARN_ON(!irqs_disabled());
-
-	data = this_cpu_read(runtime_data);
-	ghcb = &data->ghcb_page;
-
-	if (state->ghcb) {
-		/* Restore GHCB from Backup */
-		*ghcb = *state->ghcb;
-		data->backup_ghcb_active = false;
-		state->ghcb = NULL;
-	} else {
-		/*
-		 * Invalidate the GHCB so a VMGEXIT instruction issued
-		 * from userspace won't appear to be valid.
-		 */
-		vc_ghcb_invalidate(ghcb);
-		data->ghcb_active = false;
-	}
-}
-
 void __head
 early_set_pages_state(unsigned long vaddr, unsigned long paddr,
 		      unsigned long npages, enum psc_op op,
arch/x86/coco/sev/Makefile +3 -3
index db3255b979bd..53e964a22759 100644
@@ -1,9 +1,9 @@
 # SPDX-License-Identifier: GPL-2.0
 
-obj-y += core.o sev-nmi.o vc-handle.o
+obj-y += core.o noinstr.o vc-handle.o
 
 # Clang 14 and older may fail to respect __no_sanitize_undefined when inlining
-UBSAN_SANITIZE_sev-nmi.o	:= n
+UBSAN_SANITIZE_noinstr.o	:= n
 
 # GCC may fail to respect __no_sanitize_address when inlining
-KASAN_SANITIZE_sev-nmi.o	:= n
+KASAN_SANITIZE_noinstr.o	:= n
arch/x86/coco/sev/noinstr.c renamed from arch/x86/coco/sev/sev-nmi.c +74 -0

77 lines not on this page · raw message

Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
#17 lore raw

PATCH v5 14/22 x86/sev: Export startup routines for later use

In reply to Ard Biesheuvel

From: Ard Biesheuvel <[email protected]>

Create aliases that expose routines that are part of the startup code to other code in the core kernel, so that they can be called later as well.

---
 arch/x86/boot/startup/exports.h | 14 ++++++++++++++
 arch/x86/kernel/vmlinux.lds.S   |  2 ++
 2 files changed, 16 insertions(+)
arch/x86/boot/startup/exports.h new file +14 -0
new file mode 100644
index 000000000000..01d2363dc445
@@ -0,0 +1,14 @@
+
+/*
+ * The symbols below are functions that are implemented by the startup code,
+ * but called at runtime by the SEV code residing in the core kernel.
+ */
+PROVIDE(early_set_pages_state		= __pi_early_set_pages_state);
+PROVIDE(early_snp_set_memory_private	= __pi_early_snp_set_memory_private);
+PROVIDE(early_snp_set_memory_shared	= __pi_early_snp_set_memory_shared);
+PROVIDE(get_hv_features			= __pi_get_hv_features);
+PROVIDE(sev_es_terminate		= __pi_sev_es_terminate);
+PROVIDE(snp_cpuid			= __pi_snp_cpuid);
+PROVIDE(snp_cpuid_get_table		= __pi_snp_cpuid_get_table);
+PROVIDE(svsm_issue_call			= __pi_svsm_issue_call);
+PROVIDE(svsm_process_result_codes	= __pi_svsm_process_result_codes);
arch/x86/kernel/vmlinux.lds.S +2 -0
index 4fa0be732af1..5d5e3a95e1f9 100644
@@ -535,3 +535,5 @@ xen_elfnote_entry_value =
 xen_elfnote_phys32_entry_value =
 	ABSOLUTE(xen_elfnote_phys32_entry) + ABSOLUTE(pvh_start_xen - LOAD_OFFSET);
 #endif
+
+#include "../boot/startup/exports.h"
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
#18 lore raw

PATCH v5 15/22 objtool: Add action to check for absence of absolute relocations

In reply to Ard Biesheuvel

From: Ard Biesheuvel <[email protected]>

The x86 startup code must not use absolute references to code or data, as it executes before the kernel virtual mapping is up.

Add an action to objtool to check all allocatable sections (with the exception of __patchable_function_entries, which uses absolute references for nebulous reasons) and raise an error if any absolute references are found.

Note that debug sections typically contain lots of absolute references too, but those are not allocatable so they will be ignored.

---
 tools/objtool/builtin-check.c           |  2 ++
 tools/objtool/check.c                   | 36 ++++++++++++++++++++
 tools/objtool/include/objtool/builtin.h |  1 +
 3 files changed, 39 insertions(+)
tools/objtool/builtin-check.c +2 -0
index 80239843e9f0..0f6b197cfcb0 100644
@@ -87,6 +87,7 @@ static const struct option check_options[] = {
 	OPT_BOOLEAN('t', "static-call", &opts.static_call, "annotate static calls"),
 	OPT_BOOLEAN('u', "uaccess", &opts.uaccess, "validate uaccess rules for SMAP"),
 	OPT_BOOLEAN(0  , "cfi", &opts.cfi, "annotate kernel control flow integrity (kCFI) function preambles"),
+	OPT_BOOLEAN(0  , "noabs", &opts.noabs, "reject absolute references in allocatable sections"),
 	OPT_CALLBACK_OPTARG(0, "dump", NULL, NULL, "orc", "dump metadata", parse_dump),
 
 	OPT_GROUP("Options:"),
@@ -162,6 +163,7 @@ static bool opts_valid(void)
 	    opts.hack_noinstr		||
 	    opts.ibt			||
 	    opts.mcount			||
+	    opts.noabs			||
 	    opts.noinstr		||
 	    opts.orc			||
 	    opts.retpoline		||
tools/objtool/check.c +36 -0
index d967ac001498..5d1d38404892 100644
@@ -4643,6 +4643,39 @@ static void disas_warned_funcs(struct objtool_file *file)
 		disas_funcs(funcs);
 }
 
+static int check_abs_references(struct objtool_file *file)
+{
+	struct section *sec;
+	struct reloc *reloc;
+	int ret = 0;
+
+	for_each_sec(file, sec) {
+		/* absolute references in non-loadable sections are fine */
+		if (!(sec->sh.sh_flags & SHF_ALLOC))
+			continue;
+
+		/* section must have an associated .rela section */
+		if (!sec->rsec)
+			continue;
+
+		/*
+		 * Special case for compiler generated metadata that is not
+		 * consumed until after boot.
+		 */
+		if (!strcmp(sec->name, "__patchable_function_entries"))
+			continue;
+
+		for_each_reloc(sec->rsec, reloc) {
+			if (reloc_type(reloc) == R_ABS64) {
+				WARN("section %s has absolute relocation at offset 0x%lx",
+				     sec->name, reloc_offset(reloc));
+				ret++;
+			}
+		}
+	}
+	return ret;
+}
+
 struct insn_chunk {
 	void *addr;
 	struct insn_chunk *next;
@@ -4776,6 +4809,9 @@ int check(struct objtool_file *file)
 			goto out;
 	}
 
+	if (opts.noabs)
+		warnings += check_abs_references(file);
+
 	if (opts.orc && nr_insns) {
 		ret = orc_create(file);
 		if (ret)
tools/objtool/include/objtool/builtin.h +1 -0
index 6b08666fa69d..ab22673862e1 100644
@@ -26,6 +26,7 @@ struct opts {
 	bool uaccess;
 	int prefix;
 	bool cfi;
+	bool noabs;
 
 	/* options: */
 	bool backtrace;
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
maintainer
#19 lore raw

Re: [PATCH v5 15/22] objtool: Add action to check for absence of absolute relocations

Quoted 21 lines · Ard Biesheuvel

index d967ac001498..5d1d38404892 100644

tools/objtool/check.c +14 -0
@@ -4643,6 +4643,39 @@ static void disas_warned_funcs(struct objtool_file *file)
 		disas_funcs(funcs);
 }
 
