Skip to contentSkip to conversation

6.1.81-rc1 review

PATCH In review Greg Kroah-Hartman 13 participants 230 messages Mar 4, 2024 – Mar 6, 2024 Page 2 of 2

Subsystems (6)
lore
Share & tools
lore
author maintainer
#137 lore raw

PATCH 133/215 x86/decompressor: Use standard calling convention for trampoline

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ard Biesheuvel <[email protected]>

commit 918a7a04e71745e99a0efc6753e587439b794b29 upstream.

Update the trampoline code so its arguments are passed via RDI and RSI, which matches the ordinary SysV calling convention for x86_64. This will allow this code to be called directly from C.

---
 arch/x86/boot/compressed/head_64.S |   27 +++++++++++++--------------
 arch/x86/boot/compressed/pgtable.h |    2 +-
 2 files changed, 14 insertions(+), 15 deletions(-)
arch/x86/boot/compressed/head_64.S +13 -14
@@ -447,9 +447,9 @@ SYM_CODE_START(startup_64)
 	movq	%r15, %rdi
 	call	paging_prepare
 
-	/* Save the trampoline address in RCX */
-	movq	%rax, %rcx
-
+	/* Pass the trampoline address and boolean flag as args #1 and #2 */
+	movq	%rax, %rdi
+	movq	%rdx, %rsi
 	leaq	TRAMPOLINE_32BIT_CODE_OFFSET(%rax), %rax
 	call	*%rax
 
@@ -549,11 +549,14 @@ SYM_FUNC_START_LOCAL_NOALIGN(.Lrelocated
 SYM_FUNC_END(.Lrelocated)
 
 /*
- * This is the 32-bit trampoline that will be copied over to low memory.
+ * This is the 32-bit trampoline that will be copied over to low memory. It
+ * will be called using the ordinary 64-bit calling convention from code
+ * running in 64-bit mode.
  *
  * Return address is at the top of the stack (might be above 4G).
- * ECX contains the base address of the trampoline memory.
- * Non zero RDX means trampoline needs to enable 5-level paging.
+ * The first argument (EDI) contains the 32-bit addressable base of the
+ * trampoline memory. A non-zero second argument (ESI) means that the
+ * trampoline needs to enable 5-level paging.
  */
 SYM_CODE_START(trampoline_32bit_src)
 	/*
@@ -600,7 +603,7 @@ SYM_CODE_START(trampoline_32bit_src)
 	movl	%eax, %cr0
 
 	/* Check what paging mode we want to be in after the trampoline */
-	testl	%edx, %edx
+	testl	%esi, %esi
 	jz	1f
 
 	/* We want 5-level paging: don't touch CR3 if it already points to 5-level page tables */
@@ -615,21 +618,17 @@ SYM_CODE_START(trampoline_32bit_src)
 	jz	3f
 2:
 	/* Point CR3 to the trampoline's new top level page table */
-	leal	TRAMPOLINE_32BIT_PGTABLE_OFFSET(%ecx), %eax
+	leal	TRAMPOLINE_32BIT_PGTABLE_OFFSET(%edi), %eax
 	movl	%eax, %cr3
 3:
 	/* Set EFER.LME=1 as a precaution in case hypervsior pulls the rug */
-	pushl	%ecx
-	pushl	%edx
 	movl	$MSR_EFER, %ecx
 	rdmsr
 	btsl	$_EFER_LME, %eax
 	/* Avoid writing EFER if no change was made (for TDX guest) */
 	jc	1f
 	wrmsr
-1:	popl	%edx
-	popl	%ecx
-
+1:
 #ifdef CONFIG_X86_MCE
 	/*
 	 * Preserve CR4.MCE if the kernel will enable #MC support.
@@ -646,7 +645,7 @@ SYM_CODE_START(trampoline_32bit_src)
 
 	/* Enable PAE and LA57 (if required) paging modes */
 	orl	$X86_CR4_PAE, %eax
-	testl	%edx, %edx
+	testl	%esi, %esi
 	jz	1f
 	orl	$X86_CR4_LA57, %eax
 1:
arch/x86/boot/compressed/pgtable.h +1 -1
@@ -14,7 +14,7 @@
 
 extern unsigned long *trampoline_32bit;
 
-extern void trampoline_32bit_src(void *return_ptr);
+extern void trampoline_32bit_src(void *trampoline, bool enable_5lvl);
 
 #endif /* __ASSEMBLER__ */
 #endif /* BOOT_COMPRESSED_PAGETABLE_H */
author maintainer
#138 lore raw

PATCH 134/215 x86/decompressor: Avoid the need for a stack in the 32-bit trampoline

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ard Biesheuvel <[email protected]>

commit bd328aa01ff77a45aeffea5fc4521854291db11f upstream.

The 32-bit trampoline no longer uses the stack for anything except performing a far return back to long mode, and preserving the caller's stack pointer value. Currently, the trampoline stack is placed in the same page that carries the trampoline code, which means this page must be mapped writable and executable, and the stack is therefore executable as well.

Replace the far return with a far jump, so that the return address can be pre-calculated and patched into the code before it is called. This removes the need for a 32-bit addressable stack entirely, and in a later patch, this will be taken advantage of by removing writable permissions from (and adding executable permissions to) the trampoline code page when booting via the EFI stub.

Note that the value of RSP still needs to be preserved explicitly across the switch into 32-bit mode, as the register may get truncated to 32 bits.

---
 arch/x86/boot/compressed/head_64.S    |   45 ++++++++++++++++++++--------------
 arch/x86/boot/compressed/pgtable.h    |    4 +--
 arch/x86/boot/compressed/pgtable_64.c |   12 ++++++++-
 3 files changed, 40 insertions(+), 21 deletions(-)
arch/x86/boot/compressed/head_64.S +27 -18
@@ -558,6 +558,7 @@ SYM_FUNC_END(.Lrelocated)
  * trampoline memory. A non-zero second argument (ESI) means that the
  * trampoline needs to enable 5-level paging.
  */
+	.section ".rodata", "a", @progbits
 SYM_CODE_START(trampoline_32bit_src)
 	/*
 	 * Preserve live 64-bit registers on the stack: this is necessary
@@ -568,13 +569,9 @@ SYM_CODE_START(trampoline_32bit_src)
 	pushq	%rbp
 	pushq	%rbx
 
-	/* Set up 32-bit addressable stack and push the old RSP value */
-	leaq	(TRAMPOLINE_32BIT_STACK_END - 8)(%rcx), %rbx
-	movq	%rsp, (%rbx)
-	movq	%rbx, %rsp
-
-	/* Take the address of the trampoline exit code */
-	leaq	.Lret(%rip), %rbx
+	/* Preserve top half of RSP in a legacy mode GPR to avoid truncation */
+	movq	%rsp, %rbx
+	shrq	$32, %rbx
 
 	/* Switch to compatibility mode (CS.L = 0 CS.D = 1) via far return */
 	pushq	$__KERNEL32_CS
@@ -582,9 +579,17 @@ SYM_CODE_START(trampoline_32bit_src)
 	pushq	%rax
 	lretq
 
+	/*
+	 * The 32-bit code below will do a far jump back to long mode and end
+	 * up here after reconfiguring the number of paging levels. First, the
+	 * stack pointer needs to be restored to its full 64-bit value before
+	 * the callee save register contents can be popped from the stack.
+	 */
 .Lret:
+	shlq	$32, %rbx
+	orq	%rbx, %rsp
+
 	/* Restore the preserved 64-bit registers */
-	movq	(%rsp), %rsp
 	popq	%rbx
 	popq	%rbp
 	popq	%r15
@@ -592,11 +597,6 @@ SYM_CODE_START(trampoline_32bit_src)
 
 	.code32
 0:
-	/* Set up data and stack segments */
-	movl	$__KERNEL_DS, %eax
-	movl	%eax, %ds
-	movl	%eax, %ss
-
 	/* Disable paging */
 	movl	%cr0, %eax
 	btrl	$X86_CR0_PG_BIT, %eax
@@ -651,18 +651,26 @@ SYM_CODE_START(trampoline_32bit_src)
 1:
 	movl	%eax, %cr4
 
-	/* Prepare the stack for far return to Long Mode */
-	pushl	$__KERNEL_CS
-	pushl	%ebx
-
 	/* Enable paging again. */
 	movl	%cr0, %eax
 	btsl	$X86_CR0_PG_BIT, %eax
 	movl	%eax, %cr0
 
-	lret
+	/*
+	 * Return to the 64-bit calling code using LJMP rather than LRET, to
+	 * avoid the need for a 32-bit addressable stack. The destination
+	 * address will be adjusted after the template code is copied into a
+	 * 32-bit addressable buffer.
+	 */
+.Ljmp:	ljmpl	$__KERNEL_CS, $(.Lret - trampoline_32bit_src)
 SYM_CODE_END(trampoline_32bit_src)
 
+/*
+ * This symbol is placed right after trampoline_32bit_src() so its address can
+ * be used to infer the size of the trampoline code.
+ */
+SYM_DATA(trampoline_ljmp_imm_offset, .word  .Ljmp + 1 - trampoline_32bit_src)
+
 	/*
          * The trampoline code has a size limit.
          * Make sure we fail to compile if the trampoline code grows
@@ -670,6 +678,7 @@ SYM_CODE_END(trampoline_32bit_src)
 	 */
 	.org	trampoline_32bit_src + TRAMPOLINE_32BIT_CODE_SIZE
 
+	.text
 SYM_FUNC_START_LOCAL_NOALIGN(.Lno_longmode)
 	/* This isn't an x86-64 CPU, so hang intentionally, we cannot continue */
 1:
arch/x86/boot/compressed/pgtable.h +2 -2
@@ -8,13 +8,13 @@
 #define TRAMPOLINE_32BIT_CODE_OFFSET	PAGE_SIZE
 #define TRAMPOLINE_32BIT_CODE_SIZE	0xA0
 
-#define TRAMPOLINE_32BIT_STACK_END	TRAMPOLINE_32BIT_SIZE
-
 #ifndef __ASSEMBLER__
 
 extern unsigned long *trampoline_32bit;
 
 extern void trampoline_32bit_src(void *trampoline, bool enable_5lvl);
 
+extern const u16 trampoline_ljmp_imm_offset;
+
 #endif /* __ASSEMBLER__ */
 #endif /* BOOT_COMPRESSED_PAGETABLE_H */
arch/x86/boot/compressed/pgtable_64.c +11 -1
@@ -109,6 +109,7 @@ static unsigned long find_trampoline_pla
 struct paging_config paging_prepare(void *rmode)
 {
 	struct paging_config paging_config = {};
+	void *tramp_code;
 
 	/* Initialize boot_params. Required for cmdline_find_option_bool(). */
 	boot_params = rmode;
@@ -148,10 +149,19 @@ struct paging_config paging_prepare(void
 	memset(trampoline_32bit, 0, TRAMPOLINE_32BIT_SIZE);
 
 	/* Copy trampoline code in place */
-	memcpy(trampoline_32bit + TRAMPOLINE_32BIT_CODE_OFFSET / sizeof(unsigned long),
+	tramp_code = memcpy(trampoline_32bit +
+			TRAMPOLINE_32BIT_CODE_OFFSET / sizeof(unsigned long),
 			&trampoline_32bit_src, TRAMPOLINE_32BIT_CODE_SIZE);
 
 	/*
+	 * Avoid the need for a stack in the 32-bit trampoline code, by using
+	 * LJMP rather than LRET to return back to long mode. LJMP takes an
+	 * immediate absolute address, which needs to be adjusted based on the
+	 * placement of the trampoline.
+	 */
+	*(u32 *)(tramp_code + trampoline_ljmp_imm_offset) += (unsigned long)tramp_code;
+
+	/*
 	 * The code below prepares page table in trampoline memory.
 	 *
 	 * The new page table will be used by trampoline code for switching
author maintainer
#139 lore raw

PATCH 135/215 x86/decompressor: Call trampoline directly from C code

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ard Biesheuvel <[email protected]>

commit 64ef578b6b6866bec012544416946533444036c8 upstream.

Instead of returning to the asm calling code to invoke the trampoline, call it straight from the C code that sets it up. That way, the struct return type is no longer needed for returning two values, and the call can be made conditional more cleanly in a subsequent patch.

This means that all callee save 64-bit registers need to be preserved and restored, as their contents may not survive the legacy mode switch.

---
 arch/x86/boot/compressed/head_64.S    |   31 +++++++++++++------------------
 arch/x86/boot/compressed/pgtable_64.c |   32 +++++++++++++-------------------
 2 files changed, 26 insertions(+), 37 deletions(-)
arch/x86/boot/compressed/head_64.S +13 -18
@@ -433,25 +433,14 @@ SYM_CODE_START(startup_64)
 #endif
 
 	/*
-	 * paging_prepare() sets up the trampoline and checks if we need to
-	 * enable 5-level paging.
-	 *
-	 * paging_prepare() returns a two-quadword structure which lands
-	 * into RDX:RAX:
-	 *   - Address of the trampoline is returned in RAX.
-	 *   - Non zero RDX means trampoline needs to enable 5-level
-	 *     paging.
+	 * configure_5level_paging() updates the number of paging levels using
+	 * a trampoline in 32-bit addressable memory if the current number does
+	 * not match the desired number.
 	 *
 	 * Pass the boot_params pointer as the first argument.
 	 */
 	movq	%r15, %rdi
-	call	paging_prepare
-
-	/* Pass the trampoline address and boolean flag as args #1 and #2 */
-	movq	%rax, %rdi
-	movq	%rdx, %rsi
-	leaq	TRAMPOLINE_32BIT_CODE_OFFSET(%rax), %rax
-	call	*%rax
+	call	configure_5level_paging
 
 	/*
 	 * cleanup_trampoline() would restore trampoline memory.
@@ -561,11 +550,14 @@ SYM_FUNC_END(.Lrelocated)
 	.section ".rodata", "a", @progbits
 SYM_CODE_START(trampoline_32bit_src)
 	/*
-	 * Preserve live 64-bit registers on the stack: this is necessary
-	 * because the architecture does not guarantee that GPRs will retain
-	 * their full 64-bit values across a 32-bit mode switch.
+	 * Preserve callee save 64-bit registers on the stack: this is
+	 * necessary because the architecture does not guarantee that GPRs will
+	 * retain their full 64-bit values across a 32-bit mode switch.
 	 */
 	pushq	%r15
+	pushq	%r14
+	pushq	%r13
+	pushq	%r12
 	pushq	%rbp
 	pushq	%rbx
 
@@ -592,6 +584,9 @@ SYM_CODE_START(trampoline_32bit_src)
 	/* Restore the preserved 64-bit registers */
 	popq	%rbx
 	popq	%rbp
+	popq	%r12
+	popq	%r13
+	popq	%r14
 	popq	%r15
 	retq
 
arch/x86/boot/compressed/pgtable_64.c +13 -19
@@ -16,11 +16,6 @@ unsigned int __section(".data") pgdir_sh
 unsigned int __section(".data") ptrs_per_p4d = 1;
 #endif
 
-struct paging_config {
-	unsigned long trampoline_start;
-	unsigned long l5_required;
-};
-
 /* Buffer to preserve trampoline memory */
 static char trampoline_save[TRAMPOLINE_32BIT_SIZE];
 
@@ -29,7 +24,7 @@ static char trampoline_save[TRAMPOLINE_3
  * purposes.
  *
  * Avoid putting the pointer into .bss as it will be cleared between
- * paging_prepare() and extract_kernel().
+ * configure_5level_paging() and extract_kernel().
  */
 unsigned long *trampoline_32bit __section(".data");
 
@@ -106,13 +101,13 @@ static unsigned long find_trampoline_pla
 	return bios_start - TRAMPOLINE_32BIT_SIZE;
 }
 
-struct paging_config paging_prepare(void *rmode)
+asmlinkage void configure_5level_paging(struct boot_params *bp)
 {
-	struct paging_config paging_config = {};
-	void *tramp_code;
+	void (*toggle_la57)(void *trampoline, bool enable_5lvl);
+	bool l5_required = false;
 
 	/* Initialize boot_params. Required for cmdline_find_option_bool(). */
-	boot_params = rmode;
+	boot_params = bp;
 
 	/*
 	 * Check if LA57 is desired and supported.
@@ -130,7 +125,7 @@ struct paging_config paging_prepare(void
 			!cmdline_find_option_bool("no5lvl") &&
 			native_cpuid_eax(0) >= 7 &&
 			(native_cpuid_ecx(7) & (1 << (X86_FEATURE_LA57 & 31)))) {
-		paging_config.l5_required = 1;
+		l5_required = true;
 
 		/* Initialize variables for 5-level paging */
 		__pgtable_l5_enabled = 1;
@@ -138,9 +133,7 @@ struct paging_config paging_prepare(void
 		ptrs_per_p4d = 512;
 	}
 
-	paging_config.trampoline_start = find_trampoline_placement();
-
-	trampoline_32bit = (unsigned long *)paging_config.trampoline_start;
+	trampoline_32bit = (unsigned long *)find_trampoline_placement();
 
 	/* Preserve trampoline memory */
 	memcpy(trampoline_save, trampoline_32bit, TRAMPOLINE_32BIT_SIZE);
@@ -149,7 +142,7 @@ struct paging_config paging_prepare(void
 	memset(trampoline_32bit, 0, TRAMPOLINE_32BIT_SIZE);
 
 	/* Copy trampoline code in place */
-	tramp_code = memcpy(trampoline_32bit +
+	toggle_la57 = memcpy(trampoline_32bit +
 			TRAMPOLINE_32BIT_CODE_OFFSET / sizeof(unsigned long),
 			&trampoline_32bit_src, TRAMPOLINE_32BIT_CODE_SIZE);
 
@@ -159,7 +152,8 @@ struct paging_config paging_prepare(void
 	 * immediate absolute address, which needs to be adjusted based on the
 	 * placement of the trampoline.
 	 */
-	*(u32 *)(tramp_code + trampoline_ljmp_imm_offset) += (unsigned long)tramp_code;
+	*(u32 *)((u8 *)toggle_la57 + trampoline_ljmp_imm_offset) +=
+						(unsigned long)toggle_la57;
 
 	/*
 	 * The code below prepares page table in trampoline memory.
@@ -175,10 +169,10 @@ struct paging_config paging_prepare(void
 	 * We are not going to use the page table in trampoline memory if we
 	 * are already in the desired paging mode.
 	 */
-	if (paging_config.l5_required == !!(native_read_cr4() & X86_CR4_LA57))
+	if (l5_required == !!(native_read_cr4() & X86_CR4_LA57))
 		goto out;
 
-	if (paging_config.l5_required) {
+	if (l5_required) {
 		/*
 		 * For 4- to 5-level paging transition, set up current CR3 as
 		 * the first and the only entry in a new top-level page table.
@@ -201,7 +195,7 @@ struct paging_config paging_prepare(void
 	}
 
 out:
-	return paging_config;
+	toggle_la57(trampoline_32bit, l5_required);
 }
 
 void cleanup_trampoline(void *pgtable)
author maintainer
#140 lore raw

PATCH 136/215 x86/decompressor: Only call the trampoline when changing paging levels

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ard Biesheuvel <[email protected]>

commit f97b67a773cd84bd8b55c0a0ec32448a87fc56bb upstream.

Since the current and desired number of paging levels are known when the trampoline is being prepared, avoid calling the trampoline at all if it is clear that calling it is not going to result in a change to the number of paging levels.

Given that the CPU is already running in long mode, the PAE and LA57 settings are necessarily consistent with the currently active page tables, and other fields in CR4 will be initialized by the startup code in the kernel proper. So limit the manipulation of CR4 to toggling the LA57 bit, which is the only thing that really needs doing at this point in the boot. This also means that there is no need to pass the value of l5_required to toggle_la57(), as it will not be called unless CR4.LA57 needs to toggle.

---
 arch/x86/boot/compressed/head_64.S    |   45 +++-------------------------------
 arch/x86/boot/compressed/pgtable_64.c |   22 ++++++----------
 2 files changed, 13 insertions(+), 54 deletions(-)
arch/x86/boot/compressed/head_64.S +4 -41
@@ -390,10 +390,6 @@ SYM_CODE_START(startup_64)
 	 * For the trampoline, we need the top page table to reside in lower
 	 * memory as we don't have a way to load 64-bit values into CR3 in
 	 * 32-bit mode.
-	 *
-	 * We go though the trampoline even if we don't have to: if we're
-	 * already in a desired paging mode. This way the trampoline code gets
-	 * tested on every boot.
 	 */
 
 	/* Make sure we have GDT with 32-bit code segment */
@@ -544,8 +540,7 @@ SYM_FUNC_END(.Lrelocated)
  *
  * Return address is at the top of the stack (might be above 4G).
  * The first argument (EDI) contains the 32-bit addressable base of the
- * trampoline memory. A non-zero second argument (ESI) means that the
- * trampoline needs to enable 5-level paging.
+ * trampoline memory.
  */
 	.section ".rodata", "a", @progbits
 SYM_CODE_START(trampoline_32bit_src)
@@ -597,25 +592,10 @@ SYM_CODE_START(trampoline_32bit_src)
 	btrl	$X86_CR0_PG_BIT, %eax
 	movl	%eax, %cr0
 
-	/* Check what paging mode we want to be in after the trampoline */
-	testl	%esi, %esi
-	jz	1f
-
-	/* We want 5-level paging: don't touch CR3 if it already points to 5-level page tables */
-	movl	%cr4, %eax
-	testl	$X86_CR4_LA57, %eax
-	jnz	3f
-	jmp	2f
-1:
-	/* We want 4-level paging: don't touch CR3 if it already points to 4-level page tables */
-	movl	%cr4, %eax
-	testl	$X86_CR4_LA57, %eax
-	jz	3f
-2:
 	/* Point CR3 to the trampoline's new top level page table */
 	leal	TRAMPOLINE_32BIT_PGTABLE_OFFSET(%edi), %eax
 	movl	%eax, %cr3
-3:
+
 	/* Set EFER.LME=1 as a precaution in case hypervsior pulls the rug */
 	movl	$MSR_EFER, %ecx
 	rdmsr
@@ -624,26 +604,9 @@ SYM_CODE_START(trampoline_32bit_src)
 	jc	1f
 	wrmsr
 1:
-#ifdef CONFIG_X86_MCE
-	/*
-	 * Preserve CR4.MCE if the kernel will enable #MC support.
-	 * Clearing MCE may fault in some environments (that also force #MC
-	 * support). Any machine check that occurs before #MC support is fully
-	 * configured will crash the system regardless of the CR4.MCE value set
-	 * here.
-	 */
+	/* Toggle CR4.LA57 */
 	movl	%cr4, %eax
-	andl	$X86_CR4_MCE, %eax
-#else
-	movl	$0, %eax
-#endif
-
-	/* Enable PAE and LA57 (if required) paging modes */
-	orl	$X86_CR4_PAE, %eax
-	testl	%esi, %esi
-	jz	1f
-	orl	$X86_CR4_LA57, %eax
-1:
+	btcl	$X86_CR4_LA57_BIT, %eax
 	movl	%eax, %cr4
 
 	/* Enable paging again. */
arch/x86/boot/compressed/pgtable_64.c +9 -13
@@ -103,7 +103,7 @@ static unsigned long find_trampoline_pla
 
 asmlinkage void configure_5level_paging(struct boot_params *bp)
 {
-	void (*toggle_la57)(void *trampoline, bool enable_5lvl);
+	void (*toggle_la57)(void *trampoline);
 	bool l5_required = false;
 
 	/* Initialize boot_params. Required for cmdline_find_option_bool(). */
@@ -133,6 +133,13 @@ asmlinkage void configure_5level_paging(
 		ptrs_per_p4d = 512;
 	}
 
+	/*
+	 * The trampoline will not be used if the paging mode is already set to
+	 * the desired one.
+	 */
+	if (l5_required == !!(native_read_cr4() & X86_CR4_LA57))
+		return;
+
 	trampoline_32bit = (unsigned long *)find_trampoline_placement();
 
 	/* Preserve trampoline memory */
@@ -160,18 +167,8 @@ asmlinkage void configure_5level_paging(
 	 *
 	 * The new page table will be used by trampoline code for switching
 	 * from 4- to 5-level paging or vice versa.
-	 *
-	 * If switching is not required, the page table is unused: trampoline
-	 * code wouldn't touch CR3.
 	 */
 
-	/*
-	 * We are not going to use the page table in trampoline memory if we
-	 * are already in the desired paging mode.
-	 */
-	if (l5_required == !!(native_read_cr4() & X86_CR4_LA57))
-		goto out;
-
 	if (l5_required) {
 		/*
 		 * For 4- to 5-level paging transition, set up current CR3 as
@@ -194,8 +191,7 @@ asmlinkage void configure_5level_paging(
 		       (void *)src, PAGE_SIZE);
 	}
 
-out:
-	toggle_la57(trampoline_32bit, l5_required);
+	toggle_la57(trampoline_32bit);
 }
 
 void cleanup_trampoline(void *pgtable)
author maintainer
#141 lore raw

PATCH 137/215 x86/decompressor: Pass pgtable address to trampoline directly

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ard Biesheuvel <[email protected]>

commit cb83cece57e1889109dd73ea08ee338668c9d1b8 upstream.

The only remaining use of the trampoline address by the trampoline itself is deriving the page table address from it, and this involves adding an offset of 0x0. So simplify this, and pass the new CR3 value directly.

This makes the fact that the page table happens to be at the start of the trampoline allocation an implementation detail of the caller.

---
 arch/x86/boot/compressed/head_64.S    |    8 ++++----
 arch/x86/boot/compressed/pgtable.h    |    2 --
 arch/x86/boot/compressed/pgtable_64.c |    9 ++++-----
 3 files changed, 8 insertions(+), 11 deletions(-)
arch/x86/boot/compressed/head_64.S +4 -4
@@ -539,8 +539,9 @@ SYM_FUNC_END(.Lrelocated)
  * running in 64-bit mode.
  *
  * Return address is at the top of the stack (might be above 4G).
- * The first argument (EDI) contains the 32-bit addressable base of the
- * trampoline memory.
+ * The first argument (EDI) contains the address of the temporary PGD level
+ * page table in 32-bit addressable memory which will be programmed into
+ * register CR3.
  */
 	.section ".rodata", "a", @progbits
 SYM_CODE_START(trampoline_32bit_src)
@@ -593,8 +594,7 @@ SYM_CODE_START(trampoline_32bit_src)
 	movl	%eax, %cr0
 
 	/* Point CR3 to the trampoline's new top level page table */
-	leal	TRAMPOLINE_32BIT_PGTABLE_OFFSET(%edi), %eax
-	movl	%eax, %cr3
+	movl	%edi, %cr3
 
 	/* Set EFER.LME=1 as a precaution in case hypervsior pulls the rug */
 	movl	$MSR_EFER, %ecx
arch/x86/boot/compressed/pgtable.h +0 -2
@@ -3,8 +3,6 @@
 
 #define TRAMPOLINE_32BIT_SIZE		(2 * PAGE_SIZE)
 
-#define TRAMPOLINE_32BIT_PGTABLE_OFFSET	0
-
 #define TRAMPOLINE_32BIT_CODE_OFFSET	PAGE_SIZE
 #define TRAMPOLINE_32BIT_CODE_SIZE	0xA0
 
arch/x86/boot/compressed/pgtable_64.c +4 -5
@@ -103,7 +103,7 @@ static unsigned long find_trampoline_pla
 
 asmlinkage void configure_5level_paging(struct boot_params *bp)
 {
-	void (*toggle_la57)(void *trampoline);
+	void (*toggle_la57)(void *cr3);
 	bool l5_required = false;
 
 	/* Initialize boot_params. Required for cmdline_find_option_bool(). */
@@ -174,7 +174,7 @@ asmlinkage void configure_5level_paging(
 		 * For 4- to 5-level paging transition, set up current CR3 as
 		 * the first and the only entry in a new top-level page table.
 		 */
-		trampoline_32bit[TRAMPOLINE_32BIT_PGTABLE_OFFSET] = __native_read_cr3() | _PAGE_TABLE_NOENC;
+		*trampoline_32bit = __native_read_cr3() | _PAGE_TABLE_NOENC;
 	} else {
 		unsigned long src;
 
@@ -187,8 +187,7 @@ asmlinkage void configure_5level_paging(
 		 * may be above 4G.
 		 */
 		src = *(unsigned long *)__native_read_cr3() & PAGE_MASK;
-		memcpy(trampoline_32bit + TRAMPOLINE_32BIT_PGTABLE_OFFSET / sizeof(unsigned long),
-		       (void *)src, PAGE_SIZE);
+		memcpy(trampoline_32bit, (void *)src, PAGE_SIZE);
 	}
 
 	toggle_la57(trampoline_32bit);
@@ -198,7 +197,7 @@ void cleanup_trampoline(void *pgtable)
 {
 	void *trampoline_pgtable;
 
-	trampoline_pgtable = trampoline_32bit + TRAMPOLINE_32BIT_PGTABLE_OFFSET / sizeof(unsigned long);
+	trampoline_pgtable = trampoline_32bit;
 
 	/*
 	 * Move the top level page table out of trampoline memory,
author maintainer
#142 lore raw

PATCH 138/215 x86/decompressor: Merge trampoline cleanup with switching code

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ard Biesheuvel <[email protected]>

commit 03dda95137d3247564854ad9032c0354273a159d upstream.

Now that the trampoline setup code and the actual invocation of it are all done from the C routine, the trampoline cleanup can be merged into it as well, instead of returning to asm just to call another C function.

---
 arch/x86/boot/compressed/head_64.S    |   14 ++++----------
 arch/x86/boot/compressed/pgtable_64.c |   18 ++++--------------
 2 files changed, 8 insertions(+), 24 deletions(-)
arch/x86/boot/compressed/head_64.S +4 -10
@@ -433,20 +433,14 @@ SYM_CODE_START(startup_64)
 	 * a trampoline in 32-bit addressable memory if the current number does
 	 * not match the desired number.
 	 *
-	 * Pass the boot_params pointer as the first argument.
+	 * Pass the boot_params pointer as the first argument. The second
+	 * argument is the relocated address of the page table to use instead
+	 * of the page table in trampoline memory (if required).
 	 */
 	movq	%r15, %rdi
+	leaq	rva(top_pgtable)(%rbx), %rsi
 	call	configure_5level_paging
 
-	/*
-	 * cleanup_trampoline() would restore trampoline memory.
-	 *
-	 * RDI is address of the page table to use instead of page table
-	 * in trampoline memory (if required).
-	 */
-	leaq	rva(top_pgtable)(%rbx), %rdi
-	call	cleanup_trampoline
-
 	/* Zero EFLAGS */
 	pushq	$0
 	popfq
arch/x86/boot/compressed/pgtable_64.c +4 -14
@@ -101,7 +101,7 @@ static unsigned long find_trampoline_pla
 	return bios_start - TRAMPOLINE_32BIT_SIZE;
 }
 
-asmlinkage void configure_5level_paging(struct boot_params *bp)
+asmlinkage void configure_5level_paging(struct boot_params *bp, void *pgtable)
 {
 	void (*toggle_la57)(void *cr3);
 	bool l5_required = false;
@@ -191,22 +191,12 @@ asmlinkage void configure_5level_paging(
 	}
 
 	toggle_la57(trampoline_32bit);
-}
-
-void cleanup_trampoline(void *pgtable)
-{
-	void *trampoline_pgtable;
-
-	trampoline_pgtable = trampoline_32bit;
 
 	/*
-	 * Move the top level page table out of trampoline memory,
-	 * if it's there.
+	 * Move the top level page table out of trampoline memory.
 	 */
-	if ((void *)__native_read_cr3() == trampoline_pgtable) {
-		memcpy(pgtable, trampoline_pgtable, PAGE_SIZE);
-		native_write_cr3((unsigned long)pgtable);
-	}
+	memcpy(pgtable, trampoline_32bit, PAGE_SIZE);
+	native_write_cr3((unsigned long)pgtable);
 
 	/* Restore trampoline memory */
 	memcpy(trampoline_32bit, trampoline_save, TRAMPOLINE_32BIT_SIZE);
author maintainer
#143 lore raw

PATCH 139/215 x86/decompressor: Move global symbol references to C code

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ard Biesheuvel <[email protected]>

commit 24388292e2d7fae79a0d4183cc91716b851299cf upstream.

It is no longer necessary to be cautious when referring to global variables in the position independent decompressor code, now that it is built using PIE codegen and makes an assertion in the linker script that no GOT entries exist (which would require adjustment for the actual runtime load address of the decompressor binary).

This means global variables can be referenced directly from C code, instead of having to pass their runtime addresses into C routines from asm code, which needs to happen at each call site. Do so for the code that will be called directly from the EFI stub after a subsequent patch, and avoid the need to duplicate this logic a third time.

---
 arch/x86/boot/compressed/head_32.S |    8 --------
 arch/x86/boot/compressed/head_64.S |   10 ++--------
 arch/x86/boot/compressed/misc.c    |   16 +++++++++-------
 3 files changed, 11 insertions(+), 23 deletions(-)
arch/x86/boot/compressed/head_32.S +0 -8
@@ -179,13 +179,7 @@ SYM_FUNC_START_LOCAL_NOALIGN(.Lrelocated
  */
 	/* push arguments for extract_kernel: */
 
-	pushl	output_len@GOTOFF(%ebx)	/* decompressed length, end of relocs */
 	pushl	%ebp			/* output address */
-	pushl	input_len@GOTOFF(%ebx)	/* input_len */
-	leal	input_data@GOTOFF(%ebx), %eax
-	pushl	%eax			/* input_data */
-	leal	boot_heap@GOTOFF(%ebx), %eax
-	pushl	%eax			/* heap area */
 	pushl	%esi			/* real mode pointer */
 	call	extract_kernel		/* returns kernel entry point in %eax */
 	addl	$24, %esp
@@ -213,8 +207,6 @@ SYM_DATA_END_LABEL(gdt, SYM_L_LOCAL, gdt
  */
 	.bss
 	.balign 4
-boot_heap:
-	.fill BOOT_HEAP_SIZE, 1, 0
 boot_stack:
 	.fill BOOT_STACK_SIZE, 1, 0
 boot_stack_end:
arch/x86/boot/compressed/head_64.S +2 -8
@@ -511,13 +511,9 @@ SYM_FUNC_START_LOCAL_NOALIGN(.Lrelocated
 /*
  * Do the extraction, and jump to the new kernel..
  */
-	/* pass struct boot_params pointer */
+	/* pass struct boot_params pointer and output target address */
 	movq	%r15, %rdi
-	leaq	boot_heap(%rip), %rsi	/* malloc area for uncompression */
-	leaq	input_data(%rip), %rdx  /* input_data */
-	movl	input_len(%rip), %ecx	/* input_len */
-	movq	%rbp, %r8		/* output target address */
-	movl	output_len(%rip), %r9d	/* decompressed length, end of relocs */
+	movq	%rbp, %rsi
 	call	extract_kernel		/* returns kernel entry point in %rax */
 
 /*
@@ -675,8 +671,6 @@ SYM_DATA_END_LABEL(boot_idt, SYM_L_GLOBA
  */
 	.bss
 	.balign 4
-SYM_DATA_LOCAL(boot_heap,	.fill BOOT_HEAP_SIZE, 1, 0)
-
 SYM_DATA_START_LOCAL(boot_stack)
 	.fill BOOT_STACK_SIZE, 1, 0
 	.balign 16
arch/x86/boot/compressed/misc.c +9 -7
@@ -330,6 +330,11 @@ static size_t parse_elf(void *output)
 	return ehdr.e_entry - LOAD_PHYSICAL_ADDR;
 }
 
+static u8 boot_heap[BOOT_HEAP_SIZE] __aligned(4);
+
+extern unsigned char input_data[];
+extern unsigned int input_len, output_len;
+
 /*
  * The compressed kernel image (ZO), has been moved so that its position
  * is against the end of the buffer used to hold the uncompressed kernel
@@ -347,14 +352,11 @@ static size_t parse_elf(void *output)
  *             |-------uncompressed kernel image---------|
  *
  */
-asmlinkage __visible void *extract_kernel(void *rmode, memptr heap,
-				  unsigned char *input_data,
-				  unsigned long input_len,
-				  unsigned char *output,
-				  unsigned long output_len)
+asmlinkage __visible void *extract_kernel(void *rmode, unsigned char *output)
 {
 	const unsigned long kernel_total_size = VO__end - VO__text;
 	unsigned long virt_addr = LOAD_PHYSICAL_ADDR;
+	memptr heap = (memptr)boot_heap;
 	unsigned long needed_size;
 	size_t entry_offset;
 
@@ -412,7 +414,7 @@ asmlinkage __visible void *extract_kerne
 	 * entries. This ensures the full mapped area is usable RAM
 	 * and doesn't include any reserved areas.
 	 */
-	needed_size = max(output_len, kernel_total_size);
+	needed_size = max_t(unsigned long, output_len, kernel_total_size);
 #ifdef CONFIG_X86_64
 	needed_size = ALIGN(needed_size, MIN_KERNEL_ALIGN);
 #endif
@@ -443,7 +445,7 @@ asmlinkage __visible void *extract_kerne
 #ifdef CONFIG_X86_64
 	if (heap > 0x3fffffffffffUL)
 		error("Destination address too large");
-	if (virt_addr + max(output_len, kernel_total_size) > KERNEL_IMAGE_SIZE)
+	if (virt_addr + needed_size > KERNEL_IMAGE_SIZE)
 		error("Destination virtual address is beyond the kernel mapping area");
 #else
 	if (heap > ((-__PAGE_OFFSET-(128<<20)-1) & 0x7fffffff))
author maintainer
#144 lore raw

PATCH 140/215 decompress: Use 8 byte alignment

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ard Biesheuvel <[email protected]>

commit 8217ad0a435ff06d651d7298ea8ae8d72388179e upstream.

The ZSTD decompressor requires malloc() allocations to be 8 byte aligned, so ensure that this the case.

---
 include/linux/decompress/mm.h |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
include/linux/decompress/mm.h +1 -1
@@ -48,7 +48,7 @@ MALLOC_VISIBLE void *malloc(int size)
 	if (!malloc_ptr)
 		malloc_ptr = free_mem_ptr;
 
-	malloc_ptr = (malloc_ptr + 3) & ~3;     /* Align */
+	malloc_ptr = (malloc_ptr + 7) & ~7;     /* Align */
 
 	p = (void *)malloc_ptr;
 	malloc_ptr += size;
author maintainer
#145 lore raw

PATCH 141/215 drm/amd/display: Increase frame warning limit with KASAN or KCSAN in dml

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Alex Deucher <[email protected]>

commit 5b750b22530fe53bf7fd6a30baacd53ada26911b upstream.

Does the same thing as:
commit 6740ec97bcdb ("drm/amd/display: Increase frame warning limit with KASAN or KCSAN in dml2")

Fixes 67e38874b85b ("drm/amd/display: Increase num voltage states to 40")
Cc Alvin Lee <[email protected]>
Cc Hamza Mahfooz <[email protected]>
Cc Samson Tam <[email protected]>
Cc Harry Wentland <[email protected]>
Cc Nathan Chancellor <[email protected]>
---
 drivers/gpu/drm/amd/display/dc/dml/Makefile |    4 ++++
 1 file changed, 4 insertions(+)
drivers/gpu/drm/amd/display/dc/dml/Makefile +4 -0
@@ -51,8 +51,12 @@ endif
 endif
 
 ifneq ($(CONFIG_FRAME_WARN),0)
+ifeq ($(filter y,$(CONFIG_KASAN)$(CONFIG_KCSAN)),y)
+frame_warn_flag := -Wframe-larger-than=3072
+else
 frame_warn_flag := -Wframe-larger-than=2048
 endif
+endif
 
 CFLAGS_$(AMDDALPATH)/dc/dml/display_mode_lib.o := $(dml_ccflags)
 
author maintainer
#146 lore raw

PATCH 142/215 NFS: Fix data corruption caused by congestion.

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: "NeilBrown" <[email protected]>

when AOP_WRITEPAGE_ACTIVATE is returned (as NFS does when it detects congestion) it is important that the page is redirtied. nfs_writepage_locked() doesn't do this, so files can become corrupted as writes can be lost.

Note that this is not needed in v6.8 as AOP_WRITEPAGE_ACTIVATE cannot be returned. It is needed for kernels v5.18..v6.7. From 6.3 onward the patch is different as it needs to mention "folio", not "page".

Reported-and-tested-by Jacek Tomaka <[email protected]>
Fixes 6df25e58532b ("nfs: remove reliance on bdi congestion")
---
 fs/nfs/write.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
fs/nfs/write.c +3 -1
index f41d24b54fd1..6a0606668417 100644
@@ -667,8 +667,10 @@ static int nfs_writepage_locked(struct page *page,
 	int err;
 
 	if (wbc->sync_mode == WB_SYNC_NONE &&
-	    NFS_SERVER(inode)->write_congested)
+	    NFS_SERVER(inode)->write_congested) {
+		redirty_page_for_writepage(wbc, page);
 		return AOP_WRITEPAGE_ACTIVATE;
+	}
 
 	nfs_inc_stats(inode, NFSIOS_VFSWRITEPAGE);
 	nfs_pageio_init_write(&pgio, inode, 0,
Signature · 1 line
2.43.0
author maintainer
#147 lore raw

PATCH 143/215 NFSD: Simplify READ_PLUS

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Anna Schumaker <[email protected]>

[ Upstream commit eeadcb75794516839078c28b3730132aeb700ce6 ]

Chuck had suggested reverting READ_PLUS so it returns a single DATA segment covering the requested read range. This prepares the server for a future "sparse read" function so support can easily be added without needing to rip out the old READ_PLUS code at the same time.