+static int check_abs_references(struct objtool_file *file)
+{
+	struct section *sec;
+	struct reloc *reloc;
+	int ret = 0;
+
+	for_each_sec(file, sec) {
+		/* absolute references in non-loadable sections are fine */
+		if (!(sec->sh.sh_flags & SHF_ALLOC))
+			continue;
+
+		/* section must have an associated .rela section */
+		if (!sec->rsec)
+			continue;

tools/objtool/check.c static void disas_warned_funcs(struct objtool_file *file) · line 4669

Quoted 10 lines
tools/objtool/check.c +10 -0
+
+		/*
+		 * Special case for compiler generated metadata that is not
+		 * consumed until after boot.
+		 */
+		if (!strcmp(sec->name, "__patchable_function_entries"))
+			continue;
+
+		for_each_reloc(sec->rsec, reloc) {
+			if (reloc_type(reloc) == R_ABS64) {

This should probably also check R_ABS32. Yes, your current only user is x86_64 so R_ABS64 covers things, but we're getting more and more archs using objtool. No reason this check shouldn't also work on PPC32 for example.

Quoted 8 lines
tools/objtool/check.c +8 -0
+				WARN("section %s has absolute relocation at offset 0x%lx",
+				     sec->name, reloc_offset(reloc));
+				ret++;
+			}
+		}
+	}
+	return ret;
+}
maintainer
#20 lore raw

Re: [PATCH v5 15/22] objtool: Add action to check for absence of absolute relocations

Quoted 30 lines · Peter Zijlstra
Quoted 28 lines · Ard Biesheuvel

index d967ac001498..5d1d38404892 100644

tools/objtool/check.c +21 -0
@@ -4643,6 +4643,39 @@ static void disas_warned_funcs(struct objtool_file *file)
              disas_funcs(funcs);
 }
 
+static int check_abs_references(struct objtool_file *file)
+{
+     struct section *sec;
+     struct reloc *reloc;
+     int ret = 0;
+
+     for_each_sec(file, sec) {
+             /* absolute references in non-loadable sections are fine */
+             if (!(sec->sh.sh_flags & SHF_ALLOC))
+                     continue;
+
+             /* section must have an associated .rela section */
+             if (!sec->rsec)
+                     continue;
+
+             /*
+              * Special case for compiler generated metadata that is not
+              * consumed until after boot.
+              */
+             if (!strcmp(sec->name, "__patchable_function_entries"))
+                     continue;
Quoted 9 lines

tools/objtool/check.c static void disas_warned_funcs(struct objtool_file *file) · line 4669

Quoted 3 lines
tools/objtool/check.c +3 -0
+
+             for_each_reloc(sec->rsec, reloc) {
+                     if (reloc_type(reloc) == R_ABS64) {

This should probably also check R_ABS32. Yes, your current only user is x86_64 so R_ABS64 covers things, but we're getting more and more archs using objtool. No reason this check shouldn't also work on PPC32 for example.

Yeah, I was unsure about this.

This check is sufficient to ensure that PIC code is not emitted with, e.g., global variables with absolute addresses, etc. So the R_ABS64 check here is only a check whether any relocations of the native pointer size are present (but no R_ABS_NATIVE exists at this point)

For robustness, we should actually check for all absolute relocations here, including R_X86_64_32S, which is not abstracted into a R_ABSxx type for objtool.

So perhaps this needs an arch hook where x86_64 can implement it as

bool arch_is_abs_reloc(reloc)
{
switch (reloc_type(reloc)) {
case R_X86_64_32:
case R_X86_64_32S:
case R_X86_64_64:

      return true;

}
return false;
}

and the default just compares against R_ABS32 / R_ABS64 depending on the word size?

maintainer
#21 lore raw

Re: [PATCH v5 15/22] objtool: Add action to check for absence of absolute relocations

Quoted 17 lines · Ard Biesheuvel

For robustness, we should actually check for all absolute relocations here, including R_X86_64_32S, which is not abstracted into a R_ABSxx type for objtool.

So perhaps this needs an arch hook where x86_64 can implement it as

bool arch_is_abs_reloc(reloc)

{
   switch (reloc_type(reloc)) {
   case R_X86_64_32:
   case R_X86_64_32S:
   case R_X86_64_64:
      return true;
   }
   return false;
}
Quoted 2 lines

and the default just compares against R_ABS32 / R_ABS64 depending on the word size?

Yes, an arch hook like that makes sense. Perhaps make the signature:

bool arch_is_abs_reloc(struct elf *, struct reloc *);

Because the word size comes from elf_addr_size().

maintainer
#22 lore raw

Re: [PATCH v5 15/22] objtool: Add action to check for absence of absolute relocations

Quoted 19 lines · Peter Zijlstra
Quoted 17 lines · Ard Biesheuvel

For robustness, we should actually check for all absolute relocations here, including R_X86_64_32S, which is not abstracted into a R_ABSxx type for objtool.

So perhaps this needs an arch hook where x86_64 can implement it as

bool arch_is_abs_reloc(reloc)

{
   switch (reloc_type(reloc)) {
   case R_X86_64_32:
   case R_X86_64_32S:
   case R_X86_64_64:
      return true;
   }
   return false;
}
Quoted 8 lines
Quoted 2 lines

and the default just compares against R_ABS32 / R_ABS64 depending on the word size?

Yes, an arch hook like that makes sense. Perhaps make the signature:

bool arch_is_abs_reloc(struct elf *, struct reloc *);

Because the word size comes from elf_addr_size().

We already have an arch_pc_relative_reloc(), please try to keep the naming consistent.

Signature · 1 line
Josh
#23 lore raw

PATCH v5 16/22 x86/boot: Check startup code for absence of absolute relocations

In reply to Ard Biesheuvel

From: Ard Biesheuvel <[email protected]>

Invoke objtool on each startup code object individually to check for the absence of absolute relocations. This is needed because this code will be invoked from the 1:1 mapping of memory before those absolute virtual addresses (which are derived from the kernel virtual base address provided to the linker and possibly shifted at boot) are mapped.

Only objects built under arch/x86/boot/startup/ have this restriction, and once they have been incorporated into vmlinux.o, this distinction is difficult to make. So force the invocation of objtool for each object file individually, even if objtool is deferred to vmlinux.o for the rest of the build. In the latter case, only pass --noabs and nothing else; otherwise, append it to the existing objtool command line.

---
 arch/x86/boot/startup/Makefile | 8 ++++++++
 1 file changed, 8 insertions(+)
arch/x86/boot/startup/Makefile +8 -0
index b514f7e81332..32737f4ab5a8 100644
@@ -19,6 +19,7 @@ KCOV_INSTRUMENT	:= n
 
 obj-$(CONFIG_X86_64)		+= gdt_idt.o map_kernel.o
 obj-$(CONFIG_AMD_MEM_ENCRYPT)	+= sme.o sev-startup.o
+pi-objs				:= $(patsubst %.o,$(obj)/%.o,$(obj-y))
 
 lib-$(CONFIG_X86_64)		+= la57toggle.o
 lib-$(CONFIG_EFI_MIXED)		+= efi-mixed.o
@@ -28,3 +29,10 @@ lib-$(CONFIG_EFI_MIXED)		+= efi-mixed.o
 # to be linked into the decompressor or the EFI stub but not vmlinux
 #
 $(patsubst %.o,$(obj)/%.o,$(lib-y)): OBJECT_FILES_NON_STANDARD := y
+
+#
+# Invoke objtool for each object individually to check for absolute
+# relocations, even if other objtool actions are being deferred.
+#
+$(pi-objs): objtool-enabled	= 1
+$(pi-objs): objtool-args	= $(if $(delay-objtool),,$(objtool-args-y)) --noabs
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
#24 lore raw

PATCH v5 17/22 x86/boot: Revert "Reject absolute references in .head.text"

In reply to Ard Biesheuvel

From: Ard Biesheuvel <[email protected]>

This reverts commit

faf0ed487415 ("x86/boot: Reject absolute references in .head.text")

The startup code is checked directly for the absence of absolute symbol references, so checking the .head.text section in the relocs tool is no longer needed.

---
 arch/x86/tools/relocs.c | 8 +-------
 1 file changed, 1 insertion(+), 7 deletions(-)
arch/x86/tools/relocs.c +1 -7
index 5778bc498415..e5a2b9a912d1 100644
@@ -740,10 +740,10 @@ static void walk_relocs(int (*process)(struct section *sec, Elf_Rel *rel,
 static int do_reloc64(struct section *sec, Elf_Rel *rel, ElfW(Sym) *sym,
 		      const char *symname)
 {
-	int headtext = !strcmp(sec_name(sec->shdr.sh_info), ".head.text");
 	unsigned r_type = ELF64_R_TYPE(rel->r_info);
 	ElfW(Addr) offset = rel->r_offset;
 	int shn_abs = (sym->st_shndx == SHN_ABS) && !is_reloc(S_REL, symname);
+
 	if (sym->st_shndx == SHN_UNDEF)
 		return 0;
 
@@ -783,12 +783,6 @@ static int do_reloc64(struct section *sec, Elf_Rel *rel, ElfW(Sym) *sym,
 			break;
 		}
 
-		if (headtext) {
-			die("Absolute reference to symbol '%s' not permitted in .head.text\n",
-			    symname);
-			break;
-		}
-
 		/*
 		 * Relocation offsets for 64 bit kernels are output
 		 * as 32 bits and sign extended back to 64 bits when
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
#25 lore raw

PATCH v5 18/22 x86/kbuild: Incorporate boot/startup/ via Kbuild makefile

In reply to Ard Biesheuvel

From: Ard Biesheuvel <[email protected]>

Using core-y is not the correct way to get kbuild to descend into arch/x86/boot/startup. For instance, building an individual object does not work as expected when the pattern rule is local to the Makefile

  $ make arch/x86/boot/startup/map_kernel.pi.o
    GEN     Makefile
    CALL    /home/ardb/linux/scripts/checksyscalls.sh
    DESCEND objtool
    INSTALL libsubcmd_headers
  make[3]: *** No rule to make target 'arch/x86/boot/startup/map_kernel.pi.o'.  Stop.
  make[2]: *** [/home/ardb/linux/scripts/Makefile.build:461: arch/x86] Error 2
  make[1]: *** [/home/ardb/linux/Makefile:2011: .] Error 2
  make: *** [/home/ardb/linux/Makefile:248: __sub-make] Error 2

So use obj-y from arch.x86/Kbuild instead, which makes things work as expected.

---
 arch/x86/Kbuild   | 2 ++
 arch/x86/Makefile | 1 -
 2 files changed, 2 insertions(+), 1 deletion(-)
arch/x86/Kbuild +2 -0
index f7fb3d88c57b..36b985d0e7bf 100644
@@ -3,6 +3,8 @@
 # Branch profiling isn't noinstr-safe.  Disable it for arch/x86/*
 subdir-ccflags-$(CONFIG_TRACE_BRANCH_PROFILING) += -DDISABLE_BRANCH_PROFILING
 
+obj-y += boot/startup/
+
 obj-$(CONFIG_ARCH_HAS_CC_PLATFORM) += coco/
 
 obj-y += entry/
arch/x86/Makefile +0 -1
index 1913d342969b..9b76e77ff7f7 100644
@@ -286,7 +286,6 @@ archprepare: $(cpufeaturemasks.hdr)
 ###
 # Kernel objects
 
-core-y  += arch/x86/boot/startup/
 libs-y  += arch/x86/lib/
 
 # drivers-y are linked after core-y
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
#26 lore raw

PATCH v5 19/22 x86/boot: Create a confined code area for startup code

In reply to Ard Biesheuvel

From: Ard Biesheuvel <[email protected]>

In order to be able to have tight control over which code may execute from the early 1:1 mapping of memory, but still link vmlinux as a single executable, prefix all symbol references in startup code with __pi_, and invoke it from outside using the __pi_ prefix.

Use objtool to check that no absolute symbol references are present in the startup code, as these cannot be used from code running from the 1:1 mapping.

---
 arch/x86/boot/startup/Makefile     | 14 ++++++++++++++
 arch/x86/boot/startup/sev-shared.c |  4 +---
 arch/x86/boot/startup/sme.c        |  1 -
 arch/x86/coco/sev/core.c           |  2 +-
 arch/x86/include/asm/setup.h       |  1 +
 arch/x86/include/asm/sev.h         |  1 +
 arch/x86/kernel/head64.c           |  2 +-
 arch/x86/kernel/head_64.S          |  8 ++++----
 arch/x86/mm/mem_encrypt_boot.S     |  6 +++---
 tools/objtool/check.c              |  3 ++-
 10 files changed, 28 insertions(+), 14 deletions(-)
arch/x86/boot/startup/Makefile +14 -0
index 32737f4ab5a8..e8fdf020b422 100644
@@ -4,6 +4,7 @@ KBUILD_AFLAGS		+= -D__DISABLE_EXPORTS
 KBUILD_CFLAGS		+= -D__DISABLE_EXPORTS -mcmodel=small -fPIC \
 			   -Os -DDISABLE_BRANCH_PROFILING \
 			   $(DISABLE_STACKLEAK_PLUGIN) \
+			   $(DISABLE_LATENT_ENTROPY_PLUGIN) \
 			   -fno-stack-protector -D__NO_FORTIFY \
 			   -fno-jump-tables \
 			   -include $(srctree)/include/linux/hidden.h
@@ -36,3 +37,16 @@ $(patsubst %.o,$(obj)/%.o,$(lib-y)): OBJECT_FILES_NON_STANDARD := y
 #
 $(pi-objs): objtool-enabled	= 1
 $(pi-objs): objtool-args	= $(if $(delay-objtool),,$(objtool-args-y)) --noabs
+
+#
+# Confine the startup code by prefixing all symbols with __pi_ (for position
+# independent). This ensures that startup code can only call other startup
+# code, or code that has explicitly been made accessible to it via a symbol
+# alias.
+#
+$(obj)/%.pi.o: OBJCOPYFLAGS := --prefix-symbols=__pi_
+$(obj)/%.pi.o: $(obj)/%.o FORCE
+	$(call if_changed,objcopy)
+
+targets	+= $(obj-y)
+obj-y	:= $(patsubst %.o,%.pi.o,$(obj-y))
arch/x86/boot/startup/sev-shared.c +1 -3
index cc14daf816e8..b60d546e74a7 100644
@@ -11,9 +11,7 @@
 
 #include <asm/setup_data.h>
 
-#ifndef __BOOT_COMPRESSED
-#define error(v)			pr_err(v)
-#else
+#ifdef __BOOT_COMPRESSED
 #undef WARN
 #define WARN(condition, format...) (!!(condition))
 #endif
arch/x86/boot/startup/sme.c +0 -1
index 70ea1748c0a7..eb6a758ba660 100644
@@ -567,7 +567,6 @@ void __head sme_enable(struct boot_params *bp)
 