---
 fs/nfsd/nfs4xdr.c |  139 ++++++++++++------------------------------------------
 1 file changed, 32 insertions(+), 107 deletions(-)
fs/nfsd/nfs4xdr.c +32 -107
@@ -4777,79 +4777,37 @@ nfsd4_encode_offload_status(struct nfsd4
 
 static __be32
 nfsd4_encode_read_plus_data(struct nfsd4_compoundres *resp,
-			    struct nfsd4_read *read,
-			    unsigned long *maxcount, u32 *eof,
-			    loff_t *pos)
+			    struct nfsd4_read *read)
 {
-	struct xdr_stream *xdr = resp->xdr;
+	bool splice_ok = test_bit(RQ_SPLICE_OK, &resp->rqstp->rq_flags);
 	struct file *file = read->rd_nf->nf_file;
-	int starting_len = xdr->buf->len;
-	loff_t hole_pos;
-	__be32 nfserr;
-	__be32 *p, tmp;
-	__be64 tmp64;
-
-	hole_pos = pos ? *pos : vfs_llseek(file, read->rd_offset, SEEK_HOLE);
-	if (hole_pos > read->rd_offset)
-		*maxcount = min_t(unsigned long, *maxcount, hole_pos - read->rd_offset);
-	*maxcount = min_t(unsigned long, *maxcount, (xdr->buf->buflen - xdr->buf->len));
+	struct xdr_stream *xdr = resp->xdr;
+	unsigned long maxcount;
+	__be32 nfserr, *p;
 
 	/* Content type, offset, byte count */
 	p = xdr_reserve_space(xdr, 4 + 8 + 4);
 	if (!p)
-		return nfserr_resource;
+		return nfserr_io;
+	if (resp->xdr->buf->page_len && splice_ok) {
+		WARN_ON_ONCE(splice_ok);
+		return nfserr_serverfault;
+	}
 
-	read->rd_vlen = xdr_reserve_space_vec(xdr, resp->rqstp->rq_vec, *maxcount);
-	if (read->rd_vlen < 0)
-		return nfserr_resource;
+	maxcount = min_t(unsigned long, read->rd_length,
+			 (xdr->buf->buflen - xdr->buf->len));
 
-	nfserr = nfsd_readv(resp->rqstp, read->rd_fhp, file, read->rd_offset,
-			    resp->rqstp->rq_vec, read->rd_vlen, maxcount, eof);
+	if (file->f_op->splice_read && splice_ok)
+		nfserr = nfsd4_encode_splice_read(resp, read, file, maxcount);
+	else
+		nfserr = nfsd4_encode_readv(resp, read, file, maxcount);
 	if (nfserr)
 		return nfserr;
-	xdr_truncate_encode(xdr, starting_len + 16 + xdr_align_size(*maxcount));
-
-	tmp = htonl(NFS4_CONTENT_DATA);
-	write_bytes_to_xdr_buf(xdr->buf, starting_len,      &tmp,   4);
-	tmp64 = cpu_to_be64(read->rd_offset);
-	write_bytes_to_xdr_buf(xdr->buf, starting_len + 4,  &tmp64, 8);
-	tmp = htonl(*maxcount);
-	write_bytes_to_xdr_buf(xdr->buf, starting_len + 12, &tmp,   4);
-
-	tmp = xdr_zero;
-	write_bytes_to_xdr_buf(xdr->buf, starting_len + 16 + *maxcount, &tmp,
-			       xdr_pad_size(*maxcount));
-	return nfs_ok;
-}
-
-static __be32
-nfsd4_encode_read_plus_hole(struct nfsd4_compoundres *resp,
-			    struct nfsd4_read *read,
-			    unsigned long *maxcount, u32 *eof)
-{
-	struct file *file = read->rd_nf->nf_file;
-	loff_t data_pos = vfs_llseek(file, read->rd_offset, SEEK_DATA);
-	loff_t f_size = i_size_read(file_inode(file));
-	unsigned long count;
-	__be32 *p;
-
-	if (data_pos == -ENXIO)
-		data_pos = f_size;
-	else if (data_pos <= read->rd_offset || (data_pos < f_size && data_pos % PAGE_SIZE))
-		return nfsd4_encode_read_plus_data(resp, read, maxcount, eof, &f_size);
-	count = data_pos - read->rd_offset;
 
-	/* Content type, offset, byte count */
-	p = xdr_reserve_space(resp->xdr, 4 + 8 + 8);
-	if (!p)
-		return nfserr_resource;
-
-	*p++ = htonl(NFS4_CONTENT_HOLE);
+	*p++ = cpu_to_be32(NFS4_CONTENT_DATA);
 	p = xdr_encode_hyper(p, read->rd_offset);
-	p = xdr_encode_hyper(p, count);
+	*p = cpu_to_be32(read->rd_length);
 
-	*eof = (read->rd_offset + count) >= f_size;
-	*maxcount = min_t(unsigned long, count, *maxcount);
 	return nfs_ok;
 }
 
@@ -4857,69 +4815,36 @@ static __be32
 nfsd4_encode_read_plus(struct nfsd4_compoundres *resp, __be32 nfserr,
 		       struct nfsd4_read *read)
 {
-	unsigned long maxcount, count;
+	struct file *file = read->rd_nf->nf_file;
 	struct xdr_stream *xdr = resp->xdr;
-	struct file *file;
 	int starting_len = xdr->buf->len;
-	int last_segment = xdr->buf->len;
-	int segments = 0;
-	__be32 *p, tmp;
-	bool is_data;
-	loff_t pos;
-	u32 eof;
+	u32 segments = 0;
+	__be32 *p;
 
 	if (nfserr)
 		return nfserr;
-	file = read->rd_nf->nf_file;
 
 	/* eof flag, segment count */
 	p = xdr_reserve_space(xdr, 4 + 4);
 	if (!p)
-		return nfserr_resource;
+		return nfserr_io;
 	xdr_commit_encode(xdr);
 
-	maxcount = min_t(unsigned long, read->rd_length,
-			 (xdr->buf->buflen - xdr->buf->len));
-	count    = maxcount;
-
-	eof = read->rd_offset >= i_size_read(file_inode(file));
-	if (eof)
+	read->rd_eof = read->rd_offset >= i_size_read(file_inode(file));
+	if (read->rd_eof)
 		goto out;
 
-	pos = vfs_llseek(file, read->rd_offset, SEEK_HOLE);
-	is_data = pos > read->rd_offset;
-
-	while (count > 0 && !eof) {
-		maxcount = count;
-		if (is_data)
-			nfserr = nfsd4_encode_read_plus_data(resp, read, &maxcount, &eof,
-						segments == 0 ? &pos : NULL);
-		else
-			nfserr = nfsd4_encode_read_plus_hole(resp, read, &maxcount, &eof);
-		if (nfserr)
-			goto out;
-		count -= maxcount;
-		read->rd_offset += maxcount;
-		is_data = !is_data;
-		last_segment = xdr->buf->len;
-		segments++;
-	}
-
-out:
-	if (nfserr && segments == 0)
+	nfserr = nfsd4_encode_read_plus_data(resp, read);
+	if (nfserr) {
 		xdr_truncate_encode(xdr, starting_len);
-	else {
-		if (nfserr) {
-			xdr_truncate_encode(xdr, last_segment);
-			nfserr = nfs_ok;
-			eof = 0;
-		}
-		tmp = htonl(eof);
-		write_bytes_to_xdr_buf(xdr->buf, starting_len,     &tmp, 4);
-		tmp = htonl(segments);
-		write_bytes_to_xdr_buf(xdr->buf, starting_len + 4, &tmp, 4);
+		return nfserr;
 	}
 
+	segments++;
+
+out:
+	p = xdr_encode_bool(p, read->rd_eof);
+	*p = cpu_to_be32(segments);
 	return nfserr;
 }
 
author maintainer
#148 lore raw

PATCH 144/215 NFSD: Remove redundant assignment to variable host_err

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Colin Ian King <[email protected]>

[ Upstream commit 69eed23baf877bbb1f14d7f4df54f89807c9ee2a ]

Variable host_err is assigned a value that is never read, it is being re-assigned a value in every different execution path in the following switch statement. The assignment is redundant and can be removed.

Cleans up clang-scan warning:
warning: Value stored to 'host_err' is never read [deadcode.DeadStores]

---
 fs/nfsd/vfs.c |    1 -
 1 file changed, 1 deletion(-)
fs/nfsd/vfs.c +0 -1
@@ -1317,7 +1317,6 @@ nfsd_create_locked(struct svc_rqst *rqst
 		iap->ia_mode &= ~current_umask();
 
 	err = 0;
-	host_err = 0;
 	switch (type) {
 	case S_IFREG:
 		host_err = vfs_create(&init_user_ns, dirp, dchild, iap->ia_mode, true);
author maintainer
#149 lore raw

PATCH 145/215 nfsd: ignore requests to disable unsupported versions

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jeff Layton <[email protected]>

[ Upstream commit 8e823bafff2308753d430566256c83d8085952da ]

The kernel currently errors out if you attempt to enable or disable a version that it doesn't recognize. Change it to ignore attempts to disable an unrecognized version. If we don't support it, then there is no harm in doing so.

---
 fs/nfsd/nfsctl.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
fs/nfsd/nfsctl.c +3 -1
@@ -601,7 +601,9 @@ static ssize_t __write_versions(struct f
 				}
 				break;
 			default:
-				return -EINVAL;
+				/* Ignore requests to disable non-existent versions */
+				if (cmd == NFSD_SET)
+					return -EINVAL;
 			}
 			vers += len + 1;
 		} while ((len = qword_get(&mesg, vers, size)) > 0);
author maintainer
#150 lore raw

PATCH 146/215 nfsd: move nfserrno() to vfs.c

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jeff Layton <[email protected]>

[ Upstream commit cb12fae1c34b1fa7eaae92c5aadc72d86d7fae19 ]

nfserrno() is common to all nfs versions, but nfsproc.c is specifically for NFSv2. Move it to vfs.c, and the prototype to vfs.h.

While we're in here, remove the #ifdef EDQUOT check in this function. It's apparently a holdover from the initial merge of the nfsd code in 1997. No other place in the kernel checks that that symbol is defined before using it, so I think we can dispense with it here.

---
 fs/nfsd/blocklayout.c    |    1 
 fs/nfsd/blocklayoutxdr.c |    1 
 fs/nfsd/export.h         |    1 
 fs/nfsd/flexfilelayout.c |    1 
 fs/nfsd/nfs4idmap.c      |    1 
 fs/nfsd/nfsproc.c        |   62 ----------------------------------------------
 fs/nfsd/vfs.c            |   63 +++++++++++++++++++++++++++++++++++++++++++++++
 fs/nfsd/vfs.h            |    1 
 8 files changed, 68 insertions(+), 63 deletions(-)
fs/nfsd/blocklayout.c +1 -0
@@ -12,6 +12,7 @@
 #include "blocklayoutxdr.h"
 #include "pnfs.h"
 #include "filecache.h"
+#include "vfs.h"
 
 #define NFSDDBG_FACILITY	NFSDDBG_PNFS
 
fs/nfsd/blocklayoutxdr.c +1 -0
@@ -9,6 +9,7 @@
 
 #include "nfsd.h"
 #include "blocklayoutxdr.h"
+#include "vfs.h"
 
 #define NFSDDBG_FACILITY	NFSDDBG_PNFS
 
fs/nfsd/export.h +0 -1
@@ -115,7 +115,6 @@ struct svc_export *	rqst_find_fsidzero_e
 int			exp_rootfh(struct net *, struct auth_domain *,
 					char *path, struct knfsd_fh *, int maxsize);
 __be32			exp_pseudoroot(struct svc_rqst *, struct svc_fh *);
-__be32			nfserrno(int errno);
 
 static inline void exp_put(struct svc_export *exp)
 {
fs/nfsd/flexfilelayout.c +1 -0
@@ -15,6 +15,7 @@
 
 #include "flexfilelayoutxdr.h"
 #include "pnfs.h"
+#include "vfs.h"
 
 #define NFSDDBG_FACILITY	NFSDDBG_PNFS
 
fs/nfsd/nfs4idmap.c +1 -0
@@ -41,6 +41,7 @@
 #include "idmap.h"
 #include "nfsd.h"
 #include "netns.h"
+#include "vfs.h"
 
 /*
  * Turn off idmapping when using AUTH_SYS.
fs/nfsd/nfsproc.c +0 -62

65 lines not on this page · raw message

fs/nfsd/vfs.c +63 -0

69 lines not on this page · raw message

fs/nfsd/vfs.h +1 -0
@@ -60,6 +60,7 @@ static inline void nfsd_attrs_free(struc
 	posix_acl_release(attrs->na_dpacl);
 }
 
+__be32		nfserrno (int errno);
 int		nfsd_cross_mnt(struct svc_rqst *rqstp, struct dentry **dpp,
 		                struct svc_export **expp);
 __be32		nfsd_lookup(struct svc_rqst *, struct svc_fh *,
author maintainer
#151 lore raw

PATCH 147/215 nfsd: allow disabling NFSv2 at compile time

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jeff Layton <[email protected]>

[ Upstream commit 2f3a4b2ac2f28b9be78ad21f401f31e263845214 ]

rpc.nfsd stopped supporting NFSv2 a year ago. Take the next logical step toward deprecating it and allow NFSv2 support to be compiled out.

Add a new CONFIG_NFSD_V2 option that can be turned off and rework the CONFIG_NFSD_V?_ACL option dependencies. Add a description that discourages enabling it.

Also, change the description of CONFIG_NFSD to state that the always-on version is now 3 instead of 2.

Finally, add an #ifdef around "case 2:" in __write_versions. When NFSv2 is disabled at compile time, this should make the kernel ignore attempts to disable it at runtime, but still error out when trying to enable it.

---
 fs/nfsd/Kconfig  |   19 +++++++++++++++----
 fs/nfsd/Makefile |    5 +++--
 fs/nfsd/nfsctl.c |    2 ++
 fs/nfsd/nfsd.h   |    3 +--
 fs/nfsd/nfssvc.c |    6 ++++++
 5 files changed, 27 insertions(+), 8 deletions(-)
fs/nfsd/Kconfig +15 -4
@@ -8,6 +8,7 @@ config NFSD
 	select SUNRPC
 	select EXPORTFS
 	select NFS_ACL_SUPPORT if NFSD_V2_ACL
+	select NFS_ACL_SUPPORT if NFSD_V3_ACL
 	depends on MULTIUSER
 	help
 	  Choose Y here if you want to allow other computers to access
@@ -26,19 +27,29 @@ config NFSD
 
 	  Below you can choose which versions of the NFS protocol are
 	  available to clients mounting the NFS server on this system.
-	  Support for NFS version 2 (RFC 1094) is always available when
+	  Support for NFS version 3 (RFC 1813) is always available when
 	  CONFIG_NFSD is selected.
 
 	  If unsure, say N.
 
-config NFSD_V2_ACL
-	bool
+config NFSD_V2
+	bool "NFS server support for NFS version 2 (DEPRECATED)"
 	depends on NFSD
+	default n
+	help
+	  NFSv2 (RFC 1094) was the first publicly-released version of NFS.
+	  Unless you are hosting ancient (1990's era) NFS clients, you don't
+	  need this.
+
+	  If unsure, say N.
+
+config NFSD_V2_ACL
+	bool "NFS server support for the NFSv2 ACL protocol extension"
+	depends on NFSD_V2
 
 config NFSD_V3_ACL
 	bool "NFS server support for the NFSv3 ACL protocol extension"
 	depends on NFSD
-	select NFSD_V2_ACL
 	help
 	  Solaris NFS servers support an auxiliary NFSv3 ACL protocol that
 	  never became an official part of the NFS version 3 protocol.
fs/nfsd/Makefile +3 -2
@@ -10,9 +10,10 @@ obj-$(CONFIG_NFSD)	+= nfsd.o
 # this one should be compiled first, as the tracing macros can easily blow up
 nfsd-y			+= trace.o
 
-nfsd-y 			+= nfssvc.o nfsctl.o nfsproc.o nfsfh.o vfs.o \
-			   export.o auth.o lockd.o nfscache.o nfsxdr.o \
+nfsd-y 			+= nfssvc.o nfsctl.o nfsfh.o vfs.o \
+			   export.o auth.o lockd.o nfscache.o \
 			   stats.o filecache.o nfs3proc.o nfs3xdr.o
+nfsd-$(CONFIG_NFSD_V2) += nfsproc.o nfsxdr.o
 nfsd-$(CONFIG_NFSD_V2_ACL) += nfs2acl.o
 nfsd-$(CONFIG_NFSD_V3_ACL) += nfs3acl.o
 nfsd-$(CONFIG_NFSD_V4)	+= nfs4proc.o nfs4xdr.o nfs4state.o nfs4idmap.o \
fs/nfsd/nfsctl.c +2 -0
@@ -581,7 +581,9 @@ static ssize_t __write_versions(struct f
 
 			cmd = sign == '-' ? NFSD_CLEAR : NFSD_SET;
 			switch(num) {
+#ifdef CONFIG_NFSD_V2
 			case 2:
+#endif
 			case 3:
 				nfsd_vers(nn, num, cmd);
 				break;
fs/nfsd/nfsd.h +1 -2
@@ -64,8 +64,7 @@ struct readdir_cd {
 
 
 extern struct svc_program	nfsd_program;
-extern const struct svc_version	nfsd_version2, nfsd_version3,
-				nfsd_version4;
+extern const struct svc_version	nfsd_version2, nfsd_version3, nfsd_version4;
 extern struct mutex		nfsd_mutex;
 extern spinlock_t		nfsd_drc_lock;
 extern unsigned long		nfsd_drc_max_mem;
fs/nfsd/nfssvc.c +6 -0
@@ -91,8 +91,12 @@ unsigned long	nfsd_drc_mem_used;
 #if defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL)
 static struct svc_stat	nfsd_acl_svcstats;
 static const struct svc_version *nfsd_acl_version[] = {
+# if defined(CONFIG_NFSD_V2_ACL)
 	[2] = &nfsd_acl_version2,
+# endif
+# if defined(CONFIG_NFSD_V3_ACL)
 	[3] = &nfsd_acl_version3,
+# endif
 };
 
 #define NFSD_ACL_MINVERS            2
@@ -116,7 +120,9 @@ static struct svc_stat	nfsd_acl_svcstats
 #endif /* defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL) */
 
 static const struct svc_version *nfsd_version[] = {
+#if defined(CONFIG_NFSD_V2)
 	[2] = &nfsd_version2,
+#endif
 	[3] = &nfsd_version3,
 #if defined(CONFIG_NFSD_V4)
 	[4] = &nfsd_version4,
author maintainer
#152 lore raw

PATCH 148/215 exportfs: use pr_debug for unreachable debug statements

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: David Disseldorp <[email protected]>

[ Upstream commit 427505ffeaa464f683faba945a88d3e3248f6979 ]

expfs.c has a bunch of dprintk statements which are unusable due to:
#define dprintk(fmt, args...) do{}while(0)
Use pr_debug so that they can be enabled dynamically. Also make some minor changes to the debug statements to fix some incorrect types, and remove __func__ which can be handled by dynamic debug separately.

---
 fs/exportfs/expfs.c |    8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
fs/exportfs/expfs.c +4 -4
@@ -18,7 +18,7 @@
 #include <linux/sched.h>
 #include <linux/cred.h>
 
-#define dprintk(fmt, args...) do{}while(0)
+#define dprintk(fmt, args...) pr_debug(fmt, ##args)
 
 
 static int get_name(const struct path *path, char *name, struct dentry *child);
@@ -132,8 +132,8 @@ static struct dentry *reconnect_one(stru
 	inode_unlock(dentry->d_inode);
 
 	if (IS_ERR(parent)) {
-		dprintk("%s: get_parent of %ld failed, err %d\n",
-			__func__, dentry->d_inode->i_ino, PTR_ERR(parent));
+		dprintk("get_parent of %lu failed, err %ld\n",
+			dentry->d_inode->i_ino, PTR_ERR(parent));
 		return parent;
 	}
 
@@ -147,7 +147,7 @@ static struct dentry *reconnect_one(stru
 	dprintk("%s: found name: %s\n", __func__, nbuf);
 	tmp = lookup_one_unlocked(mnt_user_ns(mnt), nbuf, parent, strlen(nbuf));
 	if (IS_ERR(tmp)) {
-		dprintk("%s: lookup failed: %d\n", __func__, PTR_ERR(tmp));
+		dprintk("lookup failed: %ld\n", PTR_ERR(tmp));
 		err = PTR_ERR(tmp);
 		goto out_err;
 	}
author maintainer
#153 lore raw

PATCH 149/215 NFSD: Flesh out a documenting comment for filecache.c

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Chuck Lever <[email protected]>

[ Upstream commit b3276c1f5b268ff56622e9e125b792b4c3dc03ac ]

Record what we've learned recently about the NFSD filecache in a documenting comment so our future selves don't forget what all this is for.

---
 fs/nfsd/filecache.c |   24 ++++++++++++++++++++++++
 1 file changed, 24 insertions(+)
fs/nfsd/filecache.c +24 -0
@@ -2,6 +2,30 @@
  * Open file cache.
  *
  * (c) 2015 - Jeff Layton <[email protected]>
+ *
+ * An nfsd_file object is a per-file collection of open state that binds
+ * together:
+ *   - a struct file *
+ *   - a user credential
+ *   - a network namespace
+ *   - a read-ahead context
+ *   - monitoring for writeback errors
+ *
+ * nfsd_file objects are reference-counted. Consumers acquire a new
+ * object via the nfsd_file_acquire API. They manage their interest in
+ * the acquired object, and hence the object's reference count, via
+ * nfsd_file_get and nfsd_file_put. There are two varieties of nfsd_file
+ * object:
+ *
+ *  * non-garbage-collected: When a consumer wants to precisely control
+ *    the lifetime of a file's open state, it acquires a non-garbage-
+ *    collected nfsd_file. The final nfsd_file_put releases the open
+ *    state immediately.
+ *
+ *  * garbage-collected: When a consumer does not control the lifetime
+ *    of open state, it acquires a garbage-collected nfsd_file. The
+ *    final nfsd_file_put allows the open state to linger for a period
+ *    during which it may be re-used.
  */
 
 #include <linux/hash.h>
author maintainer
#154 lore raw

PATCH 150/215 NFSD: Clean up nfs4_preprocess_stateid_op() call sites

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Chuck Lever <[email protected]>

[ Upstream commit eeff73f7c1c583f79a401284f46c619294859310 ]

Remove the lame-duck dprintk()s around nfs4_preprocess_stateid_op() call sites.

---
 fs/nfsd/nfs4proc.c |   31 +++++++------------------------
 1 file changed, 7 insertions(+), 24 deletions(-)
fs/nfsd/nfs4proc.c +7 -24
@@ -943,12 +943,7 @@ nfsd4_read(struct svc_rqst *rqstp, struc
 	status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh,
 					&read->rd_stateid, RD_STATE,
 					&read->rd_nf, NULL);
-	if (status) {
-		dprintk("NFSD: nfsd4_read: couldn't process stateid!\n");
-		goto out;
-	}
-	status = nfs_ok;
-out:
+
 	read->rd_rqstp = rqstp;
 	read->rd_fhp = &cstate->current_fh;
 	return status;
@@ -1117,10 +1112,8 @@ nfsd4_setattr(struct svc_rqst *rqstp, st
 		status = nfs4_preprocess_stateid_op(rqstp, cstate,
 				&cstate->current_fh, &setattr->sa_stateid,
 				WR_STATE, NULL, NULL);
-		if (status) {
-			dprintk("NFSD: nfsd4_setattr: couldn't process stateid!\n");
+		if (status)
 			return status;
-		}
 	}
 	err = fh_want_write(&cstate->current_fh);
 	if (err)
@@ -1170,10 +1163,8 @@ nfsd4_write(struct svc_rqst *rqstp, stru
 			       write->wr_offset, cnt);
 	status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh,
 						stateid, WR_STATE, &nf, NULL);
-	if (status) {
-		dprintk("NFSD: nfsd4_write: couldn't process stateid!\n");
+	if (status)
 		return status;
-	}
 
 	write->wr_how_written = write->wr_stable_how;
 
@@ -1204,17 +1195,13 @@ nfsd4_verify_copy(struct svc_rqst *rqstp
 
 	status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->save_fh,
 					    src_stateid, RD_STATE, src, NULL);
-	if (status) {
-		dprintk("NFSD: %s: couldn't process src stateid!\n", __func__);
+	if (status)
 		goto out;
-	}
 
 	status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh,
 					    dst_stateid, WR_STATE, dst, NULL);
-	if (status) {
-		dprintk("NFSD: %s: couldn't process dst stateid!\n", __func__);
+	if (status)
 		goto out_put_src;
-	}
 
 	/* fix up for NFS-specific error code */
 	if (!S_ISREG(file_inode((*src)->nf_file)->i_mode) ||
@@ -1935,10 +1922,8 @@ nfsd4_fallocate(struct svc_rqst *rqstp,
 	status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh,
 					    &fallocate->falloc_stateid,
 					    WR_STATE, &nf, NULL);
-	if (status != nfs_ok) {
-		dprintk("NFSD: nfsd4_fallocate: couldn't process stateid!\n");
+	if (status != nfs_ok)
 		return status;
-	}
 
 	status = nfsd4_vfs_fallocate(rqstp, &cstate->current_fh, nf->nf_file,
 				     fallocate->falloc_offset,
@@ -1994,10 +1979,8 @@ nfsd4_seek(struct svc_rqst *rqstp, struc
 	status = nfs4_preprocess_stateid_op(rqstp, cstate, &cstate->current_fh,
 					    &seek->seek_stateid,
 					    RD_STATE, &nf, NULL);
-	if (status) {
-		dprintk("NFSD: nfsd4_seek: couldn't process stateid!\n");
+	if (status)
 		return status;
-	}
 
 	switch (seek->seek_whence) {
 	case NFS4_CONTENT_DATA:
author maintainer
#155 lore raw

PATCH 151/215 NFSD: Trace stateids returned via DELEGRETURN

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Chuck Lever <[email protected]>

[ Upstream commit 20eee313ff4b8a7e71ae9560f5c4ba27cd763005 ]

Handing out a delegation stateid is recorded with the nfsd_deleg_read tracepoint, but there isn't a matching tracepoint for recording when the stateid is returned.

---
 fs/nfsd/nfs4state.c |    1 +
 fs/nfsd/trace.h     |    1 +
 2 files changed, 2 insertions(+)
fs/nfsd/nfs4state.c +1 -0
@@ -6935,6 +6935,7 @@ nfsd4_delegreturn(struct svc_rqst *rqstp
 	if (status)
 		goto put_stateid;
 
+	trace_nfsd_deleg_return(stateid);
 	wake_up_var(d_inode(cstate->current_fh.fh_dentry));
 	destroy_delegation(dp);
 put_stateid:
fs/nfsd/trace.h +1 -0
@@ -604,6 +604,7 @@ DEFINE_STATEID_EVENT(layout_recall_relea
 
 DEFINE_STATEID_EVENT(open);
 DEFINE_STATEID_EVENT(deleg_read);
+DEFINE_STATEID_EVENT(deleg_return);
 DEFINE_STATEID_EVENT(deleg_recall);
 
 DECLARE_EVENT_CLASS(nfsd_stateseqid_class,
author maintainer
#156 lore raw

PATCH 152/215 NFSD: Trace delegation revocations

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Chuck Lever <[email protected]>

[ Upstream commit a1c74569bbde91299f24535abf711be5c84df9de ]

Delegation revocation is an exceptional event that is not otherwise visible externally (eg, no network traffic is emitted). Generate a trace record when it occurs so that revocation can be observed or other activity can be triggered. Example:

nfsd-1104 [005] 1912.002544: nfsd_stid_revoke: client 633c9343:4e82788d stateid 00000003:00000001 ref=2 type=DELEG

Trace infrastructure is provided for subsequent additional tracing related to nfs4_stid activity.

---
 fs/nfsd/nfs4state.c |    2 +
 fs/nfsd/trace.h     |   55 ++++++++++++++++++++++++++++++++++++++++++++++++++++
 2 files changed, 57 insertions(+)
fs/nfsd/nfs4state.c +2 -0
@@ -1367,6 +1367,8 @@ static void revoke_delegation(struct nfs
 
 	WARN_ON(!list_empty(&dp->dl_recall_lru));
 
+	trace_nfsd_stid_revoke(&dp->dl_stid);
+
 	if (clp->cl_minorversion) {
 		spin_lock(&clp->cl_lock);
 		dp->dl_stid.sc_type = NFS4_REVOKED_DELEG_STID;
fs/nfsd/trace.h +55 -0

61 lines not on this page · raw message

author maintainer
#157 lore raw

PATCH 153/215 NFSD: Use const pointers as parameters to fh_ helpers

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Chuck Lever <[email protected]>

[ Upstream commit b48f8056c034f28dd54668399f1d22be421b0bef ]

Enable callers to use const pointers where they are able to.

---
 fs/nfsd/nfsfh.h |   10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)
fs/nfsd/nfsfh.h +6 -4
@@ -220,7 +220,7 @@ __be32	fh_update(struct svc_fh *);
 void	fh_put(struct svc_fh *);
 
 static __inline__ struct svc_fh *
-fh_copy(struct svc_fh *dst, struct svc_fh *src)
+fh_copy(struct svc_fh *dst, const struct svc_fh *src)
 {
 	WARN_ON(src->fh_dentry);
 
@@ -229,7 +229,7 @@ fh_copy(struct svc_fh *dst, struct svc_f
 }
 
 static inline void
-fh_copy_shallow(struct knfsd_fh *dst, struct knfsd_fh *src)
+fh_copy_shallow(struct knfsd_fh *dst, const struct knfsd_fh *src)
 {
 	dst->fh_size = src->fh_size;
 	memcpy(&dst->fh_raw, &src->fh_raw, src->fh_size);
@@ -243,7 +243,8 @@ fh_init(struct svc_fh *fhp, int maxsize)
 	return fhp;
 }
 
-static inline bool fh_match(struct knfsd_fh *fh1, struct knfsd_fh *fh2)
+static inline bool fh_match(const struct knfsd_fh *fh1,
+			    const struct knfsd_fh *fh2)
 {
 	if (fh1->fh_size != fh2->fh_size)
 		return false;
@@ -252,7 +253,8 @@ static inline bool fh_match(struct knfsd
 	return true;
 }
 
-static inline bool fh_fsid_match(struct knfsd_fh *fh1, struct knfsd_fh *fh2)
+static inline bool fh_fsid_match(const struct knfsd_fh *fh1,
+				 const struct knfsd_fh *fh2)
 {
 	if (fh1->fh_fsid_type != fh2->fh_fsid_type)
 		return false;
author maintainer
#158 lore raw

PATCH 154/215 NFSD: Update file_hashtbl() helpers

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Chuck Lever <[email protected]>

[ Upstream commit 3fe828caddd81e68e9d29353c6e9285a658ca056 ]

Enable callers to use const pointers for type safety.

---
 fs/nfsd/nfs4state.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
fs/nfsd/nfs4state.c +2 -2
@@ -721,7 +721,7 @@ static unsigned int ownerstr_hashval(str
 #define FILE_HASH_BITS                   8
 #define FILE_HASH_SIZE                  (1 << FILE_HASH_BITS)
 
-static unsigned int file_hashval(struct svc_fh *fh)
+static unsigned int file_hashval(const struct svc_fh *fh)
 {
 	struct inode *inode = d_inode(fh->fh_dentry);
 
@@ -4687,7 +4687,7 @@ move_to_close_lru(struct nfs4_ol_stateid
 
 /* search file_hashtbl[] for file */
 static struct nfs4_file *
-find_file_locked(struct svc_fh *fh, unsigned int hashval)
+find_file_locked(const struct svc_fh *fh, unsigned int hashval)
 {
 	struct nfs4_file *fp;
 
author maintainer
#159 lore raw

PATCH 155/215 NFSD: Clean up nfsd4_init_file()

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Chuck Lever <[email protected]>

[ Upstream commit 81a21fa3e7fdecb3c5b97014f0fc5a17d5806cae ]

Name this function more consistently. I'm going to use nfsd4_file_ and nfsd4_file_hash_ for these helpers.

Change the @fh parameter to be const pointer for better type safety.

Finally, move the hash insertion operation to the caller. This is typical for most other "init_object" type helpers, and it is where most of the other nfs4_file hash table operations are located.

---
 fs/nfsd/nfs4state.c |   10 ++++------
 1 file changed, 4 insertions(+), 6 deletions(-)
fs/nfsd/nfs4state.c +4 -6
@@ -4278,11 +4278,9 @@ static struct nfs4_file *nfsd4_alloc_fil
 }
 
 /* OPEN Share state helper functions */
-static void nfsd4_init_file(struct svc_fh *fh, unsigned int hashval,
-				struct nfs4_file *fp)
-{
-	lockdep_assert_held(&state_lock);
 
+static void nfsd4_file_init(const struct svc_fh *fh, struct nfs4_file *fp)
+{
 	refcount_set(&fp->fi_ref, 1);
 	spin_lock_init(&fp->fi_lock);
 	INIT_LIST_HEAD(&fp->fi_stateids);
@@ -4300,7 +4298,6 @@ static void nfsd4_init_file(struct svc_f
 	INIT_LIST_HEAD(&fp->fi_lo_states);
 	atomic_set(&fp->fi_lo_recalls, 0);
 #endif
-	hlist_add_head_rcu(&fp->fi_hash, &file_hashtbl[hashval]);
 }
 
 void
@@ -4718,7 +4715,8 @@ static struct nfs4_file *insert_file(str
 			fp->fi_aliased = alias_found = true;
 	}
 	if (likely(ret == NULL)) {
-		nfsd4_init_file(fh, hashval, new);
+		nfsd4_file_init(fh, new);
+		hlist_add_head_rcu(&new->fi_hash, &file_hashtbl[hashval]);
 		new->fi_aliased = alias_found;
 		ret = new;
 	}
author maintainer
#160 lore raw

PATCH 156/215 NFSD: Add a nfsd4_file_hash_remove() helper

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Chuck Lever <[email protected]>

[ Upstream commit 3341678f2fd6106055cead09e513fad6950a0d19 ]

Refactor to relocate hash deletion operation to a helper function that is close to most other nfs4_file data structure operations.

The "noinline" annotation will become useful in a moment when the hlist_del_rcu() is replaced with a more complex rhash remove operation. It also guarantees that hash remove operations can be traced with "-p function -l remove_nfs4_file_locked".

This also simplifies the organization of forward declarations: the to-be-added rhashtable and its param structure will be defined /after/ put_nfs4_file().

---
 fs/nfsd/nfs4state.c |    8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)
fs/nfsd/nfs4state.c +7 -1
@@ -84,6 +84,7 @@ static bool check_for_locks(struct nfs4_
 static void nfs4_free_ol_stateid(struct nfs4_stid *stid);
 void nfsd4_end_grace(struct nfsd_net *nn);
 static void _free_cpntf_state_locked(struct nfsd_net *nn, struct nfs4_cpntf_state *cps);
+static void nfsd4_file_hash_remove(struct nfs4_file *fi);
 
 /* Locking: */
 
@@ -591,7 +592,7 @@ put_nfs4_file(struct nfs4_file *fi)
 	might_lock(&state_lock);
 
 	if (refcount_dec_and_lock(&fi->fi_ref, &state_lock)) {
-		hlist_del_rcu(&fi->fi_hash);
+		nfsd4_file_hash_remove(fi);
 		spin_unlock(&state_lock);
 		WARN_ON_ONCE(!list_empty(&fi->fi_clnt_odstate));
 		WARN_ON_ONCE(!list_empty(&fi->fi_delegations));
@@ -4750,6 +4751,11 @@ find_or_add_file(struct nfs4_file *new,
 	return insert_file(new, fh, hashval);
 }
 
+static noinline_for_stack void nfsd4_file_hash_remove(struct nfs4_file *fi)
+{
+	hlist_del_rcu(&fi->fi_hash);
+}
+
 /*
  * Called to check deny when READ with all zero stateid or
  * WRITE with all zero or all one stateid
author maintainer
#161 lore raw

PATCH 157/215 NFSD: Clean up find_or_add_file()

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Chuck Lever <[email protected]>

[ Upstream commit 9270fc514ba7d415636b23bcb937573a1ce54f6a ]

Remove the call to find_file_locked() in insert_nfs4_file(). Tracing shows that over 99% of these calls return NULL. Thus it is not worth the expense of the extra bucket list traversal. insert_file() already deals correctly with the case where the item is already in the hash bucket.

Since nfsd4_file_hash_insert() is now just a wrapper around insert_file(), move the meat of insert_file() into nfsd4_file_hash_insert() and get rid of it.

---
 fs/nfsd/nfs4state.c |   64 ++++++++++++++++++++++------------------------------
 1 file changed, 28 insertions(+), 36 deletions(-)
fs/nfsd/nfs4state.c +28 -36
@@ -4699,24 +4699,42 @@ find_file_locked(const struct svc_fh *fh
 	return NULL;
 }
 
-static struct nfs4_file *insert_file(struct nfs4_file *new, struct svc_fh *fh,
-				     unsigned int hashval)
+static struct nfs4_file * find_file(struct svc_fh *fh)
 {
 	struct nfs4_file *fp;
+	unsigned int hashval = file_hashval(fh);
+
+	rcu_read_lock();
+	fp = find_file_locked(fh, hashval);
+	rcu_read_unlock();
+	return fp;
+}
+
+/*
+ * On hash insertion, identify entries with the same inode but
+ * distinct filehandles. They will all be in the same hash bucket
+ * because nfs4_file's are hashed by the address in the fi_inode
+ * field.
+ */
+static noinline_for_stack struct nfs4_file *
+nfsd4_file_hash_insert(struct nfs4_file *new, const struct svc_fh *fhp)
+{
+	unsigned int hashval = file_hashval(fhp);
 	struct nfs4_file *ret = NULL;
 	bool alias_found = false;
+	struct nfs4_file *fi;
 
 	spin_lock(&state_lock);
-	hlist_for_each_entry_rcu(fp, &file_hashtbl[hashval], fi_hash,
+	hlist_for_each_entry_rcu(fi, &file_hashtbl[hashval], fi_hash,
 				 lockdep_is_held(&state_lock)) {
-		if (fh_match(&fp->fi_fhandle, &fh->fh_handle)) {
-			if (refcount_inc_not_zero(&fp->fi_ref))
-				ret = fp;
-		} else if (d_inode(fh->fh_dentry) == fp->fi_inode)
-			fp->fi_aliased = alias_found = true;
+		if (fh_match(&fi->fi_fhandle, &fhp->fh_handle)) {
+			if (refcount_inc_not_zero(&fi->fi_ref))
+				ret = fi;
+		} else if (d_inode(fhp->fh_dentry) == fi->fi_inode)
+			fi->fi_aliased = alias_found = true;
 	}
 	if (likely(ret == NULL)) {
-		nfsd4_file_init(fh, new);
+		nfsd4_file_init(fhp, new);
 		hlist_add_head_rcu(&new->fi_hash, &file_hashtbl[hashval]);
 		new->fi_aliased = alias_found;
 		ret = new;
@@ -4725,32 +4743,6 @@ static struct nfs4_file *insert_file(str
 	return ret;
 }
 
-static struct nfs4_file * find_file(struct svc_fh *fh)
-{
-	struct nfs4_file *fp;
-	unsigned int hashval = file_hashval(fh);
-
-	rcu_read_lock();
-	fp = find_file_locked(fh, hashval);
-	rcu_read_unlock();
-	return fp;
-}
-
-static struct nfs4_file *
-find_or_add_file(struct nfs4_file *new, struct svc_fh *fh)
-{
-	struct nfs4_file *fp;
-	unsigned int hashval = file_hashval(fh);
-
-	rcu_read_lock();
-	fp = find_file_locked(fh, hashval);
-	rcu_read_unlock();
-	if (fp)
-		return fp;
-
-	return insert_file(new, fh, hashval);
-}
-
 static noinline_for_stack void nfsd4_file_hash_remove(struct nfs4_file *fi)
 {
 	hlist_del_rcu(&fi->fi_hash);
@@ -5661,7 +5653,7 @@ nfsd4_process_open2(struct svc_rqst *rqs
 	 * and check for delegations in the process of being recalled.
 	 * If not found, create the nfs4_file struct
 	 */
-	fp = find_or_add_file(open->op_file, current_fh);
+	fp = nfsd4_file_hash_insert(open->op_file, current_fh);
 	if (fp != open->op_file) {
 		status = nfs4_check_deleg(cl, open, &dp);
 		if (status)
author maintainer
#162 lore raw

PATCH 158/215 NFSD: Refactor find_file()

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Chuck Lever <[email protected]>

[ Upstream commit 15424748001a9b5ea62b3e6ad45f0a8b27f01df9 ]

find_file() is now the only caller of find_file_locked(), so just fold these two together.

---
 fs/nfsd/nfs4state.c |   36 +++++++++++++++---------------------
 1 file changed, 15 insertions(+), 21 deletions(-)
fs/nfsd/nfs4state.c +15 -21
@@ -4683,31 +4683,24 @@ move_to_close_lru(struct nfs4_ol_stateid
 		nfs4_put_stid(&last->st_stid);
 }
 
-/* search file_hashtbl[] for file */
-static struct nfs4_file *
-find_file_locked(const struct svc_fh *fh, unsigned int hashval)
+static noinline_for_stack struct nfs4_file *
+nfsd4_file_hash_lookup(const struct svc_fh *fhp)
 {
-	struct nfs4_file *fp;
+	unsigned int hashval = file_hashval(fhp);
+	struct nfs4_file *fi;
 
-	hlist_for_each_entry_rcu(fp, &file_hashtbl[hashval], fi_hash,
-				lockdep_is_held(&state_lock)) {
-		if (fh_match(&fp->fi_fhandle, &fh->fh_handle)) {
-			if (refcount_inc_not_zero(&fp->fi_ref))
-				return fp;
+	rcu_read_lock();
+	hlist_for_each_entry_rcu(fi, &file_hashtbl[hashval], fi_hash,
+				 lockdep_is_held(&state_lock)) {
+		if (fh_match(&fi->fi_fhandle, &fhp->fh_handle)) {
+			if (refcount_inc_not_zero(&fi->fi_ref)) {
+				rcu_read_unlock();
+				return fi;
+			}
 		}
 	}
-	return NULL;
-}
-
-static struct nfs4_file * find_file(struct svc_fh *fh)
-{
-	struct nfs4_file *fp;
-	unsigned int hashval = file_hashval(fh);
-
-	rcu_read_lock();
-	fp = find_file_locked(fh, hashval);
 	rcu_read_unlock();
-	return fp;
+	return NULL;
 }
 
 /*
@@ -4758,9 +4751,10 @@ nfs4_share_conflict(struct svc_fh *curre
 	struct nfs4_file *fp;
 	__be32 ret = nfs_ok;
 
-	fp = find_file(current_fh);
+	fp = nfsd4_file_hash_lookup(current_fh);
 	if (!fp)
 		return ret;
+
 	/* Check for conflicting share reservations */
 	spin_lock(&fp->fi_lock);
 	if (fp->fi_share_deny & deny_type)
author maintainer
#163 lore raw

PATCH 159/215 NFSD: Use rhashtable for managing nfs4_file objects

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Chuck Lever <[email protected]>

[ Upstream commit d47b295e8d76a4d69f0e2ea0cd8a79c9d3488280 ]

fh_match() is costly, especially when filehandles are large (as is the case for NFSv4). It needs to be used sparingly when searching data structures. Unfortunately, with common workloads, I see multiple thousands of objects stored in file_hashtbl[], which has just 256 buckets, making its bucket hash chains quite lengthy.