 #ifdef CONFIG_MITIGATION_PAGE_TABLE_ISOLATION
 /* Local version for startup code, which never operates on user page tables */
-__weak
 pgd_t __pti_set_user_pgtbl(pgd_t *pgdp, pgd_t pgd)
 {
 	return pgd;
arch/x86/coco/sev/core.c +1 -1
index be89f0a4a28f..fbed1651f6d8 100644
@@ -272,7 +272,7 @@ static int svsm_perform_call_protocol(struct svsm_call *call)
 
 	do {
 		ret = ghcb ? svsm_perform_ghcb_protocol(ghcb, call)
-			   : svsm_perform_msr_protocol(call);
+			   : __pi_svsm_perform_msr_protocol(call);
 	} while (ret == -EAGAIN);
 
 	if (sev_cfg.ghcbs_initialized)
arch/x86/include/asm/setup.h +1 -0
index 692af46603a1..914eb32581c7 100644
@@ -53,6 +53,7 @@ extern void i386_reserve_resources(void);
 extern unsigned long __startup_64(unsigned long p2v_offset, struct boot_params *bp);
 extern void startup_64_setup_gdt_idt(void);
 extern void startup_64_load_idt(void *vc_handler);
+extern void __pi_startup_64_load_idt(void *vc_handler);
 extern void early_setup_idt(void);
 extern void __init do_early_exception(struct pt_regs *regs, int trapnr);
 
arch/x86/include/asm/sev.h +1 -0
index f3acbfcdca9a..2d61b13e1810 100644
@@ -551,6 +551,7 @@ struct cpuid_leaf {
 };
 
 int svsm_perform_msr_protocol(struct svsm_call *call);
+int __pi_svsm_perform_msr_protocol(struct svsm_call *call);
 int snp_cpuid(void (*cpuid_hv)(void *ctx, struct cpuid_leaf *),
 	      void *ctx, struct cpuid_leaf *leaf);
 
arch/x86/kernel/head64.c +1 -1
index 1bc40d0785ee..fd28b53dbac5 100644
@@ -319,5 +319,5 @@ void early_setup_idt(void)
 		handler = vc_boot_ghcb;
 	}
 
-	startup_64_load_idt(handler);
+	__pi_startup_64_load_idt(handler);
 }
arch/x86/kernel/head_64.S +4 -4
index 3e9b3a3bd039..d219963ecb60 100644
@@ -71,7 +71,7 @@ SYM_CODE_START_NOALIGN(startup_64)
 	xorl	%edx, %edx
 	wrmsr
 
-	call	startup_64_setup_gdt_idt
+	call	__pi_startup_64_setup_gdt_idt
 
 	/* Now switch to __KERNEL_CS so IRET works reliably */
 	pushq	$__KERNEL_CS
@@ -91,7 +91,7 @@ SYM_CODE_START_NOALIGN(startup_64)
 	 * subsequent code. Pass the boot_params pointer as the first argument.
 	 */
 	movq	%r15, %rdi
-	call	sme_enable
+	call	__pi_sme_enable
 #endif
 
 	/* Sanitize CPU configuration */
@@ -111,7 +111,7 @@ SYM_CODE_START_NOALIGN(startup_64)
 	 * programmed into CR3.
 	 */
 	movq	%r15, %rsi
-	call	__startup_64
+	call	__pi___startup_64
 
 	/* Form the CR3 value being sure to include the CR3 modifier */
 	leaq	early_top_pgt(%rip), %rcx
@@ -562,7 +562,7 @@ SYM_CODE_START_NOALIGN(vc_no_ghcb)
 	/* Call C handler */
 	movq    %rsp, %rdi
 	movq	ORIG_RAX(%rsp), %rsi
-	call    do_vc_no_ghcb
+	call    __pi_do_vc_no_ghcb
 
 	/* Unwind pt_regs */
 	POP_REGS
arch/x86/mm/mem_encrypt_boot.S +3 -3
index f8a33b25ae86..edbf9c998848 100644
@@ -16,7 +16,7 @@
 
 	.text
 	.code64
-SYM_FUNC_START(sme_encrypt_execute)
+SYM_FUNC_START(__pi_sme_encrypt_execute)
 
 	/*
 	 * Entry parameters:
@@ -69,9 +69,9 @@ SYM_FUNC_START(sme_encrypt_execute)
 	ANNOTATE_UNRET_SAFE
 	ret
 	int3
-SYM_FUNC_END(sme_encrypt_execute)
+SYM_FUNC_END(__pi_sme_encrypt_execute)
 
-SYM_FUNC_START(__enc_copy)
+SYM_FUNC_START_LOCAL(__enc_copy)
 	ANNOTATE_NOENDBR
 /*
  * Routine used to encrypt memory in place.
tools/objtool/check.c +2 -1
index 5d1d38404892..f43bd598d928 100644
@@ -3563,7 +3563,8 @@ static int validate_branch(struct objtool_file *file, struct symbol *func,
 		if (func && insn_func(insn) && func != insn_func(insn)->pfunc) {
 			/* Ignore KCFI type preambles, which always fall through */
 			if (!strncmp(func->name, "__cfi_", 6) ||
-			    !strncmp(func->name, "__pfx_", 6))
+			    !strncmp(func->name, "__pfx_", 6) ||
+			    !strncmp(func->name, "__pi___pfx_", 11))
 				return 0;
 
 			if (file->ignore_unreachables)
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
#27 lore raw

PATCH v5 20/22 efistub/x86: Remap inittext read-execute when needed

In reply to Ard Biesheuvel

From: Ard Biesheuvel <[email protected]>

Recent EFI x86 systems are more strict when it comes to mapping boot images, and require that mappings are either read-write or read-execute.

Now that the boot code is being cleaned up and refactored, most of it is being moved into .init.text [where it arguably belongs] but that implies that when booting on such strict EFI firmware, we need to take care to map .init.text (and the .altinstr_aux section that follows it) read-execute as well.

---
 arch/x86/boot/compressed/Makefile       | 2 +-
 arch/x86/boot/compressed/misc.c         | 2 ++
 arch/x86/include/asm/boot.h             | 2 ++
 arch/x86/kernel/vmlinux.lds.S           | 2 ++
 drivers/firmware/efi/libstub/x86-stub.c | 4 +++-
 5 files changed, 10 insertions(+), 2 deletions(-)
arch/x86/boot/compressed/Makefile +1 -1
index 3a38fdcdb9bd..74657589264d 100644
@@ -73,7 +73,7 @@ LDFLAGS_vmlinux += -T
 hostprogs	:= mkpiggy
 HOST_EXTRACFLAGS += -I$(srctree)/tools/include
 
-sed-voffset := -e 's/^\([0-9a-fA-F]*\) [ABbCDGRSTtVW] \(_text\|__start_rodata\|__bss_start\|_end\)$$/\#define VO_\2 _AC(0x\1,UL)/p'
+sed-voffset := -e 's/^\([0-9a-fA-F]*\) [ABbCDGRSTtVW] \(_text\|__start_rodata\|_sinittext\|__inittext_end\|__bss_start\|_end\)$$/\#define VO_\2 _AC(0x\1,UL)/p'
 
 quiet_cmd_voffset = VOFFSET $@
       cmd_voffset = $(NM) $< | sed -n $(sed-voffset) > $@
arch/x86/boot/compressed/misc.c +2 -0
index 94b5991da001..0f41ca0e52c0 100644
@@ -332,6 +332,8 @@ static size_t parse_elf(void *output)
 }
 
 const unsigned long kernel_text_size = VO___start_rodata - VO__text;
+const unsigned long kernel_inittext_offset = VO__sinittext - VO__text;
+const unsigned long kernel_inittext_size = VO___inittext_end - VO__sinittext;
 const unsigned long kernel_total_size = VO__end - VO__text;
 
 static u8 boot_heap[BOOT_HEAP_SIZE] __aligned(4);
arch/x86/include/asm/boot.h +2 -0
index 02b23aa78955..f7b67cb73915 100644
@@ -82,6 +82,8 @@
 #ifndef __ASSEMBLER__
 extern unsigned int output_len;
 extern const unsigned long kernel_text_size;
+extern const unsigned long kernel_inittext_offset;
+extern const unsigned long kernel_inittext_size;
 extern const unsigned long kernel_total_size;
 
 unsigned long decompress_kernel(unsigned char *outbuf, unsigned long virt_addr,
arch/x86/kernel/vmlinux.lds.S +2 -0
index 5d5e3a95e1f9..4277efb26358 100644
@@ -227,6 +227,8 @@ SECTIONS
 	 */
 	.altinstr_aux : AT(ADDR(.altinstr_aux) - LOAD_OFFSET) {
 		*(.altinstr_aux)
+		. = ALIGN(PAGE_SIZE);
+		__inittext_end = .;
 	}
 
 	INIT_DATA_SECTION(16)
drivers/firmware/efi/libstub/x86-stub.c +3 -1
index cafc90d4caaf..0d05eac7c72b 100644
@@ -788,7 +788,9 @@ static efi_status_t efi_decompress_kernel(unsigned long *kernel_entry,
 
 	*kernel_entry = addr + entry;
 
-	return efi_adjust_memory_range_protection(addr, kernel_text_size);
+	return efi_adjust_memory_range_protection(addr, kernel_text_size) ?:
+	       efi_adjust_memory_range_protection(addr + kernel_inittext_offset,
+						  kernel_inittext_size);
 }
 
 static void __noreturn enter_kernel(unsigned long kernel_addr,
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
#28 lore raw

PATCH v5 21/22 x86/boot: Move startup code out of __head section

In reply to Ard Biesheuvel

From: Ard Biesheuvel <[email protected]>

Move startup code out of the __head section, now that this no longer has a special significance. Move everything into .text or .init.text as appropriate, so that startup code is not kept around unnecessarily.

---
 arch/x86/boot/compressed/sev.c      |  3 --
 arch/x86/boot/startup/gdt_idt.c     |  4 +--
 arch/x86/boot/startup/map_kernel.c  |  4 +--
 arch/x86/boot/startup/sev-shared.c  | 38 ++++++++++----------
 arch/x86/boot/startup/sev-startup.c | 14 ++++----
 arch/x86/boot/startup/sme.c         | 26 +++++++-------
 arch/x86/include/asm/init.h         |  6 ----
 arch/x86/kernel/head_32.S           |  2 +-
 arch/x86/kernel/head_64.S           |  2 +-
 arch/x86/platform/pvh/head.S        |  2 +-
 10 files changed, 46 insertions(+), 55 deletions(-)
arch/x86/boot/compressed/sev.c +0 -3
index faa6cc2f9990..a7af906145e8 100644
@@ -32,9 +32,6 @@ struct ghcb *boot_ghcb;
 #undef __init
 #define __init
 
-#undef __head
-#define __head
-
 #define __BOOT_COMPRESSED
 
 u8 snp_vmpl;
arch/x86/boot/startup/gdt_idt.c +2 -2
index a3112a69b06a..d16102abdaec 100644
@@ -24,7 +24,7 @@
 static gate_desc bringup_idt_table[NUM_EXCEPTION_VECTORS] __page_aligned_data;
 
 /* This may run while still in the direct mapping */
-void __head startup_64_load_idt(void *vc_handler)
+void startup_64_load_idt(void *vc_handler)
 {
 	struct desc_ptr desc = {
 		.address = (unsigned long)rip_rel_ptr(bringup_idt_table),
@@ -46,7 +46,7 @@ void __head startup_64_load_idt(void *vc_handler)
 /*
  * Setup boot CPU state needed before kernel switches to virtual addresses.
  */
-void __head startup_64_setup_gdt_idt(void)
+void __init startup_64_setup_gdt_idt(void)
 {
 	struct gdt_page *gp = rip_rel_ptr((void *)(__force unsigned long)&gdt_page);
 	void *handler = NULL;
arch/x86/boot/startup/map_kernel.c +2 -2
index 332dbe6688c4..83ba98d61572 100644
@@ -30,7 +30,7 @@ static inline bool check_la57_support(void)
 	return true;
 }
 
-static unsigned long __head sme_postprocess_startup(struct boot_params *bp,
+static unsigned long __init sme_postprocess_startup(struct boot_params *bp,
 						    pmdval_t *pmd,
 						    unsigned long p2v_offset)
 {
@@ -84,7 +84,7 @@ static unsigned long __head sme_postprocess_startup(struct boot_params *bp,
  * the 1:1 mapping of memory. Kernel virtual addresses can be determined by
  * subtracting p2v_offset from the RIP-relative address.
  */
-unsigned long __head __startup_64(unsigned long p2v_offset,
+unsigned long __init __startup_64(unsigned long p2v_offset,
 				  struct boot_params *bp)
 {
 	pmd_t (*early_pgts)[PTRS_PER_PMD] = rip_rel_ptr(early_dynamic_pgts);
arch/x86/boot/startup/sev-shared.c +19 -19
index b60d546e74a7..4e36a81d8c18 100644
@@ -29,7 +29,7 @@ static u32 cpuid_std_range_max __ro_after_init;
 static u32 cpuid_hyp_range_max __ro_after_init;
 static u32 cpuid_ext_range_max __ro_after_init;
 
-void __head __noreturn
+void __noreturn
 sev_es_terminate(unsigned int set, unsigned int reason)
 {
 	u64 val = GHCB_MSR_TERM_REQ;
@@ -48,7 +48,7 @@ sev_es_terminate(unsigned int set, unsigned int reason)
 /*
  * The hypervisor features are available from GHCB version 2 onward.
  */
-u64 get_hv_features(void)
+u64 __init get_hv_features(void)
 {
 	u64 val;
 
@@ -218,7 +218,7 @@ const struct snp_cpuid_table *snp_cpuid_get_table(void)
  *
  * Return: XSAVE area size on success, 0 otherwise.
  */
-static u32 __head snp_cpuid_calc_xsave_size(u64 xfeatures_en, bool compacted)
+static u32 snp_cpuid_calc_xsave_size(u64 xfeatures_en, bool compacted)
 {
 	const struct snp_cpuid_table *cpuid_table = snp_cpuid_get_table();
 	u64 xfeatures_found = 0;
@@ -254,7 +254,7 @@ static u32 __head snp_cpuid_calc_xsave_size(u64 xfeatures_en, bool compacted)
 	return xsave_size;
 }
 
-static bool __head
+static bool
 snp_cpuid_get_validated_func(struct cpuid_leaf *leaf)
 {
 	const struct snp_cpuid_table *cpuid_table = snp_cpuid_get_table();
@@ -296,8 +296,8 @@ static void snp_cpuid_hv_msr(void *ctx, struct cpuid_leaf *leaf)
 		sev_es_terminate(SEV_TERM_SET_LINUX, GHCB_TERM_CPUID_HV);
 }
 
-static int __head snp_cpuid_postprocess(void (*cpuid_fn)(void *ctx, struct cpuid_leaf *),
-					void *ctx, struct cpuid_leaf *leaf)
+static int snp_cpuid_postprocess(void (*cpuid_fn)(void *ctx, struct cpuid_leaf *),
+				 void *ctx, struct cpuid_leaf *leaf)
 {
 	struct cpuid_leaf leaf_hv = *leaf;
 