Walking long hash chains with the state_lock held blocks other activity that needs that lock. Sizable hash chains are a common occurrance once the server has handed out some delegations, for example -- IIUC, each delegated file is held open on the server by an nfs4_file object.

To help mitigate the cost of searching with fh_match(), replace the nfs4_file hash table with an rhashtable, which can dynamically resize its bucket array to minimize hash chain length.

The result of this modification is an improvement in the latency of NFSv4 operations, and the reduction of nfsd CPU utilization due to eliminating the cost of multiple calls to fh_match() and reducing the CPU cache misses incurred while walking long hash chains in the nfs4_file hash table.

---
 fs/nfsd/nfs4state.c |   97 +++++++++++++++++++++++++++++++++-------------------
 fs/nfsd/state.h     |    5 --
 2 files changed, 63 insertions(+), 39 deletions(-)
fs/nfsd/nfs4state.c +62 -35
@@ -44,7 +44,9 @@
 #include <linux/jhash.h>
 #include <linux/string_helpers.h>
 #include <linux/fsnotify.h>
+#include <linux/rhashtable.h>
 #include <linux/nfs_ssc.h>
+
 #include "xdr4.h"
 #include "xdr4cb.h"
 #include "vfs.h"
@@ -589,11 +591,8 @@ static void nfsd4_free_file_rcu(struct r
 void
 put_nfs4_file(struct nfs4_file *fi)
 {
-	might_lock(&state_lock);
-
-	if (refcount_dec_and_lock(&fi->fi_ref, &state_lock)) {
+	if (refcount_dec_and_test(&fi->fi_ref)) {
 		nfsd4_file_hash_remove(fi);
-		spin_unlock(&state_lock);
 		WARN_ON_ONCE(!list_empty(&fi->fi_clnt_odstate));
 		WARN_ON_ONCE(!list_empty(&fi->fi_delegations));
 		call_rcu(&fi->fi_rcu, nfsd4_free_file_rcu);
@@ -718,19 +717,20 @@ static unsigned int ownerstr_hashval(str
 	return ret & OWNER_HASH_MASK;
 }
 
-/* hash table for nfs4_file */
-#define FILE_HASH_BITS                   8
-#define FILE_HASH_SIZE                  (1 << FILE_HASH_BITS)
-
-static unsigned int file_hashval(const struct svc_fh *fh)
-{
-	struct inode *inode = d_inode(fh->fh_dentry);
+static struct rhltable nfs4_file_rhltable ____cacheline_aligned_in_smp;
 
-	/* XXX: why not (here & in file cache) use inode? */
-	return (unsigned int)hash_long(inode->i_ino, FILE_HASH_BITS);
-}
+static const struct rhashtable_params nfs4_file_rhash_params = {
+	.key_len		= sizeof_field(struct nfs4_file, fi_inode),
+	.key_offset		= offsetof(struct nfs4_file, fi_inode),
+	.head_offset		= offsetof(struct nfs4_file, fi_rlist),
 
-static struct hlist_head file_hashtbl[FILE_HASH_SIZE];
+	/*
+	 * Start with a single page hash table to reduce resizing churn
+	 * on light workloads.
+	 */
+	.min_size		= 256,
+	.automatic_shrinking	= true,
+};
 
 /*
  * Check if courtesy clients have conflicting access and resolve it if possible
@@ -4686,12 +4686,14 @@ move_to_close_lru(struct nfs4_ol_stateid
 static noinline_for_stack struct nfs4_file *
 nfsd4_file_hash_lookup(const struct svc_fh *fhp)
 {
-	unsigned int hashval = file_hashval(fhp);
+	struct inode *inode = d_inode(fhp->fh_dentry);
+	struct rhlist_head *tmp, *list;
 	struct nfs4_file *fi;
 
 	rcu_read_lock();
-	hlist_for_each_entry_rcu(fi, &file_hashtbl[hashval], fi_hash,
-				 lockdep_is_held(&state_lock)) {
+	list = rhltable_lookup(&nfs4_file_rhltable, &inode,
+			       nfs4_file_rhash_params);
+	rhl_for_each_entry_rcu(fi, tmp, list, fi_rlist) {
 		if (fh_match(&fi->fi_fhandle, &fhp->fh_handle)) {
 			if (refcount_inc_not_zero(&fi->fi_ref)) {
 				rcu_read_unlock();
@@ -4705,40 +4707,56 @@ nfsd4_file_hash_lookup(const struct svc_
 
 /*
  * On hash insertion, identify entries with the same inode but
- * distinct filehandles. They will all be in the same hash bucket
- * because nfs4_file's are hashed by the address in the fi_inode
- * field.
+ * distinct filehandles. They will all be on the list returned
+ * by rhltable_lookup().
+ *
+ * inode->i_lock prevents racing insertions from adding an entry
+ * for the same inode/fhp pair twice.
  */
 static noinline_for_stack struct nfs4_file *
 nfsd4_file_hash_insert(struct nfs4_file *new, const struct svc_fh *fhp)
 {
-	unsigned int hashval = file_hashval(fhp);
+	struct inode *inode = d_inode(fhp->fh_dentry);
+	struct rhlist_head *tmp, *list;
 	struct nfs4_file *ret = NULL;
 	bool alias_found = false;
 	struct nfs4_file *fi;
+	int err;
 
-	spin_lock(&state_lock);
-	hlist_for_each_entry_rcu(fi, &file_hashtbl[hashval], fi_hash,
-				 lockdep_is_held(&state_lock)) {
+	rcu_read_lock();
+	spin_lock(&inode->i_lock);
+
+	list = rhltable_lookup(&nfs4_file_rhltable, &inode,
+			       nfs4_file_rhash_params);
+	rhl_for_each_entry_rcu(fi, tmp, list, fi_rlist) {
 		if (fh_match(&fi->fi_fhandle, &fhp->fh_handle)) {
 			if (refcount_inc_not_zero(&fi->fi_ref))
 				ret = fi;
-		} else if (d_inode(fhp->fh_dentry) == fi->fi_inode)
+		} else
 			fi->fi_aliased = alias_found = true;
 	}
-	if (likely(ret == NULL)) {
-		nfsd4_file_init(fhp, new);
-		hlist_add_head_rcu(&new->fi_hash, &file_hashtbl[hashval]);
-		new->fi_aliased = alias_found;
-		ret = new;
-	}
-	spin_unlock(&state_lock);
+	if (ret)
+		goto out_unlock;
+
+	nfsd4_file_init(fhp, new);
+	err = rhltable_insert(&nfs4_file_rhltable, &new->fi_rlist,
+			      nfs4_file_rhash_params);
+	if (err)
+		goto out_unlock;
+
+	new->fi_aliased = alias_found;
+	ret = new;
+
+out_unlock:
+	spin_unlock(&inode->i_lock);
+	rcu_read_unlock();
 	return ret;
 }
 
 static noinline_for_stack void nfsd4_file_hash_remove(struct nfs4_file *fi)
 {
-	hlist_del_rcu(&fi->fi_hash);
+	rhltable_remove(&nfs4_file_rhltable, &fi->fi_rlist,
+			nfs4_file_rhash_params);
 }
 
 /*
@@ -5648,6 +5666,8 @@ nfsd4_process_open2(struct svc_rqst *rqs
 	 * If not found, create the nfs4_file struct
 	 */
 	fp = nfsd4_file_hash_insert(open->op_file, current_fh);
+	if (unlikely(!fp))
+		return nfserr_jukebox;
 	if (fp != open->op_file) {
 		status = nfs4_check_deleg(cl, open, &dp);
 		if (status)
@@ -8064,10 +8084,16 @@ nfs4_state_start(void)
 {
 	int ret;
 
-	ret = nfsd4_create_callback_queue();
+	ret = rhltable_init(&nfs4_file_rhltable, &nfs4_file_rhash_params);
 	if (ret)
 		return ret;
 
+	ret = nfsd4_create_callback_queue();
+	if (ret) {
+		rhltable_destroy(&nfs4_file_rhltable);
+		return ret;
+	}
+
 	set_max_delegations();
 	return 0;
 }
@@ -8098,6 +8124,7 @@ nfs4_state_shutdown_net(struct net *net)
 
 	nfsd4_client_tracking_exit(net);
 	nfs4_state_destroy_net(net);
+	rhltable_destroy(&nfs4_file_rhltable);
 #ifdef CONFIG_NFSD_V4_2_INTER_SSC
 	nfsd4_ssc_shutdown_umount(nn);
 #endif
fs/nfsd/state.h +1 -4
@@ -536,16 +536,13 @@ struct nfs4_clnt_odstate {
  * inode can have multiple filehandles associated with it, so there is
  * (potentially) a many to one relationship between this struct and struct
  * inode.
- *
- * These are hashed by filehandle in the file_hashtbl, which is protected by
- * the global state_lock spinlock.
  */
 struct nfs4_file {
 	refcount_t		fi_ref;
 	struct inode *		fi_inode;
 	bool			fi_aliased;
 	spinlock_t		fi_lock;
-	struct hlist_node       fi_hash;	/* hash on fi_fhandle */
+	struct rhlist_head	fi_rlist;
 	struct list_head        fi_stateids;
 	union {
 		struct list_head	fi_delegations;
author maintainer
#164 lore raw

PATCH 160/215 NFSD: Fix licensing header in filecache.c

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Chuck Lever <[email protected]>

[ Upstream commit 3f054211b29c0fa06dfdcab402c795fd7e906be1 ]

Add a missing SPDX header.

---
 fs/nfsd/filecache.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
fs/nfsd/filecache.c +2 -1
@@ -1,5 +1,6 @@
+// SPDX-License-Identifier: GPL-2.0
 /*
- * Open file cache.
+ * The NFSD open file cache.
  *
  * (c) 2015 - Jeff Layton <[email protected]>
  *
author maintainer
#165 lore raw

PATCH 161/215 filelock: add a new locks_inode_context accessor function

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jeff Layton <[email protected]>

[ Upstream commit 401a8b8fd5acd51582b15238d72a8d0edd580e9f ]

There are a number of places in the kernel that are accessing the inode->i_flctx field without smp_load_acquire. This is required to ensure that the caller doesn't see a partially-initialized structure.

Add a new accessor function for it to make this clear and convert all of the relevant accesses in locks.c to use it. Also, convert locks_free_lock_context to use the helper as well instead of just doing a "bare" assignment.

Stable-dep-of: 77c67530e1f9 ("nfsd: use locks_inode_context helper")

---
 fs/locks.c         |   24 ++++++++++++------------
 include/linux/fs.h |   14 ++++++++++++++
 2 files changed, 26 insertions(+), 12 deletions(-)
fs/locks.c +12 -12
@@ -175,7 +175,7 @@ locks_get_lock_context(struct inode *ino
 	struct file_lock_context *ctx;
 
 	/* paired with cmpxchg() below */
-	ctx = smp_load_acquire(&inode->i_flctx);
+	ctx = locks_inode_context(inode);
 	if (likely(ctx) || type == F_UNLCK)
 		goto out;
 
@@ -194,7 +194,7 @@ locks_get_lock_context(struct inode *ino
 	 */
 	if (cmpxchg(&inode->i_flctx, NULL, ctx)) {
 		kmem_cache_free(flctx_cache, ctx);
-		ctx = smp_load_acquire(&inode->i_flctx);
+		ctx = locks_inode_context(inode);
 	}
 out:
 	trace_locks_get_lock_context(inode, type, ctx);
@@ -247,7 +247,7 @@ locks_check_ctx_file_list(struct file *f
 void
 locks_free_lock_context(struct inode *inode)
 {
-	struct file_lock_context *ctx = inode->i_flctx;
+	struct file_lock_context *ctx = locks_inode_context(inode);
 
 	if (unlikely(ctx)) {
 		locks_check_ctx_lists(inode);
@@ -891,7 +891,7 @@ posix_test_lock(struct file *filp, struc
 	void *owner;
 	void (*func)(void);
 
-	ctx = smp_load_acquire(&inode->i_flctx);
+	ctx = locks_inode_context(inode);
 	if (!ctx || list_empty_careful(&ctx->flc_posix)) {
 		fl->fl_type = F_UNLCK;
 		return;
@@ -1483,7 +1483,7 @@ int __break_lease(struct inode *inode, u
 	new_fl->fl_flags = type;
 
 	/* typically we will check that ctx is non-NULL before calling */
-	ctx = smp_load_acquire(&inode->i_flctx);
+	ctx = locks_inode_context(inode);
 	if (!ctx) {
 		WARN_ON_ONCE(1);
 		goto free_lock;
@@ -1588,7 +1588,7 @@ void lease_get_mtime(struct inode *inode
 	struct file_lock_context *ctx;
 	struct file_lock *fl;
 
-	ctx = smp_load_acquire(&inode->i_flctx);
+	ctx = locks_inode_context(inode);
 	if (ctx && !list_empty_careful(&ctx->flc_lease)) {
 		spin_lock(&ctx->flc_lock);
 		fl = list_first_entry_or_null(&ctx->flc_lease,
@@ -1634,7 +1634,7 @@ int fcntl_getlease(struct file *filp)
 	int type = F_UNLCK;
 	LIST_HEAD(dispose);
 
-	ctx = smp_load_acquire(&inode->i_flctx);
+	ctx = locks_inode_context(inode);
 	if (ctx && !list_empty_careful(&ctx->flc_lease)) {
 		percpu_down_read(&file_rwsem);
 		spin_lock(&ctx->flc_lock);
@@ -1823,7 +1823,7 @@ static int generic_delete_lease(struct f
 	struct file_lock_context *ctx;
 	LIST_HEAD(dispose);
 
-	ctx = smp_load_acquire(&inode->i_flctx);
+	ctx = locks_inode_context(inode);
 	if (!ctx) {
 		trace_generic_delete_lease(inode, NULL);
 		return error;
@@ -2562,7 +2562,7 @@ void locks_remove_posix(struct file *fil
 	 * posix_lock_file().  Another process could be setting a lock on this
 	 * file at the same time, but we wouldn't remove that lock anyway.
 	 */
-	ctx =  smp_load_acquire(&inode->i_flctx);
+	ctx = locks_inode_context(inode);
 	if (!ctx || list_empty(&ctx->flc_posix))
 		return;
 
@@ -2635,7 +2635,7 @@ void locks_remove_file(struct file *filp
 {
 	struct file_lock_context *ctx;
 
-	ctx = smp_load_acquire(&locks_inode(filp)->i_flctx);
+	ctx = locks_inode_context(locks_inode(filp));
 	if (!ctx)
 		return;
 
@@ -2682,7 +2682,7 @@ bool vfs_inode_has_locks(struct inode *i
 	struct file_lock_context *ctx;
 	bool ret;
 
-	ctx = smp_load_acquire(&inode->i_flctx);
+	ctx = locks_inode_context(inode);
 	if (!ctx)
 		return false;
 
@@ -2863,7 +2863,7 @@ void show_fd_locks(struct seq_file *f,
 	struct file_lock_context *ctx;
 	int id = 0;
 
-	ctx = smp_load_acquire(&inode->i_flctx);
+	ctx = locks_inode_context(inode);
 	if (!ctx)
 		return;
 
include/linux/fs.h +14 -0
@@ -1189,6 +1189,13 @@ extern void show_fd_locks(struct seq_fil
 			 struct file *filp, struct files_struct *files);
 extern bool locks_owner_has_blockers(struct file_lock_context *flctx,
 			fl_owner_t owner);
+
+static inline struct file_lock_context *
+locks_inode_context(const struct inode *inode)
+{
+	return smp_load_acquire(&inode->i_flctx);
+}
+
 #else /* !CONFIG_FILE_LOCKING */
 static inline int fcntl_getlk(struct file *file, unsigned int cmd,
 			      struct flock __user *user)
@@ -1334,6 +1341,13 @@ static inline bool locks_owner_has_block
 {
 	return false;
 }
+
+static inline struct file_lock_context *
+locks_inode_context(const struct inode *inode)
+{
+	return NULL;
+}
+
 #endif /* !CONFIG_FILE_LOCKING */
 
 static inline struct inode *file_inode(const struct file *f)
author maintainer
#166 lore raw

PATCH 162/215 lockd: use locks_inode_context helper

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jeff Layton <[email protected]>

[ Upstream commit 98b41ffe0afdfeaa1439a5d6bd2db4a94277e31b ]

lockd currently doesn't access i_flctx safely. This requires a smp_load_acquire, as the pointer is set via cmpxchg (a release operation).

---
 fs/lockd/svcsubs.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
fs/lockd/svcsubs.c +2 -2
@@ -210,7 +210,7 @@ nlm_traverse_locks(struct nlm_host *host
 {
 	struct inode	 *inode = nlmsvc_file_inode(file);
 	struct file_lock *fl;
-	struct file_lock_context *flctx = inode->i_flctx;
+	struct file_lock_context *flctx = locks_inode_context(inode);
 	struct nlm_host	 *lockhost;
 
 	if (!flctx || list_empty_careful(&flctx->flc_posix))
@@ -265,7 +265,7 @@ nlm_file_inuse(struct nlm_file *file)
 {
 	struct inode	 *inode = nlmsvc_file_inode(file);
 	struct file_lock *fl;
-	struct file_lock_context *flctx = inode->i_flctx;
+	struct file_lock_context *flctx = locks_inode_context(inode);
 
 	if (file->f_count || !list_empty(&file->f_blocks) || file->f_shares)
 		return 1;
author maintainer
#167 lore raw

PATCH 163/215 nfsd: use locks_inode_context helper

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jeff Layton <[email protected]>

[ Upstream commit 77c67530e1f95ac25c7075635f32f04367380894 ]

nfsd currently doesn't access i_flctx safely everywhere. This requires a smp_load_acquire, as the pointer is set via cmpxchg (a release operation).

---
 fs/nfsd/nfs4state.c |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
fs/nfsd/nfs4state.c +3 -3
@@ -4784,7 +4784,7 @@ nfs4_share_conflict(struct svc_fh *curre
 
 static bool nfsd4_deleg_present(const struct inode *inode)
 {
-	struct file_lock_context *ctx = smp_load_acquire(&inode->i_flctx);
+	struct file_lock_context *ctx = locks_inode_context(inode);
 
 	return ctx && !list_empty_careful(&ctx->flc_lease);
 }
@@ -5944,7 +5944,7 @@ nfs4_lockowner_has_blockers(struct nfs4_
 
 	list_for_each_entry(stp, &lo->lo_owner.so_stateids, st_perstateowner) {
 		nf = stp->st_stid.sc_file;
-		ctx = nf->fi_inode->i_flctx;
+		ctx = locks_inode_context(nf->fi_inode);
 		if (!ctx)
 			continue;
 		if (locks_owner_has_blockers(ctx, lo))
@@ -7761,7 +7761,7 @@ check_for_locks(struct nfs4_file *fp, st
 	}
 
 	inode = locks_inode(nf->nf_file);
-	flctx = inode->i_flctx;
+	flctx = locks_inode_context(inode);
 
 	if (flctx && !list_empty_careful(&flctx->flc_posix)) {
 		spin_lock(&flctx->flc_lock);
author maintainer
#168 lore raw

PATCH 164/215 nfsd: fix up the filecache laundrette scheduling

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jeff Layton <[email protected]>

[ Upstream commit 22ae4c114f77b55a4c5036e8f70409a0799a08f8 ]

We don't really care whether there are hashed entries when it comes to scheduling the laundrette. They might all be non-gc entries, after all. We only want to schedule it if there are entries on the LRU.

Switch to using list_lru_count, and move the check into nfsd_file_gc_worker. The other callsite in nfsd_file_put doesn't need to count entries, since it only schedules the laundrette after adding an entry to the LRU.

---
 fs/nfsd/filecache.c |   12 +++++-------
 1 file changed, 5 insertions(+), 7 deletions(-)
fs/nfsd/filecache.c +5 -7
@@ -211,12 +211,9 @@ static const struct rhashtable_params nf
 static void
 nfsd_file_schedule_laundrette(void)
 {
-	if ((atomic_read(&nfsd_file_rhash_tbl.nelems) == 0) ||
-	    test_bit(NFSD_FILE_CACHE_UP, &nfsd_file_flags) == 0)
-		return;
-
-	queue_delayed_work(system_wq, &nfsd_filecache_laundrette,
-			NFSD_LAUNDRETTE_DELAY);
+	if (test_bit(NFSD_FILE_CACHE_UP, &nfsd_file_flags))
+		queue_delayed_work(system_wq, &nfsd_filecache_laundrette,
+				   NFSD_LAUNDRETTE_DELAY);
 }
 
 static void
@@ -614,7 +611,8 @@ static void
 nfsd_file_gc_worker(struct work_struct *work)
 {
 	nfsd_file_gc();
-	nfsd_file_schedule_laundrette();
+	if (list_lru_count(&nfsd_file_lru))
+		nfsd_file_schedule_laundrette();
 }
 
 static unsigned long
author maintainer
#169 lore raw

PATCH 165/215 NFSD: Use struct_size() helper in alloc_session()

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Xiu Jianfeng <[email protected]>

[ Upstream commit 85a0d0c9a58002ef7d1bf5e3ea630f4fbd42a4f0 ]

Use struct_size() helper to simplify the code, no functional changes.

---
 fs/nfsd/nfs4state.c |    9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)
fs/nfsd/nfs4state.c +4 -5
@@ -1834,13 +1834,12 @@ static struct nfsd4_session *alloc_sessi
 	int numslots = fattrs->maxreqs;
 	int slotsize = slot_bytes(fattrs);
 	struct nfsd4_session *new;
-	int mem, i;
+	int i;
 
-	BUILD_BUG_ON(NFSD_MAX_SLOTS_PER_SESSION * sizeof(struct nfsd4_slot *)
-			+ sizeof(struct nfsd4_session) > PAGE_SIZE);
-	mem = numslots * sizeof(struct nfsd4_slot *);
+	BUILD_BUG_ON(struct_size(new, se_slots, NFSD_MAX_SLOTS_PER_SESSION)
+		     > PAGE_SIZE);
 
-	new = kzalloc(sizeof(*new) + mem, GFP_KERNEL);
+	new = kzalloc(struct_size(new, se_slots, numslots), GFP_KERNEL);
 	if (!new)
 		return NULL;
 	/* allocate each struct nfsd4_slot and data cache in one piece */
author maintainer
#170 lore raw

PATCH 166/215 lockd: set missing fl_flags field when retrieving args

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jeff Layton <[email protected]>

[ Upstream commit 75c7940d2a86d3f1b60a0a265478cb8fc887b970 ]

---
 fs/lockd/svc4proc.c |    1 +
 fs/lockd/svcproc.c  |    1 +
 2 files changed, 2 insertions(+)
fs/lockd/svc4proc.c +1 -0
@@ -52,6 +52,7 @@ nlm4svc_retrieve_args(struct svc_rqst *r
 		*filp = file;
 
 		/* Set up the missing parts of the file_lock structure */
+		lock->fl.fl_flags = FL_POSIX;
 		lock->fl.fl_file  = file->f_file[mode];
 		lock->fl.fl_pid = current->tgid;
 		lock->fl.fl_start = (loff_t)lock->lock_start;
fs/lockd/svcproc.c +1 -0
@@ -77,6 +77,7 @@ nlmsvc_retrieve_args(struct svc_rqst *rq
 
 		/* Set up the missing parts of the file_lock structure */
 		mode = lock_to_openmode(&lock->fl);
+		lock->fl.fl_flags = FL_POSIX;
 		lock->fl.fl_file  = file->f_file[mode];
 		lock->fl.fl_pid = current->tgid;
 		lock->fl.fl_lmops = &nlmsvc_lock_operations;
author maintainer
#171 lore raw

PATCH 167/215 lockd: ensure we use the correct file descriptor when unlocking

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jeff Layton <[email protected]>

[ Upstream commit 69efce009f7df888e1fede3cb2913690eb829f52 ]

Shared locks are set on O_RDONLY descriptors and exclusive locks are set on O_WRONLY ones. nlmsvc_unlock however calls vfs_lock_file twice, once for each descriptor, but it doesn't reset fl_file. Ensure that it does.

---
 fs/lockd/svclock.c |   10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)
fs/lockd/svclock.c +6 -4
@@ -659,11 +659,13 @@ nlmsvc_unlock(struct net *net, struct nl
 	nlmsvc_cancel_blocked(net, file, lock);
 
 	lock->fl.fl_type = F_UNLCK;
-	if (file->f_file[O_RDONLY])
-		error = vfs_lock_file(file->f_file[O_RDONLY], F_SETLK,
+	lock->fl.fl_file = file->f_file[O_RDONLY];
+	if (lock->fl.fl_file)
+		error = vfs_lock_file(lock->fl.fl_file, F_SETLK,
 					&lock->fl, NULL);
-	if (file->f_file[O_WRONLY])
-		error = vfs_lock_file(file->f_file[O_WRONLY], F_SETLK,
+	lock->fl.fl_file = file->f_file[O_WRONLY];
+	if (lock->fl.fl_file)
+		error |= vfs_lock_file(lock->fl.fl_file, F_SETLK,
 					&lock->fl, NULL);
 
 	return (error < 0)? nlm_lck_denied_nolocks : nlm_granted;
author maintainer
#172 lore raw

PATCH 168/215 lockd: fix file selection in nlmsvc_cancel_blocked

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jeff Layton <[email protected]>

[ Upstream commit 9f27783b4dd235ef3c8dbf69fc6322777450323c ]

We currently do a lock_to_openmode call based on the arguments from the NLM_UNLOCK call, but that will always set the fl_type of the lock to F_UNLCK, and the O_RDONLY descriptor is always chosen.

Fix it to use the file_lock from the block instead.

---
 fs/lockd/svclock.c |    7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
fs/lockd/svclock.c +4 -3
@@ -699,9 +699,10 @@ nlmsvc_cancel_blocked(struct net *net, s
 	block = nlmsvc_lookup_block(file, lock);
 	mutex_unlock(&file->f_mutex);
 	if (block != NULL) {
-		mode = lock_to_openmode(&lock->fl);
-		vfs_cancel_lock(block->b_file->f_file[mode],
-				&block->b_call->a_args.lock.fl);
+		struct file_lock *fl = &block->b_call->a_args.lock.fl;
+
+		mode = lock_to_openmode(fl);
+		vfs_cancel_lock(block->b_file->f_file[mode], fl);
 		status = nlmsvc_unlink_block(block);
 		nlmsvc_release_block(block);
 	}
author maintainer
#173 lore raw

PATCH 169/215 trace: Relocate event helper files

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Chuck Lever <[email protected]>

[ Upstream commit 247c01ff5f8d66e62a404c91733be52fecb8b7f6 ]

Steven Rostedt says:

Quoted 5 lines

The include/trace/events/ directory should only hold files that are to create events, not headers that hold helper functions.

Can you please move them out of include/trace/events/ as that directory is "special" in the creation of events.

Stable-dep-of: 638593be55c0 ("NFSD: add CB_RECALL_ANY tracepoints")

---
 MAINTAINERS                                           | 7 +++++++
 drivers/infiniband/core/cm_trace.h                    | 2 +-
 drivers/infiniband/core/cma_trace.h                   | 2 +-
 fs/nfs/nfs4trace.h                                    | 6 +++---
 fs/nfs/nfstrace.h                                     | 6 +++---
 include/trace/events/rpcgss.h                         | 2 +-
 include/trace/events/rpcrdma.h                        | 4 ++--
 include/trace/events/sunrpc.h                         | 2 +-
 include/trace/{events => misc}/fs.h                   | 0
 include/trace/{events => misc}/nfs.h                  | 0
 include/trace/{events => misc}/rdma.h                 | 0
 include/trace/{events/sunrpc_base.h => misc/sunrpc.h} | 0
 MAINTAINERS                         |    7 
 drivers/infiniband/core/cm_trace.h  |    2 
 drivers/infiniband/core/cma_trace.h |    2 
 fs/nfs/nfs4trace.h                  |    6 
 fs/nfs/nfstrace.h                   |    6 
 include/trace/events/fs.h           |  122 -----------
 include/trace/events/nfs.h          |  375 ------------------------------------
 include/trace/events/rdma.h         |  168 ----------------
 include/trace/events/rpcgss.h       |    2 
 include/trace/events/rpcrdma.h      |    4 
 include/trace/events/sunrpc.h       |    2 
 include/trace/events/sunrpc_base.h  |   18 -
 include/trace/misc/fs.h             |  122 +++++++++++
 include/trace/misc/nfs.h            |  375 ++++++++++++++++++++++++++++++++++++
 include/trace/misc/rdma.h           |  168 ++++++++++++++++
 include/trace/misc/sunrpc.h         |   18 +
 16 files changed, 702 insertions(+), 695 deletions(-)
 rename include/trace/{events => misc}/fs.h (100%)
 rename include/trace/{events => misc}/nfs.h (100%)
 rename include/trace/{events => misc}/rdma.h (100%)
 rename include/trace/{events/sunrpc_base.h => misc/sunrpc.h} (100%)
MAINTAINERS +7 -0
@@ -10051,6 +10051,7 @@ F:	drivers/infiniband/
 F:	include/rdma/
 F:	include/trace/events/ib_mad.h
 F:	include/trace/events/ib_umad.h
+F:	include/trace/misc/rdma.h
 F:	include/uapi/linux/if_infiniband.h
 F:	include/uapi/rdma/
 F:	samples/bpf/ibumad_kern.c
@@ -11139,6 +11140,12 @@ F:	fs/nfs_common/
 F:	fs/nfsd/
 F:	include/linux/lockd/
 F:	include/linux/sunrpc/
+F:	include/trace/events/rpcgss.h
+F:	include/trace/events/rpcrdma.h
+F:	include/trace/events/sunrpc.h
+F:	include/trace/misc/fs.h
+F:	include/trace/misc/nfs.h
+F:	include/trace/misc/sunrpc.h
 F:	include/uapi/linux/nfsd/
 F:	include/uapi/linux/sunrpc/
 F:	net/sunrpc/
drivers/infiniband/core/cm_trace.h +1 -1
@@ -16,7 +16,7 @@
 
 #include <linux/tracepoint.h>
 #include <rdma/ib_cm.h>
-#include <trace/events/rdma.h>
+#include <trace/misc/rdma.h>
 
 /*
  * enum ib_cm_state, from include/rdma/ib_cm.h
drivers/infiniband/core/cma_trace.h +1 -1
@@ -15,7 +15,7 @@
 #define _TRACE_RDMA_CMA_H
 
 #include <linux/tracepoint.h>
-#include <trace/events/rdma.h>
+#include <trace/misc/rdma.h>
 
 
 DECLARE_EVENT_CLASS(cma_fsm_class,
fs/nfs/nfs4trace.h +3 -3
@@ -9,10 +9,10 @@
 #define _TRACE_NFS4_H
 
 #include <linux/tracepoint.h>
-#include <trace/events/sunrpc_base.h>
+#include <trace/misc/sunrpc.h>
 
-#include <trace/events/fs.h>
-#include <trace/events/nfs.h>
+#include <trace/misc/fs.h>
+#include <trace/misc/nfs.h>
 
 #define show_nfs_fattr_flags(valid) \
 	__print_flags((unsigned long)valid, "|", \
fs/nfs/nfstrace.h +3 -3
@@ -11,9 +11,9 @@
 #include <linux/tracepoint.h>
 #include <linux/iversion.h>
 
-#include <trace/events/fs.h>
-#include <trace/events/nfs.h>
-#include <trace/events/sunrpc_base.h>
+#include <trace/misc/fs.h>
+#include <trace/misc/nfs.h>
+#include <trace/misc/sunrpc.h>
 
 #define nfs_show_cache_validity(v) \
 	__print_flags(v, "|", \
include/trace/events/fs.h deleted +0 -122

122 lines not on this page · raw message

include/trace/events/nfs.h deleted +0 -375

375 lines not on this page · raw message

include/trace/events/rdma.h deleted +0 -168

168 lines not on this page · raw message

include/trace/events/rpcgss.h +1 -1
@@ -13,7 +13,7 @@
 
 #include <linux/tracepoint.h>
 
-#include <trace/events/sunrpc_base.h>
+#include <trace/misc/sunrpc.h>
 
 /**
  ** GSS-API related trace events
include/trace/events/rpcrdma.h +2 -2
@@ -15,8 +15,8 @@
 #include <linux/tracepoint.h>
 #include <rdma/ib_cm.h>
 
-#include <trace/events/rdma.h>
-#include <trace/events/sunrpc_base.h>
+#include <trace/misc/rdma.h>
+#include <trace/misc/sunrpc.h>
 
 /**
  ** Event classes
include/trace/events/sunrpc.h +1 -1
@@ -14,7 +14,7 @@
 #include <linux/net.h>
 #include <linux/tracepoint.h>
 
-#include <trace/events/sunrpc_base.h>
+#include <trace/misc/sunrpc.h>
 
 TRACE_DEFINE_ENUM(SOCK_STREAM);
 TRACE_DEFINE_ENUM(SOCK_DGRAM);
include/trace/events/sunrpc_base.h deleted +0 -18
@@ -1,18 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-/*
- * Copyright (c) 2021 Oracle and/or its affiliates.
- *
- * Common types and format specifiers for sunrpc.
- */
-
-#if !defined(_TRACE_SUNRPC_BASE_H)
-#define _TRACE_SUNRPC_BASE_H
-
-#include <linux/tracepoint.h>
-
-#define SUNRPC_TRACE_PID_SPECIFIER	"%08x"
-#define SUNRPC_TRACE_CLID_SPECIFIER	"%08x"
-#define SUNRPC_TRACE_TASK_SPECIFIER \
-	"task:" SUNRPC_TRACE_PID_SPECIFIER "@" SUNRPC_TRACE_CLID_SPECIFIER
-
-#endif /* _TRACE_SUNRPC_BASE_H */
include/trace/misc/fs.h new file +122 -0

122 lines not on this page · raw message

include/trace/misc/nfs.h new file +375 -0

375 lines not on this page · raw message

include/trace/misc/rdma.h new file +168 -0

168 lines not on this page · raw message

include/trace/misc/sunrpc.h new file +18 -0
@@ -0,0 +1,18 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Copyright (c) 2021 Oracle and/or its affiliates.
+ *
+ * Common types and format specifiers for sunrpc.
+ */
+
+#if !defined(_TRACE_SUNRPC_BASE_H)
+#define _TRACE_SUNRPC_BASE_H
+
+#include <linux/tracepoint.h>
+
+#define SUNRPC_TRACE_PID_SPECIFIER	"%08x"
+#define SUNRPC_TRACE_CLID_SPECIFIER	"%08x"
+#define SUNRPC_TRACE_TASK_SPECIFIER \
+	"task:" SUNRPC_TRACE_PID_SPECIFIER "@" SUNRPC_TRACE_CLID_SPECIFIER
+
+#endif /* _TRACE_SUNRPC_BASE_H */
author maintainer
#174 lore raw

PATCH 170/215 NFSD: refactoring courtesy_client_reaper to a generic low memory shrinker

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Dai Ngo <[email protected]>

[ Upstream commit a1049eb47f20b9eabf9afb218578fff16b4baca6 ]

Refactoring courtesy_client_reaper to generic low memory shrinker so it can be used for other purposes.

---
 fs/nfsd/nfs4state.c |   25 ++++++++++++++++---------
 1 file changed, 16 insertions(+), 9 deletions(-)
fs/nfsd/nfs4state.c +16 -9
@@ -4362,7 +4362,7 @@ out:
 }
 
 static unsigned long
-nfsd_courtesy_client_count(struct shrinker *shrink, struct shrink_control *sc)
+nfsd4_state_shrinker_count(struct shrinker *shrink, struct shrink_control *sc)
 {
 	int cnt;
 	struct nfsd_net *nn = container_of(shrink,
@@ -4375,7 +4375,7 @@ nfsd_courtesy_client_count(struct shrink
 }
 
 static unsigned long
-nfsd_courtesy_client_scan(struct shrinker *shrink, struct shrink_control *sc)
+nfsd4_state_shrinker_scan(struct shrinker *shrink, struct shrink_control *sc)
 {
 	return SHRINK_STOP;
 }
@@ -4402,8 +4402,8 @@ nfsd4_init_leases_net(struct nfsd_net *n
 	nn->nfs4_max_clients = max_t(int, max_clients, NFS4_CLIENTS_PER_GB);
 
 	atomic_set(&nn->nfsd_courtesy_clients, 0);
-	nn->nfsd_client_shrinker.scan_objects = nfsd_courtesy_client_scan;
-	nn->nfsd_client_shrinker.count_objects = nfsd_courtesy_client_count;
+	nn->nfsd_client_shrinker.scan_objects = nfsd4_state_shrinker_scan;
+	nn->nfsd_client_shrinker.count_objects = nfsd4_state_shrinker_count;
 	nn->nfsd_client_shrinker.seeks = DEFAULT_SEEKS;
 	return register_shrinker(&nn->nfsd_client_shrinker, "nfsd-client");
 }
@@ -6171,17 +6171,24 @@ laundromat_main(struct work_struct *laun
 }
 
 static void
-courtesy_client_reaper(struct work_struct *reaper)
+courtesy_client_reaper(struct nfsd_net *nn)
 {
 	struct list_head reaplist;
-	struct delayed_work *dwork = to_delayed_work(reaper);
-	struct nfsd_net *nn = container_of(dwork, struct nfsd_net,
-					nfsd_shrinker_work);
 
 	nfs4_get_courtesy_client_reaplist(nn, &reaplist);
 	nfs4_process_client_reaplist(&reaplist);
 }
 
+static void
+nfsd4_state_shrinker_worker(struct work_struct *work)
+{
+	struct delayed_work *dwork = to_delayed_work(work);
+	struct nfsd_net *nn = container_of(dwork, struct nfsd_net,
+				nfsd_shrinker_work);
+
+	courtesy_client_reaper(nn);
+}
+
 static inline __be32 nfs4_check_fh(struct svc_fh *fhp, struct nfs4_stid *stp)
 {
 	if (!fh_match(&fhp->fh_handle, &stp->sc_file->fi_fhandle))
@@ -8007,7 +8014,7 @@ static int nfs4_state_create_net(struct
 	INIT_LIST_HEAD(&nn->blocked_locks_lru);
 
 	INIT_DELAYED_WORK(&nn->laundromat_work, laundromat_main);
-	INIT_DELAYED_WORK(&nn->nfsd_shrinker_work, courtesy_client_reaper);
+	INIT_DELAYED_WORK(&nn->nfsd_shrinker_work, nfsd4_state_shrinker_worker);
 	get_net(net);
 
 	return 0;
author maintainer
#175 lore raw

PATCH 171/215 NFSD: add support for sending CB_RECALL_ANY

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Dai Ngo <[email protected]>

[ Upstream commit 3959066b697b5dfbb7141124ae9665337d4bc638 ]

Add XDR encode and decode function for CB_RECALL_ANY.

---
 fs/nfsd/nfs4callback.c |   72 +++++++++++++++++++++++++++++++++++++++++++++++++
 fs/nfsd/state.h        |    1 
 fs/nfsd/xdr4.h         |    5 +++
 fs/nfsd/xdr4cb.h       |    6 ++++
 4 files changed, 84 insertions(+)
fs/nfsd/nfs4callback.c +72 -0
@@ -76,6 +76,17 @@ static __be32 *xdr_encode_empty_array(__
  * 1 Protocol"
  */
 
+static void encode_uint32(struct xdr_stream *xdr, u32 n)
+{
+	WARN_ON_ONCE(xdr_stream_encode_u32(xdr, n) < 0);
+}
+
+static void encode_bitmap4(struct xdr_stream *xdr, const __u32 *bitmap,
+			   size_t len)
+{
+	WARN_ON_ONCE(xdr_stream_encode_uint32_array(xdr, bitmap, len) < 0);
+}
+
 /*
  *	nfs_cb_opnum4
  *
@@ -329,6 +340,24 @@ static void encode_cb_recall4args(struct
 }
 
 /*
+ * CB_RECALLANY4args
+ *
+ *	struct CB_RECALLANY4args {
+ *		uint32_t	craa_objects_to_keep;
+ *		bitmap4		craa_type_mask;
+ *	};
+ */
+static void
+encode_cb_recallany4args(struct xdr_stream *xdr,
+	struct nfs4_cb_compound_hdr *hdr, struct nfsd4_cb_recall_any *ra)
+{
+	encode_nfs_cb_opnum4(xdr, OP_CB_RECALL_ANY);
+	encode_uint32(xdr, ra->ra_keep);
+	encode_bitmap4(xdr, ra->ra_bmval, ARRAY_SIZE(ra->ra_bmval));
+	hdr->nops++;
+}
+
+/*
  * CB_SEQUENCE4args
  *
  *	struct CB_SEQUENCE4args {
@@ -482,6 +511,26 @@ static void nfs4_xdr_enc_cb_recall(struc
 	encode_cb_nops(&hdr);
 }
 
+/*
+ * 20.6. Operation 8: CB_RECALL_ANY - Keep Any N Recallable Objects
+ */
+static void
+nfs4_xdr_enc_cb_recall_any(struct rpc_rqst *req,
+		struct xdr_stream *xdr, const void *data)
+{
+	const struct nfsd4_callback *cb = data;
+	struct nfsd4_cb_recall_any *ra;
+	struct nfs4_cb_compound_hdr hdr = {
+		.ident = cb->cb_clp->cl_cb_ident,
+		.minorversion = cb->cb_clp->cl_minorversion,
+	};
+
+	ra = container_of(cb, struct nfsd4_cb_recall_any, ra_cb);
+	encode_cb_compound4args(xdr, &hdr);
+	encode_cb_sequence4args(xdr, cb, &hdr);
+	encode_cb_recallany4args(xdr, &hdr, ra);
+	encode_cb_nops(&hdr);
+}
 
 /*
  * NFSv4.0 and NFSv4.1 XDR decode functions
@@ -520,6 +569,28 @@ static int nfs4_xdr_dec_cb_recall(struct
 	return decode_cb_op_status(xdr, OP_CB_RECALL, &cb->cb_status);
 }
 
+/*
+ * 20.6. Operation 8: CB_RECALL_ANY - Keep Any N Recallable Objects
+ */
+static int
+nfs4_xdr_dec_cb_recall_any(struct rpc_rqst *rqstp,
+				  struct xdr_stream *xdr,
+				  void *data)
+{
+	struct nfsd4_callback *cb = data;
+	struct nfs4_cb_compound_hdr hdr;
+	int status;
+
+	status = decode_cb_compound4res(xdr, &hdr);
+	if (unlikely(status))
+		return status;
+	status = decode_cb_sequence4res(xdr, cb);
+	if (unlikely(status || cb->cb_seq_status))
+		return status;
+	status =  decode_cb_op_status(xdr, OP_CB_RECALL_ANY, &cb->cb_status);
+	return status;
+}
+
 #ifdef CONFIG_NFSD_PNFS
 /*
  * CB_LAYOUTRECALL4args
@@ -783,6 +854,7 @@ static const struct rpc_procinfo nfs4_cb
 #endif
 	PROC(CB_NOTIFY_LOCK,	COMPOUND,	cb_notify_lock,	cb_notify_lock),
 	PROC(CB_OFFLOAD,	COMPOUND,	cb_offload,	cb_offload),
+	PROC(CB_RECALL_ANY,	COMPOUND,	cb_recall_any,	cb_recall_any),
 };
 
 static unsigned int nfs4_cb_counts[ARRAY_SIZE(nfs4_cb_procedures)];
fs/nfsd/state.h +1 -0
@@ -636,6 +636,7 @@ enum nfsd4_cb_op {
 	NFSPROC4_CLNT_CB_OFFLOAD,
 	NFSPROC4_CLNT_CB_SEQUENCE,
 	NFSPROC4_CLNT_CB_NOTIFY_LOCK,
+	NFSPROC4_CLNT_CB_RECALL_ANY,
 };
 
 /* Returns true iff a is later than b: */
fs/nfsd/xdr4.h +5 -0
@@ -896,5 +896,10 @@ struct nfsd4_operation {
 			union nfsd4_op_u *);
 };
 
+struct nfsd4_cb_recall_any {
+	struct nfsd4_callback	ra_cb;
+	u32			ra_keep;
+	u32			ra_bmval[1];
+};
 
 #endif
fs/nfsd/xdr4cb.h +6 -0
@@ -48,3 +48,9 @@
 #define NFS4_dec_cb_offload_sz		(cb_compound_dec_hdr_sz  +      \
 					cb_sequence_dec_sz +            \
 					op_dec_sz)
+#define NFS4_enc_cb_recall_any_sz	(cb_compound_enc_hdr_sz +       \
+					cb_sequence_enc_sz +            \
+					1 + 1 + 1)
+#define NFS4_dec_cb_recall_any_sz	(cb_compound_dec_hdr_sz  +      \
+					cb_sequence_dec_sz +            \
+					op_dec_sz)
author maintainer
#176 lore raw

PATCH 172/215 NFSD: add delegation reaper to react to low memory condition

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Dai Ngo <[email protected]>

[ Upstream commit 44df6f439a1790a5f602e3842879efa88f346672 ]

The delegation reaper is called by nfsd memory shrinker's on the 'count' callback. It scans the client list and sends the courtesy CB_RECALL_ANY to the clients that hold delegations.