@@ -391,8 +391,8 @@ static int __head snp_cpuid_postprocess(void (*cpuid_fn)(void *ctx, struct cpuid
  * Returns -EOPNOTSUPP if feature not enabled. Any other non-zero return value
  * should be treated as fatal by caller.
  */
-int __head snp_cpuid(void (*cpuid_fn)(void *ctx, struct cpuid_leaf *), void *ctx,
-		     struct cpuid_leaf *leaf)
+int snp_cpuid(void (*cpuid_fn)(void *ctx, struct cpuid_leaf *), void *ctx,
+	      struct cpuid_leaf *leaf)
 {
 	const struct snp_cpuid_table *cpuid_table = snp_cpuid_get_table();
 
@@ -434,7 +434,7 @@ int __head snp_cpuid(void (*cpuid_fn)(void *ctx, struct cpuid_leaf *), void *ctx
  * page yet, so it only supports the MSR based communication with the
  * hypervisor and only the CPUID exit-code.
  */
-void __head do_vc_no_ghcb(struct pt_regs *regs, unsigned long exit_code)
+void do_vc_no_ghcb(struct pt_regs *regs, unsigned long exit_code)
 {
 	unsigned int subfn = lower_bits(regs->cx, 32);
 	unsigned int fn = lower_bits(regs->ax, 32);
@@ -510,7 +510,7 @@ struct cc_setup_data {
  * Search for a Confidential Computing blob passed in as a setup_data entry
  * via the Linux Boot Protocol.
  */
-static __head
+static __init
 struct cc_blob_sev_info *find_cc_blob_setup_data(struct boot_params *bp)
 {
 	struct cc_setup_data *sd = NULL;
@@ -538,7 +538,7 @@ struct cc_blob_sev_info *find_cc_blob_setup_data(struct boot_params *bp)
  * mapping needs to be updated in sync with all the changes to virtual memory
  * layout and related mapping facilities throughout the boot process.
  */
-static void __head setup_cpuid_table(const struct cc_blob_sev_info *cc_info)
+static void __init setup_cpuid_table(const struct cc_blob_sev_info *cc_info)
 {
 	const struct snp_cpuid_table *cpuid_table_fw, *cpuid_table;
 	int i;
@@ -566,7 +566,7 @@ static void __head setup_cpuid_table(const struct cc_blob_sev_info *cc_info)
 	}
 }
 
-static int __head svsm_call_msr_protocol(struct svsm_call *call)
+static int svsm_call_msr_protocol(struct svsm_call *call)
 {
 	int ret;
 
@@ -577,8 +577,8 @@ static int __head svsm_call_msr_protocol(struct svsm_call *call)
 	return ret;
 }
 
-static void __head svsm_pval_4k_page(unsigned long paddr, bool validate,
-				     struct svsm_ca *caa, u64 caa_pa)
+static void svsm_pval_4k_page(unsigned long paddr, bool validate,
+			      struct svsm_ca *caa, u64 caa_pa)
 {
 	struct svsm_pvalidate_call *pc;
 	struct svsm_call call = {};
@@ -618,8 +618,8 @@ static void __head svsm_pval_4k_page(unsigned long paddr, bool validate,
 	native_local_irq_restore(flags);
 }
 
-static void __head pvalidate_4k_page(unsigned long vaddr, unsigned long paddr,
-				     bool validate, struct svsm_ca *caa, u64 caa_pa)
+static void pvalidate_4k_page(unsigned long vaddr, unsigned long paddr,
+			      bool validate, struct svsm_ca *caa, u64 caa_pa)
 {
 	int ret;
 
@@ -632,8 +632,8 @@ static void __head pvalidate_4k_page(unsigned long vaddr, unsigned long paddr,
 	}
 }
 
-static void __head __page_state_change(unsigned long vaddr, unsigned long paddr,
-				       enum psc_op op, struct svsm_ca *caa, u64 caa_pa)
+static void __page_state_change(unsigned long vaddr, unsigned long paddr,
+				enum psc_op op, struct svsm_ca *caa, u64 caa_pa)
 {
 	u64 val, msr;
 
@@ -671,7 +671,7 @@ static void __head __page_state_change(unsigned long vaddr, unsigned long paddr,
  * Maintain the GPA of the SVSM Calling Area (CA) in order to utilize the SVSM
  * services needed when not running in VMPL0.
  */
-static bool __head svsm_setup_ca(const struct cc_blob_sev_info *cc_info,
+static bool __init svsm_setup_ca(const struct cc_blob_sev_info *cc_info,
 				 void *page)
 {
 	struct snp_secrets_page *secrets_page;
arch/x86/boot/startup/sev-startup.c +7 -7
index e9238149f2cf..1fdf196f9fad 100644
@@ -44,7 +44,7 @@
 /* Include code shared with pre-decompression boot stage */
 #include "sev-shared.c"
 
-void __head
+void __init
 early_set_pages_state(unsigned long vaddr, unsigned long paddr,
 		      unsigned long npages, enum psc_op op,
 		      struct svsm_ca *caa, u64 caa_pa)
@@ -64,7 +64,7 @@ early_set_pages_state(unsigned long vaddr, unsigned long paddr,
 	}
 }
 
-void __head early_snp_set_memory_private(unsigned long vaddr, unsigned long paddr,
+void __init early_snp_set_memory_private(unsigned long vaddr, unsigned long paddr,
 					 unsigned long npages)
 {
 	/*
@@ -84,7 +84,7 @@ void __head early_snp_set_memory_private(unsigned long vaddr, unsigned long padd
 			      rip_rel_ptr(&boot_svsm_ca_page), boot_svsm_caa_pa);
 }
 
-void __head early_snp_set_memory_shared(unsigned long vaddr, unsigned long paddr,
+void __init early_snp_set_memory_shared(unsigned long vaddr, unsigned long paddr,
 					unsigned long npages)
 {
 	/*
@@ -114,7 +114,7 @@ void __head early_snp_set_memory_shared(unsigned long vaddr, unsigned long paddr
  *
  * Scan for the blob in that order.
  */
-static __head struct cc_blob_sev_info *find_cc_blob(struct boot_params *bp)
+static struct cc_blob_sev_info *__init find_cc_blob(struct boot_params *bp)
 {
 	struct cc_blob_sev_info *cc_info;
 
@@ -140,7 +140,7 @@ static __head struct cc_blob_sev_info *find_cc_blob(struct boot_params *bp)
 	return cc_info;
 }
 
-static __head void svsm_setup(struct cc_blob_sev_info *cc_info)
+static void __init svsm_setup(struct cc_blob_sev_info *cc_info)
 {
 	struct snp_secrets_page *secrets = (void *)cc_info->secrets_phys;
 	struct svsm_call call = {};
@@ -181,7 +181,7 @@ static __head void svsm_setup(struct cc_blob_sev_info *cc_info)
 	boot_svsm_caa_pa = pa;
 }
 
-bool __head snp_init(struct boot_params *bp)
+bool __init snp_init(struct boot_params *bp)
 {
 	struct cc_blob_sev_info *cc_info;
 
@@ -210,7 +210,7 @@ bool __head snp_init(struct boot_params *bp)
 	return true;
 }
 
-void __head __noreturn snp_abort(void)
+void __init __noreturn snp_abort(void)
 {
 	sev_es_terminate(SEV_TERM_SET_GEN, GHCB_SNP_UNSUPPORTED);
 }
arch/x86/boot/startup/sme.c +13 -13
index eb6a758ba660..39e7e9d18974 100644
@@ -91,7 +91,7 @@ struct sme_populate_pgd_data {
  */
 static char sme_workarea[2 * PMD_SIZE] __section(".init.scratch");
 
-static void __head sme_clear_pgd(struct sme_populate_pgd_data *ppd)
+static void __init sme_clear_pgd(struct sme_populate_pgd_data *ppd)
 {
 	unsigned long pgd_start, pgd_end, pgd_size;
 	pgd_t *pgd_p;
@@ -106,7 +106,7 @@ static void __head sme_clear_pgd(struct sme_populate_pgd_data *ppd)
 	memset(pgd_p, 0, pgd_size);
 }
 
-static pud_t __head *sme_prepare_pgd(struct sme_populate_pgd_data *ppd)
+static pud_t __init *sme_prepare_pgd(struct sme_populate_pgd_data *ppd)
 {
 	pgd_t *pgd;
 	p4d_t *p4d;
@@ -143,7 +143,7 @@ static pud_t __head *sme_prepare_pgd(struct sme_populate_pgd_data *ppd)
 	return pud;
 }
 
-static void __head sme_populate_pgd_large(struct sme_populate_pgd_data *ppd)
+static void __init sme_populate_pgd_large(struct sme_populate_pgd_data *ppd)
 {
 	pud_t *pud;
 	pmd_t *pmd;
@@ -159,7 +159,7 @@ static void __head sme_populate_pgd_large(struct sme_populate_pgd_data *ppd)
 	set_pmd(pmd, __pmd(ppd->paddr | ppd->pmd_flags));
 }
 
-static void __head sme_populate_pgd(struct sme_populate_pgd_data *ppd)
+static void __init sme_populate_pgd(struct sme_populate_pgd_data *ppd)
 {
 	pud_t *pud;
 	pmd_t *pmd;
@@ -185,7 +185,7 @@ static void __head sme_populate_pgd(struct sme_populate_pgd_data *ppd)
 		set_pte(pte, __pte(ppd->paddr | ppd->pte_flags));
 }
 
-static void __head __sme_map_range_pmd(struct sme_populate_pgd_data *ppd)
+static void __init __sme_map_range_pmd(struct sme_populate_pgd_data *ppd)
 {
 	while (ppd->vaddr < ppd->vaddr_end) {
 		sme_populate_pgd_large(ppd);
@@ -195,7 +195,7 @@ static void __head __sme_map_range_pmd(struct sme_populate_pgd_data *ppd)
 	}
 }
 
-static void __head __sme_map_range_pte(struct sme_populate_pgd_data *ppd)
+static void __init __sme_map_range_pte(struct sme_populate_pgd_data *ppd)
 {
 	while (ppd->vaddr < ppd->vaddr_end) {
 		sme_populate_pgd(ppd);
@@ -205,7 +205,7 @@ static void __head __sme_map_range_pte(struct sme_populate_pgd_data *ppd)
 	}
 }
 
-static void __head __sme_map_range(struct sme_populate_pgd_data *ppd,
+static void __init __sme_map_range(struct sme_populate_pgd_data *ppd,
 				   pmdval_t pmd_flags, pteval_t pte_flags)
 {
 	unsigned long vaddr_end;
@@ -229,22 +229,22 @@ static void __head __sme_map_range(struct sme_populate_pgd_data *ppd,
 	__sme_map_range_pte(ppd);
 }
 
-static void __head sme_map_range_encrypted(struct sme_populate_pgd_data *ppd)
+static void __init sme_map_range_encrypted(struct sme_populate_pgd_data *ppd)
 {
 	__sme_map_range(ppd, PMD_FLAGS_ENC, PTE_FLAGS_ENC);
 }
 
-static void __head sme_map_range_decrypted(struct sme_populate_pgd_data *ppd)
+static void __init sme_map_range_decrypted(struct sme_populate_pgd_data *ppd)
 {
 	__sme_map_range(ppd, PMD_FLAGS_DEC, PTE_FLAGS_DEC);
 }
 
-static void __head sme_map_range_decrypted_wp(struct sme_populate_pgd_data *ppd)
+static void __init sme_map_range_decrypted_wp(struct sme_populate_pgd_data *ppd)
 {
 	__sme_map_range(ppd, PMD_FLAGS_DEC_WP, PTE_FLAGS_DEC_WP);
 }
 
-static unsigned long __head sme_pgtable_calc(unsigned long len)
+static unsigned long __init sme_pgtable_calc(unsigned long len)
 {
 	unsigned long entries = 0, tables = 0;
 
@@ -281,7 +281,7 @@ static unsigned long __head sme_pgtable_calc(unsigned long len)
 	return entries + tables;
 }
 
-void __head sme_encrypt_kernel(struct boot_params *bp)
+void __init sme_encrypt_kernel(struct boot_params *bp)
 {
 	unsigned long workarea_start, workarea_end, workarea_len;
 	unsigned long execute_start, execute_end, execute_len;
@@ -485,7 +485,7 @@ void __head sme_encrypt_kernel(struct boot_params *bp)
 	native_write_cr3(__native_read_cr3());
 }
 
-void __head sme_enable(struct boot_params *bp)
+void __init sme_enable(struct boot_params *bp)
 {
 	unsigned int eax, ebx, ecx, edx;
 	unsigned long feature_mask;
arch/x86/include/asm/init.h +0 -6
index 8b1b1abcef15..01ccdd168df0 100644
@@ -2,12 +2,6 @@
 #ifndef _ASM_X86_INIT_H
 #define _ASM_X86_INIT_H
 
-#if defined(CONFIG_CC_IS_CLANG) && CONFIG_CLANG_VERSION < 170000
-#define __head	__section(".head.text") __no_sanitize_undefined __no_stack_protector
-#else
-#define __head	__section(".head.text") __no_sanitize_undefined
-#endif
-
 struct x86_mapping_info {
 	void *(*alloc_pgt_page)(void *); /* allocate buf for page table */
 	void (*free_pgt_page)(void *, void *); /* free buf for page table */
arch/x86/kernel/head_32.S +1 -1
index 76743dfad6ab..437effb1ef03 100644
@@ -61,7 +61,7 @@ RESERVE_BRK(pagetables, INIT_MAP_SIZE)
  * any particular GDT layout, because we load our own as soon as we
  * can.
  */
-__HEAD
+	__INIT
 SYM_CODE_START(startup_32)
 	movl pa(initial_stack),%ecx
 	
arch/x86/kernel/head_64.S +1 -1
index d219963ecb60..21816b48537c 100644
@@ -33,7 +33,7 @@
  * because we need identity-mapped pages.
  */
 
-	__HEAD
+	__INIT
 	.code64
 SYM_CODE_START_NOALIGN(startup_64)
 	UNWIND_HINT_END_OF_STACK
arch/x86/platform/pvh/head.S +1 -1
index 1d78e5631bb8..344030c1a81d 100644
@@ -24,7 +24,7 @@
 #include <asm/nospec-branch.h>
 #include <xen/interface/elfnote.h>
 
-	__HEAD
+	__INIT
 
 /*
  * Entry point for PVH guests.
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
#29 lore raw

PATCH v5 22/22 x86/boot: Get rid of the .head.text section

In reply to Ard Biesheuvel

From: Ard Biesheuvel <[email protected]>

The .head.text section is now empty, so it can be dropped from the linker script.

---
 arch/x86/kernel/vmlinux.lds.S | 5 -----
 1 file changed, 5 deletions(-)
arch/x86/kernel/vmlinux.lds.S +0 -5
index 4277efb26358..d7af4a64c211 100644
@@ -160,11 +160,6 @@ SECTIONS
 
 	} :text = 0xcccccccc
 
-	/* bootstrapping code */
-	.head.text : AT(ADDR(.head.text) - LOAD_OFFSET) {
-		HEAD_TEXT
-	} :text = 0xcccccccc
-
 	/* End of text section, which should occupy whole number of pages */
 	_etext = .;
 	. = ALIGN(PAGE_SIZE);
Signature · 1 line
2.50.0.727.gbf7dc18ff4-goog
#30 lore raw
In reply to Ard Biesheuvel
Quoted 1 line · Ard Biesheuvel

From: Ard Biesheuvel <[email protected]>

Hi Ard,

I tried to apply this to tip/master but ran into conflicts. What commit is the series based on?