To avoid flooding the clients with CB_RECALL_ANY requests, the delegation reaper sends only one CB_RECALL_ANY request to each client per 5 seconds.

[ cel: moved definition of RCA4_TYPE_MASK_RDATA_DLG ]
---
 fs/nfsd/nfs4state.c  |   88 ++++++++++++++++++++++++++++++++++++++++++++++++---
 fs/nfsd/state.h      |    5 ++
 include/linux/nfs4.h |   13 +++++++
 3 files changed, 102 insertions(+), 4 deletions(-)
fs/nfsd/nfs4state.c +84 -4
@@ -2145,6 +2145,7 @@ static void __free_client(struct kref *k
 	kfree(clp->cl_nii_domain.data);
 	kfree(clp->cl_nii_name.data);
 	idr_destroy(&clp->cl_stateids);
+	kfree(clp->cl_ra);
 	kmem_cache_free(client_slab, clp);
 }
 
@@ -2872,6 +2873,36 @@ static const struct tree_descr client_fi
 	[3] = {""},
 };
 
+static int
+nfsd4_cb_recall_any_done(struct nfsd4_callback *cb,
+				struct rpc_task *task)
+{
+	switch (task->tk_status) {
+	case -NFS4ERR_DELAY:
+		rpc_delay(task, 2 * HZ);
+		return 0;
+	default:
+		return 1;
+	}
+}
+
+static void
+nfsd4_cb_recall_any_release(struct nfsd4_callback *cb)
+{
+	struct nfs4_client *clp = cb->cb_clp;
+	struct nfsd_net *nn = net_generic(clp->net, nfsd_net_id);
+
+	spin_lock(&nn->client_lock);
+	clear_bit(NFSD4_CLIENT_CB_RECALL_ANY, &clp->cl_flags);
+	put_client_renew_locked(clp);
+	spin_unlock(&nn->client_lock);
+}
+
+static const struct nfsd4_callback_ops nfsd4_cb_recall_any_ops = {
+	.done		= nfsd4_cb_recall_any_done,
+	.release	= nfsd4_cb_recall_any_release,
+};
+
 static struct nfs4_client *create_client(struct xdr_netobj name,
 		struct svc_rqst *rqstp, nfs4_verifier *verf)
 {
@@ -2909,6 +2940,14 @@ static struct nfs4_client *create_client
 		free_client(clp);
 		return NULL;
 	}
+	clp->cl_ra = kzalloc(sizeof(*clp->cl_ra), GFP_KERNEL);
+	if (!clp->cl_ra) {
+		free_client(clp);
+		return NULL;
+	}
+	clp->cl_ra_time = 0;
+	nfsd4_init_cb(&clp->cl_ra->ra_cb, clp, &nfsd4_cb_recall_any_ops,
+			NFSPROC4_CLNT_CB_RECALL_ANY);
 	return clp;
 }
 
@@ -4364,14 +4403,16 @@ out:
 static unsigned long
 nfsd4_state_shrinker_count(struct shrinker *shrink, struct shrink_control *sc)
 {
-	int cnt;
+	int count;
 	struct nfsd_net *nn = container_of(shrink,
 			struct nfsd_net, nfsd_client_shrinker);
 
-	cnt = atomic_read(&nn->nfsd_courtesy_clients);
-	if (cnt > 0)
+	count = atomic_read(&nn->nfsd_courtesy_clients);
+	if (!count)
+		count = atomic_long_read(&num_delegations);
+	if (count)
 		mod_delayed_work(laundry_wq, &nn->nfsd_shrinker_work, 0);
-	return (unsigned long)cnt;
+	return (unsigned long)count;
 }
 
 static unsigned long
@@ -6180,6 +6221,44 @@ courtesy_client_reaper(struct nfsd_net *
 }
 
 static void
+deleg_reaper(struct nfsd_net *nn)
+{
+	struct list_head *pos, *next;
+	struct nfs4_client *clp;
+	struct list_head cblist;
+
+	INIT_LIST_HEAD(&cblist);
+	spin_lock(&nn->client_lock);
+	list_for_each_safe(pos, next, &nn->client_lru) {
+		clp = list_entry(pos, struct nfs4_client, cl_lru);
+		if (clp->cl_state != NFSD4_ACTIVE ||
+			list_empty(&clp->cl_delegations) ||
+			atomic_read(&clp->cl_delegs_in_recall) ||
+			test_bit(NFSD4_CLIENT_CB_RECALL_ANY, &clp->cl_flags) ||
+			(ktime_get_boottime_seconds() -
+				clp->cl_ra_time < 5)) {
+			continue;
+		}
+		list_add(&clp->cl_ra_cblist, &cblist);
+
+		/* release in nfsd4_cb_recall_any_release */
+		atomic_inc(&clp->cl_rpc_users);
+		set_bit(NFSD4_CLIENT_CB_RECALL_ANY, &clp->cl_flags);
+		clp->cl_ra_time = ktime_get_boottime_seconds();
+	}
+	spin_unlock(&nn->client_lock);
+
+	while (!list_empty(&cblist)) {
+		clp = list_first_entry(&cblist, struct nfs4_client,
+					cl_ra_cblist);
+		list_del_init(&clp->cl_ra_cblist);
+		clp->cl_ra->ra_keep = 0;
+		clp->cl_ra->ra_bmval[0] = BIT(RCA4_TYPE_MASK_RDATA_DLG);
+		nfsd4_run_cb(&clp->cl_ra->ra_cb);
+	}
+}
+
+static void
 nfsd4_state_shrinker_worker(struct work_struct *work)
 {
 	struct delayed_work *dwork = to_delayed_work(work);
@@ -6187,6 +6266,7 @@ nfsd4_state_shrinker_worker(struct work_
 				nfsd_shrinker_work);
 
 	courtesy_client_reaper(nn);
+	deleg_reaper(nn);
 }
 
 static inline __be32 nfs4_check_fh(struct svc_fh *fhp, struct nfs4_stid *stp)
fs/nfsd/state.h +5 -0
@@ -368,6 +368,7 @@ struct nfs4_client {
 #define NFSD4_CLIENT_UPCALL_LOCK	(5)	/* upcall serialization */
 #define NFSD4_CLIENT_CB_FLAG_MASK	(1 << NFSD4_CLIENT_CB_UPDATE | \
 					 1 << NFSD4_CLIENT_CB_KILL)
+#define NFSD4_CLIENT_CB_RECALL_ANY	(6)
 	unsigned long		cl_flags;
 	const struct cred	*cl_cb_cred;
 	struct rpc_clnt		*cl_cb_client;
@@ -411,6 +412,10 @@ struct nfs4_client {
 
 	unsigned int		cl_state;
 	atomic_t		cl_delegs_in_recall;
+
+	struct nfsd4_cb_recall_any	*cl_ra;
+	time64_t		cl_ra_time;
+	struct list_head	cl_ra_cblist;
 };
 
 /* struct nfs4_client_reset
include/linux/nfs4.h +13 -0
@@ -732,4 +732,17 @@ enum nfs4_setxattr_options {
 	SETXATTR4_CREATE	= 1,
 	SETXATTR4_REPLACE	= 2,
 };
+
+enum {
+	RCA4_TYPE_MASK_RDATA_DLG	= 0,
+	RCA4_TYPE_MASK_WDATA_DLG	= 1,
+	RCA4_TYPE_MASK_DIR_DLG		= 2,
+	RCA4_TYPE_MASK_FILE_LAYOUT	= 3,
+	RCA4_TYPE_MASK_BLK_LAYOUT	= 4,
+	RCA4_TYPE_MASK_OBJ_LAYOUT_MIN	= 8,
+	RCA4_TYPE_MASK_OBJ_LAYOUT_MAX	= 9,
+	RCA4_TYPE_MASK_OTHER_LAYOUT_MIN	= 12,
+	RCA4_TYPE_MASK_OTHER_LAYOUT_MAX	= 15,
+};
+
 #endif
author maintainer
#177 lore raw

PATCH 173/215 NFSD: add CB_RECALL_ANY tracepoints

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Dai Ngo <[email protected]>

[ Upstream commit 638593be55c0b37a1930038460a9918215d5c24b ]

Add tracepoints to trace start and end of CB_RECALL_ANY operation.

[ cel: added show_rca_mask() macro ]
---
 fs/nfsd/nfs4state.c      |    2 +
 fs/nfsd/trace.h          |   50 +++++++++++++++++++++++++++++++++++++++++++++++
 include/trace/misc/nfs.h |   12 +++++++++++
 3 files changed, 64 insertions(+)
fs/nfsd/nfs4state.c +2 -0
@@ -2877,6 +2877,7 @@ static int
 nfsd4_cb_recall_any_done(struct nfsd4_callback *cb,
 				struct rpc_task *task)
 {
+	trace_nfsd_cb_recall_any_done(cb, task);
 	switch (task->tk_status) {
 	case -NFS4ERR_DELAY:
 		rpc_delay(task, 2 * HZ);
@@ -6254,6 +6255,7 @@ deleg_reaper(struct nfsd_net *nn)
 		list_del_init(&clp->cl_ra_cblist);
 		clp->cl_ra->ra_keep = 0;
 		clp->cl_ra->ra_bmval[0] = BIT(RCA4_TYPE_MASK_RDATA_DLG);
+		trace_nfsd_cb_recall_any(clp->cl_ra);
 		nfsd4_run_cb(&clp->cl_ra->ra_cb);
 	}
 }
fs/nfsd/trace.h +50 -0

71 lines not on this page · raw message

include/trace/misc/nfs.h +12 -0
@@ -360,6 +360,18 @@ TRACE_DEFINE_ENUM(IOMODE_ANY);
 		{ IOMODE_RW,			"RW" }, \
 		{ IOMODE_ANY,			"ANY" })
 
+#define show_rca_mask(x) \
+	__print_flags(x, "|", \
+		{ BIT(RCA4_TYPE_MASK_RDATA_DLG),	"RDATA_DLG" }, \
+		{ BIT(RCA4_TYPE_MASK_WDATA_DLG),	"WDATA_DLG" }, \
+		{ BIT(RCA4_TYPE_MASK_DIR_DLG),		"DIR_DLG" }, \
+		{ BIT(RCA4_TYPE_MASK_FILE_LAYOUT),	"FILE_LAYOUT" }, \
+		{ BIT(RCA4_TYPE_MASK_BLK_LAYOUT),	"BLK_LAYOUT" }, \
+		{ BIT(RCA4_TYPE_MASK_OBJ_LAYOUT_MIN),	"OBJ_LAYOUT_MIN" }, \
+		{ BIT(RCA4_TYPE_MASK_OBJ_LAYOUT_MAX),	"OBJ_LAYOUT_MAX" }, \
+		{ BIT(RCA4_TYPE_MASK_OTHER_LAYOUT_MIN),	"OTHER_LAYOUT_MIN" }, \
+		{ BIT(RCA4_TYPE_MASK_OTHER_LAYOUT_MAX),	"OTHER_LAYOUT_MAX" })
+
 #define show_nfs4_seq4_status(x) \
 	__print_flags(x, "|", \
 		{ SEQ4_STATUS_CB_PATH_DOWN,		"CB_PATH_DOWN" }, \
author maintainer
#178 lore raw

PATCH 174/215 NFSD: Use only RQ_DROPME to signal the need to drop a reply

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Chuck Lever <[email protected]>

[ Upstream commit 9315564747cb6a570e99196b3a4880fb817635fd ]

Clean up: NFSv2 has the only two usages of rpc_drop_reply in the NFSD code base. Since NFSv2 is going away at some point, replace these in order to simplify the "drop this reply?" check in nfsd_dispatch().

---
 fs/nfsd/nfsproc.c |    4 ++--
 fs/nfsd/nfssvc.c  |    2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)
fs/nfsd/nfsproc.c +2 -2
@@ -211,7 +211,7 @@ nfsd_proc_read(struct svc_rqst *rqstp)
 	if (resp->status == nfs_ok)
 		resp->status = fh_getattr(&resp->fh, &resp->stat);
 	else if (resp->status == nfserr_jukebox)
-		return rpc_drop_reply;
+		__set_bit(RQ_DROPME, &rqstp->rq_flags);
 	return rpc_success;
 }
 
@@ -246,7 +246,7 @@ nfsd_proc_write(struct svc_rqst *rqstp)
 	if (resp->status == nfs_ok)
 		resp->status = fh_getattr(&resp->fh, &resp->stat);
 	else if (resp->status == nfserr_jukebox)
-		return rpc_drop_reply;
+		__set_bit(RQ_DROPME, &rqstp->rq_flags);
 	return rpc_success;
 }
 
fs/nfsd/nfssvc.c +1 -1
@@ -1071,7 +1071,7 @@ int nfsd_dispatch(struct svc_rqst *rqstp
 
 	nfs_reply = xdr_inline_decode(&rqstp->rq_res_stream, 0);
 	*statp = proc->pc_func(rqstp);
-	if (*statp == rpc_drop_reply || test_bit(RQ_DROPME, &rqstp->rq_flags))
+	if (test_bit(RQ_DROPME, &rqstp->rq_flags))
 		goto out_update_drop;
 
 	if (!proc->pc_encode(rqstp, &rqstp->rq_res_stream))
author maintainer
#179 lore raw

PATCH 175/215 NFSD: Avoid clashing function prototypes

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Kees Cook <[email protected]>

[ Upstream commit e78e274eb22d966258a3845acc71d3c5b8ee2ea8 ]

When built with Control Flow Integrity, function prototypes between caller and function declaration must match. These mismatches are visible at compile time with the new -Wcast-function-type-strict in Clang[1].

There were 97 warnings produced by NFS. For example:

fs/nfsd/nfs4xdr.c:2228:17: warning: cast from '__be32 (*)(struct nfsd4_compoundargs *, struct nfsd4_access *)' (aka 'unsigned int (*)(struct nfsd4_compoundargs *, struct nfsd4_access *)') to 'nfsd4_dec' (aka 'unsigned int (*)(struct nfsd4_compoundargs *, void *)') converts to incompatible function type [-Wcast-function-type-strict]
[OP_ACCESS] = (nfsd4_dec)nfsd4_decode_access,
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

The enc/dec callbacks were defined as passing "void *" as the second argument, but were being implicitly cast to a new type. Replace the argument with union nfsd4_op_u, and perform explicit member selection in the function body. There are no resulting binary differences.

Changes were made mechanically using the following Coccinelle script, with minor by-hand fixes for members that didn't already match their existing argument name:

@find@
identifier func;
type T, opsT;
identifier ops, N;
@@

 opsT ops[] = {
        [N] = (T) func,

};

@already_void@
identifier find.func;
identifier name;
@@

func(...,
-void
+union nfsd4_op_u
*name)

 {
        ...

}

@proto depends on !already_void@
identifier find.func;
type T;
identifier name;
position p;
@@

func@p(...,
T name

 ) {
        ...

}

@script:python get_member@
type_name << proto.T;
member;
@@

coccinelle.member = cocci.make_ident(type_name.split("_", 1)[1].split(' ',1)[0])

@convert@
identifier find.func;
type proto.T;
identifier proto.name;
position proto.p;
identifier get_member.member;
@@

func@p(...,
- T name
+ union nfsd4_op_u *u
) {

+       T name = &u->member;
        ...

}

@cast@
identifier find.func;
type T, opsT;
identifier ops, N;
@@

 opsT ops[] = {
        [N] =

- (T)
func,
};

---
 fs/nfsd/nfs4xdr.c |  632 ++++++++++++++++++++++++++++++++----------------------
 1 file changed, 377 insertions(+), 255 deletions(-)
fs/nfsd/nfs4xdr.c +377 -255
@@ -770,16 +770,18 @@ nfsd4_decode_cb_sec(struct nfsd4_compoun
 
 static __be32
 nfsd4_decode_access(struct nfsd4_compoundargs *argp,
-		    struct nfsd4_access *access)
+		    union nfsd4_op_u *u)
 {
+	struct nfsd4_access *access = &u->access;
 	if (xdr_stream_decode_u32(argp->xdr, &access->ac_req_access) < 0)
 		return nfserr_bad_xdr;
 	return nfs_ok;
 }
 
 static __be32
-nfsd4_decode_close(struct nfsd4_compoundargs *argp, struct nfsd4_close *close)
+nfsd4_decode_close(struct nfsd4_compoundargs *argp, union nfsd4_op_u *u)
 {
+	struct nfsd4_close *close = &u->close;
 	if (xdr_stream_decode_u32(argp->xdr, &close->cl_seqid) < 0)
 		return nfserr_bad_xdr;
 	return nfsd4_decode_stateid4(argp, &close->cl_stateid);
@@ -787,8 +789,9 @@ nfsd4_decode_close(struct nfsd4_compound
 
 
 static __be32
-nfsd4_decode_commit(struct nfsd4_compoundargs *argp, struct nfsd4_commit *commit)
+nfsd4_decode_commit(struct nfsd4_compoundargs *argp, union nfsd4_op_u *u)
 {
+	struct nfsd4_commit *commit = &u->commit;
 	if (xdr_stream_decode_u64(argp->xdr, &commit->co_offset) < 0)
 		return nfserr_bad_xdr;
 	if (xdr_stream_decode_u32(argp->xdr, &commit->co_count) < 0)
@@ -798,8 +801,9 @@ nfsd4_decode_commit(struct nfsd4_compoun
 }
 
 static __be32
-nfsd4_decode_create(struct nfsd4_compoundargs *argp, struct nfsd4_create *create)
+nfsd4_decode_create(struct nfsd4_compoundargs *argp, union nfsd4_op_u *u)
 {
+	struct nfsd4_create *create = &u->create;
 	__be32 *p, status;
 
 	memset(create, 0, sizeof(*create));
@@ -844,22 +848,25 @@ nfsd4_decode_create(struct nfsd4_compoun
 }
 
 static inline __be32
-nfsd4_decode_delegreturn(struct nfsd4_compoundargs *argp, struct nfsd4_delegreturn *dr)
+nfsd4_decode_delegreturn(struct nfsd4_compoundargs *argp, union nfsd4_op_u *u)
 {
+	struct nfsd4_delegreturn *dr = &u->delegreturn;
 	return nfsd4_decode_stateid4(argp, &dr->dr_stateid);
 }
 
 static inline __be32
-nfsd4_decode_getattr(struct nfsd4_compoundargs *argp, struct nfsd4_getattr *getattr)
+nfsd4_decode_getattr(struct nfsd4_compoundargs *argp, union nfsd4_op_u *u)
 {
+	struct nfsd4_getattr *getattr = &u->getattr;
 	memset(getattr, 0, sizeof(*getattr));
 	return nfsd4_decode_bitmap4(argp, getattr->ga_bmval,
 				    ARRAY_SIZE(getattr->ga_bmval));
 }
 
 static __be32
-nfsd4_decode_link(struct nfsd4_compoundargs *argp, struct nfsd4_link *link)
+nfsd4_decode_link(struct nfsd4_compoundargs *argp, union nfsd4_op_u *u)
 {
+	struct nfsd4_link *link = &u->link;
 	memset(link, 0, sizeof(*link));
 	return nfsd4_decode_component4(argp, &link->li_name, &link->li_namelen);
 }
@@ -907,8 +914,9 @@ nfsd4_decode_locker4(struct nfsd4_compou
 }
 
 static __be32
-nfsd4_decode_lock(struct nfsd4_compoundargs *argp, struct nfsd4_lock *lock)
+nfsd4_decode_lock(struct nfsd4_compoundargs *argp, union nfsd4_op_u *u)
 {
+	struct nfsd4_lock *lock = &u->lock;
 	memset(lock, 0, sizeof(*lock));
 	if (xdr_stream_decode_u32(argp->xdr, &lock->lk_type) < 0)
 		return nfserr_bad_xdr;
@@ -924,8 +932,9 @@ nfsd4_decode_lock(struct nfsd4_compounda
 }
 
 static __be32
-nfsd4_decode_lockt(struct nfsd4_compoundargs *argp, struct nfsd4_lockt *lockt)
+nfsd4_decode_lockt(struct nfsd4_compoundargs *argp, union nfsd4_op_u *u)
 {
+	struct nfsd4_lockt *lockt = &u->lockt;
 	memset(lockt, 0, sizeof(*lockt));
 	if (xdr_stream_decode_u32(argp->xdr, &lockt->lt_type) < 0)
 		return nfserr_bad_xdr;
@@ -940,8 +949,9 @@ nfsd4_decode_lockt(struct nfsd4_compound
 }
 
 static __be32
-nfsd4_decode_locku(struct nfsd4_compoundargs *argp, struct nfsd4_locku *locku)
+nfsd4_decode_locku(struct nfsd4_compoundargs *argp, union nfsd4_op_u *u)
 {
+	struct nfsd4_locku *locku = &u->locku;
 	__be32 status;
 
 	if (xdr_stream_decode_u32(argp->xdr, &locku->lu_type) < 0)
@@ -962,8 +972,9 @@ nfsd4_decode_locku(struct nfsd4_compound
 }
 
 static __be32
-nfsd4_decode_lookup(struct nfsd4_compoundargs *argp, struct nfsd4_lookup *lookup)
+nfsd4_decode_lookup(struct nfsd4_compoundargs *argp, union nfsd4_op_u *u)
 {
+	struct nfsd4_lookup *lookup = &u->lookup;
 	return nfsd4_decode_component4(argp, &lookup->lo_name, &lookup->lo_len);
 }
 
@@ -1143,8 +1154,9 @@ nfsd4_decode_open_claim4(struct nfsd4_co
 }
 
 static __be32
-nfsd4_decode_open(struct nfsd4_compoundargs *argp, struct nfsd4_open *open)
+nfsd4_decode_open(struct nfsd4_compoundargs *argp, union nfsd4_op_u *u)
 {
+	struct nfsd4_open *open = &u->open;
 	__be32 status;
 	u32 dummy;
 
@@ -1171,8 +1183,10 @@ nfsd4_decode_open(struct nfsd4_compounda
 }
 
 static __be32
-nfsd4_decode_open_confirm(struct nfsd4_compoundargs *argp, struct nfsd4_open_confirm *open_conf)
+nfsd4_decode_open_confirm(struct nfsd4_compoundargs *argp,
+			  union nfsd4_op_u *u)
 {
+	struct nfsd4_open_confirm *open_conf = &u->open_confirm;
 	__be32 status;
 
 	if (argp->minorversion >= 1)
@@ -1190,8 +1204,10 @@ nfsd4_decode_open_confirm(struct nfsd4_c
 }
 
 static __be32
-nfsd4_decode_open_downgrade(struct nfsd4_compoundargs *argp, struct nfsd4_open_downgrade *open_down)
+nfsd4_decode_open_downgrade(struct nfsd4_compoundargs *argp,
+			    union nfsd4_op_u *u)
 {
+	struct nfsd4_open_downgrade *open_down = &u->open_downgrade;
 	__be32 status;
 
 	memset(open_down, 0, sizeof(*open_down));
@@ -1209,8 +1225,9 @@ nfsd4_decode_open_downgrade(struct nfsd4
 }
 
 static __be32
-nfsd4_decode_putfh(struct nfsd4_compoundargs *argp, struct nfsd4_putfh *putfh)
+nfsd4_decode_putfh(struct nfsd4_compoundargs *argp, union nfsd4_op_u *u)
 {
+	struct nfsd4_putfh *putfh = &u->putfh;
 	__be32 *p;
 
 	if (xdr_stream_decode_u32(argp->xdr, &putfh->pf_fhlen) < 0)
@@ -1229,7 +1246,7 @@ nfsd4_decode_putfh(struct nfsd4_compound
 }
 
 static __be32
-nfsd4_decode_putpubfh(struct nfsd4_compoundargs *argp, void *p)
+nfsd4_decode_putpubfh(struct nfsd4_compoundargs *argp, union nfsd4_op_u *p)
 {
 	if (argp->minorversion == 0)
 		return nfs_ok;
@@ -1237,8 +1254,9 @@ nfsd4_decode_putpubfh(struct nfsd4_compo
 }
 
 static __be32
-nfsd4_decode_read(struct nfsd4_compoundargs *argp, struct nfsd4_read *read)
+nfsd4_decode_read(struct nfsd4_compoundargs *argp, union nfsd4_op_u *u)
 {
+	struct nfsd4_read *read = &u->read;
 	__be32 status;
 
 	memset(read, 0, sizeof(*read));
@@ -1254,8 +1272,9 @@ nfsd4_decode_read(struct nfsd4_compounda
 }
 
 static __be32
-nfsd4_decode_readdir(struct nfsd4_compoundargs *argp, struct nfsd4_readdir *readdir)
+nfsd4_decode_readdir(struct nfsd4_compoundargs *argp, union nfsd4_op_u *u)
 {
+	struct nfsd4_readdir *readdir = &u->readdir;
 	__be32 status;
 
 	memset(readdir, 0, sizeof(*readdir));
@@ -1276,15 +1295,17 @@ nfsd4_decode_readdir(struct nfsd4_compou
 }
 
 static __be32
-nfsd4_decode_remove(struct nfsd4_compoundargs *argp, struct nfsd4_remove *remove)
+nfsd4_decode_remove(struct nfsd4_compoundargs *argp, union nfsd4_op_u *u)
 {
+	struct nfsd4_remove *remove = &u->remove;
 	memset(&remove->rm_cinfo, 0, sizeof(remove->rm_cinfo));
 	return nfsd4_decode_component4(argp, &remove->rm_name, &remove->rm_namelen);
 }
 
 static __be32
-nfsd4_decode_rename(struct nfsd4_compoundargs *argp, struct nfsd4_rename *rename)
+nfsd4_decode_rename(struct nfsd4_compoundargs *argp, union nfsd4_op_u *u)
 {
+	struct nfsd4_rename *rename = &u->rename;
 	__be32 status;
 
 	memset(rename, 0, sizeof(*rename));
@@ -1295,22 +1316,25 @@ nfsd4_decode_rename(struct nfsd4_compoun
 }
 
 static __be32
-nfsd4_decode_renew(struct nfsd4_compoundargs *argp, clientid_t *clientid)
+nfsd4_decode_renew(struct nfsd4_compoundargs *argp, union nfsd4_op_u *u)
 {
+	clientid_t *clientid = &u->renew;
 	return nfsd4_decode_clientid4(argp, clientid);
 }
 
 static __be32
 nfsd4_decode_secinfo(struct nfsd4_compoundargs *argp,
-		     struct nfsd4_secinfo *secinfo)
+		     union nfsd4_op_u *u)
 {
+	struct nfsd4_secinfo *secinfo = &u->secinfo;
 	secinfo->si_exp = NULL;
 	return nfsd4_decode_component4(argp, &secinfo->si_name, &secinfo->si_namelen);
 }
 
 static __be32
-nfsd4_decode_setattr(struct nfsd4_compoundargs *argp, struct nfsd4_setattr *setattr)
+nfsd4_decode_setattr(struct nfsd4_compoundargs *argp, union nfsd4_op_u *u)
 {
+	struct nfsd4_setattr *setattr = &u->setattr;
 	__be32 status;
 
 	memset(setattr, 0, sizeof(*setattr));
@@ -1324,8 +1348,9 @@ nfsd4_decode_setattr(struct nfsd4_compou
 }
 
 static __be32
-nfsd4_decode_setclientid(struct nfsd4_compoundargs *argp, struct nfsd4_setclientid *setclientid)
+nfsd4_decode_setclientid(struct nfsd4_compoundargs *argp, union nfsd4_op_u *u)
 {
+	struct nfsd4_setclientid *setclientid = &u->setclientid;
 	__be32 *p, status;
 
 	memset(setclientid, 0, sizeof(*setclientid));
@@ -1367,8 +1392,10 @@ nfsd4_decode_setclientid(struct nfsd4_co
 }
 
 static __be32
-nfsd4_decode_setclientid_confirm(struct nfsd4_compoundargs *argp, struct nfsd4_setclientid_confirm *scd_c)
+nfsd4_decode_setclientid_confirm(struct nfsd4_compoundargs *argp,
+				 union nfsd4_op_u *u)
 {
+	struct nfsd4_setclientid_confirm *scd_c = &u->setclientid_confirm;
 	__be32 status;
 
 	if (argp->minorversion >= 1)
@@ -1382,8 +1409,9 @@ nfsd4_decode_setclientid_confirm(struct
 
 /* Also used for NVERIFY */
 static __be32
-nfsd4_decode_verify(struct nfsd4_compoundargs *argp, struct nfsd4_verify *verify)
+nfsd4_decode_verify(struct nfsd4_compoundargs *argp, union nfsd4_op_u *u)
 {
+	struct nfsd4_verify *verify = &u->verify;
 	__be32 *p, status;
 
 	memset(verify, 0, sizeof(*verify));
@@ -1409,8 +1437,9 @@ nfsd4_decode_verify(struct nfsd4_compoun
 }
 
 static __be32
-nfsd4_decode_write(struct nfsd4_compoundargs *argp, struct nfsd4_write *write)
+nfsd4_decode_write(struct nfsd4_compoundargs *argp, union nfsd4_op_u *u)
 {
+	struct nfsd4_write *write = &u->write;
 	__be32 status;
 
 	status = nfsd4_decode_stateid4(argp, &write->wr_stateid);
@@ -1434,8 +1463,10 @@ nfsd4_decode_write(struct nfsd4_compound
 }
 
 static __be32
-nfsd4_decode_release_lockowner(struct nfsd4_compoundargs *argp, struct nfsd4_release_lockowner *rlockowner)
+nfsd4_decode_release_lockowner(struct nfsd4_compoundargs *argp,
+			       union nfsd4_op_u *u)
 {
+	struct nfsd4_release_lockowner *rlockowner = &u->release_lockowner;
 	__be32 status;
 
 	if (argp->minorversion >= 1)
@@ -1452,16 +1483,20 @@ nfsd4_decode_release_lockowner(struct nf
 	return nfs_ok;
 }
 
-static __be32 nfsd4_decode_backchannel_ctl(struct nfsd4_compoundargs *argp, struct nfsd4_backchannel_ctl *bc)
+static __be32 nfsd4_decode_backchannel_ctl(struct nfsd4_compoundargs *argp,
+					   union nfsd4_op_u *u)
 {
+	struct nfsd4_backchannel_ctl *bc = &u->backchannel_ctl;
 	memset(bc, 0, sizeof(*bc));
 	if (xdr_stream_decode_u32(argp->xdr, &bc->bc_cb_program) < 0)
 		return nfserr_bad_xdr;
 	return nfsd4_decode_cb_sec(argp, &bc->bc_cb_sec);
 }
 
-static __be32 nfsd4_decode_bind_conn_to_session(struct nfsd4_compoundargs *argp, struct nfsd4_bind_conn_to_session *bcts)
+static __be32 nfsd4_decode_bind_conn_to_session(struct nfsd4_compoundargs *argp,
+						union nfsd4_op_u *u)
 {
+	struct nfsd4_bind_conn_to_session *bcts = &u->bind_conn_to_session;
 	u32 use_conn_in_rdma_mode;
 	__be32 status;
 
@@ -1603,8 +1638,9 @@ nfsd4_decode_nfs_impl_id4(struct nfsd4_c
 
 static __be32
 nfsd4_decode_exchange_id(struct nfsd4_compoundargs *argp,
-			 struct nfsd4_exchange_id *exid)
+			 union nfsd4_op_u *u)
 {
+	struct nfsd4_exchange_id *exid = &u->exchange_id;
 	__be32 status;
 
 	memset(exid, 0, sizeof(*exid));
@@ -1656,8 +1692,9 @@ nfsd4_decode_channel_attrs4(struct nfsd4
 
 static __be32
 nfsd4_decode_create_session(struct nfsd4_compoundargs *argp,
-			    struct nfsd4_create_session *sess)
+			    union nfsd4_op_u *u)
 {
+	struct nfsd4_create_session *sess = &u->create_session;
 	__be32 status;
 
 	memset(sess, 0, sizeof(*sess));
@@ -1681,23 +1718,26 @@ nfsd4_decode_create_session(struct nfsd4
 
 static __be32
 nfsd4_decode_destroy_session(struct nfsd4_compoundargs *argp,
-			     struct nfsd4_destroy_session *destroy_session)
+			     union nfsd4_op_u *u)
 {
+	struct nfsd4_destroy_session *destroy_session = &u->destroy_session;
 	return nfsd4_decode_sessionid4(argp, &destroy_session->sessionid);
 }
 
 static __be32
 nfsd4_decode_free_stateid(struct nfsd4_compoundargs *argp,
-			  struct nfsd4_free_stateid *free_stateid)
+			  union nfsd4_op_u *u)
 {
+	struct nfsd4_free_stateid *free_stateid = &u->free_stateid;
 	return nfsd4_decode_stateid4(argp, &free_stateid->fr_stateid);
 }
 
 #ifdef CONFIG_NFSD_PNFS
 static __be32
 nfsd4_decode_getdeviceinfo(struct nfsd4_compoundargs *argp,
-		struct nfsd4_getdeviceinfo *gdev)
+		union nfsd4_op_u *u)
 {
+	struct nfsd4_getdeviceinfo *gdev = &u->getdeviceinfo;
 	__be32 status;
 
 	memset(gdev, 0, sizeof(*gdev));
@@ -1717,8 +1757,9 @@ nfsd4_decode_getdeviceinfo(struct nfsd4_
 
 static __be32
 nfsd4_decode_layoutcommit(struct nfsd4_compoundargs *argp,
-			  struct nfsd4_layoutcommit *lcp)
+			  union nfsd4_op_u *u)
 {
+	struct nfsd4_layoutcommit *lcp = &u->layoutcommit;
 	__be32 *p, status;
 
 	memset(lcp, 0, sizeof(*lcp));
@@ -1753,8 +1794,9 @@ nfsd4_decode_layoutcommit(struct nfsd4_c
 
 static __be32
 nfsd4_decode_layoutget(struct nfsd4_compoundargs *argp,
-		struct nfsd4_layoutget *lgp)
+		union nfsd4_op_u *u)
 {
+	struct nfsd4_layoutget *lgp = &u->layoutget;
 	__be32 status;
 
 	memset(lgp, 0, sizeof(*lgp));
@@ -1781,8 +1823,9 @@ nfsd4_decode_layoutget(struct nfsd4_comp
 
 static __be32
 nfsd4_decode_layoutreturn(struct nfsd4_compoundargs *argp,
-		struct nfsd4_layoutreturn *lrp)
+		union nfsd4_op_u *u)
 {
+	struct nfsd4_layoutreturn *lrp = &u->layoutreturn;
 	memset(lrp, 0, sizeof(*lrp));
 	if (xdr_stream_decode_bool(argp->xdr, &lrp->lr_reclaim) < 0)
 		return nfserr_bad_xdr;
@@ -1795,8 +1838,9 @@ nfsd4_decode_layoutreturn(struct nfsd4_c
 #endif /* CONFIG_NFSD_PNFS */
 
 static __be32 nfsd4_decode_secinfo_no_name(struct nfsd4_compoundargs *argp,
-					   struct nfsd4_secinfo_no_name *sin)
+					   union nfsd4_op_u *u)
 {
+	struct nfsd4_secinfo_no_name *sin = &u->secinfo_no_name;
 	if (xdr_stream_decode_u32(argp->xdr, &sin->sin_style) < 0)
 		return nfserr_bad_xdr;
 
@@ -1806,8 +1850,9 @@ static __be32 nfsd4_decode_secinfo_no_na
 
 static __be32
 nfsd4_decode_sequence(struct nfsd4_compoundargs *argp,
-		      struct nfsd4_sequence *seq)
+		      union nfsd4_op_u *u)
 {
+	struct nfsd4_sequence *seq = &u->sequence;
 	__be32 *p, status;
 
 	status = nfsd4_decode_sessionid4(argp, &seq->sessionid);
@@ -1826,8 +1871,10 @@ nfsd4_decode_sequence(struct nfsd4_compo
 }
 
 static __be32
-nfsd4_decode_test_stateid(struct nfsd4_compoundargs *argp, struct nfsd4_test_stateid *test_stateid)
+nfsd4_decode_test_stateid(struct nfsd4_compoundargs *argp,
+			  union nfsd4_op_u *u)
 {
+	struct nfsd4_test_stateid *test_stateid = &u->test_stateid;
 	struct nfsd4_test_stateid_id *stateid;
 	__be32 status;
 	u32 i;
@@ -1852,14 +1899,16 @@ nfsd4_decode_test_stateid(struct nfsd4_c
 }
 
 static __be32 nfsd4_decode_destroy_clientid(struct nfsd4_compoundargs *argp,
-					    struct nfsd4_destroy_clientid *dc)
+					    union nfsd4_op_u *u)
 {
+	struct nfsd4_destroy_clientid *dc = &u->destroy_clientid;
 	return nfsd4_decode_clientid4(argp, &dc->clientid);
 }
 
 static __be32 nfsd4_decode_reclaim_complete(struct nfsd4_compoundargs *argp,
-					    struct nfsd4_reclaim_complete *rc)
+					    union nfsd4_op_u *u)
 {
+	struct nfsd4_reclaim_complete *rc = &u->reclaim_complete;
 	if (xdr_stream_decode_bool(argp->xdr, &rc->rca_one_fs) < 0)
 		return nfserr_bad_xdr;
 	return nfs_ok;
@@ -1867,8 +1916,9 @@ static __be32 nfsd4_decode_reclaim_compl
 
 static __be32
 nfsd4_decode_fallocate(struct nfsd4_compoundargs *argp,
-		       struct nfsd4_fallocate *fallocate)
+		       union nfsd4_op_u *u)
 {
+	struct nfsd4_fallocate *fallocate = &u->allocate;
 	__be32 status;
 
 	status = nfsd4_decode_stateid4(argp, &fallocate->falloc_stateid);
@@ -1924,8 +1974,9 @@ static __be32 nfsd4_decode_nl4_server(st
 }
 
 static __be32
-nfsd4_decode_copy(struct nfsd4_compoundargs *argp, struct nfsd4_copy *copy)
+nfsd4_decode_copy(struct nfsd4_compoundargs *argp, union nfsd4_op_u *u)
 {
+	struct nfsd4_copy *copy = &u->copy;
 	u32 consecutive, i, count, sync;
 	struct nl4_server *ns_dummy;
 	__be32 status;
@@ -1982,8 +2033,9 @@ nfsd4_decode_copy(struct nfsd4_compounda
 
 static __be32
 nfsd4_decode_copy_notify(struct nfsd4_compoundargs *argp,
-			 struct nfsd4_copy_notify *cn)
+			 union nfsd4_op_u *u)
 {
+	struct nfsd4_copy_notify *cn = &u->copy_notify;
 	__be32 status;
 
 	memset(cn, 0, sizeof(*cn));
@@ -2002,16 +2054,18 @@ nfsd4_decode_copy_notify(struct nfsd4_co
 
 static __be32
 nfsd4_decode_offload_status(struct nfsd4_compoundargs *argp,
-			    struct nfsd4_offload_status *os)
+			    union nfsd4_op_u *u)
 {
+	struct nfsd4_offload_status *os = &u->offload_status;
 	os->count = 0;
 	os->status = 0;
 	return nfsd4_decode_stateid4(argp, &os->stateid);
 }
 
 static __be32
-nfsd4_decode_seek(struct nfsd4_compoundargs *argp, struct nfsd4_seek *seek)
+nfsd4_decode_seek(struct nfsd4_compoundargs *argp, union nfsd4_op_u *u)
 {
+	struct nfsd4_seek *seek = &u->seek;
 	__be32 status;
 
 	status = nfsd4_decode_stateid4(argp, &seek->seek_stateid);
@@ -2028,8 +2082,9 @@ nfsd4_decode_seek(struct nfsd4_compounda
 }
 
 static __be32
-nfsd4_decode_clone(struct nfsd4_compoundargs *argp, struct nfsd4_clone *clone)
+nfsd4_decode_clone(struct nfsd4_compoundargs *argp, union nfsd4_op_u *u)
 {
+	struct nfsd4_clone *clone = &u->clone;
 	__be32 status;
 
 	status = nfsd4_decode_stateid4(argp, &clone->cl_src_stateid);
@@ -2154,8 +2209,9 @@ nfsd4_decode_xattr_name(struct nfsd4_com
  */
 static __be32
 nfsd4_decode_getxattr(struct nfsd4_compoundargs *argp,
-		      struct nfsd4_getxattr *getxattr)
+		      union nfsd4_op_u *u)
 {
+	struct nfsd4_getxattr *getxattr = &u->getxattr;
 	__be32 status;
 	u32 maxcount;
 