Thanks,
Tom

Quoted 144 lines

This series implements a strict separation between startup code and ordinary code, where startup code is built in a way that tolerates being invoked from the initial 1:1 mapping of memory.

The existing approach of emitting this code into .head.text and checking for absolute relocations in that section is not 100% safe, and produces diagnostics that are sometimes difficult to interpret. [0]

Instead, rely on symbol prefixes, similar to how this is implemented for the EFI stub and for the startup code in the arm64 port. This ensures that startup code can only call other startup code, unless a special symbol alias is emitted that exposes a non-startup routine to the startup code.

This is somewhat intrusive, as there are many data objects that are referenced both by startup code and by ordinary code, and an alias needs to be emitted for each of those. If startup code references anything that has not been made available to it explicitly, a build time link error will occur.

This ultimately allows the .head.text section to be dropped entirely, as it no longer has a special significance. Instead, code that only executes at boot is emitted into .init.text as it should.

The majority of changes is around early SEV code. The main issue is that its use of GHCB pages and SVSM calling areas in code that may run from both the 1:1 mapping and the kernel virtual mapping is problematic as it relies on __pa() to perform VA to PA translations, which are ambiguous in this context. Also, __pa() pulls in non-trivial instrumented code when CONFIG_DEBUG_VIRTUAL=y and so it is better to avoid VA to PA translations altogether in the startup code.

Changes since v4:
- Incorporate feedback from Tom, and add a couple of RBs
- Drop patch that moved the MSR save/restore out of the early page state change helper - this is less efficient but likely negligible in practice
- Drop patch that unified the SEV-SNP hypervisor feature check, which was identified by Nikunj as the one breaking SEV-SNP boot.

Changes since RFT/v3:
- Rebase onto tip/master
- Incorporate Borislav's feedback on v3
- Switch to objtool to check for absolute references in startup code
- Remap inittext R-X when running on EFI implementations that require strict R-X/RW- separation
- Include a kbuild fix to incorporate arch/x86/boot/startup/ in the right manner
- For now, omit the LA57 changes that remove the problematic early 5-level paging checks. We can revisit this once there is agreement on the approach.

Changes since RFT/v2:
- Rebase onto tip/x86/boot and drop the patches from the previous revision that have been applied in the meantime.
- Omit the pgtable_l5_enabled() changes for now, and just expose PIC aliases for the variables in question - this can be sorted later.
- Don't use the boot SVSM calling area in snp_kexec_finish(), but pass down the correct per-CPU one to the early page state API.
- Rename arch/x86/coco/sev/sev-noinstr.o to arch/x86/coco/sev/noinstr.o
- Further reduce the amount of SEV code that needs to be constructed in a special way.

Change since RFC/v1:
- Include a major disentanglement/refactor of the SEV-SNP startup code, so that only code that really needs to run from the 1:1 mapping is included in the startup/ code

- Incorporate some early notes from Ingo

Cc Borislav Petkov <[email protected]>
Cc Ingo Molnar <[email protected]>
Cc Kevin Loughlin <[email protected]>
Cc Tom Lendacky <[email protected]>
Cc Josh Poimboeuf <[email protected]>
Cc Peter Zijlstra <[email protected]>
Cc Nikunj A Dadhania <[email protected]>

[0] https://lore.kernel.org/all/CAHk-=wj7k9nvJn6cpa3-5Ciwn2RGyE605BMkjWE4MqnvC9E92A@mail.gmail.com/

Ard Biesheuvel (22):
x86/sev: Separate MSR and GHCB based snp_cpuid() via a callback x86/sev: Use MSR protocol for remapping SVSM calling area x86/sev: Use MSR protocol only for early SVSM PVALIDATE call x86/sev: Run RMPADJUST on SVSM calling area page to test VMPL x86/sev: Move GHCB page based HV communication out of startup code x86/sev: Avoid global variable to store virtual address of SVSM area x86/sev: Share implementation of MSR-based page state change x86/sev: Pass SVSM calling area down to early page state change API x86/sev: Use boot SVSM CA for all startup and init code x86/boot: Drop redundant RMPADJUST in SEV SVSM presence check x86/boot: Provide PIC aliases for 5-level paging related constants x86/sev: Provide PIC aliases for SEV related data objects x86/sev: Move __sev_[get|put]_ghcb() into separate noinstr object x86/sev: Export startup routines for later use
objtool: Add action to check for absence of absolute relocations x86/boot: Check startup code for absence of absolute relocations x86/boot: Revert "Reject absolute references in .head.text" x86/kbuild: Incorporate boot/startup/ via Kbuild makefile x86/boot: Create a confined code area for startup code efistub/x86: Remap inittext read-execute when needed x86/boot: Move startup code out of __head section x86/boot: Get rid of the .head.text section

 arch/x86/Kbuild                            |   2 +
 arch/x86/Makefile                          |   1 -
 arch/x86/boot/compressed/Makefile          |   2 +-
 arch/x86/boot/compressed/misc.c            |   2 +
 arch/x86/boot/compressed/sev-handle-vc.c   |   3 +
 arch/x86/boot/compressed/sev.c             | 108 +------
 arch/x86/boot/startup/Makefile             |  22 ++
 arch/x86/boot/startup/exports.h            |  14 +
 arch/x86/boot/startup/gdt_idt.c            |   4 +-
 arch/x86/boot/startup/map_kernel.c         |   4 +-
 arch/x86/boot/startup/sev-shared.c         | 317 ++++++--------------
 arch/x86/boot/startup/sev-startup.c        | 196 ++----------
 arch/x86/boot/startup/sme.c                |  27 +-
 arch/x86/coco/sev/Makefile                 |   6 +-
 arch/x86/coco/sev/core.c                   | 169 ++++++++---
 arch/x86/coco/sev/{sev-nmi.c => noinstr.c} |  74 +++++
 arch/x86/coco/sev/vc-handle.c              |   2 +
 arch/x86/coco/sev/vc-shared.c              | 143 ++++++++-
 arch/x86/include/asm/boot.h                |   2 +
 arch/x86/include/asm/init.h                |   6 -
 arch/x86/include/asm/setup.h               |   1 +
 arch/x86/include/asm/sev-internal.h        |  27 +-
 arch/x86/include/asm/sev.h                 |  17 +-
 arch/x86/kernel/head64.c                   |   5 +-
 arch/x86/kernel/head_32.S                  |   2 +-
 arch/x86/kernel/head_64.S                  |  10 +-
 arch/x86/kernel/vmlinux.lds.S              |   9 +-
 arch/x86/mm/mem_encrypt_amd.c              |   6 -
 arch/x86/mm/mem_encrypt_boot.S             |   6 +-
 arch/x86/platform/pvh/head.S               |   2 +-
 arch/x86/tools/relocs.c                    |   8 +-
 drivers/firmware/efi/libstub/x86-stub.c    |   4 +-
 tools/objtool/builtin-check.c              |   2 +
 tools/objtool/check.c                      |  39 ++-
 tools/objtool/include/objtool/builtin.h    |   1 +
 35 files changed, 620 insertions(+), 623 deletions(-)
 create mode 100644 arch/x86/boot/startup/exports.h
 rename arch/x86/coco/sev/{sev-nmi.c => noinstr.c} (61%)