@@ -2173,8 +2229,9 @@ nfsd4_decode_getxattr(struct nfsd4_compo
 
 static __be32
 nfsd4_decode_setxattr(struct nfsd4_compoundargs *argp,
-		      struct nfsd4_setxattr *setxattr)
+		      union nfsd4_op_u *u)
 {
+	struct nfsd4_setxattr *setxattr = &u->setxattr;
 	u32 flags, maxcount, size;
 	__be32 status;
 
@@ -2214,8 +2271,9 @@ nfsd4_decode_setxattr(struct nfsd4_compo
 
 static __be32
 nfsd4_decode_listxattrs(struct nfsd4_compoundargs *argp,
-			struct nfsd4_listxattrs *listxattrs)
+			union nfsd4_op_u *u)
 {
+	struct nfsd4_listxattrs *listxattrs = &u->listxattrs;
 	u32 maxcount;
 
 	memset(listxattrs, 0, sizeof(*listxattrs));
@@ -2245,113 +2303,114 @@ nfsd4_decode_listxattrs(struct nfsd4_com
 
 static __be32
 nfsd4_decode_removexattr(struct nfsd4_compoundargs *argp,
-			 struct nfsd4_removexattr *removexattr)
+			 union nfsd4_op_u *u)
 {
+	struct nfsd4_removexattr *removexattr = &u->removexattr;
 	memset(removexattr, 0, sizeof(*removexattr));
 	return nfsd4_decode_xattr_name(argp, &removexattr->rmxa_name);
 }
 
 static __be32
-nfsd4_decode_noop(struct nfsd4_compoundargs *argp, void *p)
+nfsd4_decode_noop(struct nfsd4_compoundargs *argp, union nfsd4_op_u *p)
 {
 	return nfs_ok;
 }
 
 static __be32
-nfsd4_decode_notsupp(struct nfsd4_compoundargs *argp, void *p)
+nfsd4_decode_notsupp(struct nfsd4_compoundargs *argp, union nfsd4_op_u *p)
 {
 	return nfserr_notsupp;
 }
 
-typedef __be32(*nfsd4_dec)(struct nfsd4_compoundargs *argp, void *);
+typedef __be32(*nfsd4_dec)(struct nfsd4_compoundargs *argp, union nfsd4_op_u *u);
 
 static const nfsd4_dec nfsd4_dec_ops[] = {
-	[OP_ACCESS]		= (nfsd4_dec)nfsd4_decode_access,
-	[OP_CLOSE]		= (nfsd4_dec)nfsd4_decode_close,
-	[OP_COMMIT]		= (nfsd4_dec)nfsd4_decode_commit,
-	[OP_CREATE]		= (nfsd4_dec)nfsd4_decode_create,
-	[OP_DELEGPURGE]		= (nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_DELEGRETURN]	= (nfsd4_dec)nfsd4_decode_delegreturn,
-	[OP_GETATTR]		= (nfsd4_dec)nfsd4_decode_getattr,
-	[OP_GETFH]		= (nfsd4_dec)nfsd4_decode_noop,
-	[OP_LINK]		= (nfsd4_dec)nfsd4_decode_link,
-	[OP_LOCK]		= (nfsd4_dec)nfsd4_decode_lock,
-	[OP_LOCKT]		= (nfsd4_dec)nfsd4_decode_lockt,
-	[OP_LOCKU]		= (nfsd4_dec)nfsd4_decode_locku,
-	[OP_LOOKUP]		= (nfsd4_dec)nfsd4_decode_lookup,
-	[OP_LOOKUPP]		= (nfsd4_dec)nfsd4_decode_noop,
-	[OP_NVERIFY]		= (nfsd4_dec)nfsd4_decode_verify,
-	[OP_OPEN]		= (nfsd4_dec)nfsd4_decode_open,
-	[OP_OPENATTR]		= (nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_OPEN_CONFIRM]	= (nfsd4_dec)nfsd4_decode_open_confirm,
-	[OP_OPEN_DOWNGRADE]	= (nfsd4_dec)nfsd4_decode_open_downgrade,
-	[OP_PUTFH]		= (nfsd4_dec)nfsd4_decode_putfh,
-	[OP_PUTPUBFH]		= (nfsd4_dec)nfsd4_decode_putpubfh,
-	[OP_PUTROOTFH]		= (nfsd4_dec)nfsd4_decode_noop,
-	[OP_READ]		= (nfsd4_dec)nfsd4_decode_read,
-	[OP_READDIR]		= (nfsd4_dec)nfsd4_decode_readdir,
-	[OP_READLINK]		= (nfsd4_dec)nfsd4_decode_noop,
-	[OP_REMOVE]		= (nfsd4_dec)nfsd4_decode_remove,
-	[OP_RENAME]		= (nfsd4_dec)nfsd4_decode_rename,
-	[OP_RENEW]		= (nfsd4_dec)nfsd4_decode_renew,
-	[OP_RESTOREFH]		= (nfsd4_dec)nfsd4_decode_noop,
-	[OP_SAVEFH]		= (nfsd4_dec)nfsd4_decode_noop,
-	[OP_SECINFO]		= (nfsd4_dec)nfsd4_decode_secinfo,
-	[OP_SETATTR]		= (nfsd4_dec)nfsd4_decode_setattr,
-	[OP_SETCLIENTID]	= (nfsd4_dec)nfsd4_decode_setclientid,
-	[OP_SETCLIENTID_CONFIRM] = (nfsd4_dec)nfsd4_decode_setclientid_confirm,
-	[OP_VERIFY]		= (nfsd4_dec)nfsd4_decode_verify,
-	[OP_WRITE]		= (nfsd4_dec)nfsd4_decode_write,
-	[OP_RELEASE_LOCKOWNER]	= (nfsd4_dec)nfsd4_decode_release_lockowner,
+	[OP_ACCESS]		= nfsd4_decode_access,
+	[OP_CLOSE]		= nfsd4_decode_close,
+	[OP_COMMIT]		= nfsd4_decode_commit,
+	[OP_CREATE]		= nfsd4_decode_create,
+	[OP_DELEGPURGE]		= nfsd4_decode_notsupp,
+	[OP_DELEGRETURN]	= nfsd4_decode_delegreturn,
+	[OP_GETATTR]		= nfsd4_decode_getattr,
+	[OP_GETFH]		= nfsd4_decode_noop,
+	[OP_LINK]		= nfsd4_decode_link,
+	[OP_LOCK]		= nfsd4_decode_lock,
+	[OP_LOCKT]		= nfsd4_decode_lockt,
+	[OP_LOCKU]		= nfsd4_decode_locku,
+	[OP_LOOKUP]		= nfsd4_decode_lookup,
+	[OP_LOOKUPP]		= nfsd4_decode_noop,
+	[OP_NVERIFY]		= nfsd4_decode_verify,
+	[OP_OPEN]		= nfsd4_decode_open,
+	[OP_OPENATTR]		= nfsd4_decode_notsupp,
+	[OP_OPEN_CONFIRM]	= nfsd4_decode_open_confirm,
+	[OP_OPEN_DOWNGRADE]	= nfsd4_decode_open_downgrade,
+	[OP_PUTFH]		= nfsd4_decode_putfh,
+	[OP_PUTPUBFH]		= nfsd4_decode_putpubfh,
+	[OP_PUTROOTFH]		= nfsd4_decode_noop,
+	[OP_READ]		= nfsd4_decode_read,
+	[OP_READDIR]		= nfsd4_decode_readdir,
+	[OP_READLINK]		= nfsd4_decode_noop,
+	[OP_REMOVE]		= nfsd4_decode_remove,
+	[OP_RENAME]		= nfsd4_decode_rename,
+	[OP_RENEW]		= nfsd4_decode_renew,
+	[OP_RESTOREFH]		= nfsd4_decode_noop,
+	[OP_SAVEFH]		= nfsd4_decode_noop,
+	[OP_SECINFO]		= nfsd4_decode_secinfo,
+	[OP_SETATTR]		= nfsd4_decode_setattr,
+	[OP_SETCLIENTID]	= nfsd4_decode_setclientid,
+	[OP_SETCLIENTID_CONFIRM] = nfsd4_decode_setclientid_confirm,
+	[OP_VERIFY]		= nfsd4_decode_verify,
+	[OP_WRITE]		= nfsd4_decode_write,
+	[OP_RELEASE_LOCKOWNER]	= nfsd4_decode_release_lockowner,
 
 	/* new operations for NFSv4.1 */
-	[OP_BACKCHANNEL_CTL]	= (nfsd4_dec)nfsd4_decode_backchannel_ctl,
-	[OP_BIND_CONN_TO_SESSION]= (nfsd4_dec)nfsd4_decode_bind_conn_to_session,
-	[OP_EXCHANGE_ID]	= (nfsd4_dec)nfsd4_decode_exchange_id,
-	[OP_CREATE_SESSION]	= (nfsd4_dec)nfsd4_decode_create_session,
-	[OP_DESTROY_SESSION]	= (nfsd4_dec)nfsd4_decode_destroy_session,
-	[OP_FREE_STATEID]	= (nfsd4_dec)nfsd4_decode_free_stateid,
-	[OP_GET_DIR_DELEGATION]	= (nfsd4_dec)nfsd4_decode_notsupp,
+	[OP_BACKCHANNEL_CTL]	= nfsd4_decode_backchannel_ctl,
+	[OP_BIND_CONN_TO_SESSION] = nfsd4_decode_bind_conn_to_session,
+	[OP_EXCHANGE_ID]	= nfsd4_decode_exchange_id,
+	[OP_CREATE_SESSION]	= nfsd4_decode_create_session,
+	[OP_DESTROY_SESSION]	= nfsd4_decode_destroy_session,
+	[OP_FREE_STATEID]	= nfsd4_decode_free_stateid,
+	[OP_GET_DIR_DELEGATION]	= nfsd4_decode_notsupp,
 #ifdef CONFIG_NFSD_PNFS
-	[OP_GETDEVICEINFO]	= (nfsd4_dec)nfsd4_decode_getdeviceinfo,
-	[OP_GETDEVICELIST]	= (nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_LAYOUTCOMMIT]	= (nfsd4_dec)nfsd4_decode_layoutcommit,
-	[OP_LAYOUTGET]		= (nfsd4_dec)nfsd4_decode_layoutget,
-	[OP_LAYOUTRETURN]	= (nfsd4_dec)nfsd4_decode_layoutreturn,
+	[OP_GETDEVICEINFO]	= nfsd4_decode_getdeviceinfo,
+	[OP_GETDEVICELIST]	= nfsd4_decode_notsupp,
+	[OP_LAYOUTCOMMIT]	= nfsd4_decode_layoutcommit,
+	[OP_LAYOUTGET]		= nfsd4_decode_layoutget,
+	[OP_LAYOUTRETURN]	= nfsd4_decode_layoutreturn,
 #else
-	[OP_GETDEVICEINFO]	= (nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_GETDEVICELIST]	= (nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_LAYOUTCOMMIT]	= (nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_LAYOUTGET]		= (nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_LAYOUTRETURN]	= (nfsd4_dec)nfsd4_decode_notsupp,
+	[OP_GETDEVICEINFO]	= nfsd4_decode_notsupp,
+	[OP_GETDEVICELIST]	= nfsd4_decode_notsupp,
+	[OP_LAYOUTCOMMIT]	= nfsd4_decode_notsupp,
+	[OP_LAYOUTGET]		= nfsd4_decode_notsupp,
+	[OP_LAYOUTRETURN]	= nfsd4_decode_notsupp,
 #endif
-	[OP_SECINFO_NO_NAME]	= (nfsd4_dec)nfsd4_decode_secinfo_no_name,
-	[OP_SEQUENCE]		= (nfsd4_dec)nfsd4_decode_sequence,
-	[OP_SET_SSV]		= (nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_TEST_STATEID]	= (nfsd4_dec)nfsd4_decode_test_stateid,
-	[OP_WANT_DELEGATION]	= (nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_DESTROY_CLIENTID]	= (nfsd4_dec)nfsd4_decode_destroy_clientid,
-	[OP_RECLAIM_COMPLETE]	= (nfsd4_dec)nfsd4_decode_reclaim_complete,
+	[OP_SECINFO_NO_NAME]	= nfsd4_decode_secinfo_no_name,
+	[OP_SEQUENCE]		= nfsd4_decode_sequence,
+	[OP_SET_SSV]		= nfsd4_decode_notsupp,
+	[OP_TEST_STATEID]	= nfsd4_decode_test_stateid,
+	[OP_WANT_DELEGATION]	= nfsd4_decode_notsupp,
+	[OP_DESTROY_CLIENTID]	= nfsd4_decode_destroy_clientid,
+	[OP_RECLAIM_COMPLETE]	= nfsd4_decode_reclaim_complete,
 
 	/* new operations for NFSv4.2 */
-	[OP_ALLOCATE]		= (nfsd4_dec)nfsd4_decode_fallocate,
-	[OP_COPY]		= (nfsd4_dec)nfsd4_decode_copy,
-	[OP_COPY_NOTIFY]	= (nfsd4_dec)nfsd4_decode_copy_notify,
-	[OP_DEALLOCATE]		= (nfsd4_dec)nfsd4_decode_fallocate,
-	[OP_IO_ADVISE]		= (nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_LAYOUTERROR]	= (nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_LAYOUTSTATS]	= (nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_OFFLOAD_CANCEL]	= (nfsd4_dec)nfsd4_decode_offload_status,
-	[OP_OFFLOAD_STATUS]	= (nfsd4_dec)nfsd4_decode_offload_status,
-	[OP_READ_PLUS]		= (nfsd4_dec)nfsd4_decode_read,
-	[OP_SEEK]		= (nfsd4_dec)nfsd4_decode_seek,
-	[OP_WRITE_SAME]		= (nfsd4_dec)nfsd4_decode_notsupp,
-	[OP_CLONE]		= (nfsd4_dec)nfsd4_decode_clone,
+	[OP_ALLOCATE]		= nfsd4_decode_fallocate,
+	[OP_COPY]		= nfsd4_decode_copy,
+	[OP_COPY_NOTIFY]	= nfsd4_decode_copy_notify,
+	[OP_DEALLOCATE]		= nfsd4_decode_fallocate,
+	[OP_IO_ADVISE]		= nfsd4_decode_notsupp,
+	[OP_LAYOUTERROR]	= nfsd4_decode_notsupp,
+	[OP_LAYOUTSTATS]	= nfsd4_decode_notsupp,
+	[OP_OFFLOAD_CANCEL]	= nfsd4_decode_offload_status,
+	[OP_OFFLOAD_STATUS]	= nfsd4_decode_offload_status,
+	[OP_READ_PLUS]		= nfsd4_decode_read,
+	[OP_SEEK]		= nfsd4_decode_seek,
+	[OP_WRITE_SAME]		= nfsd4_decode_notsupp,
+	[OP_CLONE]		= nfsd4_decode_clone,
 	/* RFC 8276 extended atributes operations */
-	[OP_GETXATTR]		= (nfsd4_dec)nfsd4_decode_getxattr,
-	[OP_SETXATTR]		= (nfsd4_dec)nfsd4_decode_setxattr,
-	[OP_LISTXATTRS]		= (nfsd4_dec)nfsd4_decode_listxattrs,
-	[OP_REMOVEXATTR]	= (nfsd4_dec)nfsd4_decode_removexattr,
+	[OP_GETXATTR]		= nfsd4_decode_getxattr,
+	[OP_SETXATTR]		= nfsd4_decode_setxattr,
+	[OP_LISTXATTRS]		= nfsd4_decode_listxattrs,
+	[OP_REMOVEXATTR]	= nfsd4_decode_removexattr,
 };
 
 static inline bool
@@ -3643,8 +3702,10 @@ nfsd4_encode_stateid(struct xdr_stream *
 }
 
 static __be32
-nfsd4_encode_access(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_access *access)
+nfsd4_encode_access(struct nfsd4_compoundres *resp, __be32 nfserr,
+		    union nfsd4_op_u *u)
 {
+	struct nfsd4_access *access = &u->access;
 	struct xdr_stream *xdr = resp->xdr;
 	__be32 *p;
 
@@ -3656,8 +3717,10 @@ nfsd4_encode_access(struct nfsd4_compoun
 	return 0;
 }
 
-static __be32 nfsd4_encode_bind_conn_to_session(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_bind_conn_to_session *bcts)
+static __be32 nfsd4_encode_bind_conn_to_session(struct nfsd4_compoundres *resp, __be32 nfserr,
+						union nfsd4_op_u *u)
 {
+	struct nfsd4_bind_conn_to_session *bcts = &u->bind_conn_to_session;
 	struct xdr_stream *xdr = resp->xdr;
 	__be32 *p;
 
@@ -3673,8 +3736,10 @@ static __be32 nfsd4_encode_bind_conn_to_
 }
 
 static __be32
-nfsd4_encode_close(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_close *close)
+nfsd4_encode_close(struct nfsd4_compoundres *resp, __be32 nfserr,
+		   union nfsd4_op_u *u)
 {
+	struct nfsd4_close *close = &u->close;
 	struct xdr_stream *xdr = resp->xdr;
 
 	return nfsd4_encode_stateid(xdr, &close->cl_stateid);
@@ -3682,8 +3747,10 @@ nfsd4_encode_close(struct nfsd4_compound
 
 
 static __be32
-nfsd4_encode_commit(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_commit *commit)
+nfsd4_encode_commit(struct nfsd4_compoundres *resp, __be32 nfserr,
+		    union nfsd4_op_u *u)
 {
+	struct nfsd4_commit *commit = &u->commit;
 	struct xdr_stream *xdr = resp->xdr;
 	__be32 *p;
 
@@ -3696,8 +3763,10 @@ nfsd4_encode_commit(struct nfsd4_compoun
 }
 
 static __be32
-nfsd4_encode_create(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_create *create)
+nfsd4_encode_create(struct nfsd4_compoundres *resp, __be32 nfserr,
+		    union nfsd4_op_u *u)
 {
+	struct nfsd4_create *create = &u->create;
 	struct xdr_stream *xdr = resp->xdr;
 	__be32 *p;
 
@@ -3710,8 +3779,10 @@ nfsd4_encode_create(struct nfsd4_compoun
 }
 
 static __be32
-nfsd4_encode_getattr(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_getattr *getattr)
+nfsd4_encode_getattr(struct nfsd4_compoundres *resp, __be32 nfserr,
+		     union nfsd4_op_u *u)
 {
+	struct nfsd4_getattr *getattr = &u->getattr;
 	struct svc_fh *fhp = getattr->ga_fhp;
 	struct xdr_stream *xdr = resp->xdr;
 
@@ -3720,8 +3791,10 @@ nfsd4_encode_getattr(struct nfsd4_compou
 }
 
 static __be32
-nfsd4_encode_getfh(struct nfsd4_compoundres *resp, __be32 nfserr, struct svc_fh **fhpp)
+nfsd4_encode_getfh(struct nfsd4_compoundres *resp, __be32 nfserr,
+		   union nfsd4_op_u *u)
 {
+	struct svc_fh **fhpp = &u->getfh;
 	struct xdr_stream *xdr = resp->xdr;
 	struct svc_fh *fhp = *fhpp;
 	unsigned int len;
@@ -3775,8 +3848,10 @@ again:
 }
 
 static __be32
-nfsd4_encode_lock(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_lock *lock)
+nfsd4_encode_lock(struct nfsd4_compoundres *resp, __be32 nfserr,
+		  union nfsd4_op_u *u)
 {
+	struct nfsd4_lock *lock = &u->lock;
 	struct xdr_stream *xdr = resp->xdr;
 
 	if (!nfserr)
@@ -3788,8 +3863,10 @@ nfsd4_encode_lock(struct nfsd4_compoundr
 }
 
 static __be32
-nfsd4_encode_lockt(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_lockt *lockt)
+nfsd4_encode_lockt(struct nfsd4_compoundres *resp, __be32 nfserr,
+		   union nfsd4_op_u *u)
 {
+	struct nfsd4_lockt *lockt = &u->lockt;
 	struct xdr_stream *xdr = resp->xdr;
 
 	if (nfserr == nfserr_denied)
@@ -3798,8 +3875,10 @@ nfsd4_encode_lockt(struct nfsd4_compound
 }
 
 static __be32
-nfsd4_encode_locku(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_locku *locku)
+nfsd4_encode_locku(struct nfsd4_compoundres *resp, __be32 nfserr,
+		   union nfsd4_op_u *u)
 {
+	struct nfsd4_locku *locku = &u->locku;
 	struct xdr_stream *xdr = resp->xdr;
 
 	return nfsd4_encode_stateid(xdr, &locku->lu_stateid);
@@ -3807,8 +3886,10 @@ nfsd4_encode_locku(struct nfsd4_compound
 
 
 static __be32
-nfsd4_encode_link(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_link *link)
+nfsd4_encode_link(struct nfsd4_compoundres *resp, __be32 nfserr,
+		  union nfsd4_op_u *u)
 {
+	struct nfsd4_link *link = &u->link;
 	struct xdr_stream *xdr = resp->xdr;
 	__be32 *p;
 
@@ -3821,8 +3902,10 @@ nfsd4_encode_link(struct nfsd4_compoundr
 
 
 static __be32
-nfsd4_encode_open(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_open *open)
+nfsd4_encode_open(struct nfsd4_compoundres *resp, __be32 nfserr,
+		  union nfsd4_op_u *u)
 {
+	struct nfsd4_open *open = &u->open;
 	struct xdr_stream *xdr = resp->xdr;
 	__be32 *p;
 
@@ -3915,16 +3998,20 @@ nfsd4_encode_open(struct nfsd4_compoundr
 }
 
 static __be32
-nfsd4_encode_open_confirm(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_open_confirm *oc)
+nfsd4_encode_open_confirm(struct nfsd4_compoundres *resp, __be32 nfserr,
+			  union nfsd4_op_u *u)
 {
+	struct nfsd4_open_confirm *oc = &u->open_confirm;
 	struct xdr_stream *xdr = resp->xdr;
 
 	return nfsd4_encode_stateid(xdr, &oc->oc_resp_stateid);
 }
 
 static __be32
-nfsd4_encode_open_downgrade(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_open_downgrade *od)
+nfsd4_encode_open_downgrade(struct nfsd4_compoundres *resp, __be32 nfserr,
+			    union nfsd4_op_u *u)
 {
+	struct nfsd4_open_downgrade *od = &u->open_downgrade;
 	struct xdr_stream *xdr = resp->xdr;
 
 	return nfsd4_encode_stateid(xdr, &od->od_stateid);
@@ -4023,8 +4110,9 @@ static __be32 nfsd4_encode_readv(struct
 
 static __be32
 nfsd4_encode_read(struct nfsd4_compoundres *resp, __be32 nfserr,
-		  struct nfsd4_read *read)
+		  union nfsd4_op_u *u)
 {
+	struct nfsd4_read *read = &u->read;
 	bool splice_ok = test_bit(RQ_SPLICE_OK, &resp->rqstp->rq_flags);
 	unsigned long maxcount;
 	struct xdr_stream *xdr = resp->xdr;
@@ -4065,8 +4153,10 @@ nfsd4_encode_read(struct nfsd4_compoundr
 }
 
 static __be32
-nfsd4_encode_readlink(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_readlink *readlink)
+nfsd4_encode_readlink(struct nfsd4_compoundres *resp, __be32 nfserr,
+		      union nfsd4_op_u *u)
 {
+	struct nfsd4_readlink *readlink = &u->readlink;
 	__be32 *p, *maxcount_p, zero = xdr_zero;
 	struct xdr_stream *xdr = resp->xdr;
 	int length_offset = xdr->buf->len;
@@ -4110,8 +4200,10 @@ out_err:
 }
 
 static __be32
-nfsd4_encode_readdir(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_readdir *readdir)
+nfsd4_encode_readdir(struct nfsd4_compoundres *resp, __be32 nfserr,
+		     union nfsd4_op_u *u)
 {
+	struct nfsd4_readdir *readdir = &u->readdir;
 	int maxcount;
 	int bytes_left;
 	loff_t offset;
@@ -4201,8 +4293,10 @@ err_no_verf:
 }
 
 static __be32
-nfsd4_encode_remove(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_remove *remove)
+nfsd4_encode_remove(struct nfsd4_compoundres *resp, __be32 nfserr,
+		    union nfsd4_op_u *u)
 {
+	struct nfsd4_remove *remove = &u->remove;
 	struct xdr_stream *xdr = resp->xdr;
 	__be32 *p;
 
@@ -4214,8 +4308,10 @@ nfsd4_encode_remove(struct nfsd4_compoun
 }
 
 static __be32
-nfsd4_encode_rename(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_rename *rename)
+nfsd4_encode_rename(struct nfsd4_compoundres *resp, __be32 nfserr,
+		    union nfsd4_op_u *u)
 {
+	struct nfsd4_rename *rename = &u->rename;
 	struct xdr_stream *xdr = resp->xdr;
 	__be32 *p;
 
@@ -4297,8 +4393,9 @@ nfsd4_do_encode_secinfo(struct xdr_strea
 
 static __be32
 nfsd4_encode_secinfo(struct nfsd4_compoundres *resp, __be32 nfserr,
-		     struct nfsd4_secinfo *secinfo)
+		     union nfsd4_op_u *u)
 {
+	struct nfsd4_secinfo *secinfo = &u->secinfo;
 	struct xdr_stream *xdr = resp->xdr;
 
 	return nfsd4_do_encode_secinfo(xdr, secinfo->si_exp);
@@ -4306,8 +4403,9 @@ nfsd4_encode_secinfo(struct nfsd4_compou
 
 static __be32
 nfsd4_encode_secinfo_no_name(struct nfsd4_compoundres *resp, __be32 nfserr,
-		     struct nfsd4_secinfo_no_name *secinfo)
+		     union nfsd4_op_u *u)
 {
+	struct nfsd4_secinfo_no_name *secinfo = &u->secinfo_no_name;
 	struct xdr_stream *xdr = resp->xdr;
 
 	return nfsd4_do_encode_secinfo(xdr, secinfo->sin_exp);
@@ -4318,8 +4416,10 @@ nfsd4_encode_secinfo_no_name(struct nfsd
  * regardless of the error status.
  */
 static __be32
-nfsd4_encode_setattr(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_setattr *setattr)
+nfsd4_encode_setattr(struct nfsd4_compoundres *resp, __be32 nfserr,
+		     union nfsd4_op_u *u)
 {
+	struct nfsd4_setattr *setattr = &u->setattr;
 	struct xdr_stream *xdr = resp->xdr;
 	__be32 *p;
 
@@ -4342,8 +4442,10 @@ nfsd4_encode_setattr(struct nfsd4_compou
 }
 
 static __be32
-nfsd4_encode_setclientid(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_setclientid *scd)
+nfsd4_encode_setclientid(struct nfsd4_compoundres *resp, __be32 nfserr,
+			 union nfsd4_op_u *u)
 {
+	struct nfsd4_setclientid *scd = &u->setclientid;
 	struct xdr_stream *xdr = resp->xdr;
 	__be32 *p;
 
@@ -4366,8 +4468,10 @@ nfsd4_encode_setclientid(struct nfsd4_co
 }
 
 static __be32
-nfsd4_encode_write(struct nfsd4_compoundres *resp, __be32 nfserr, struct nfsd4_write *write)
+nfsd4_encode_write(struct nfsd4_compoundres *resp, __be32 nfserr,
+		   union nfsd4_op_u *u)
 {
+	struct nfsd4_write *write = &u->write;
 	struct xdr_stream *xdr = resp->xdr;
 	__be32 *p;
 
@@ -4383,8 +4487,9 @@ nfsd4_encode_write(struct nfsd4_compound
 
 static __be32
 nfsd4_encode_exchange_id(struct nfsd4_compoundres *resp, __be32 nfserr,
-			 struct nfsd4_exchange_id *exid)
+			 union nfsd4_op_u *u)
 {
+	struct nfsd4_exchange_id *exid = &u->exchange_id;
 	struct xdr_stream *xdr = resp->xdr;
 	__be32 *p;
 	char *major_id;
@@ -4461,8 +4566,9 @@ nfsd4_encode_exchange_id(struct nfsd4_co
 
 static __be32
 nfsd4_encode_create_session(struct nfsd4_compoundres *resp, __be32 nfserr,
-			    struct nfsd4_create_session *sess)
+			    union nfsd4_op_u *u)
 {
+	struct nfsd4_create_session *sess = &u->create_session;
 	struct xdr_stream *xdr = resp->xdr;
 	__be32 *p;
 
@@ -4514,8 +4620,9 @@ nfsd4_encode_create_session(struct nfsd4
 
 static __be32
 nfsd4_encode_sequence(struct nfsd4_compoundres *resp, __be32 nfserr,
-		      struct nfsd4_sequence *seq)
+		      union nfsd4_op_u *u)
 {
+	struct nfsd4_sequence *seq = &u->sequence;
 	struct xdr_stream *xdr = resp->xdr;
 	__be32 *p;
 
@@ -4537,8 +4644,9 @@ nfsd4_encode_sequence(struct nfsd4_compo
 
 static __be32
 nfsd4_encode_test_stateid(struct nfsd4_compoundres *resp, __be32 nfserr,
-			  struct nfsd4_test_stateid *test_stateid)
+			  union nfsd4_op_u *u)
 {
+	struct nfsd4_test_stateid *test_stateid = &u->test_stateid;
 	struct xdr_stream *xdr = resp->xdr;
 	struct nfsd4_test_stateid_id *stateid, *next;
 	__be32 *p;
@@ -4558,8 +4666,9 @@ nfsd4_encode_test_stateid(struct nfsd4_c
 #ifdef CONFIG_NFSD_PNFS
 static __be32
 nfsd4_encode_getdeviceinfo(struct nfsd4_compoundres *resp, __be32 nfserr,
-		struct nfsd4_getdeviceinfo *gdev)
+		union nfsd4_op_u *u)
 {
+	struct nfsd4_getdeviceinfo *gdev = &u->getdeviceinfo;
 	struct xdr_stream *xdr = resp->xdr;
 	const struct nfsd4_layout_ops *ops;
 	u32 starting_len = xdr->buf->len, needed_len;
@@ -4611,8 +4720,9 @@ toosmall:
 
 static __be32
 nfsd4_encode_layoutget(struct nfsd4_compoundres *resp, __be32 nfserr,
-		struct nfsd4_layoutget *lgp)
+		union nfsd4_op_u *u)
 {
+	struct nfsd4_layoutget *lgp = &u->layoutget;
 	struct xdr_stream *xdr = resp->xdr;
 	const struct nfsd4_layout_ops *ops;
 	__be32 *p;
@@ -4638,8 +4748,9 @@ nfsd4_encode_layoutget(struct nfsd4_comp
 
 static __be32
 nfsd4_encode_layoutcommit(struct nfsd4_compoundres *resp, __be32 nfserr,
-			  struct nfsd4_layoutcommit *lcp)
+			  union nfsd4_op_u *u)
 {
+	struct nfsd4_layoutcommit *lcp = &u->layoutcommit;
 	struct xdr_stream *xdr = resp->xdr;
 	__be32 *p;
 
@@ -4659,8 +4770,9 @@ nfsd4_encode_layoutcommit(struct nfsd4_c
 
 static __be32
 nfsd4_encode_layoutreturn(struct nfsd4_compoundres *resp, __be32 nfserr,
-		struct nfsd4_layoutreturn *lrp)
+		union nfsd4_op_u *u)
 {
+	struct nfsd4_layoutreturn *lrp = &u->layoutreturn;
 	struct xdr_stream *xdr = resp->xdr;
 	__be32 *p;
 
@@ -4745,8 +4857,9 @@ nfsd42_encode_nl4_server(struct nfsd4_co
 
 static __be32
 nfsd4_encode_copy(struct nfsd4_compoundres *resp, __be32 nfserr,
-		  struct nfsd4_copy *copy)
+		  union nfsd4_op_u *u)
 {
+	struct nfsd4_copy *copy = &u->copy;
 	__be32 *p;
 
 	nfserr = nfsd42_encode_write_res(resp, &copy->cp_res,
@@ -4762,8 +4875,9 @@ nfsd4_encode_copy(struct nfsd4_compoundr
 
 static __be32
 nfsd4_encode_offload_status(struct nfsd4_compoundres *resp, __be32 nfserr,
-			    struct nfsd4_offload_status *os)
+			    union nfsd4_op_u *u)
 {
+	struct nfsd4_offload_status *os = &u->offload_status;
 	struct xdr_stream *xdr = resp->xdr;
 	__be32 *p;
 
@@ -4813,8 +4927,9 @@ nfsd4_encode_read_plus_data(struct nfsd4
 
 static __be32
 nfsd4_encode_read_plus(struct nfsd4_compoundres *resp, __be32 nfserr,
-		       struct nfsd4_read *read)
+		       union nfsd4_op_u *u)
 {
+	struct nfsd4_read *read = &u->read;
 	struct file *file = read->rd_nf->nf_file;
 	struct xdr_stream *xdr = resp->xdr;
 	int starting_len = xdr->buf->len;
@@ -4850,8 +4965,9 @@ out:
 
 static __be32
 nfsd4_encode_copy_notify(struct nfsd4_compoundres *resp, __be32 nfserr,
-			 struct nfsd4_copy_notify *cn)
+			 union nfsd4_op_u *u)
 {
+	struct nfsd4_copy_notify *cn = &u->copy_notify;
 	struct xdr_stream *xdr = resp->xdr;
 	__be32 *p;
 
@@ -4885,8 +5001,9 @@ nfsd4_encode_copy_notify(struct nfsd4_co
 
 static __be32
 nfsd4_encode_seek(struct nfsd4_compoundres *resp, __be32 nfserr,
-		  struct nfsd4_seek *seek)
+		  union nfsd4_op_u *u)
 {
+	struct nfsd4_seek *seek = &u->seek;
 	__be32 *p;
 
 	p = xdr_reserve_space(resp->xdr, 4 + 8);
@@ -4897,7 +5014,8 @@ nfsd4_encode_seek(struct nfsd4_compoundr
 }
 
 static __be32
-nfsd4_encode_noop(struct nfsd4_compoundres *resp, __be32 nfserr, void *p)
+nfsd4_encode_noop(struct nfsd4_compoundres *resp, __be32 nfserr,
+		  union nfsd4_op_u *p)
 {
 	return nfserr;
 }
@@ -4948,8 +5066,9 @@ nfsd4_vbuf_to_stream(struct xdr_stream *
 
 static __be32
 nfsd4_encode_getxattr(struct nfsd4_compoundres *resp, __be32 nfserr,
-		      struct nfsd4_getxattr *getxattr)
+		      union nfsd4_op_u *u)
 {
+	struct nfsd4_getxattr *getxattr = &u->getxattr;
 	struct xdr_stream *xdr = resp->xdr;
 	__be32 *p, err;
 
@@ -4972,8 +5091,9 @@ nfsd4_encode_getxattr(struct nfsd4_compo
 
 static __be32
 nfsd4_encode_setxattr(struct nfsd4_compoundres *resp, __be32 nfserr,
-		      struct nfsd4_setxattr *setxattr)
+		      union nfsd4_op_u *u)
 {
+	struct nfsd4_setxattr *setxattr = &u->setxattr;
 	struct xdr_stream *xdr = resp->xdr;
 	__be32 *p;
 
@@ -5013,8 +5133,9 @@ nfsd4_listxattr_validate_cookie(struct n
 
 static __be32
 nfsd4_encode_listxattrs(struct nfsd4_compoundres *resp, __be32 nfserr,
-			struct nfsd4_listxattrs *listxattrs)
+			union nfsd4_op_u *u)
 {
+	struct nfsd4_listxattrs *listxattrs = &u->listxattrs;
 	struct xdr_stream *xdr = resp->xdr;
 	u32 cookie_offset, count_offset, eof;
 	u32 left, xdrleft, slen, count;
@@ -5124,8 +5245,9 @@ out:
 
 static __be32
 nfsd4_encode_removexattr(struct nfsd4_compoundres *resp, __be32 nfserr,
-			 struct nfsd4_removexattr *removexattr)
+			 union nfsd4_op_u *u)
 {
+	struct nfsd4_removexattr *removexattr = &u->removexattr;
 	struct xdr_stream *xdr = resp->xdr;
 	__be32 *p;
 
@@ -5137,7 +5259,7 @@ nfsd4_encode_removexattr(struct nfsd4_co
 	return 0;
 }
 
-typedef __be32(* nfsd4_enc)(struct nfsd4_compoundres *, __be32, void *);
+typedef __be32(*nfsd4_enc)(struct nfsd4_compoundres *, __be32, union nfsd4_op_u *u);
 
 /*
  * Note: nfsd4_enc_ops vector is shared for v4.0 and v4.1
@@ -5145,93 +5267,93 @@ typedef __be32(* nfsd4_enc)(struct nfsd4
  * done in the decoding phase.
  */
 static const nfsd4_enc nfsd4_enc_ops[] = {
-	[OP_ACCESS]		= (nfsd4_enc)nfsd4_encode_access,
-	[OP_CLOSE]		= (nfsd4_enc)nfsd4_encode_close,
-	[OP_COMMIT]		= (nfsd4_enc)nfsd4_encode_commit,
-	[OP_CREATE]		= (nfsd4_enc)nfsd4_encode_create,
-	[OP_DELEGPURGE]		= (nfsd4_enc)nfsd4_encode_noop,
-	[OP_DELEGRETURN]	= (nfsd4_enc)nfsd4_encode_noop,
-	[OP_GETATTR]		= (nfsd4_enc)nfsd4_encode_getattr,
-	[OP_GETFH]		= (nfsd4_enc)nfsd4_encode_getfh,
-	[OP_LINK]		= (nfsd4_enc)nfsd4_encode_link,
-	[OP_LOCK]		= (nfsd4_enc)nfsd4_encode_lock,
-	[OP_LOCKT]		= (nfsd4_enc)nfsd4_encode_lockt,
-	[OP_LOCKU]		= (nfsd4_enc)nfsd4_encode_locku,
-	[OP_LOOKUP]		= (nfsd4_enc)nfsd4_encode_noop,
-	[OP_LOOKUPP]		= (nfsd4_enc)nfsd4_encode_noop,
-	[OP_NVERIFY]		= (nfsd4_enc)nfsd4_encode_noop,
-	[OP_OPEN]		= (nfsd4_enc)nfsd4_encode_open,
-	[OP_OPENATTR]		= (nfsd4_enc)nfsd4_encode_noop,
-	[OP_OPEN_CONFIRM]	= (nfsd4_enc)nfsd4_encode_open_confirm,
-	[OP_OPEN_DOWNGRADE]	= (nfsd4_enc)nfsd4_encode_open_downgrade,
-	[OP_PUTFH]		= (nfsd4_enc)nfsd4_encode_noop,
-	[OP_PUTPUBFH]		= (nfsd4_enc)nfsd4_encode_noop,
-	[OP_PUTROOTFH]		= (nfsd4_enc)nfsd4_encode_noop,
-	[OP_READ]		= (nfsd4_enc)nfsd4_encode_read,
-	[OP_READDIR]		= (nfsd4_enc)nfsd4_encode_readdir,
-	[OP_READLINK]		= (nfsd4_enc)nfsd4_encode_readlink,
-	[OP_REMOVE]		= (nfsd4_enc)nfsd4_encode_remove,
-	[OP_RENAME]		= (nfsd4_enc)nfsd4_encode_rename,
-	[OP_RENEW]		= (nfsd4_enc)nfsd4_encode_noop,
-	[OP_RESTOREFH]		= (nfsd4_enc)nfsd4_encode_noop,
-	[OP_SAVEFH]		= (nfsd4_enc)nfsd4_encode_noop,
-	[OP_SECINFO]		= (nfsd4_enc)nfsd4_encode_secinfo,
-	[OP_SETATTR]		= (nfsd4_enc)nfsd4_encode_setattr,
-	[OP_SETCLIENTID]	= (nfsd4_enc)nfsd4_encode_setclientid,
-	[OP_SETCLIENTID_CONFIRM] = (nfsd4_enc)nfsd4_encode_noop,
-	[OP_VERIFY]		= (nfsd4_enc)nfsd4_encode_noop,
-	[OP_WRITE]		= (nfsd4_enc)nfsd4_encode_write,
-	[OP_RELEASE_LOCKOWNER]	= (nfsd4_enc)nfsd4_encode_noop,
+	[OP_ACCESS]		= nfsd4_encode_access,
+	[OP_CLOSE]		= nfsd4_encode_close,
+	[OP_COMMIT]		= nfsd4_encode_commit,
+	[OP_CREATE]		= nfsd4_encode_create,
+	[OP_DELEGPURGE]		= nfsd4_encode_noop,
+	[OP_DELEGRETURN]	= nfsd4_encode_noop,
+	[OP_GETATTR]		= nfsd4_encode_getattr,
+	[OP_GETFH]		= nfsd4_encode_getfh,
+	[OP_LINK]		= nfsd4_encode_link,
+	[OP_LOCK]		= nfsd4_encode_lock,
+	[OP_LOCKT]		= nfsd4_encode_lockt,
+	[OP_LOCKU]		= nfsd4_encode_locku,
+	[OP_LOOKUP]		= nfsd4_encode_noop,
+	[OP_LOOKUPP]		= nfsd4_encode_noop,
+	[OP_NVERIFY]		= nfsd4_encode_noop,
+	[OP_OPEN]		= nfsd4_encode_open,
+	[OP_OPENATTR]		= nfsd4_encode_noop,
+	[OP_OPEN_CONFIRM]	= nfsd4_encode_open_confirm,
+	[OP_OPEN_DOWNGRADE]	= nfsd4_encode_open_downgrade,
+	[OP_PUTFH]		= nfsd4_encode_noop,
+	[OP_PUTPUBFH]		= nfsd4_encode_noop,
+	[OP_PUTROOTFH]		= nfsd4_encode_noop,
+	[OP_READ]		= nfsd4_encode_read,
+	[OP_READDIR]		= nfsd4_encode_readdir,
+	[OP_READLINK]		= nfsd4_encode_readlink,
+	[OP_REMOVE]		= nfsd4_encode_remove,
+	[OP_RENAME]		= nfsd4_encode_rename,
+	[OP_RENEW]		= nfsd4_encode_noop,
+	[OP_RESTOREFH]		= nfsd4_encode_noop,
+	[OP_SAVEFH]		= nfsd4_encode_noop,
+	[OP_SECINFO]		= nfsd4_encode_secinfo,
+	[OP_SETATTR]		= nfsd4_encode_setattr,
+	[OP_SETCLIENTID]	= nfsd4_encode_setclientid,
+	[OP_SETCLIENTID_CONFIRM] = nfsd4_encode_noop,
+	[OP_VERIFY]		= nfsd4_encode_noop,
+	[OP_WRITE]		= nfsd4_encode_write,
+	[OP_RELEASE_LOCKOWNER]	= nfsd4_encode_noop,
 
 	/* NFSv4.1 operations */
-	[OP_BACKCHANNEL_CTL]	= (nfsd4_enc)nfsd4_encode_noop,
-	[OP_BIND_CONN_TO_SESSION] = (nfsd4_enc)nfsd4_encode_bind_conn_to_session,
-	[OP_EXCHANGE_ID]	= (nfsd4_enc)nfsd4_encode_exchange_id,
-	[OP_CREATE_SESSION]	= (nfsd4_enc)nfsd4_encode_create_session,
-	[OP_DESTROY_SESSION]	= (nfsd4_enc)nfsd4_encode_noop,
-	[OP_FREE_STATEID]	= (nfsd4_enc)nfsd4_encode_noop,
-	[OP_GET_DIR_DELEGATION]	= (nfsd4_enc)nfsd4_encode_noop,
+	[OP_BACKCHANNEL_CTL]	= nfsd4_encode_noop,
+	[OP_BIND_CONN_TO_SESSION] = nfsd4_encode_bind_conn_to_session,
+	[OP_EXCHANGE_ID]	= nfsd4_encode_exchange_id,
+	[OP_CREATE_SESSION]	= nfsd4_encode_create_session,
+	[OP_DESTROY_SESSION]	= nfsd4_encode_noop,
+	[OP_FREE_STATEID]	= nfsd4_encode_noop,
+	[OP_GET_DIR_DELEGATION]	= nfsd4_encode_noop,
 #ifdef CONFIG_NFSD_PNFS
-	[OP_GETDEVICEINFO]	= (nfsd4_enc)nfsd4_encode_getdeviceinfo,
-	[OP_GETDEVICELIST]	= (nfsd4_enc)nfsd4_encode_noop,
-	[OP_LAYOUTCOMMIT]	= (nfsd4_enc)nfsd4_encode_layoutcommit,
-	[OP_LAYOUTGET]		= (nfsd4_enc)nfsd4_encode_layoutget,
-	[OP_LAYOUTRETURN]	= (nfsd4_enc)nfsd4_encode_layoutreturn,
+	[OP_GETDEVICEINFO]	= nfsd4_encode_getdeviceinfo,
+	[OP_GETDEVICELIST]	= nfsd4_encode_noop,
+	[OP_LAYOUTCOMMIT]	= nfsd4_encode_layoutcommit,
+	[OP_LAYOUTGET]		= nfsd4_encode_layoutget,
+	[OP_LAYOUTRETURN]	= nfsd4_encode_layoutreturn,
 #else
-	[OP_GETDEVICEINFO]	= (nfsd4_enc)nfsd4_encode_noop,
-	[OP_GETDEVICELIST]	= (nfsd4_enc)nfsd4_encode_noop,
-	[OP_LAYOUTCOMMIT]	= (nfsd4_enc)nfsd4_encode_noop,
-	[OP_LAYOUTGET]		= (nfsd4_enc)nfsd4_encode_noop,
-	[OP_LAYOUTRETURN]	= (nfsd4_enc)nfsd4_encode_noop,
+	[OP_GETDEVICEINFO]	= nfsd4_encode_noop,
+	[OP_GETDEVICELIST]	= nfsd4_encode_noop,
+	[OP_LAYOUTCOMMIT]	= nfsd4_encode_noop,
+	[OP_LAYOUTGET]		= nfsd4_encode_noop,
+	[OP_LAYOUTRETURN]	= nfsd4_encode_noop,
 #endif
-	[OP_SECINFO_NO_NAME]	= (nfsd4_enc)nfsd4_encode_secinfo_no_name,
-	[OP_SEQUENCE]		= (nfsd4_enc)nfsd4_encode_sequence,
-	[OP_SET_SSV]		= (nfsd4_enc)nfsd4_encode_noop,
-	[OP_TEST_STATEID]	= (nfsd4_enc)nfsd4_encode_test_stateid,
-	[OP_WANT_DELEGATION]	= (nfsd4_enc)nfsd4_encode_noop,
-	[OP_DESTROY_CLIENTID]	= (nfsd4_enc)nfsd4_encode_noop,
-	[OP_RECLAIM_COMPLETE]	= (nfsd4_enc)nfsd4_encode_noop,
+	[OP_SECINFO_NO_NAME]	= nfsd4_encode_secinfo_no_name,
+	[OP_SEQUENCE]		= nfsd4_encode_sequence,
+	[OP_SET_SSV]		= nfsd4_encode_noop,
+	[OP_TEST_STATEID]	= nfsd4_encode_test_stateid,
+	[OP_WANT_DELEGATION]	= nfsd4_encode_noop,
+	[OP_DESTROY_CLIENTID]	= nfsd4_encode_noop,
+	[OP_RECLAIM_COMPLETE]	= nfsd4_encode_noop,
 
 	/* NFSv4.2 operations */
-	[OP_ALLOCATE]		= (nfsd4_enc)nfsd4_encode_noop,
-	[OP_COPY]		= (nfsd4_enc)nfsd4_encode_copy,
-	[OP_COPY_NOTIFY]	= (nfsd4_enc)nfsd4_encode_copy_notify,
-	[OP_DEALLOCATE]		= (nfsd4_enc)nfsd4_encode_noop,
-	[OP_IO_ADVISE]		= (nfsd4_enc)nfsd4_encode_noop,
-	[OP_LAYOUTERROR]	= (nfsd4_enc)nfsd4_encode_noop,
-	[OP_LAYOUTSTATS]	= (nfsd4_enc)nfsd4_encode_noop,
-	[OP_OFFLOAD_CANCEL]	= (nfsd4_enc)nfsd4_encode_noop,
-	[OP_OFFLOAD_STATUS]	= (nfsd4_enc)nfsd4_encode_offload_status,
-	[OP_READ_PLUS]		= (nfsd4_enc)nfsd4_encode_read_plus,
-	[OP_SEEK]		= (nfsd4_enc)nfsd4_encode_seek,
-	[OP_WRITE_SAME]		= (nfsd4_enc)nfsd4_encode_noop,
-	[OP_CLONE]		= (nfsd4_enc)nfsd4_encode_noop,
+	[OP_ALLOCATE]		= nfsd4_encode_noop,
+	[OP_COPY]		= nfsd4_encode_copy,
+	[OP_COPY_NOTIFY]	= nfsd4_encode_copy_notify,
+	[OP_DEALLOCATE]		= nfsd4_encode_noop,
+	[OP_IO_ADVISE]		= nfsd4_encode_noop,
+	[OP_LAYOUTERROR]	= nfsd4_encode_noop,
+	[OP_LAYOUTSTATS]	= nfsd4_encode_noop,
+	[OP_OFFLOAD_CANCEL]	= nfsd4_encode_noop,
+	[OP_OFFLOAD_STATUS]	= nfsd4_encode_offload_status,
+	[OP_READ_PLUS]		= nfsd4_encode_read_plus,
+	[OP_SEEK]		= nfsd4_encode_seek,
+	[OP_WRITE_SAME]		= nfsd4_encode_noop,
+	[OP_CLONE]		= nfsd4_encode_noop,
 
 	/* RFC 8276 extended atributes operations */
-	[OP_GETXATTR]		= (nfsd4_enc)nfsd4_encode_getxattr,
-	[OP_SETXATTR]		= (nfsd4_enc)nfsd4_encode_setxattr,
-	[OP_LISTXATTRS]		= (nfsd4_enc)nfsd4_encode_listxattrs,
-	[OP_REMOVEXATTR]	= (nfsd4_enc)nfsd4_encode_removexattr,
+	[OP_GETXATTR]		= nfsd4_encode_getxattr,
+	[OP_SETXATTR]		= nfsd4_encode_setxattr,
+	[OP_LISTXATTRS]		= nfsd4_encode_listxattrs,
+	[OP_REMOVEXATTR]	= nfsd4_encode_removexattr,
 };
 
 /*
author maintainer
#180 lore raw

PATCH 176/215 NFSD: Use set_bit(RQ_DROPME)

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Chuck Lever <[email protected]>

[ Upstream commit 5304930dbae82d259bcf7e5611db7c81e7a42eff ]

The premise that "Once an svc thread is scheduled and executing an RPC, no other processes will touch svc_rqst::rq_flags" is false. svc_xprt_enqueue() examines the RQ_BUSY flag in scheduled nfsd threads when determining which thread to wake up next.

Fixes 9315564747cb ("NFSD: Use only RQ_DROPME to signal the need to drop a reply")
---
 fs/nfsd/nfsproc.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
fs/nfsd/nfsproc.c +2 -2
@@ -211,7 +211,7 @@ nfsd_proc_read(struct svc_rqst *rqstp)
 	if (resp->status == nfs_ok)
 		resp->status = fh_getattr(&resp->fh, &resp->stat);
 	else if (resp->status == nfserr_jukebox)
-		__set_bit(RQ_DROPME, &rqstp->rq_flags);
+		set_bit(RQ_DROPME, &rqstp->rq_flags);
 	return rpc_success;
 }
 
@@ -246,7 +246,7 @@ nfsd_proc_write(struct svc_rqst *rqstp)
 	if (resp->status == nfs_ok)
 		resp->status = fh_getattr(&resp->fh, &resp->stat);
 	else if (resp->status == nfserr_jukebox)
-		__set_bit(RQ_DROPME, &rqstp->rq_flags);
+		set_bit(RQ_DROPME, &rqstp->rq_flags);
 	return rpc_success;
 }
 
author maintainer
#181 lore raw

PATCH 177/215 NFSD: register/unregister of nfsd-client shrinker at nfsd startup/shutdown time

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Dai Ngo <[email protected]>

[ Upstream commit f385f7d244134246f984975ed34cd75f77de479f ]

Currently the nfsd-client shrinker is registered and unregistered at the time the nfsd module is loaded and unloaded. The problem with this is the shrinker is being registered before all of the relevant fields in nfsd_net are initialized when nfsd is started. This can lead to an oops when memory is low and the shrinker is called while nfsd is not running.

This patch moves the register/unregister of nfsd-client shrinker from module load/unload time to nfsd startup/shutdown time.

Fixes 44df6f439a17 ("NFSD: add delegation reaper to react to low memory condition")
---
 fs/nfsd/nfs4state.c |   22 +++++++++++-----------
 fs/nfsd/nfsctl.c    |    7 +------
 fs/nfsd/nfsd.h      |    6 ++----
 3 files changed, 14 insertions(+), 21 deletions(-)
fs/nfsd/nfs4state.c +11 -11
@@ -4422,7 +4422,7 @@ nfsd4_state_shrinker_scan(struct shrinke
 	return SHRINK_STOP;
 }
 
-int
+void
 nfsd4_init_leases_net(struct nfsd_net *nn)
 {
 	struct sysinfo si;
@@ -4444,16 +4444,6 @@ nfsd4_init_leases_net(struct nfsd_net *n
 	nn->nfs4_max_clients = max_t(int, max_clients, NFS4_CLIENTS_PER_GB);
 
 	atomic_set(&nn->nfsd_courtesy_clients, 0);
-	nn->nfsd_client_shrinker.scan_objects = nfsd4_state_shrinker_scan;
-	nn->nfsd_client_shrinker.count_objects = nfsd4_state_shrinker_count;
-	nn->nfsd_client_shrinker.seeks = DEFAULT_SEEKS;
-	return register_shrinker(&nn->nfsd_client_shrinker, "nfsd-client");
-}
-
-void
-nfsd4_leases_net_shutdown(struct nfsd_net *nn)
-{
-	unregister_shrinker(&nn->nfsd_client_shrinker);
 }
 
 static void init_nfs4_replay(struct nfs4_replay *rp)
@@ -8099,8 +8089,17 @@ static int nfs4_state_create_net(struct
 	INIT_DELAYED_WORK(&nn->nfsd_shrinker_work, nfsd4_state_shrinker_worker);
 	get_net(net);
 
+	nn->nfsd_client_shrinker.scan_objects = nfsd4_state_shrinker_scan;
+	nn->nfsd_client_shrinker.count_objects = nfsd4_state_shrinker_count;
+	nn->nfsd_client_shrinker.seeks = DEFAULT_SEEKS;
+
+	if (register_shrinker(&nn->nfsd_client_shrinker, "nfsd-client"))
+		goto err_shrinker;
 	return 0;
 
+err_shrinker:
+	put_net(net);
+	kfree(nn->sessionid_hashtbl);
 err_sessionid:
 	kfree(nn->unconf_id_hashtbl);
 err_unconf_id:
@@ -8193,6 +8192,7 @@ nfs4_state_shutdown_net(struct net *net)
 	struct list_head *pos, *next, reaplist;
 	struct nfsd_net *nn = net_generic(net, nfsd_net_id);
 
+	unregister_shrinker(&nn->nfsd_client_shrinker);
 	cancel_delayed_work_sync(&nn->laundromat_work);
 	locks_end_grace(&nn->nfsd4_manager);
 
fs/nfsd/nfsctl.c +1 -6
@@ -1452,9 +1452,7 @@ static __net_init int nfsd_init_net(stru
 		goto out_idmap_error;
 	nn->nfsd_versions = NULL;
 	nn->nfsd4_minorversions = NULL;
-	retval = nfsd4_init_leases_net(nn);
-	if (retval)
-		goto out_drc_error;
+	nfsd4_init_leases_net(nn);
 	retval = nfsd_reply_cache_init(nn);
 	if (retval)
 		goto out_cache_error;
@@ -1464,8 +1462,6 @@ static __net_init int nfsd_init_net(stru
 	return 0;
 
 out_cache_error:
-	nfsd4_leases_net_shutdown(nn);
-out_drc_error:
 	nfsd_idmap_shutdown(net);
 out_idmap_error:
 	nfsd_export_shutdown(net);
@@ -1481,7 +1477,6 @@ static __net_exit void nfsd_exit_net(str
 	nfsd_idmap_shutdown(net);
 	nfsd_export_shutdown(net);
 	nfsd_netns_free_versions(net_generic(net, nfsd_net_id));
-	nfsd4_leases_net_shutdown(nn);
 }
 
 static struct pernet_operations nfsd_net_ops = {
fs/nfsd/nfsd.h +2 -4
@@ -504,8 +504,7 @@ extern void unregister_cld_notifier(void
 extern void nfsd4_ssc_init_umount_work(struct nfsd_net *nn);
 #endif
 
-extern int nfsd4_init_leases_net(struct nfsd_net *nn);
-extern void nfsd4_leases_net_shutdown(struct nfsd_net *nn);
+extern void nfsd4_init_leases_net(struct nfsd_net *nn);
 
 #else /* CONFIG_NFSD_V4 */
 static inline int nfsd4_is_junction(struct dentry *dentry)
@@ -513,8 +512,7 @@ static inline int nfsd4_is_junction(stru
 	return 0;
 }
 
-static inline int nfsd4_init_leases_net(struct nfsd_net *nn) { return 0; };
-static inline void nfsd4_leases_net_shutdown(struct nfsd_net *nn) {};
+static inline void nfsd4_init_leases_net(struct nfsd_net *nn) { };
 
 #define register_cld_notifier() 0
 #define unregister_cld_notifier() do { } while(0)
author maintainer
#182 lore raw

PATCH 178/215 NFSD: replace delayed_work with work_struct for nfsd_client_shrinker

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Dai Ngo <[email protected]>

[ Upstream commit 7c24fa225081f31bc6da6a355c1ba801889ab29a ]

Since nfsd4_state_shrinker_count always calls mod_delayed_work with 0 delay, we can replace delayed_work with work_struct to save some space and overhead.

Also add the call to cancel_work after unregister the shrinker in nfs4_state_shutdown_net.

---
 fs/nfsd/netns.h     |    2 +-
 fs/nfsd/nfs4state.c |    8 ++++----
 2 files changed, 5 insertions(+), 5 deletions(-)
fs/nfsd/netns.h +1 -1
@@ -195,7 +195,7 @@ struct nfsd_net {
 
 	atomic_t		nfsd_courtesy_clients;
 	struct shrinker		nfsd_client_shrinker;
-	struct delayed_work	nfsd_shrinker_work;
+	struct work_struct	nfsd_shrinker_work;
 };
 
 /* Simple check to find out if a given net was properly initialized */
fs/nfsd/nfs4state.c +4 -4
@@ -4412,7 +4412,7 @@ nfsd4_state_shrinker_count(struct shrink
 	if (!count)
 		count = atomic_long_read(&num_delegations);
 	if (count)
-		mod_delayed_work(laundry_wq, &nn->nfsd_shrinker_work, 0);
+		queue_work(laundry_wq, &nn->nfsd_shrinker_work);
 	return (unsigned long)count;
 }
 
@@ -6253,8 +6253,7 @@ deleg_reaper(struct nfsd_net *nn)
 static void
 nfsd4_state_shrinker_worker(struct work_struct *work)
 {
-	struct delayed_work *dwork = to_delayed_work(work);
-	struct nfsd_net *nn = container_of(dwork, struct nfsd_net,
+	struct nfsd_net *nn = container_of(work, struct nfsd_net,
 				nfsd_shrinker_work);
 
 	courtesy_client_reaper(nn);
@@ -8086,7 +8085,7 @@ static int nfs4_state_create_net(struct
 	INIT_LIST_HEAD(&nn->blocked_locks_lru);
 
 	INIT_DELAYED_WORK(&nn->laundromat_work, laundromat_main);
-	INIT_DELAYED_WORK(&nn->nfsd_shrinker_work, nfsd4_state_shrinker_worker);
+	INIT_WORK(&nn->nfsd_shrinker_work, nfsd4_state_shrinker_worker);
 	get_net(net);
 
 	nn->nfsd_client_shrinker.scan_objects = nfsd4_state_shrinker_scan;
@@ -8193,6 +8192,7 @@ nfs4_state_shutdown_net(struct net *net)
 	struct nfsd_net *nn = net_generic(net, nfsd_net_id);
 
 	unregister_shrinker(&nn->nfsd_client_shrinker);
+	cancel_work(&nn->nfsd_shrinker_work);
 	cancel_delayed_work_sync(&nn->laundromat_work);
 	locks_end_grace(&nn->nfsd4_manager);
 
author maintainer
#183 lore raw

PATCH 179/215 nfsd: dont destroy global nfs4_file table in per-net shutdown

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jeff Layton <[email protected]>

[ Upstream commit 4102db175b5d884d133270fdbd0e59111ce688fc ]

The nfs4_file table is global, so shutting it down when a containerized nfsd is shut down is wrong and can lead to double-frees. Tear down the nfs4_file_rhltable in nfs4_state_shutdown instead of nfs4_state_shutdown_net.

Fixes d47b295e8d76 ("NFSD: Use rhashtable for managing nfs4_file objects")
---
 fs/nfsd/nfs4state.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
fs/nfsd/nfs4state.c +1 -1
@@ -8212,7 +8212,6 @@ nfs4_state_shutdown_net(struct net *net)
 
 	nfsd4_client_tracking_exit(net);
 	nfs4_state_destroy_net(net);
-	rhltable_destroy(&nfs4_file_rhltable);
 #ifdef CONFIG_NFSD_V4_2_INTER_SSC
 	nfsd4_ssc_shutdown_umount(nn);
 #endif
@@ -8222,6 +8221,7 @@ void
 nfs4_state_shutdown(void)
 {
 	nfsd4_destroy_callback_queue();
+	rhltable_destroy(&nfs4_file_rhltable);
 }
 
 static void
author maintainer
#184 lore raw

PATCH 180/215 arm64: efi: Limit allocations to 48-bit addressable physical region

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ard Biesheuvel <[email protected]>

From: Ard Biesheuvel <[email protected]>

[ Commit a37dac5c5dcfe0f1fd58513c16cdbc280a47f628 upstream ]

The UEFI spec does not mention or reason about the configured size of the virtual address space at all, but it does mention that all memory should be identity mapped using a page size of 4 KiB.

This means that a LPA2 capable system that has any system memory outside of the 48-bit addressable physical range and follows the spec to the letter may serve page allocation requests from regions of memory that the kernel cannot access unless it was built with LPA2 support and enables it at runtime.

So let's ensure that all page allocations are limited to the 48-bit range.

---
 arch/arm64/include/asm/efi.h               |    1 +
 drivers/firmware/efi/libstub/alignedmem.c  |    2 ++
 drivers/firmware/efi/libstub/arm64-stub.c  |    5 +++--
 drivers/firmware/efi/libstub/efistub.h     |    4 ++++
 drivers/firmware/efi/libstub/mem.c         |    2 ++
 drivers/firmware/efi/libstub/randomalloc.c |    2 +-
 6 files changed, 13 insertions(+), 3 deletions(-)
arch/arm64/include/asm/efi.h +1 -0
@@ -103,6 +103,7 @@ static inline void free_screen_info(stru
 }
 
 #define EFI_ALLOC_ALIGN		SZ_64K
+#define EFI_ALLOC_LIMIT		((1UL << 48) - 1)
 
 /*
  * On ARM systems, virtually remapped UEFI runtime services are set up in two
drivers/firmware/efi/libstub/alignedmem.c +2 -0
@@ -29,6 +29,8 @@ efi_status_t efi_allocate_pages_aligned(
 	efi_status_t status;
 	int slack;
 
+	max = min(max, EFI_ALLOC_LIMIT);
+
 	if (align < EFI_ALLOC_ALIGN)
 		align = EFI_ALLOC_ALIGN;
 
drivers/firmware/efi/libstub/arm64-stub.c +3 -2
@@ -191,10 +191,11 @@ efi_status_t handle_kernel_image(unsigne
 	if (status != EFI_SUCCESS) {
 		if (!check_image_region((u64)_text, kernel_memsize)) {
 			efi_err("FIRMWARE BUG: Image BSS overlaps adjacent EFI memory region\n");
-		} else if (IS_ALIGNED((u64)_text, min_kimg_align)) {
+		} else if (IS_ALIGNED((u64)_text, min_kimg_align) &&
+			   (u64)_end < EFI_ALLOC_LIMIT) {
 			/*
 			 * Just execute from wherever we were loaded by the
-			 * UEFI PE/COFF loader if the alignment is suitable.
+			 * UEFI PE/COFF loader if the placement is suitable.
 			 */
 			*image_addr = (u64)_text;
 			*reserve_size = 0;
drivers/firmware/efi/libstub/efistub.h +4 -0
@@ -29,6 +29,10 @@
 #define EFI_ALLOC_ALIGN		EFI_PAGE_SIZE
 #endif
 
+#ifndef EFI_ALLOC_LIMIT
+#define EFI_ALLOC_LIMIT		ULONG_MAX
+#endif
+
 extern bool efi_nochunk;
 extern bool efi_nokaslr;
 extern int efi_loglevel;
drivers/firmware/efi/libstub/mem.c +2 -0
@@ -89,6 +89,8 @@ efi_status_t efi_allocate_pages(unsigned
 	efi_physical_addr_t alloc_addr;
 	efi_status_t status;
 
+	max = min(max, EFI_ALLOC_LIMIT);
+
 	if (EFI_ALLOC_ALIGN > EFI_PAGE_SIZE)
 		return efi_allocate_pages_aligned(size, addr, max,
 						  EFI_ALLOC_ALIGN,
drivers/firmware/efi/libstub/randomalloc.c +1 -1
@@ -29,7 +29,7 @@ static unsigned long get_entry_num_slots
 		return 0;
 
 	region_end = min(md->phys_addr + md->num_pages * EFI_PAGE_SIZE - 1,
-			 (u64)ULONG_MAX);
+			 (u64)EFI_ALLOC_LIMIT);
 	if (region_end < size)
 		return 0;
 
author maintainer
#185 lore raw

PATCH 181/215 efi: efivars: prevent double registration

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ard Biesheuvel <[email protected]>

From: Johan Hovold <[email protected]>

[ Commit 0217a40d7ba6e71d7f3422fbe89b436e8ee7ece7 upstream ]

Add the missing sanity check to efivars_register() so that it is no longer possible to override an already registered set of efivar ops (without first deregistering them).

This can help debug initialisation ordering issues where drivers have so far unknowingly been relying on overriding the generic ops.

---
 drivers/firmware/efi/vars.c |   13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)
drivers/firmware/efi/vars.c +11 -2
@@ -66,19 +66,28 @@ int efivars_register(struct efivars *efi
 		     const struct efivar_operations *ops,
 		     struct kobject *kobject)
 {
+	int rv;
+
 	if (down_interruptible(&efivars_lock))
 		return -EINTR;
 
+	if (__efivars) {
+		pr_warn("efivars already registered\n");
+		rv = -EBUSY;
+		goto out;
+	}
+
 	efivars->ops = ops;
 	efivars->kobject = kobject;
 
 	__efivars = efivars;
 
 	pr_info("Registered efivars operations\n");
-
+	rv = 0;
+out:
 	up(&efivars_lock);
 
-	return 0;
+	return rv;
 }
 EXPORT_SYMBOL_GPL(efivars_register);
 
author maintainer
#186 lore raw

PATCH 182/215 x86/efistub: Simplify and clean up handover entry code

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ard Biesheuvel <[email protected]>

From: Ard Biesheuvel <[email protected]>

[ Commit df9215f15206c2a81909ccf60f21d170801dce38 upstream ]

Now that the EFI entry code in assembler is only used by the optional and deprecated EFI handover protocol, and given that the EFI stub C code no longer returns to it, most of it can simply be dropped.

While at it, clarify the symbol naming, by merging efi_main() and efi_stub_entry(), making the latter the shared entry point for all different boot modes that enter via the EFI stub.

The efi32_stub_entry() and efi64_stub_entry() names are referenced explicitly by the tooling that populates the setup header, so these must be retained, but can be emitted as aliases of efi_stub_entry() where appropriate.

---
 Documentation/x86/boot.rst              |    2 +-
 arch/x86/boot/compressed/efi_mixed.S    |   22 ++++++++++++----------
 arch/x86/boot/compressed/head_32.S      |   11 -----------
 arch/x86/boot/compressed/head_64.S      |   12 ++----------
 drivers/firmware/efi/libstub/x86-stub.c |   20 ++++++++++++++++----
 5 files changed, 31 insertions(+), 36 deletions(-)
Documentation/x86/boot.rst +1 -1
@@ -1416,7 +1416,7 @@ execution context provided by the EFI fi
 
 The function prototype for the handover entry point looks like this::
 
-    efi_main(void *handle, efi_system_table_t *table, struct boot_params *bp)
+    efi_stub_entry(void *handle, efi_system_table_t *table, struct boot_params *bp)
 
 'handle' is the EFI image handle passed to the boot loader by the EFI
 firmware, 'table' is the EFI system table - these are the first two
arch/x86/boot/compressed/efi_mixed.S +12 -10
@@ -26,8 +26,8 @@
  * When booting in 64-bit mode on 32-bit EFI firmware, startup_64_mixed_mode()
  * is the first thing that runs after switching to long mode. Depending on
  * whether the EFI handover protocol or the compat entry point was used to
- * enter the kernel, it will either branch to the 64-bit EFI handover
- * entrypoint at offset 0x390 in the image, or to the 64-bit EFI PE/COFF
+ * enter the kernel, it will either branch to the common 64-bit EFI stub
+ * entrypoint efi_stub_entry() directly, or via the 64-bit EFI PE/COFF
  * entrypoint efi_pe_entry(). In the former case, the bootloader must provide a
  * struct bootparams pointer as the third argument, so the presence of such a
  * pointer is used to disambiguate.
@@ -37,21 +37,23 @@
  *  | efi32_pe_entry   |---->|            |            |       +-----------+--+
  *  +------------------+     |            |     +------+----------------+  |
  *                           | startup_32 |---->| startup_64_mixed_mode |  |
- *  +------------------+     |            |     +------+----------------+  V
- *  | efi32_stub_entry |---->|            |            |     +------------------+
- *  +------------------+     +------------+            +---->| efi64_stub_entry |
- *                                                           +-------------+----+
- *                           +------------+     +----------+               |
- *                           | startup_64 |<----| efi_main |<--------------+
- *                           +------------+     +----------+
+ *  +------------------+     |            |     +------+----------------+  |
+ *  | efi32_stub_entry |---->|            |            |                   |
+ *  +------------------+     +------------+            |                   |
+ *                                                     V                   |
+ *                           +------------+     +----------------+         |
+ *                           | startup_64 |<----| efi_stub_entry |<--------+
+ *                           +------------+     +----------------+
  */
 SYM_FUNC_START(startup_64_mixed_mode)
 	lea	efi32_boot_args(%rip), %rdx
 	mov	0(%rdx), %edi
 	mov	4(%rdx), %esi
+#ifdef CONFIG_EFI_HANDOVER_PROTOCOL
 	mov	8(%rdx), %edx		// saved bootparams pointer
 	test	%edx, %edx
-	jnz	efi64_stub_entry
+	jnz	efi_stub_entry
+#endif
 	/*
 	 * efi_pe_entry uses MS calling convention, which requires 32 bytes of
 	 * shadow space on the stack even if all arguments are passed in
arch/x86/boot/compressed/head_32.S +0 -11
@@ -150,17 +150,6 @@ SYM_FUNC_START(startup_32)
 	jmp	*%eax
 SYM_FUNC_END(startup_32)
 
-#ifdef CONFIG_EFI_STUB
-SYM_FUNC_START(efi32_stub_entry)
-	add	$0x4, %esp
-	movl	8(%esp), %esi	/* save boot_params pointer */
-	call	efi_main
-	/* efi_main returns the possibly relocated address of startup_32 */
-	jmp	*%eax
-SYM_FUNC_END(efi32_stub_entry)
-SYM_FUNC_ALIAS(efi_stub_entry, efi32_stub_entry)
-#endif
-
 	.text
 SYM_FUNC_START_LOCAL_NOALIGN(.Lrelocated)
 
arch/x86/boot/compressed/head_64.S +2 -10
@@ -474,19 +474,11 @@ SYM_CODE_START(startup_64)
 	jmp	*%rax
 SYM_CODE_END(startup_64)
 
-#ifdef CONFIG_EFI_STUB
-#ifdef CONFIG_EFI_HANDOVER_PROTOCOL
+#if IS_ENABLED(CONFIG_EFI_MIXED) && IS_ENABLED(CONFIG_EFI_HANDOVER_PROTOCOL)
 	.org 0x390
-#endif
 SYM_FUNC_START(efi64_stub_entry)
-	and	$~0xf, %rsp			/* realign the stack */
-	movq	%rdx, %rbx			/* save boot_params pointer */
-	call	efi_main
-	movq	%rbx,%rsi
-	leaq	rva(startup_64)(%rax), %rax
-	jmp	*%rax
+	jmp	efi_stub_entry
 SYM_FUNC_END(efi64_stub_entry)
-SYM_FUNC_ALIAS(efi_stub_entry, efi64_stub_entry)
 #endif
 
 	.text
drivers/firmware/efi/libstub/x86-stub.c +16 -4
@@ -774,9 +774,9 @@ static void __noreturn enter_kernel(unsi
  * return.  On failure, it will exit to the firmware via efi_exit() instead of
  * returning.
  */
-asmlinkage unsigned long efi_main(efi_handle_t handle,
-				  efi_system_table_t *sys_table_arg,
-				  struct boot_params *boot_params)
+void __noreturn efi_stub_entry(efi_handle_t handle,
+			       efi_system_table_t *sys_table_arg,
+			       struct boot_params *boot_params)
 {
 	unsigned long bzimage_addr = (unsigned long)startup_32;
 	unsigned long buffer_start, buffer_end;
@@ -919,7 +919,19 @@ asmlinkage unsigned long efi_main(efi_ha
 
 	enter_kernel(bzimage_addr, boot_params);
 fail:
-	efi_err("efi_main() failed!\n");
+	efi_err("efi_stub_entry() failed!\n");
 
 	efi_exit(handle, status);
 }
+
+#ifdef CONFIG_EFI_HANDOVER_PROTOCOL
+#ifndef CONFIG_EFI_MIXED
+extern __alias(efi_stub_entry)
+void efi32_stub_entry(efi_handle_t handle, efi_system_table_t *sys_table_arg,
+		      struct boot_params *boot_params);
+
+extern __alias(efi_stub_entry)
+void efi64_stub_entry(efi_handle_t handle, efi_system_table_t *sys_table_arg,
+		      struct boot_params *boot_params);
+#endif
+#endif
author maintainer
#187 lore raw

PATCH 183/215 x86/decompressor: Avoid magic offsets for EFI handover entrypoint

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ard Biesheuvel <[email protected]>

From: Ard Biesheuvel <[email protected]>

[ Commit 12792064587623065250069d1df980e2c9ac3e67 upstream ]

The native 32-bit or 64-bit EFI handover protocol entrypoint offset relative to the respective startup_32/64 address is described in boot_params as handover_offset, so that the special Linux/x86 aware EFI loader can find it there.

When mixed mode is enabled, this single field has to describe this offset for both the 32-bit and 64-bit entrypoints, so their respective relative offsets have to be identical. Given that startup_32 and startup_64 are 0x200 bytes apart, and the EFI handover entrypoint resides at a fixed offset, the 32-bit and 64-bit versions of those entrypoints must be exactly 0x200 bytes apart as well.

Currently, hard-coded fixed offsets are used to ensure this, but it is sufficient to emit the 64-bit entrypoint 0x200 bytes after the 32-bit one, wherever it happens to reside. This allows this code (which is now EFI mixed mode specific) to be moved into efi_mixed.S and out of the startup code in head_64.S.

---
 arch/x86/boot/compressed/efi_mixed.S |   20 +++++++++++++++++++-
 arch/x86/boot/compressed/head_64.S   |   18 ------------------
 2 files changed, 19 insertions(+), 19 deletions(-)
arch/x86/boot/compressed/efi_mixed.S +19 -1
@@ -146,6 +146,16 @@ SYM_FUNC_START(__efi64_thunk)
 SYM_FUNC_END(__efi64_thunk)
 
 	.code32
+#ifdef CONFIG_EFI_HANDOVER_PROTOCOL
+SYM_FUNC_START(efi32_stub_entry)
+	add	$0x4, %esp		/* Discard return address */
+	popl	%ecx
+	popl	%edx
+	popl	%esi
+	jmp	efi32_entry
+SYM_FUNC_END(efi32_stub_entry)
+#endif
+
 /*
  * EFI service pointer must be in %edi.
  *
@@ -226,7 +236,7 @@ SYM_FUNC_END(efi_enter32)
  * stub may still exit and return to the firmware using the Exit() EFI boot
  * service.]
  */
-SYM_FUNC_START(efi32_entry)
+SYM_FUNC_START_LOCAL(efi32_entry)
 	call	1f
 1:	pop	%ebx
 
@@ -326,6 +336,14 @@ SYM_FUNC_START(efi32_pe_entry)
 	RET
 SYM_FUNC_END(efi32_pe_entry)
 
+#ifdef CONFIG_EFI_HANDOVER_PROTOCOL
+	.org	efi32_stub_entry + 0x200
+	.code64
+SYM_FUNC_START_NOALIGN(efi64_stub_entry)
+	jmp	efi_stub_entry
+SYM_FUNC_END(efi64_stub_entry)
+#endif
+
 	.section ".rodata"
 	/* EFI loaded image protocol GUID */
 	.balign 4
arch/x86/boot/compressed/head_64.S +0 -18
@@ -286,17 +286,6 @@ SYM_FUNC_START(startup_32)
 	lret
 SYM_FUNC_END(startup_32)
 
-#if IS_ENABLED(CONFIG_EFI_MIXED) && IS_ENABLED(CONFIG_EFI_HANDOVER_PROTOCOL)
-	.org 0x190
-SYM_FUNC_START(efi32_stub_entry)
-	add	$0x4, %esp		/* Discard return address */
-	popl	%ecx
-	popl	%edx
-	popl	%esi
-	jmp	efi32_entry
-SYM_FUNC_END(efi32_stub_entry)
-#endif
-
 	.code64
 	.org 0x200
 SYM_CODE_START(startup_64)
@@ -474,13 +463,6 @@ SYM_CODE_START(startup_64)
 	jmp	*%rax
 SYM_CODE_END(startup_64)
 
-#if IS_ENABLED(CONFIG_EFI_MIXED) && IS_ENABLED(CONFIG_EFI_HANDOVER_PROTOCOL)
-	.org 0x390
-SYM_FUNC_START(efi64_stub_entry)
-	jmp	efi_stub_entry
-SYM_FUNC_END(efi64_stub_entry)
-#endif
-
 	.text
 SYM_FUNC_START_LOCAL_NOALIGN(.Lrelocated)
 
author maintainer
#188 lore raw

PATCH 184/215 x86/efistub: Clear BSS in EFI handover protocol entrypoint

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ard Biesheuvel <[email protected]>

From: Ard Biesheuvel <[email protected]>

[ Commit d7156b986d4cc0657fa6dc05c9fcf51c3d55a0fe upstream ]

The so-called EFI handover protocol is value-add from the distros that permits a loader to simply copy a PE kernel image into memory and call an alternative entrypoint that is described by an embedded boot_params structure.

Most implementations of this protocol do not bother to check the PE header for minimum alignment, section placement, etc, and therefore also don't clear the image's BSS, or even allocate enough memory for it.

Allocating more memory on the fly is rather difficult, but at least clear the BSS region explicitly when entering in this manner, so that the EFI stub code does not get confused by global variables that were not zero-initialized correctly.

When booting in mixed mode, this BSS clearing must occur before any global state is created, so clear it in the 32-bit asm entry point.

---
 arch/x86/boot/compressed/efi_mixed.S    |   14 +++++++++++++-
 drivers/firmware/efi/libstub/x86-stub.c |   13 +++++++++++--
 2 files changed, 24 insertions(+), 3 deletions(-)
arch/x86/boot/compressed/efi_mixed.S +13 -1
@@ -148,6 +148,18 @@ SYM_FUNC_END(__efi64_thunk)
 	.code32
 #ifdef CONFIG_EFI_HANDOVER_PROTOCOL
 SYM_FUNC_START(efi32_stub_entry)
+	call	1f
+1:	popl	%ecx
+
+	/* Clear BSS */
+	xorl	%eax, %eax
+	leal	(_bss - 1b)(%ecx), %edi
+	leal	(_ebss - 1b)(%ecx), %ecx
+	subl	%edi, %ecx
+	shrl	$2, %ecx
+	cld
+	rep	stosl
+
 	add	$0x4, %esp		/* Discard return address */
 	popl	%ecx
 	popl	%edx
@@ -340,7 +352,7 @@ SYM_FUNC_END(efi32_pe_entry)
 	.org	efi32_stub_entry + 0x200
 	.code64
 SYM_FUNC_START_NOALIGN(efi64_stub_entry)
-	jmp	efi_stub_entry
+	jmp	efi_handover_entry
 SYM_FUNC_END(efi64_stub_entry)
 #endif
 
drivers/firmware/efi/libstub/x86-stub.c +11 -2
@@ -925,12 +925,21 @@ fail:
 }
 
 #ifdef CONFIG_EFI_HANDOVER_PROTOCOL
+void efi_handover_entry(efi_handle_t handle, efi_system_table_t *sys_table_arg,
+			struct boot_params *boot_params)
+{
+	extern char _bss[], _ebss[];
+
+	memset(_bss, 0, _ebss - _bss);
+	efi_stub_entry(handle, sys_table_arg, boot_params);
+}
+
 #ifndef CONFIG_EFI_MIXED
-extern __alias(efi_stub_entry)
+extern __alias(efi_handover_entry)
 void efi32_stub_entry(efi_handle_t handle, efi_system_table_t *sys_table_arg,
 		      struct boot_params *boot_params);
 
-extern __alias(efi_stub_entry)
+extern __alias(efi_handover_entry)
 void efi64_stub_entry(efi_handle_t handle, efi_system_table_t *sys_table_arg,
 		      struct boot_params *boot_params);
 #endif
author maintainer
#189 lore raw

PATCH 185/215 efi/libstub: Add memory attribute protocol definitions

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ard Biesheuvel <[email protected]>

From: Evgeniy Baskov <[email protected]>

[ Commit 79729f26b074a5d2722c27fa76cc45ef721e65cd upstream ]

EFI_MEMORY_ATTRIBUTE_PROTOCOL servers as a better alternative to DXE services for setting memory attributes in EFI Boot Services environment. This protocol is better since it is a part of UEFI specification itself and not UEFI PI specification like DXE services.

Add EFI_MEMORY_ATTRIBUTE_PROTOCOL definitions.
Support mixed mode properly for its calls.

---
 arch/x86/include/asm/efi.h             |    7 +++++++
 drivers/firmware/efi/libstub/efistub.h |   20 ++++++++++++++++++++
 include/linux/efi.h                    |    1 +
 3 files changed, 28 insertions(+)
arch/x86/include/asm/efi.h +7 -0
@@ -325,6 +325,13 @@ static inline u32 efi64_convert_status(e
 #define __efi64_argmap_set_memory_space_attributes(phys, size, flags) \
 	(__efi64_split(phys), __efi64_split(size), __efi64_split(flags))
 
+/* Memory Attribute Protocol */
+#define __efi64_argmap_set_memory_attributes(protocol, phys, size, flags) \
+	((protocol), __efi64_split(phys), __efi64_split(size), __efi64_split(flags))
+
+#define __efi64_argmap_clear_memory_attributes(protocol, phys, size, flags) \
+	((protocol), __efi64_split(phys), __efi64_split(size), __efi64_split(flags))
+
 /*
  * The macros below handle the plumbing for the argument mapping. To add a
  * mapping for a specific EFI method, simply define a macro
drivers/firmware/efi/libstub/efistub.h +20 -0
@@ -419,6 +419,26 @@ union efi_dxe_services_table {
 	} mixed_mode;
 };
 
+typedef union efi_memory_attribute_protocol efi_memory_attribute_protocol_t;
+
+union efi_memory_attribute_protocol {
+	struct {
+		efi_status_t (__efiapi *get_memory_attributes)(
+			efi_memory_attribute_protocol_t *, efi_physical_addr_t, u64, u64 *);
+
+		efi_status_t (__efiapi *set_memory_attributes)(
+			efi_memory_attribute_protocol_t *, efi_physical_addr_t, u64, u64);
+
+		efi_status_t (__efiapi *clear_memory_attributes)(
+			efi_memory_attribute_protocol_t *, efi_physical_addr_t, u64, u64);
+	};
+	struct {
+		u32 get_memory_attributes;
+		u32 set_memory_attributes;
+		u32 clear_memory_attributes;
+	} mixed_mode;
+};
+
 typedef union efi_uga_draw_protocol efi_uga_draw_protocol_t;
 
 union efi_uga_draw_protocol {
include/linux/efi.h +1 -0
@@ -390,6 +390,7 @@ void efi_native_runtime_setup(void);
 #define EFI_RT_PROPERTIES_TABLE_GUID		EFI_GUID(0xeb66918a, 0x7eef, 0x402a,  0x84, 0x2e, 0x93, 0x1d, 0x21, 0xc3, 0x8a, 0xe9)
 #define EFI_DXE_SERVICES_TABLE_GUID		EFI_GUID(0x05ad34ba, 0x6f02, 0x4214,  0x95, 0x2e, 0x4d, 0xa0, 0x39, 0x8e, 0x2b, 0xb9)
 #define EFI_SMBIOS_PROTOCOL_GUID		EFI_GUID(0x03583ff6, 0xcb36, 0x4940,  0x94, 0x7e, 0xb9, 0xb3, 0x9f, 0x4a, 0xfa, 0xf7)
+#define EFI_MEMORY_ATTRIBUTE_PROTOCOL_GUID	EFI_GUID(0xf4560cf6, 0x40ec, 0x4b4a,  0xa1, 0x92, 0xbf, 0x1d, 0x57, 0xd0, 0xb1, 0x89)
 
 #define EFI_IMAGE_SECURITY_DATABASE_GUID	EFI_GUID(0xd719b2cb, 0x3d3a, 0x4596,  0xa3, 0xbc, 0xda, 0xd0, 0x0e, 0x67, 0x65, 0x6f)
 #define EFI_SHIM_LOCK_GUID			EFI_GUID(0x605dab50, 0xe046, 0x4300,  0xab, 0xb6, 0x3d, 0xd8, 0x10, 0xdd, 0x8b, 0x23)
author maintainer
#190 lore raw

PATCH 186/215 efi/libstub: Add limit argument to efi_random_alloc()

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ard Biesheuvel <[email protected]>

From: Ard Biesheuvel <[email protected]>

[ Commit bc5ddceff4c14494d83449ad45c985e6cd353fce upstream ]

x86 will need to limit the kernel memory allocation to the lowest 512 MiB of memory, to match the behavior of the existing bare metal KASLR physical randomization logic. So in preparation for that, add a limit parameter to efi_random_alloc() and wire it up.

---
 drivers/firmware/efi/libstub/arm64-stub.c  |    2 +-
 drivers/firmware/efi/libstub/efistub.h     |    2 +-
 drivers/firmware/efi/libstub/randomalloc.c |   10 ++++++----
 3 files changed, 8 insertions(+), 6 deletions(-)
drivers/firmware/efi/libstub/arm64-stub.c +1 -1
@@ -181,7 +181,7 @@ efi_status_t handle_kernel_image(unsigne
 		 */
 		status = efi_random_alloc(*reserve_size, min_kimg_align,
 					  reserve_addr, phys_seed,
-					  EFI_LOADER_CODE);
+					  EFI_LOADER_CODE, EFI_ALLOC_LIMIT);
 		if (status != EFI_SUCCESS)
 			efi_warn("efi_random_alloc() failed: 0x%lx\n", status);
 	} else {
drivers/firmware/efi/libstub/efistub.h +1 -1
@@ -905,7 +905,7 @@ efi_status_t efi_get_random_bytes(unsign
 
 efi_status_t efi_random_alloc(unsigned long size, unsigned long align,
 			      unsigned long *addr, unsigned long random_seed,
-			      int memory_type);
+			      int memory_type, unsigned long alloc_limit);
 
 efi_status_t efi_random_get_seed(void);
 
drivers/firmware/efi/libstub/randomalloc.c +6 -4
@@ -16,7 +16,8 @@
  */
 static unsigned long get_entry_num_slots(efi_memory_desc_t *md,
 					 unsigned long size,
-					 unsigned long align_shift)
+					 unsigned long align_shift,
+					 u64 alloc_limit)
 {
 	unsigned long align = 1UL << align_shift;
 	u64 first_slot, last_slot, region_end;
@@ -29,7 +30,7 @@ static unsigned long get_entry_num_slots
 		return 0;
 
 	region_end = min(md->phys_addr + md->num_pages * EFI_PAGE_SIZE - 1,
-			 (u64)EFI_ALLOC_LIMIT);
+			 alloc_limit);
 	if (region_end < size)
 		return 0;
 
@@ -54,7 +55,8 @@ efi_status_t efi_random_alloc(unsigned l
 			      unsigned long align,
 			      unsigned long *addr,
 			      unsigned long random_seed,
-			      int memory_type)
+			      int memory_type,
+			      unsigned long alloc_limit)
 {
 	unsigned long total_slots = 0, target_slot;
 	unsigned long total_mirrored_slots = 0;
@@ -76,7 +78,7 @@ efi_status_t efi_random_alloc(unsigned l
 		efi_memory_desc_t *md = (void *)map->map + map_offset;
 		unsigned long slots;
 
-		slots = get_entry_num_slots(md, size, ilog2(align));
+		slots = get_entry_num_slots(md, size, ilog2(align), alloc_limit);
 		MD_NUM_SLOTS(md) = slots;
 		total_slots += slots;
 		if (md->attribute & EFI_MEMORY_MORE_RELIABLE)
author maintainer
#191 lore raw

PATCH 187/215 x86/efistub: Perform 4/5 level paging switch from the stub

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ard Biesheuvel <[email protected]>

From: Ard Biesheuvel <[email protected]>

[ Commit cb380000dd23cbbf8bd7d023b51896804c1f7e68 upstream ]

In preparation for updating the EFI stub boot flow to avoid the bare metal decompressor code altogether, implement the support code for switching between 4 and 5 levels of paging before jumping to the kernel proper.

This reuses the newly refactored trampoline that the bare metal decompressor uses, but relies on EFI APIs to allocate 32-bit addressable memory and remap it with the appropriate permissions. Given that the bare metal decompressor will no longer call into the trampoline if the number of paging levels is already set correctly, it is no longer needed to remove NX restrictions from the memory range where this trampoline may end up.

---
 drivers/firmware/efi/libstub/Makefile          |    1 
 drivers/firmware/efi/libstub/efi-stub-helper.c |    2 
 drivers/firmware/efi/libstub/efistub.h         |    1 
 drivers/firmware/efi/libstub/x86-5lvl.c        |   95 +++++++++++++++++++++++++
 drivers/firmware/efi/libstub/x86-stub.c        |   40 +++-------
 drivers/firmware/efi/libstub/x86-stub.h        |   17 ++++
 6 files changed, 130 insertions(+), 26 deletions(-)
drivers/firmware/efi/libstub/Makefile +1 -0
@@ -84,6 +84,7 @@ lib-$(CONFIG_EFI_GENERIC_STUB)	+= efi-st
 lib-$(CONFIG_ARM)		+= arm32-stub.o
 lib-$(CONFIG_ARM64)		+= arm64-stub.o smbios.o
 lib-$(CONFIG_X86)		+= x86-stub.o
+lib-$(CONFIG_X86_64)		+= x86-5lvl.o
 lib-$(CONFIG_RISCV)		+= riscv-stub.o
 lib-$(CONFIG_LOONGARCH)		+= loongarch-stub.o
 
drivers/firmware/efi/libstub/efi-stub-helper.c +2 -0
@@ -216,6 +216,8 @@ efi_status_t efi_parse_options(char cons
 			efi_loglevel = CONSOLE_LOGLEVEL_QUIET;
 		} else if (!strcmp(param, "noinitrd")) {
 			efi_noinitrd = true;
+		} else if (IS_ENABLED(CONFIG_X86_64) && !strcmp(param, "no5lvl")) {
+			efi_no5lvl = true;
 		} else if (!strcmp(param, "efi") && val) {
 			efi_nochunk = parse_option_str(val, "nochunk");
 			efi_novamap |= parse_option_str(val, "novamap");
drivers/firmware/efi/libstub/efistub.h +1 -0
@@ -33,6 +33,7 @@
 #define EFI_ALLOC_LIMIT		ULONG_MAX
 #endif
 
+extern bool efi_no5lvl;
 extern bool efi_nochunk;
 extern bool efi_nokaslr;
 extern int efi_loglevel;
drivers/firmware/efi/libstub/x86-5lvl.c new file +95 -0

95 lines not on this page · raw message

drivers/firmware/efi/libstub/x86-stub.c +14 -26
@@ -17,6 +17,7 @@
 #include <asm/boot.h>
 
 #include "efistub.h"
+#include "x86-stub.h"
 
 /* Maximum physical address for 64-bit kernel with 4-level paging */
 #define MAXMEM_X86_64_4LEVEL (1ull << 46)
@@ -212,8 +213,8 @@ static void retrieve_apple_device_proper
 	}
 }
 
-static void
-adjust_memory_range_protection(unsigned long start, unsigned long size)
+void efi_adjust_memory_range_protection(unsigned long start,
+					unsigned long size)
 {
 	efi_status_t status;
 	efi_gcd_memory_space_desc_t desc;
@@ -267,35 +268,14 @@ adjust_memory_range_protection(unsigned
 	}
 }
 
-/*
- * Trampoline takes 2 pages and can be loaded in first megabyte of memory
- * with its end placed between 128k and 640k where BIOS might start.
- * (see arch/x86/boot/compressed/pgtable_64.c)
- *
- * We cannot find exact trampoline placement since memory map
- * can be modified by UEFI, and it can alter the computed address.
- */
-
-#define TRAMPOLINE_PLACEMENT_BASE ((128 - 8)*1024)
-#define TRAMPOLINE_PLACEMENT_SIZE (640*1024 - (128 - 8)*1024)
-
 extern const u8 startup_32[], startup_64[];
 
 static void
 setup_memory_protection(unsigned long image_base, unsigned long image_size)
 {
-	/*
-	 * Allow execution of possible trampoline used
-	 * for switching between 4- and 5-level page tables
-	 * and relocated kernel image.
-	 */
-
-	adjust_memory_range_protection(TRAMPOLINE_PLACEMENT_BASE,
-				       TRAMPOLINE_PLACEMENT_SIZE);
-
 #ifdef CONFIG_64BIT
 	if (image_base != (unsigned long)startup_32)
-		adjust_memory_range_protection(image_base, image_size);
+		efi_adjust_memory_range_protection(image_base, image_size);
 #else
 	/*
 	 * Clear protection flags on a whole range of possible
@@ -305,8 +285,8 @@ setup_memory_protection(unsigned long im
 	 * need to remove possible protection on relocated image
 	 * itself disregarding further relocations.
 	 */
-	adjust_memory_range_protection(LOAD_PHYSICAL_ADDR,
-				       KERNEL_IMAGE_SIZE - LOAD_PHYSICAL_ADDR);
+	efi_adjust_memory_range_protection(LOAD_PHYSICAL_ADDR,
+					   KERNEL_IMAGE_SIZE - LOAD_PHYSICAL_ADDR);
 #endif
 }
 
@@ -796,6 +776,12 @@ void __noreturn efi_stub_entry(efi_handl
 		efi_dxe_table = NULL;
 	}
 
+	status = efi_setup_5level_paging();
+	if (status != EFI_SUCCESS) {
+		efi_err("efi_setup_5level_paging() failed!\n");
+		goto fail;
+	}
+
 	/*
 	 * If the kernel isn't already loaded at a suitable address,
 	 * relocate it.
@@ -914,6 +900,8 @@ void __noreturn efi_stub_entry(efi_handl
 		goto fail;
 	}
 
+	efi_5level_switch();
+
 	if (IS_ENABLED(CONFIG_X86_64))
 		bzimage_addr += startup_64 - startup_32;
 
drivers/firmware/efi/libstub/x86-stub.h new file +17 -0
@@ -0,0 +1,17 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+
+#include <linux/efi.h>
+
+extern void trampoline_32bit_src(void *, bool);
+extern const u16 trampoline_ljmp_imm_offset;
+
+void efi_adjust_memory_range_protection(unsigned long start,
+					unsigned long size);
+
+#ifdef CONFIG_X86_64
+efi_status_t efi_setup_5level_paging(void);
+void efi_5level_switch(void);
+#else
+static inline efi_status_t efi_setup_5level_paging(void) { return EFI_SUCCESS; }
+static inline void efi_5level_switch(void) {}
+#endif
author maintainer
#192 lore raw

PATCH 188/215 x86/decompressor: Factor out kernel decompression and relocation

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ard Biesheuvel <[email protected]>

From: Ard Biesheuvel <[email protected]>

[ Commit 83381519352d6b5b3e429bf72aaab907480cb6b6 upstream ]

Factor out the decompressor sequence that invokes the decompressor, parses the ELF and applies the relocations so that it can be called directly from the EFI stub.

---
 arch/x86/boot/compressed/misc.c |   29 ++++++++++++++++++++++++-----
 arch/x86/include/asm/boot.h     |    8 ++++++++
 2 files changed, 32 insertions(+), 5 deletions(-)
arch/x86/boot/compressed/misc.c +24 -5
@@ -330,11 +330,33 @@ static size_t parse_elf(void *output)
 	return ehdr.e_entry - LOAD_PHYSICAL_ADDR;
 }
 
+const unsigned long kernel_total_size = VO__end - VO__text;
+
 static u8 boot_heap[BOOT_HEAP_SIZE] __aligned(4);
 
 extern unsigned char input_data[];
 extern unsigned int input_len, output_len;
 
+unsigned long decompress_kernel(unsigned char *outbuf, unsigned long virt_addr,
+				void (*error)(char *x))
+{
+	unsigned long entry;
+
+	if (!free_mem_ptr) {
+		free_mem_ptr     = (unsigned long)boot_heap;
+		free_mem_end_ptr = (unsigned long)boot_heap + sizeof(boot_heap);
+	}
+
+	if (__decompress(input_data, input_len, NULL, NULL, outbuf, output_len,
+			 NULL, error) < 0)
+		return ULONG_MAX;
+
+	entry = parse_elf(outbuf);
+	handle_relocations(outbuf, output_len, virt_addr);
+
+	return entry;
+}
+
 /*
  * The compressed kernel image (ZO), has been moved so that its position
  * is against the end of the buffer used to hold the uncompressed kernel
@@ -354,7 +376,6 @@ extern unsigned int input_len, output_le
  */
 asmlinkage __visible void *extract_kernel(void *rmode, unsigned char *output)
 {
-	const unsigned long kernel_total_size = VO__end - VO__text;
 	unsigned long virt_addr = LOAD_PHYSICAL_ADDR;
 	memptr heap = (memptr)boot_heap;
 	unsigned long needed_size;
@@ -457,10 +478,8 @@ asmlinkage __visible void *extract_kerne
 #endif
 
 	debug_putstr("\nDecompressing Linux... ");
-	__decompress(input_data, input_len, NULL, NULL, output, output_len,
-			NULL, error);
-	entry_offset = parse_elf(output);
-	handle_relocations(output, output_len, virt_addr);
+
+	entry_offset = decompress_kernel(output, virt_addr, error);
 
 	debug_putstr("done.\nBooting the kernel (entry_offset: 0x");
 	debug_puthex(entry_offset);
arch/x86/include/asm/boot.h +8 -0
@@ -79,4 +79,12 @@
 # define BOOT_STACK_SIZE	0x1000
 #endif
 
+#ifndef __ASSEMBLY__
+extern unsigned int output_len;
+extern const unsigned long kernel_total_size;
+
+unsigned long decompress_kernel(unsigned char *outbuf, unsigned long virt_addr,
+				void (*error)(char *x));
+#endif
+
 #endif /* _ASM_X86_BOOT_H */
author maintainer
#193 lore raw

PATCH 189/215 x86/efistub: Prefer EFI memory attributes protocol over DXE services

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ard Biesheuvel <[email protected]>

From: Ard Biesheuvel <[email protected]>

[ Commit 11078876b7a6a1b7226344fecab968945c806832 upstream ]

Currently, the EFI stub relies on DXE services in some cases to clear non-execute restrictions from page allocations that need to be executable. This is dodgy, because DXE services are not specified by UEFI but by PI, and they are not intended for consumption by OS loaders. However, no alternative existed at the time.

Now, there is a new UEFI protocol that should be used instead, so if it exists, prefer it over the DXE services calls.

---
 drivers/firmware/efi/libstub/x86-stub.c |   29 +++++++++++++++++++++--------
 1 file changed, 21 insertions(+), 8 deletions(-)
drivers/firmware/efi/libstub/x86-stub.c +21 -8
@@ -26,6 +26,7 @@ const efi_system_table_t *efi_system_tab
 const efi_dxe_services_table_t *efi_dxe_table;
 u32 image_offset __section(".data");
 static efi_loaded_image_t *image = NULL;
+static efi_memory_attribute_protocol_t *memattr;
 
 static efi_status_t
 preserve_pci_rom_image(efi_pci_io_protocol_t *pci, struct pci_setup_rom **__rom)
@@ -222,12 +223,18 @@ void efi_adjust_memory_range_protection(
 	unsigned long rounded_start, rounded_end;
 	unsigned long unprotect_start, unprotect_size;
 
-	if (efi_dxe_table == NULL)
-		return;
-
 	rounded_start = rounddown(start, EFI_PAGE_SIZE);
 	rounded_end = roundup(start + size, EFI_PAGE_SIZE);
 
+	if (memattr != NULL) {
+		efi_call_proto(memattr, clear_memory_attributes, rounded_start,
+			       rounded_end - rounded_start, EFI_MEMORY_XP);
+		return;
+	}
+
+	if (efi_dxe_table == NULL)
+		return;
+
 	/*
 	 * Don't modify memory region attributes, they are
 	 * already suitable, to lower the possibility to
@@ -758,6 +765,7 @@ void __noreturn efi_stub_entry(efi_handl
 			       efi_system_table_t *sys_table_arg,
 			       struct boot_params *boot_params)
 {
+	efi_guid_t guid = EFI_MEMORY_ATTRIBUTE_PROTOCOL_GUID;
 	unsigned long bzimage_addr = (unsigned long)startup_32;
 	unsigned long buffer_start, buffer_end;
 	struct setup_header *hdr = &boot_params->hdr;
@@ -769,13 +777,18 @@ void __noreturn efi_stub_entry(efi_handl
 	if (efi_system_table->hdr.signature != EFI_SYSTEM_TABLE_SIGNATURE)
 		efi_exit(handle, EFI_INVALID_PARAMETER);
 
-	efi_dxe_table = get_efi_config_table(EFI_DXE_SERVICES_TABLE_GUID);
-	if (efi_dxe_table &&
-	    efi_dxe_table->hdr.signature != EFI_DXE_SERVICES_TABLE_SIGNATURE) {
-		efi_warn("Ignoring DXE services table: invalid signature\n");
-		efi_dxe_table = NULL;
+	if (IS_ENABLED(CONFIG_EFI_DXE_MEM_ATTRIBUTES)) {
+		efi_dxe_table = get_efi_config_table(EFI_DXE_SERVICES_TABLE_GUID);
+		if (efi_dxe_table &&
+		    efi_dxe_table->hdr.signature != EFI_DXE_SERVICES_TABLE_SIGNATURE) {
+			efi_warn("Ignoring DXE services table: invalid signature\n");
+			efi_dxe_table = NULL;
+		}
 	}
 
+	/* grab the memory attributes protocol if it exists */
+	efi_bs_call(locate_protocol, &guid, NULL, (void **)&memattr);
+
 	status = efi_setup_5level_paging();
 	if (status != EFI_SUCCESS) {
 		efi_err("efi_setup_5level_paging() failed!\n");
author maintainer
#194 lore raw

PATCH 190/215 x86/efistub: Perform SNP feature test while running in the firmware

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ard Biesheuvel <[email protected]>

From: Ard Biesheuvel <[email protected]>

[ Commit 31c77a50992e8dd136feed7b67073bb5f1f978cc upstream ]

Before refactoring the EFI stub boot flow to avoid the legacy bare metal decompressor, duplicate the SNP feature check in the EFI stub before handing over to the kernel proper.

The SNP feature check can be performed while running under the EFI boot services, which means it can force the boot to fail gracefully and return an error to the bootloader if the loaded kernel does not implement support for all the features that the hypervisor enabled.

---
 arch/x86/boot/compressed/sev.c          |  112 ++++++++++++++++++--------------
 arch/x86/include/asm/sev.h              |    5 +
 drivers/firmware/efi/libstub/x86-stub.c |   17 ++++
 3 files changed, 88 insertions(+), 46 deletions(-)
arch/x86/boot/compressed/sev.c +66 -46
@@ -327,20 +327,25 @@ static void enforce_vmpl0(void)
  */
 #define SNP_FEATURES_PRESENT (0)
 
+u64 snp_get_unsupported_features(u64 status)
+{
+	if (!(status & MSR_AMD64_SEV_SNP_ENABLED))
+		return 0;
+
+	return status & SNP_FEATURES_IMPL_REQ & ~SNP_FEATURES_PRESENT;
+}
+
 void snp_check_features(void)
 {
 	u64 unsupported;
 
-	if (!(sev_status & MSR_AMD64_SEV_SNP_ENABLED))
-		return;
-
 	/*
 	 * Terminate the boot if hypervisor has enabled any feature lacking
 	 * guest side implementation. Pass on the unsupported features mask through
 	 * EXIT_INFO_2 of the GHCB protocol so that those features can be reported
 	 * as part of the guest boot failure.
 	 */
-	unsupported = sev_status & SNP_FEATURES_IMPL_REQ & ~SNP_FEATURES_PRESENT;
+	unsupported = snp_get_unsupported_features(sev_status);
 	if (unsupported) {
 		if (ghcb_version < 2 || (!boot_ghcb && !early_setup_ghcb()))
 			sev_es_terminate(SEV_TERM_SET_GEN, GHCB_SNP_UNSUPPORTED);
@@ -350,35 +355,22 @@ void snp_check_features(void)
 	}
 }
 
-void sev_enable(struct boot_params *bp)
+/*
+ * sev_check_cpu_support - Check for SEV support in the CPU capabilities
+ *
+ * Returns < 0 if SEV is not supported, otherwise the position of the
+ * encryption bit in the page table descriptors.
+ */
+static int sev_check_cpu_support(void)
 {
 	unsigned int eax, ebx, ecx, edx;
-	struct msr m;
-	bool snp;
-
-	/*
-	 * bp->cc_blob_address should only be set by boot/compressed kernel.
-	 * Initialize it to 0 to ensure that uninitialized values from
-	 * buggy bootloaders aren't propagated.
-	 */
-	if (bp)
-		bp->cc_blob_address = 0;
-
-	/*
-	 * Do an initial SEV capability check before snp_init() which
-	 * loads the CPUID page and the same checks afterwards are done
-	 * without the hypervisor and are trustworthy.
-	 *
-	 * If the HV fakes SEV support, the guest will crash'n'burn
-	 * which is good enough.
-	 */
 
 	/* Check for the SME/SEV support leaf */
 	eax = 0x80000000;
 	ecx = 0;
 	native_cpuid(&eax, &ebx, &ecx, &edx);
 	if (eax < 0x8000001f)
-		return;
+		return -ENODEV;
 
 	/*
 	 * Check for the SME/SEV feature:
@@ -393,6 +385,35 @@ void sev_enable(struct boot_params *bp)
 	native_cpuid(&eax, &ebx, &ecx, &edx);
 	/* Check whether SEV is supported */
 	if (!(eax & BIT(1)))
+		return -ENODEV;
+
+	return ebx & 0x3f;
+}
+
+void sev_enable(struct boot_params *bp)
+{
+	struct msr m;
+	int bitpos;
+	bool snp;
+
+	/*
+	 * bp->cc_blob_address should only be set by boot/compressed kernel.
+	 * Initialize it to 0 to ensure that uninitialized values from
+	 * buggy bootloaders aren't propagated.
+	 */
+	if (bp)
+		bp->cc_blob_address = 0;
+
+	/*
+	 * Do an initial SEV capability check before snp_init() which
+	 * loads the CPUID page and the same checks afterwards are done
+	 * without the hypervisor and are trustworthy.
+	 *
+	 * If the HV fakes SEV support, the guest will crash'n'burn
+	 * which is good enough.
+	 */
+
+	if (sev_check_cpu_support() < 0)
 		return;
 
 	/*
@@ -403,26 +424,8 @@ void sev_enable(struct boot_params *bp)
 
 	/* Now repeat the checks with the SNP CPUID table. */
 
-	/* Recheck the SME/SEV support leaf */
-	eax = 0x80000000;
-	ecx = 0;
-	native_cpuid(&eax, &ebx, &ecx, &edx);
-	if (eax < 0x8000001f)
-		return;
-
-	/*
-	 * Recheck for the SME/SEV feature:
-	 *   CPUID Fn8000_001F[EAX]
-	 *   - Bit 0 - Secure Memory Encryption support
-	 *   - Bit 1 - Secure Encrypted Virtualization support
-	 *   CPUID Fn8000_001F[EBX]
-	 *   - Bits 5:0 - Pagetable bit position used to indicate encryption
-	 */
-	eax = 0x8000001f;
-	ecx = 0;
-	native_cpuid(&eax, &ebx, &ecx, &edx);
-	/* Check whether SEV is supported */
-	if (!(eax & BIT(1))) {
+	bitpos = sev_check_cpu_support();
+	if (bitpos < 0) {
 		if (snp)
 			error("SEV-SNP support indicated by CC blob, but not CPUID.");
 		return;
@@ -454,7 +457,24 @@ void sev_enable(struct boot_params *bp)
 	if (snp && !(sev_status & MSR_AMD64_SEV_SNP_ENABLED))
 		error("SEV-SNP supported indicated by CC blob, but not SEV status MSR.");
 
-	sme_me_mask = BIT_ULL(ebx & 0x3f);
+	sme_me_mask = BIT_ULL(bitpos);
+}
+
+/*
+ * sev_get_status - Retrieve the SEV status mask
+ *
+ * Returns 0 if the CPU is not SEV capable, otherwise the value of the
+ * AMD64_SEV MSR.
+ */
+u64 sev_get_status(void)
+{
+	struct msr m;
+
+	if (sev_check_cpu_support() < 0)
+		return 0;
+
+	boot_rdmsr(MSR_AMD64_SEV, &m);
+	return m.q;
 }
 
 /* Search for Confidential Computing blob in the EFI config table. */
arch/x86/include/asm/sev.h +5 -0
@@ -202,6 +202,8 @@ void snp_set_wakeup_secondary_cpu(void);
 bool snp_init(struct boot_params *bp);
 void __init __noreturn snp_abort(void);
 int snp_issue_guest_request(u64 exit_code, struct snp_req_data *input, struct snp_guest_request_ioctl *rio);
+u64 snp_get_unsupported_features(u64 status);
+u64 sev_get_status(void);
 #else
 static inline void sev_es_ist_enter(struct pt_regs *regs) { }
 static inline void sev_es_ist_exit(void) { }
@@ -225,6 +227,9 @@ static inline int snp_issue_guest_reques
 {
 	return -ENOTTY;
 }
+
+static inline u64 snp_get_unsupported_features(u64 status) { return 0; }
+static inline u64 sev_get_status(void) { return 0; }
 #endif
 
 #endif
drivers/firmware/efi/libstub/x86-stub.c +17 -0
@@ -15,6 +15,7 @@
 #include <asm/setup.h>
 #include <asm/desc.h>
 #include <asm/boot.h>
+#include <asm/sev.h>
 
 #include "efistub.h"
 #include "x86-stub.h"
@@ -747,6 +748,19 @@ static efi_status_t exit_boot(struct boo
 	return EFI_SUCCESS;
 }
 
+static bool have_unsupported_snp_features(void)
+{
+	u64 unsupported;
+
+	unsupported = snp_get_unsupported_features(sev_get_status());
+	if (unsupported) {
+		efi_err("Unsupported SEV-SNP features detected: 0x%llx\n",
+			unsupported);
+		return true;
+	}
+	return false;
+}
+
 static void __noreturn enter_kernel(unsigned long kernel_addr,
 				    struct boot_params *boot_params)
 {
@@ -777,6 +791,9 @@ void __noreturn efi_stub_entry(efi_handl
 	if (efi_system_table->hdr.signature != EFI_SYSTEM_TABLE_SIGNATURE)
 		efi_exit(handle, EFI_INVALID_PARAMETER);
 
+	if (have_unsupported_snp_features())
+		efi_exit(handle, EFI_UNSUPPORTED);
+
 	if (IS_ENABLED(CONFIG_EFI_DXE_MEM_ATTRIBUTES)) {
 		efi_dxe_table = get_efi_config_table(EFI_DXE_SERVICES_TABLE_GUID);
 		if (efi_dxe_table &&
author maintainer
#195 lore raw

PATCH 191/215 x86/efistub: Avoid legacy decompressor when doing EFI boot

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ard Biesheuvel <[email protected]>

From: Ard Biesheuvel <[email protected]>

[ Commit a1b87d54f4e45ff5e0d081fb1d9db3bf1a8fb39a upstream ]

The bare metal decompressor code was never really intended to run in a hosted environment such as the EFI boot services, and does a few things that are becoming problematic in the context of EFI boot now that the logo requirements are getting tighter: EFI executables will no longer be allowed to consist of a single executable section that is mapped with read, write and execute permissions if they are intended for use in a context where Secure Boot is enabled (and where Microsoft's set of certificates is used, i.e., every x86 PC built to run Windows).

To avoid stepping on reserved memory before having inspected the E820 tables, and to ensure the correct placement when running a kernel build that is non-relocatable, the bare metal decompressor moves its own executable image to the end of the allocation that was reserved for it, in order to perform the decompression in place. This means the region in question requires both write and execute permissions, which either need to be given upfront (which EFI will no longer permit), or need to be applied on demand using the existing page fault handling framework.

However, the physical placement of the kernel is usually randomized anyway, and even if it isn't, a dedicated decompression output buffer can be allocated anywhere in memory using EFI APIs when still running in the boot services, given that EFI support already implies a relocatable kernel. This means that decompression in place is never necessary, nor is moving the compressed image from one end to the other.

Since EFI already maps all of memory 1:1, it is also unnecessary to create new page tables or handle page faults when decompressing the kernel. That means there is also no need to replace the special exception handlers for SEV. Generally, there is little need to do any of the things that the decompressor does beyond

- initialize SEV encryption, if needed,
- perform the 4/5 level paging switch, if needed,
- decompress the kernel
- relocate the kernel

So do all of this from the EFI stub code, and avoid the bare metal decompressor altogether.

---
 arch/x86/boot/compressed/Makefile       |    5 
 arch/x86/boot/compressed/efi_mixed.S    |   55 ----------
 arch/x86/boot/compressed/head_32.S      |   13 --
 arch/x86/boot/compressed/head_64.S      |   27 -----
 arch/x86/include/asm/efi.h              |    7 -
 arch/x86/include/asm/sev.h              |    2 
 drivers/firmware/efi/libstub/x86-stub.c |  166 +++++++++++++-------------------
 7 files changed, 84 insertions(+), 191 deletions(-)
arch/x86/boot/compressed/Makefile +5 -0
@@ -74,6 +74,11 @@ LDFLAGS_vmlinux += -z noexecstack
 ifeq ($(CONFIG_LD_IS_BFD),y)
 LDFLAGS_vmlinux += $(call ld-option,--no-warn-rwx-segments)
 endif
+ifeq ($(CONFIG_EFI_STUB),y)
+# ensure that the static EFI stub library will be pulled in, even if it is
+# never referenced explicitly from the startup code
+LDFLAGS_vmlinux += -u efi_pe_entry
+endif
 LDFLAGS_vmlinux += -T
 
 hostprogs	:= mkpiggy
arch/x86/boot/compressed/efi_mixed.S +0 -55

81 lines not on this page · raw message

arch/x86/boot/compressed/head_32.S +0 -13
@@ -84,19 +84,6 @@ SYM_FUNC_START(startup_32)
 
 #ifdef CONFIG_RELOCATABLE
 	leal	startup_32@GOTOFF(%edx), %ebx
-
-#ifdef CONFIG_EFI_STUB
-/*
- * If we were loaded via the EFI LoadImage service, startup_32() will be at an
- * offset to the start of the space allocated for the image. efi_pe_entry() will
- * set up image_offset to tell us where the image actually starts, so that we
- * can use the full available buffer.
- *	image_offset = startup_32 - image_base
- * Otherwise image_offset will be zero and has no effect on the calculations.
- */
-	subl    image_offset@GOTOFF(%edx), %ebx
-#endif
-
 	movl	BP_kernel_alignment(%esi), %eax
 	decl	%eax
 	addl    %eax, %ebx
arch/x86/boot/compressed/head_64.S +0 -27
@@ -138,19 +138,6 @@ SYM_FUNC_START(startup_32)
 
 #ifdef CONFIG_RELOCATABLE
 	movl	%ebp, %ebx
-
-#ifdef CONFIG_EFI_STUB
-/*
- * If we were loaded via the EFI LoadImage service, startup_32 will be at an
- * offset to the start of the space allocated for the image. efi_pe_entry will
- * set up image_offset to tell us where the image actually starts, so that we
- * can use the full available buffer.
- *	image_offset = startup_32 - image_base
- * Otherwise image_offset will be zero and has no effect on the calculations.
- */
-	subl    rva(image_offset)(%ebp), %ebx
-#endif
-
 	movl	BP_kernel_alignment(%esi), %eax
 	decl	%eax
 	addl	%eax, %ebx
@@ -327,20 +314,6 @@ SYM_CODE_START(startup_64)
 	/* Start with the delta to where the kernel will run at. */
 #ifdef CONFIG_RELOCATABLE
 	leaq	startup_32(%rip) /* - $startup_32 */, %rbp
-
-#ifdef CONFIG_EFI_STUB
-/*
- * If we were loaded via the EFI LoadImage service, startup_32 will be at an
- * offset to the start of the space allocated for the image. efi_pe_entry will
- * set up image_offset to tell us where the image actually starts, so that we
- * can use the full available buffer.
- *	image_offset = startup_32 - image_base
- * Otherwise image_offset will be zero and has no effect on the calculations.
- */
-	movl    image_offset(%rip), %eax
-	subq	%rax, %rbp
-#endif
-
 	movl	BP_kernel_alignment(%rsi), %eax
 	decl	%eax
 	addq	%rax, %rbp
arch/x86/include/asm/efi.h +5 -2
@@ -88,6 +88,8 @@ static inline void efi_fpu_end(void)
 }
 
 #ifdef CONFIG_X86_32
+#define EFI_X86_KERNEL_ALLOC_LIMIT		(SZ_512M - 1)
+
 #define arch_efi_call_virt_setup()					\
 ({									\
 	efi_fpu_begin();						\
@@ -101,8 +103,7 @@ static inline void efi_fpu_end(void)
 })
 
 #else /* !CONFIG_X86_32 */
-
-#define EFI_LOADER_SIGNATURE	"EL64"
+#define EFI_X86_KERNEL_ALLOC_LIMIT		EFI_ALLOC_LIMIT
 
 extern asmlinkage u64 __efi_call(void *fp, ...);
 
@@ -214,6 +215,8 @@ efi_status_t efi_set_virtual_address_map
 
 #ifdef CONFIG_EFI_MIXED
 
+#define EFI_ALLOC_LIMIT		(efi_is_64bit() ? ULONG_MAX : U32_MAX)
+
 #define ARCH_HAS_EFISTUB_WRAPPERS
 
 static inline bool efi_is_64bit(void)
arch/x86/include/asm/sev.h +2 -0
@@ -157,6 +157,7 @@ static __always_inline void sev_es_nmi_c
 		__sev_es_nmi_complete();
 }
 extern int __init sev_es_efi_map_ghcbs(pgd_t *pgd);
+extern void sev_enable(struct boot_params *bp);
 
 static inline int rmpadjust(unsigned long vaddr, bool rmp_psize, unsigned long attrs)
 {
@@ -210,6 +211,7 @@ static inline void sev_es_ist_exit(void)
 static inline int sev_es_setup_ap_jump_table(struct real_mode_header *rmh) { return 0; }
 static inline void sev_es_nmi_complete(void) { }
 static inline int sev_es_efi_map_ghcbs(pgd_t *pgd) { return 0; }
+static inline void sev_enable(struct boot_params *bp) { }
 static inline int pvalidate(unsigned long vaddr, bool rmp_psize, bool validate) { return 0; }
 static inline int rmpadjust(unsigned long vaddr, bool rmp_psize, unsigned long attrs) { return 0; }
 static inline void setup_ghcb(void) { }
drivers/firmware/efi/libstub/x86-stub.c +72 -94

239 lines not on this page · raw message

author maintainer
#196 lore raw

PATCH 192/215 efi/x86: Avoid physical KASLR on older Dell systems

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ard Biesheuvel <[email protected]>

From: Ard Biesheuvel <[email protected]>

[ Commit 50d7cdf7a9b1ab6f4f74a69c84e974d5dc0c1bf1 upstream ]

River reports boot hangs with v6.6 and v6.7, and the bisect points to commit

a1b87d54f4e4 ("x86/efistub: Avoid legacy decompressor when doing EFI boot")

which moves the memory allocation and kernel decompression from the legacy decompressor (which executes *after* ExitBootServices()) to the EFI stub, using boot services for allocating the memory. The memory allocation succeeds but the subsequent call to decompress_kernel() never returns, resulting in a failed boot and a hanging system.

As it turns out, this issue only occurs when physical address randomization (KASLR) is enabled, and given that this is a feature we can live without (virtual KASLR is much more important), let's disable the physical part of KASLR when booting on AMI UEFI firmware claiming to implement revision v2.0 of the specification (which was released in 2006), as this is the version these systems advertise.

Fixes a1b87d54f4e4 ("x86/efistub: Avoid legacy decompressor when doing EFI boot")
---
 drivers/firmware/efi/libstub/x86-stub.c |   31 ++++++++++++++++++++++++-------
 1 file changed, 24 insertions(+), 7 deletions(-)
drivers/firmware/efi/libstub/x86-stub.c +24 -7
@@ -273,17 +273,20 @@ void efi_adjust_memory_range_protection(
 	}
 }
 
+static efi_char16_t *efistub_fw_vendor(void)
+{
+	unsigned long vendor = efi_table_attr(efi_system_table, fw_vendor);
+
+	return (efi_char16_t *)vendor;
+}
+
 static const efi_char16_t apple[] = L"Apple";
 
 static void setup_quirks(struct boot_params *boot_params)
 {
-	efi_char16_t *fw_vendor = (efi_char16_t *)(unsigned long)
-		efi_table_attr(efi_system_table, fw_vendor);
-
-	if (!memcmp(fw_vendor, apple, sizeof(apple))) {
-		if (IS_ENABLED(CONFIG_APPLE_PROPERTIES))
-			retrieve_apple_device_properties(boot_params);
-	}
+	if (IS_ENABLED(CONFIG_APPLE_PROPERTIES) &&
+	    !memcmp(efistub_fw_vendor(), apple, sizeof(apple)))
+		retrieve_apple_device_properties(boot_params);
 }
 
 /*
@@ -759,11 +762,25 @@ static efi_status_t efi_decompress_kerne
 
 	if (IS_ENABLED(CONFIG_RANDOMIZE_BASE) && !efi_nokaslr) {
 		u64 range = KERNEL_IMAGE_SIZE - LOAD_PHYSICAL_ADDR - kernel_total_size;
+		static const efi_char16_t ami[] = L"American Megatrends";
 
 		efi_get_seed(seed, sizeof(seed));
 
 		virt_addr += (range * seed[1]) >> 32;
 		virt_addr &= ~(CONFIG_PHYSICAL_ALIGN - 1);
+
+		/*
+		 * Older Dell systems with AMI UEFI firmware v2.0 may hang
+		 * while decompressing the kernel if physical address
+		 * randomization is enabled.
+		 *
+		 * https://bugzilla.kernel.org/show_bug.cgi?id=218173
+		 */
+		if (efi_system_table->hdr.revision <= EFI_2_00_SYSTEM_TABLE_REVISION &&
+		    !memcmp(efistub_fw_vendor(), ami, sizeof(ami))) {
+			efi_debug("AMI firmware v2.0 or older detected - disabling physical KASLR\n");
+			seed[0] = 0;
+		}
 	}
 
 	status = efi_random_alloc(alloc_size, CONFIG_PHYSICAL_ALIGN, &addr,
author maintainer
#197 lore raw

PATCH 193/215 x86/efistub: Avoid placing the kernel below LOAD_PHYSICAL_ADDR

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ard Biesheuvel <[email protected]>

From: Ard Biesheuvel <[email protected]>

[ Commit 2f77465b05b1270c832b5e2ee27037672ad2a10a upstream ]

The EFI stub's kernel placement logic randomizes the physical placement of the kernel by taking all available memory into account, and picking a region at random, based on a random seed.

When KASLR is disabled, this seed is set to 0x0, and this results in the lowest available region of memory to be selected for loading the kernel, even if this is below LOAD_PHYSICAL_ADDR. Some of this memory is typically reserved for the GFP_DMA region, to accommodate masters that can only access the first 16 MiB of system memory.

Even if such devices are rare these days, we may still end up with a warning in the kernel log, as reported by Tom:

swapper/0: page allocation failure: order:10, mode:0xcc1(GFP_KERNEL|GFP_DMA), nodemask=(null),cpuset=/,mems_allowed=0

Fix this by tweaking the random allocation logic to accept a low bound on the placement, and set it to LOAD_PHYSICAL_ADDR.

Fixes a1b87d54f4e4 ("x86/efistub: Avoid legacy decompressor when doing EFI boot")
---
 drivers/firmware/efi/libstub/arm64-stub.c  |    2 +-
 drivers/firmware/efi/libstub/efistub.h     |    3 ++-
 drivers/firmware/efi/libstub/randomalloc.c |   12 +++++++-----
 drivers/firmware/efi/libstub/x86-stub.c    |    1 +
 4 files changed, 11 insertions(+), 7 deletions(-)
drivers/firmware/efi/libstub/arm64-stub.c +1 -1
@@ -181,7 +181,7 @@ efi_status_t handle_kernel_image(unsigne
 		 */
 		status = efi_random_alloc(*reserve_size, min_kimg_align,
 					  reserve_addr, phys_seed,
-					  EFI_LOADER_CODE, EFI_ALLOC_LIMIT);
+					  EFI_LOADER_CODE, 0, EFI_ALLOC_LIMIT);
 		if (status != EFI_SUCCESS)
 			efi_warn("efi_random_alloc() failed: 0x%lx\n", status);
 	} else {
drivers/firmware/efi/libstub/efistub.h +2 -1
@@ -906,7 +906,8 @@ efi_status_t efi_get_random_bytes(unsign
 
 efi_status_t efi_random_alloc(unsigned long size, unsigned long align,
 			      unsigned long *addr, unsigned long random_seed,
-			      int memory_type, unsigned long alloc_limit);
+			      int memory_type, unsigned long alloc_min,
+			      unsigned long alloc_max);
 
 efi_status_t efi_random_get_seed(void);
 
drivers/firmware/efi/libstub/randomalloc.c +7 -5
@@ -17,7 +17,7 @@
 static unsigned long get_entry_num_slots(efi_memory_desc_t *md,
 					 unsigned long size,
 					 unsigned long align_shift,
-					 u64 alloc_limit)
+					 u64 alloc_min, u64 alloc_max)
 {
 	unsigned long align = 1UL << align_shift;
 	u64 first_slot, last_slot, region_end;
@@ -30,11 +30,11 @@ static unsigned long get_entry_num_slots
 		return 0;
 
 	region_end = min(md->phys_addr + md->num_pages * EFI_PAGE_SIZE - 1,
-			 alloc_limit);
+			 alloc_max);
 	if (region_end < size)
 		return 0;
 
-	first_slot = round_up(md->phys_addr, align);
+	first_slot = round_up(max(md->phys_addr, alloc_min), align);
 	last_slot = round_down(region_end - size + 1, align);
 
 	if (first_slot > last_slot)
@@ -56,7 +56,8 @@ efi_status_t efi_random_alloc(unsigned l
 			      unsigned long *addr,
 			      unsigned long random_seed,
 			      int memory_type,
-			      unsigned long alloc_limit)
+			      unsigned long alloc_min,
+			      unsigned long alloc_max)
 {
 	unsigned long total_slots = 0, target_slot;
 	unsigned long total_mirrored_slots = 0;
@@ -78,7 +79,8 @@ efi_status_t efi_random_alloc(unsigned l
 		efi_memory_desc_t *md = (void *)map->map + map_offset;
 		unsigned long slots;
 
-		slots = get_entry_num_slots(md, size, ilog2(align), alloc_limit);
+		slots = get_entry_num_slots(md, size, ilog2(align), alloc_min,
+					    alloc_max);
 		MD_NUM_SLOTS(md) = slots;
 		total_slots += slots;
 		if (md->attribute & EFI_MEMORY_MORE_RELIABLE)
drivers/firmware/efi/libstub/x86-stub.c +1 -0
@@ -785,6 +785,7 @@ static efi_status_t efi_decompress_kerne
 
 	status = efi_random_alloc(alloc_size, CONFIG_PHYSICAL_ALIGN, &addr,
 				  seed[0], EFI_LOADER_CODE,
+				  LOAD_PHYSICAL_ADDR,
 				  EFI_X86_KERNEL_ALLOC_LIMIT);
 	if (status != EFI_SUCCESS)
 		return status;
author maintainer
#198 lore raw

PATCH 194/215 x86/boot: Rename conflicting boot_params pointer to boot_params_ptr

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ard Biesheuvel <[email protected]>

From: Ard Biesheuvel <[email protected]>

[ Commit b9e909f78e7e4b826f318cfe7bedf3ce229920e6 upstream ]

The x86 decompressor is built and linked as a separate executable, but it shares components with the kernel proper, which are either #include'd as C files, or linked into the decompresor as a static library (e.g, the EFI stub)

Both the kernel itself and the decompressor define a global symbol 'boot_params' to refer to the boot_params struct, but in the former case, it refers to the struct directly, whereas in the decompressor, it refers to a global pointer variable referring to the struct boot_params passed by the bootloader or constructed from scratch.

This ambiguity is unfortunate, and makes it impossible to assign this decompressor variable from the x86 EFI stub, given that declaring it as extern results in a clash. So rename the decompressor version (whose scope is limited) to boot_params_ptr.

[ mingo: Renamed 'boot_params_p' to 'boot_params_ptr' for clarity ]

---
 arch/x86/boot/compressed/acpi.c         |   14 +++++++-------
 arch/x86/boot/compressed/cmdline.c      |    4 ++--
 arch/x86/boot/compressed/ident_map_64.c |    7 ++++---
 arch/x86/boot/compressed/kaslr.c        |   26 +++++++++++++-------------
 arch/x86/boot/compressed/misc.c         |   24 ++++++++++++------------
 arch/x86/boot/compressed/misc.h         |    1 -
 arch/x86/boot/compressed/pgtable_64.c   |    9 ++++-----
 arch/x86/boot/compressed/sev.c          |    2 +-
 arch/x86/include/asm/boot.h             |    2 ++
 9 files changed, 45 insertions(+), 44 deletions(-)
arch/x86/boot/compressed/acpi.c +7 -7
@@ -30,13 +30,13 @@ __efi_get_rsdp_addr(unsigned long cfg_tb
 	 * Search EFI system tables for RSDP. Preferred is ACPI_20_TABLE_GUID to
 	 * ACPI_TABLE_GUID because it has more features.
 	 */
-	rsdp_addr = efi_find_vendor_table(boot_params, cfg_tbl_pa, cfg_tbl_len,
+	rsdp_addr = efi_find_vendor_table(boot_params_ptr, cfg_tbl_pa, cfg_tbl_len,
 					  ACPI_20_TABLE_GUID);
 	if (rsdp_addr)
 		return (acpi_physical_address)rsdp_addr;
 
 	/* No ACPI_20_TABLE_GUID found, fallback to ACPI_TABLE_GUID. */
-	rsdp_addr = efi_find_vendor_table(boot_params, cfg_tbl_pa, cfg_tbl_len,
+	rsdp_addr = efi_find_vendor_table(boot_params_ptr, cfg_tbl_pa, cfg_tbl_len,
 					  ACPI_TABLE_GUID);
 	if (rsdp_addr)
 		return (acpi_physical_address)rsdp_addr;
@@ -56,15 +56,15 @@ static acpi_physical_address efi_get_rsd
 	enum efi_type et;
 	int ret;
 
-	et = efi_get_type(boot_params);
+	et = efi_get_type(boot_params_ptr);
 	if (et == EFI_TYPE_NONE)
 		return 0;
 
-	systab_pa = efi_get_system_table(boot_params);
+	systab_pa = efi_get_system_table(boot_params_ptr);
 	if (!systab_pa)
 		error("EFI support advertised, but unable to locate system table.");
 
-	ret = efi_get_conf_table(boot_params, &cfg_tbl_pa, &cfg_tbl_len);
+	ret = efi_get_conf_table(boot_params_ptr, &cfg_tbl_pa, &cfg_tbl_len);
 	if (ret || !cfg_tbl_pa)
 		error("EFI config table not found.");
 
@@ -156,7 +156,7 @@ acpi_physical_address get_rsdp_addr(void
 {
 	acpi_physical_address pa;
 
-	pa = boot_params->acpi_rsdp_addr;
+	pa = boot_params_ptr->acpi_rsdp_addr;
 
 	if (!pa)
 		pa = efi_get_rsdp_addr();
@@ -210,7 +210,7 @@ static unsigned long get_acpi_srat_table
 	rsdp = (struct acpi_table_rsdp *)get_cmdline_acpi_rsdp();
 	if (!rsdp)
 		rsdp = (struct acpi_table_rsdp *)(long)
-			boot_params->acpi_rsdp_addr;
+			boot_params_ptr->acpi_rsdp_addr;
 
 	if (!rsdp)
 		return 0;
arch/x86/boot/compressed/cmdline.c +2 -2
@@ -14,9 +14,9 @@ static inline char rdfs8(addr_t addr)
 #include "../cmdline.c"
 unsigned long get_cmd_line_ptr(void)
 {
-	unsigned long cmd_line_ptr = boot_params->hdr.cmd_line_ptr;
+	unsigned long cmd_line_ptr = boot_params_ptr->hdr.cmd_line_ptr;
 
-	cmd_line_ptr |= (u64)boot_params->ext_cmd_line_ptr << 32;
+	cmd_line_ptr |= (u64)boot_params_ptr->ext_cmd_line_ptr << 32;
 
 	return cmd_line_ptr;
 }
arch/x86/boot/compressed/ident_map_64.c +4 -3
@@ -167,8 +167,9 @@ void initialize_identity_maps(void *rmod
 	 * or does not touch all the pages covering them.
 	 */
 	kernel_add_identity_map((unsigned long)_head, (unsigned long)_end);
-	boot_params = rmode;
-	kernel_add_identity_map((unsigned long)boot_params, (unsigned long)(boot_params + 1));
+	boot_params_ptr = rmode;
+	kernel_add_identity_map((unsigned long)boot_params_ptr,
+				(unsigned long)(boot_params_ptr + 1));
 	cmdline = get_cmd_line_ptr();
 	kernel_add_identity_map(cmdline, cmdline + COMMAND_LINE_SIZE);
 
@@ -176,7 +177,7 @@ void initialize_identity_maps(void *rmod
 	 * Also map the setup_data entries passed via boot_params in case they
 	 * need to be accessed by uncompressed kernel via the identity mapping.
 	 */
-	sd = (struct setup_data *)boot_params->hdr.setup_data;
+	sd = (struct setup_data *)boot_params_ptr->hdr.setup_data;
 	while (sd) {
 		unsigned long sd_addr = (unsigned long)sd;
 
arch/x86/boot/compressed/kaslr.c +13 -13
@@ -63,7 +63,7 @@ static unsigned long get_boot_seed(void)
 	unsigned long hash = 0;
 
 	hash = rotate_xor(hash, build_str, sizeof(build_str));
-	hash = rotate_xor(hash, boot_params, sizeof(*boot_params));
+	hash = rotate_xor(hash, boot_params_ptr, sizeof(*boot_params_ptr));
 
 	return hash;
 }
@@ -383,7 +383,7 @@ static void handle_mem_options(void)
 static void mem_avoid_init(unsigned long input, unsigned long input_size,
 			   unsigned long output)
 {
-	unsigned long init_size = boot_params->hdr.init_size;
+	unsigned long init_size = boot_params_ptr->hdr.init_size;
 	u64 initrd_start, initrd_size;
 	unsigned long cmd_line, cmd_line_size;
 
@@ -395,10 +395,10 @@ static void mem_avoid_init(unsigned long
 	mem_avoid[MEM_AVOID_ZO_RANGE].size = (output + init_size) - input;
 
 	/* Avoid initrd. */
-	initrd_start  = (u64)boot_params->ext_ramdisk_image << 32;
-	initrd_start |= boot_params->hdr.ramdisk_image;
-	initrd_size  = (u64)boot_params->ext_ramdisk_size << 32;
-	initrd_size |= boot_params->hdr.ramdisk_size;
+	initrd_start  = (u64)boot_params_ptr->ext_ramdisk_image << 32;
+	initrd_start |= boot_params_ptr->hdr.ramdisk_image;
+	initrd_size  = (u64)boot_params_ptr->ext_ramdisk_size << 32;
+	initrd_size |= boot_params_ptr->hdr.ramdisk_size;
 	mem_avoid[MEM_AVOID_INITRD].start = initrd_start;
 	mem_avoid[MEM_AVOID_INITRD].size = initrd_size;
 	/* No need to set mapping for initrd, it will be handled in VO. */
@@ -413,8 +413,8 @@ static void mem_avoid_init(unsigned long
 	}
 
 	/* Avoid boot parameters. */
-	mem_avoid[MEM_AVOID_BOOTPARAMS].start = (unsigned long)boot_params;
-	mem_avoid[MEM_AVOID_BOOTPARAMS].size = sizeof(*boot_params);
+	mem_avoid[MEM_AVOID_BOOTPARAMS].start = (unsigned long)boot_params_ptr;
+	mem_avoid[MEM_AVOID_BOOTPARAMS].size = sizeof(*boot_params_ptr);
 
 	/* We don't need to set a mapping for setup_data. */
 
@@ -447,7 +447,7 @@ static bool mem_avoid_overlap(struct mem
 	}
 
 	/* Avoid all entries in the setup_data linked list. */
-	ptr = (struct setup_data *)(unsigned long)boot_params->hdr.setup_data;
+	ptr = (struct setup_data *)(unsigned long)boot_params_ptr->hdr.setup_data;
 	while (ptr) {
 		struct mem_vector avoid;
 
@@ -679,7 +679,7 @@ static bool process_mem_region(struct me
 static bool
 process_efi_entries(unsigned long minimum, unsigned long image_size)
 {
-	struct efi_info *e = &boot_params->efi_info;
+	struct efi_info *e = &boot_params_ptr->efi_info;
 	bool efi_mirror_found = false;
 	struct mem_vector region;
 	efi_memory_desc_t *md;
@@ -761,8 +761,8 @@ static void process_e820_entries(unsigne
 	struct boot_e820_entry *entry;
 
 	/* Verify potential e820 positions, appending to slots list. */
-	for (i = 0; i < boot_params->e820_entries; i++) {
-		entry = &boot_params->e820_table[i];
+	for (i = 0; i < boot_params_ptr->e820_entries; i++) {
+		entry = &boot_params_ptr->e820_table[i];
 		/* Skip non-RAM entries. */
 		if (entry->type != E820_TYPE_RAM)
 			continue;
@@ -836,7 +836,7 @@ void choose_random_location(unsigned lon
 		return;
 	}
 
-	boot_params->hdr.loadflags |= KASLR_FLAG;
+	boot_params_ptr->hdr.loadflags |= KASLR_FLAG;
 
 	if (IS_ENABLED(CONFIG_X86_32))
 		mem_limit = KERNEL_IMAGE_SIZE;
arch/x86/boot/compressed/misc.c +12 -12
@@ -46,7 +46,7 @@ void *memmove(void *dest, const void *sr
 /*
  * This is set up by the setup-routine at boot-time
  */
-struct boot_params *boot_params;
+struct boot_params *boot_params_ptr;
 
 struct port_io_ops pio_ops;
 
@@ -132,8 +132,8 @@ void __putstr(const char *s)
 	if (lines == 0 || cols == 0)
 		return;
 
-	x = boot_params->screen_info.orig_x;
-	y = boot_params->screen_info.orig_y;
+	x = boot_params_ptr->screen_info.orig_x;
+	y = boot_params_ptr->screen_info.orig_y;
 
 	while ((c = *s++) != '\0') {
 		if (c == '\n') {
@@ -154,8 +154,8 @@ void __putstr(const char *s)
 		}
 	}
 
-	boot_params->screen_info.orig_x = x;
-	boot_params->screen_info.orig_y = y;
+	boot_params_ptr->screen_info.orig_x = x;
+	boot_params_ptr->screen_info.orig_y = y;
 
 	pos = (x + cols * y) * 2;	/* Update cursor position */
 	outb(14, vidport);
@@ -382,14 +382,14 @@ asmlinkage __visible void *extract_kerne
 	size_t entry_offset;
 
 	/* Retain x86 boot parameters pointer passed from startup_32/64. */
-	boot_params = rmode;
+	boot_params_ptr = rmode;
 
 	/* Clear flags intended for solely in-kernel use. */
-	boot_params->hdr.loadflags &= ~KASLR_FLAG;
+	boot_params_ptr->hdr.loadflags &= ~KASLR_FLAG;
 
-	sanitize_boot_params(boot_params);
+	sanitize_boot_params(boot_params_ptr);
 
-	if (boot_params->screen_info.orig_video_mode == 7) {
+	if (boot_params_ptr->screen_info.orig_video_mode == 7) {
 		vidmem = (char *) 0xb0000;
 		vidport = 0x3b4;
 	} else {
@@ -397,8 +397,8 @@ asmlinkage __visible void *extract_kerne
 		vidport = 0x3d4;
 	}
 
-	lines = boot_params->screen_info.orig_video_lines;
-	cols = boot_params->screen_info.orig_video_cols;
+	lines = boot_params_ptr->screen_info.orig_video_lines;
+	cols = boot_params_ptr->screen_info.orig_video_cols;
 
 	init_default_io_ops();
 
@@ -417,7 +417,7 @@ asmlinkage __visible void *extract_kerne
 	 * so that early debugging output from the RSDP parsing code can be
 	 * collected.
 	 */
-	boot_params->acpi_rsdp_addr = get_rsdp_addr();
+	boot_params_ptr->acpi_rsdp_addr = get_rsdp_addr();
 
 	debug_putstr("early console in extract_kernel\n");
 
arch/x86/boot/compressed/misc.h +0 -1
@@ -52,7 +52,6 @@ extern memptr free_mem_ptr;
 extern memptr free_mem_end_ptr;
 void *malloc(int size);
 void free(void *where);
-extern struct boot_params *boot_params;
 void __putstr(const char *s);
 void __puthex(unsigned long value);
 #define error_putstr(__x)  __putstr(__x)
arch/x86/boot/compressed/pgtable_64.c +4 -5
@@ -28,7 +28,6 @@ static char trampoline_save[TRAMPOLINE_3
  */
 unsigned long *trampoline_32bit __section(".data");
 
-extern struct boot_params *boot_params;
 int cmdline_find_option_bool(const char *option);
 
 static unsigned long find_trampoline_placement(void)
@@ -49,7 +48,7 @@ static unsigned long find_trampoline_pla
 	 *
 	 * Only look for values in the legacy ROM for non-EFI system.
 	 */
-	signature = (char *)&boot_params->efi_info.efi_loader_signature;
+	signature = (char *)&boot_params_ptr->efi_info.efi_loader_signature;
 	if (strncmp(signature, EFI32_LOADER_SIGNATURE, 4) &&
 	    strncmp(signature, EFI64_LOADER_SIGNATURE, 4)) {
 		ebda_start = *(unsigned short *)0x40e << 4;
@@ -65,10 +64,10 @@ static unsigned long find_trampoline_pla
 	bios_start = round_down(bios_start, PAGE_SIZE);
 
 	/* Find the first usable memory region under bios_start. */
-	for (i = boot_params->e820_entries - 1; i >= 0; i--) {
+	for (i = boot_params_ptr->e820_entries - 1; i >= 0; i--) {
 		unsigned long new = bios_start;
 
-		entry = &boot_params->e820_table[i];
+		entry = &boot_params_ptr->e820_table[i];
 
 		/* Skip all entries above bios_start. */
 		if (bios_start <= entry->addr)
@@ -107,7 +106,7 @@ asmlinkage void configure_5level_paging(
 	bool l5_required = false;
 
 	/* Initialize boot_params. Required for cmdline_find_option_bool(). */
-	boot_params = bp;
+	boot_params_ptr = bp;
 
 	/*
 	 * Check if LA57 is desired and supported.
arch/x86/boot/compressed/sev.c +1 -1
@@ -565,7 +565,7 @@ void sev_prep_identity_maps(unsigned lon
 	 * accessed after switchover.
 	 */
 	if (sev_snp_enabled()) {
-		unsigned long cc_info_pa = boot_params->cc_blob_address;
+		unsigned long cc_info_pa = boot_params_ptr->cc_blob_address;
 		struct cc_blob_sev_info *cc_info;
 
 		kernel_add_identity_map(cc_info_pa, cc_info_pa + sizeof(*cc_info));
arch/x86/include/asm/boot.h +2 -0
@@ -85,6 +85,8 @@ extern const unsigned long kernel_total_
 
 unsigned long decompress_kernel(unsigned char *outbuf, unsigned long virt_addr,
 				void (*error)(char *x));
+
+extern struct boot_params *boot_params_ptr;
 #endif
 
 #endif /* _ASM_X86_BOOT_H */
author maintainer
#199 lore raw

PATCH 195/215 x86/boot: efistub: Assign global boot_params variable

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ard Biesheuvel <[email protected]>

From: Ard Biesheuvel <[email protected]>

[ Commit 50dcc2e0d62e3c4a54f39673c4dc3dcde7c74d52 upstream ]

Now that the x86 EFI stub calls into some APIs exposed by the decompressor (e.g., kaslr_get_random_long()), it is necessary to ensure that the global boot_params variable is set correctly before doing so.

---
 drivers/firmware/efi/libstub/x86-stub.c |    2 ++
 1 file changed, 2 insertions(+)
drivers/firmware/efi/libstub/x86-stub.c +2 -0
@@ -827,6 +827,8 @@ void __noreturn efi_stub_entry(efi_handl
 	unsigned long kernel_entry;
 	efi_status_t status;
 
+	boot_params_ptr = boot_params;
+
 	efi_system_table = sys_table_arg;
 	/* Check if we were booted by the EFI firmware */
 	if (efi_system_table->hdr.signature != EFI_SYSTEM_TABLE_SIGNATURE)
author maintainer
#200 lore raw

PATCH 196/215 efi/x86: Fix the missing KASLR_FLAG bit in boot_params->hdr.loadflags

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ard Biesheuvel <[email protected]>

From: Yuntao Wang <[email protected]>

[ Commit 01638431c465741e071ab34acf3bef3c2570f878 upstream ]

When KASLR is enabled, the KASLR_FLAG bit in boot_params->hdr.loadflags should be set to 1 to propagate KASLR status from compressed kernel to kernel, just as the choose_random_location() function does.

Currently, when the kernel is booted via the EFI stub, the KASLR_FLAG bit in boot_params->hdr.loadflags is not set, even though it should be. This causes some functions, such as kernel_randomize_memory(), not to execute as expected. Fix it.

Fixes a1b87d54f4e4 ("x86/efistub: Avoid legacy decompressor when doing EFI boot")
[ardb: drop 'else' branch clearing KASLR_FLAG]
---
 drivers/firmware/efi/libstub/x86-stub.c |    2 ++
 1 file changed, 2 insertions(+)
drivers/firmware/efi/libstub/x86-stub.c +2 -0
@@ -781,6 +781,8 @@ static efi_status_t efi_decompress_kerne
 			efi_debug("AMI firmware v2.0 or older detected - disabling physical KASLR\n");
 			seed[0] = 0;
 		}
+
+		boot_params_ptr->hdr.loadflags |= KASLR_FLAG;
 	}
 
 	status = efi_random_alloc(alloc_size, CONFIG_PHYSICAL_ALIGN, &addr,
author maintainer
#201 lore raw

PATCH 197/215 af_unix: Drop oob_skb ref before purging queue in GC.

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Kuniyuki Iwashima <[email protected]>

commit aa82ac51d63328714645c827775d64dbfd9941f3 upstream.

syzbot reported another task hung in __unix_gc(). [0]

The current while loop assumes that all of the left candidates have oob_skb and calling kfree_skb(oob_skb) releases the remaining candidates.

However, I missed a case that oob_skb has self-referencing fd and another fd and the latter sk is placed before the former in the candidate list. Then, the while loop never proceeds, resulting the task hung.

__unix_gc() has the same loop just before purging the collected skb, so we can call kfree_skb(oob_skb) there and let __skb_queue_purge() release all inflight sockets.

[0]:
Sending NMI from CPU 0 to CPUs 1:
NMI backtrace for cpu 1

CPU: 1 PID: 2784 Comm: kworker/u4:8 Not tainted 6.8.0-rc4-syzkaller-01028-g71b605d32017 #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/25/2024
Workqueue: events_unbound __unix_gc
RIP: 0010:__sanitizer_cov_trace_pc+0x0/0x70 kernel/kcov.c:200
Code: 89 fb e8 23 00 00 00 48 8b 3d 84 f5 1a 0c 48 89 de 5b e9 43 26 57 00 0f 1f 00 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 <f3> 0f 1e fa 48 8b 04 24 65 48 8b 0d 90 52 70 7e 65 8b 15 91 52 70
RSP: 0018:ffffc9000a17fa78 EFLAGS: 00000287
RAX: ffffffff8a0a6108 RBX: ffff88802b6c2640 RCX: ffff88802c0b3b80
RDX: 0000000000000000 RSI: 0000000000000002 RDI: 0000000000000000
RBP: ffffc9000a17fbf0 R08: ffffffff89383f1d R09: 1ffff1100ee5ff84
R10: dffffc0000000000 R11: ffffed100ee5ff85 R12: 1ffff110056d84ee
R13: ffffc9000a17fae0 R14: 0000000000000000 R15: ffffffff8f47b840
FS:  0000000000000000(0000) GS:ffff8880b9500000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007ffef5687ff8 CR3: 0000000029b34000 CR4: 00000000003506f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 <NMI>
 </NMI>
 <TASK>
 __unix_gc+0xe69/0xf40 net/unix/garbage.c:343
 process_one_work kernel/workqueue.c:2633 [inline]
 process_scheduled_works+0x913/0x1420 kernel/workqueue.c:2706
 worker_thread+0xa5f/0x1000 kernel/workqueue.c:2787
 kthread+0x2ef/0x390 kernel/kthread.c:388
 ret_from_fork+0x4b/0x80 arch/x86/kernel/process.c:147
 ret_from_fork_asm+0x1b/0x30 arch/x86/entry/entry_64.S:242
 </TASK>
Reported-and-tested-by [email protected]
Fixes 25236c91b5ab ("af_unix: Fix task hung while purging oob_skb in GC.")
---
 net/unix/garbage.c |   22 +++++++++-------------
 1 file changed, 9 insertions(+), 13 deletions(-)
net/unix/garbage.c +9 -13
@@ -284,9 +284,17 @@ void unix_gc(void)
 	 * which are creating the cycle(s).
 	 */
 	skb_queue_head_init(&hitlist);
-	list_for_each_entry(u, &gc_candidates, link)
+	list_for_each_entry(u, &gc_candidates, link) {
 		scan_children(&u->sk, inc_inflight, &hitlist);
 
+#if IS_ENABLED(CONFIG_AF_UNIX_OOB)
+		if (u->oob_skb) {
+			kfree_skb(u->oob_skb);
+			u->oob_skb = NULL;
+		}
+#endif
+	}
+
 	/* not_cycle_list contains those sockets which do not make up a
 	 * cycle.  Restore these to the inflight list.
 	 */
@@ -314,18 +322,6 @@ void unix_gc(void)
 	/* Here we are. Hitlist is filled. Die. */
 	__skb_queue_purge(&hitlist);
 
-#if IS_ENABLED(CONFIG_AF_UNIX_OOB)
-	while (!list_empty(&gc_candidates)) {
-		u = list_entry(gc_candidates.next, struct unix_sock, link);
-		if (u->oob_skb) {
-			struct sk_buff *skb = u->oob_skb;
-
-			u->oob_skb = NULL;
-			kfree_skb(skb);
-		}
-	}
-#endif
-
 	spin_lock(&unix_gc_lock);
 
 	/* There could be io_uring registered files, just push them back to
author maintainer
#202 lore raw

PATCH 198/215 phy: freescale: phy-fsl-imx8-mipi-dphy: Fix alias name to use dashes

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Alexander Stein <[email protected]>

[ Upstream commit 7936378cb6d87073163130e1e1fc1e5f76a597cf ]

Devicetree spec lists only dashes as valid characters for alias names. Table 3.2: Valid characters for alias names, Devicee Specification, Release v0.4

Fixes 3fbae284887de ("phy: freescale: phy-fsl-imx8-mipi-dphy: Add i.MX8qxp LVDS PHY mode support")
---
 drivers/phy/freescale/phy-fsl-imx8-mipi-dphy.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/phy/freescale/phy-fsl-imx8-mipi-dphy.c +1 -1
index e625b32889bfc..0928a526e2ab3 100644
@@ -706,7 +706,7 @@ static int mixel_dphy_probe(struct platform_device *pdev)
 			return ret;
 		}
 
-		priv->id = of_alias_get_id(np, "mipi_dphy");
+		priv->id = of_alias_get_id(np, "mipi-dphy");
 		if (priv->id < 0) {
 			dev_err(dev, "Failed to get phy node alias id: %d\n",
 				priv->id);
Signature · 1 line
2.43.0
author maintainer
#203 lore raw

PATCH 199/215 powerpc/pseries/iommu: IOMMU table is not initialized for kdump over SR-IOV

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Gaurav Batra <[email protected]>

[ Upstream commit 09a3c1e46142199adcee372a420b024b4fc61051 ]

When kdump kernel tries to copy dump data over SR-IOV, LPAR panics due to NULL pointer exception:

Kernel attempted to read user page (0) - exploit attempt? (uid: 0)

  BUG: Kernel NULL pointer dereference on read at 0x00000000
  Faulting instruction address: 0xc000000020847ad4
  Oops: Kernel access of bad area, sig: 11 [#1]
  LE PAGE_SIZE=64K MMU=Radix SMP NR_CPUS=2048 NUMA pSeries
  Modules linked in: mlx5_core(+) vmx_crypto pseries_wdt papr_scm libnvdimm mlxfw tls psample sunrpc fuse overlay squashfs loop
  CPU: 12 PID: 315 Comm: systemd-udevd Not tainted 6.4.0-Test102+ #12
  Hardware name: IBM,9080-HEX POWER10 (raw) 0x800200 0xf000006 of:IBM,FW1060.00 (NH1060_008) hv:phyp pSeries
  NIP:  c000000020847ad4 LR: c00000002083b2dc CTR: 00000000006cd18c
  REGS: c000000029162ca0 TRAP: 0300   Not tainted  (6.4.0-Test102+)
  MSR:  800000000280b033 <SF,VEC,VSX,EE,FP,ME,IR,DR,RI,LE>  CR: 48288244  XER: 00000008
  CFAR: c00000002083b2d8 DAR: 0000000000000000 DSISR: 40000000 IRQMASK: 1
  ...
  NIP _find_next_zero_bit+0x24/0x110
  LR  bitmap_find_next_zero_area_off+0x5c/0xe0
  Call Trace:
    dev_printk_emit+0x38/0x48 (unreliable)
    iommu_area_alloc+0xc4/0x180
    iommu_range_alloc+0x1e8/0x580
    iommu_alloc+0x60/0x130
    iommu_alloc_coherent+0x158/0x2b0
    dma_iommu_alloc_coherent+0x3c/0x50
    dma_alloc_attrs+0x170/0x1f0
    mlx5_cmd_init+0xc0/0x760 [mlx5_core]
    mlx5_function_setup+0xf0/0x510 [mlx5_core]
    mlx5_init_one+0x84/0x210 [mlx5_core]
    probe_one+0x118/0x2c0 [mlx5_core]
    local_pci_probe+0x68/0x110
    pci_call_probe+0x68/0x200
    pci_device_probe+0xbc/0x1a0
    really_probe+0x104/0x540
    __driver_probe_device+0xb4/0x230
    driver_probe_device+0x54/0x130
    __driver_attach+0x158/0x2b0
    bus_for_each_dev+0xa8/0x130
    driver_attach+0x34/0x50
    bus_add_driver+0x16c/0x300
    driver_register+0xa4/0x1b0
    __pci_register_driver+0x68/0x80
    mlx5_init+0xb8/0x100 [mlx5_core]
    do_one_initcall+0x60/0x300
    do_init_module+0x7c/0x2b0

At the time of LPAR dump, before kexec hands over control to kdump kernel, DDWs (Dynamic DMA Windows) are scanned and added to the FDT. For the SR-IOV case, default DMA window "ibm,dma-window" is removed from the FDT and DDW added, for the device.

Now, kexec hands over control to the kdump kernel.

When the kdump kernel initializes, PCI busses are scanned and IOMMU group/tables created, in pci_dma_bus_setup_pSeriesLP(). For the SR-IOV case, there is no "ibm,dma-window". The original commit: b1fc44eaa9ba, fixes the path where memory is pre-mapped (direct mapped) to the DDW. When TCEs are direct mapped, there is no need to initialize IOMMU tables.

iommu_table_setparms_lpar() only considers "ibm,dma-window" property when initiallizing IOMMU table. In the scenario where TCEs are dynamically allocated for SR-IOV, newly created IOMMU table is not initialized. Later, when the device driver tries to enter TCEs for the SR-IOV device, NULL pointer execption is thrown from iommu_area_alloc().

The fix is to initialize the IOMMU table with DDW property stored in the FDT. There are 2 points to remember:

1. For the dedicated adapter, kdump kernel would encounter both default and DDW in FDT. In this case, DDW property is used to initialize the IOMMU table.

2. A DDW could be direct or dynamic mapped. kdump kernel would initialize IOMMU table and mark the existing DDW as "dynamic". This works fine since, at the time of table initialization, iommu_table_clear() makes some space in the DDW, for some predefined number of TCEs which are needed for kdump to succeed.

Fixes b1fc44eaa9ba ("pseries/iommu/ddw: Fix kdump to work in absence of ibm,dma-window")
---
 arch/powerpc/platforms/pseries/iommu.c | 156 +++++++++++++++++--------
 1 file changed, 105 insertions(+), 51 deletions(-)
arch/powerpc/platforms/pseries/iommu.c +105 -51
index 97b026130c71b..1e5f083cdb720 100644
@@ -569,29 +569,6 @@ static void iommu_table_setparms(struct pci_controller *phb,
 
 struct iommu_table_ops iommu_table_lpar_multi_ops;
 
-/*
- * iommu_table_setparms_lpar
- *
- * Function: On pSeries LPAR systems, return TCE table info, given a pci bus.
- */
-static void iommu_table_setparms_lpar(struct pci_controller *phb,
-				      struct device_node *dn,
-				      struct iommu_table *tbl,
-				      struct iommu_table_group *table_group,
-				      const __be32 *dma_window)
-{
-	unsigned long offset, size, liobn;
-
-	of_parse_dma_window(dn, dma_window, &liobn, &offset, &size);
-
-	iommu_table_setparms_common(tbl, phb->bus->number, liobn, offset, size, IOMMU_PAGE_SHIFT_4K, NULL,
-				    &iommu_table_lpar_multi_ops);
-
-
-	table_group->tce32_start = offset;
-	table_group->tce32_size = size;
-}
-
 struct iommu_table_ops iommu_table_pseries_ops = {
 	.set = tce_build_pSeries,
 	.clear = tce_free_pSeries,
@@ -719,26 +696,71 @@ struct iommu_table_ops iommu_table_lpar_multi_ops = {
  * dynamic 64bit DMA window, walking up the device tree.
  */
 static struct device_node *pci_dma_find(struct device_node *dn,
-					const __be32 **dma_window)
+					struct dynamic_dma_window_prop *prop)
 {
-	const __be32 *dw = NULL;
+	const __be32 *default_prop = NULL;
+	const __be32 *ddw_prop = NULL;
+	struct device_node *rdn = NULL;
+	bool default_win = false, ddw_win = false;
 
 	for ( ; dn && PCI_DN(dn); dn = dn->parent) {
-		dw = of_get_property(dn, "ibm,dma-window", NULL);
-		if (dw) {
-			if (dma_window)
-				*dma_window = dw;
-			return dn;
+		default_prop = of_get_property(dn, "ibm,dma-window", NULL);
+		if (default_prop) {
+			rdn = dn;
+			default_win = true;
+		}
+		ddw_prop = of_get_property(dn, DIRECT64_PROPNAME, NULL);
+		if (ddw_prop) {
+			rdn = dn;
+			ddw_win = true;
+			break;
+		}
+		ddw_prop = of_get_property(dn, DMA64_PROPNAME, NULL);
+		if (ddw_prop) {
+			rdn = dn;
+			ddw_win = true;
+			break;
 		}
-		dw = of_get_property(dn, DIRECT64_PROPNAME, NULL);
-		if (dw)
-			return dn;
-		dw = of_get_property(dn, DMA64_PROPNAME, NULL);
-		if (dw)
-			return dn;
+
+		/* At least found default window, which is the case for normal boot */
+		if (default_win)
+			break;
 	}
 
-	return NULL;
+	/* For PCI devices there will always be a DMA window, either on the device
+	 * or parent bus
+	 */
+	WARN_ON(!(default_win | ddw_win));
+
+	/* caller doesn't want to get DMA window property */
+	if (!prop)
+		return rdn;
+
+	/* parse DMA window property. During normal system boot, only default
+	 * DMA window is passed in OF. But, for kdump, a dedicated adapter might
+	 * have both default and DDW in FDT. In this scenario, DDW takes precedence
+	 * over default window.
+	 */
+	if (ddw_win) {
+		struct dynamic_dma_window_prop *p;
+
+		p = (struct dynamic_dma_window_prop *)ddw_prop;
+		prop->liobn = p->liobn;
+		prop->dma_base = p->dma_base;
+		prop->tce_shift = p->tce_shift;
+		prop->window_shift = p->window_shift;
+	} else if (default_win) {
+		unsigned long offset, size, liobn;
+
+		of_parse_dma_window(rdn, default_prop, &liobn, &offset, &size);
+
+		prop->liobn = cpu_to_be32((u32)liobn);
+		prop->dma_base = cpu_to_be64(offset);
+		prop->tce_shift = cpu_to_be32(IOMMU_PAGE_SHIFT_4K);
+		prop->window_shift = cpu_to_be32(order_base_2(size));
+	}
+
+	return rdn;
 }
 
 static void pci_dma_bus_setup_pSeriesLP(struct pci_bus *bus)
@@ -746,17 +768,20 @@ static void pci_dma_bus_setup_pSeriesLP(struct pci_bus *bus)
 	struct iommu_table *tbl;
 	struct device_node *dn, *pdn;
 	struct pci_dn *ppci;
-	const __be32 *dma_window = NULL;
+	struct dynamic_dma_window_prop prop;
 
 	dn = pci_bus_to_OF_node(bus);
 
 	pr_debug("pci_dma_bus_setup_pSeriesLP: setting up bus %pOF\n",
 		 dn);
 
-	pdn = pci_dma_find(dn, &dma_window);
+	pdn = pci_dma_find(dn, &prop);
 
-	if (dma_window == NULL)
-		pr_debug("  no ibm,dma-window property !\n");
+	/* In PPC architecture, there will always be DMA window on bus or one of the
+	 * parent bus. During reboot, there will be ibm,dma-window property to
+	 * define DMA window. For kdump, there will at least be default window or DDW
+	 * or both.
+	 */
 
 	ppci = PCI_DN(pdn);
 
@@ -766,13 +791,24 @@ static void pci_dma_bus_setup_pSeriesLP(struct pci_bus *bus)
 	if (!ppci->table_group) {
 		ppci->table_group = iommu_pseries_alloc_group(ppci->phb->node);
 		tbl = ppci->table_group->tables[0];
-		if (dma_window) {
-			iommu_table_setparms_lpar(ppci->phb, pdn, tbl,
-						  ppci->table_group, dma_window);
 
-			if (!iommu_init_table(tbl, ppci->phb->node, 0, 0))
-				panic("Failed to initialize iommu table");
-		}
+		iommu_table_setparms_common(tbl, ppci->phb->bus->number,
+				be32_to_cpu(prop.liobn),
+				be64_to_cpu(prop.dma_base),
+				1ULL << be32_to_cpu(prop.window_shift),
+				be32_to_cpu(prop.tce_shift), NULL,
+				&iommu_table_lpar_multi_ops);
+
+		/* Only for normal boot with default window. Doesn't matter even
+		 * if we set these with DDW which is 64bit during kdump, since
+		 * these will not be used during kdump.
+		 */
+		ppci->table_group->tce32_start = be64_to_cpu(prop.dma_base);
+		ppci->table_group->tce32_size = 1 << be32_to_cpu(prop.window_shift);
+
+		if (!iommu_init_table(tbl, ppci->phb->node, 0, 0))
+			panic("Failed to initialize iommu table");
+
 		iommu_register_group(ppci->table_group,
 				pci_domain_nr(bus), 0);
 		pr_debug("  created table: %p\n", ppci->table_group);
@@ -960,6 +996,12 @@ static void find_existing_ddw_windows_named(const char *name)
 			continue;
 		}
 
+		/* If at the time of system initialization, there are DDWs in OF,
+		 * it means this is during kexec. DDW could be direct or dynamic.
+		 * We will just mark DDWs as "dynamic" since this is kdump path,
+		 * no need to worry about perforance. ddw_list_new_entry() will
+		 * set window->direct = false.
+		 */
 		window = ddw_list_new_entry(pdn, dma64);
 		if (!window) {
 			of_node_put(pdn);
@@ -1525,8 +1567,8 @@ static void pci_dma_dev_setup_pSeriesLP(struct pci_dev *dev)
 {
 	struct device_node *pdn, *dn;
 	struct iommu_table *tbl;
-	const __be32 *dma_window = NULL;
 	struct pci_dn *pci;
+	struct dynamic_dma_window_prop prop;
 
 	pr_debug("pci_dma_dev_setup_pSeriesLP: %s\n", pci_name(dev));
 
@@ -1539,7 +1581,7 @@ static void pci_dma_dev_setup_pSeriesLP(struct pci_dev *dev)
 	dn = pci_device_to_OF_node(dev);
 	pr_debug("  node is %pOF\n", dn);
 
-	pdn = pci_dma_find(dn, &dma_window);
+	pdn = pci_dma_find(dn, &prop);
 	if (!pdn || !PCI_DN(pdn)) {
 		printk(KERN_WARNING "pci_dma_dev_setup_pSeriesLP: "
 		       "no DMA window found for pci dev=%s dn=%pOF\n",
@@ -1552,8 +1594,20 @@ static void pci_dma_dev_setup_pSeriesLP(struct pci_dev *dev)
 	if (!pci->table_group) {
 		pci->table_group = iommu_pseries_alloc_group(pci->phb->node);
 		tbl = pci->table_group->tables[0];
-		iommu_table_setparms_lpar(pci->phb, pdn, tbl,
-				pci->table_group, dma_window);
+
+		iommu_table_setparms_common(tbl, pci->phb->bus->number,
+				be32_to_cpu(prop.liobn),
+				be64_to_cpu(prop.dma_base),
+				1ULL << be32_to_cpu(prop.window_shift),
+				be32_to_cpu(prop.tce_shift), NULL,
+				&iommu_table_lpar_multi_ops);
+
+		/* Only for normal boot with default window. Doesn't matter even
+		 * if we set these with DDW which is 64bit during kdump, since
+		 * these will not be used during kdump.
+		 */
+		pci->table_group->tce32_start = be64_to_cpu(prop.dma_base);
+		pci->table_group->tce32_size = 1 << be32_to_cpu(prop.window_shift);
 
 		iommu_init_table(tbl, pci->phb->node, 0, 0);
 		iommu_register_group(pci->table_group,
Signature · 1 line
2.43.0
author maintainer
#204 lore raw

PATCH 200/215 gpio: 74x164: Enable output pins after registers are reset

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Arturas Moskvinas <[email protected]>

[ Upstream commit 530b1dbd97846b110ea8a94c7cc903eca21786e5 ]

Chip outputs are enabled[1] before actual reset is performed[2] which might cause pin output value to flip flop if previous pin value was set to 1. Fix that behavior by making sure chip is fully reset before all outputs are enabled.

Flip-flop can be noticed when module is removed and inserted again and one of the pins was changed to 1 before removal. 100 microsecond flipping is noticeable on oscilloscope (100khz SPI bus).

For a properly reset chip - output is enabled around 100 microseconds (on 100khz SPI bus) later during probing process hence should be irrelevant behavioral change.

---
 drivers/gpio/gpio-74x164.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/gpio/gpio-74x164.c +2 -2
index e00c333105170..753e7be039e4d 100644
@@ -127,8 +127,6 @@ static int gen_74x164_probe(struct spi_device *spi)
 	if (IS_ERR(chip->gpiod_oe))
 		return PTR_ERR(chip->gpiod_oe);
 
-	gpiod_set_value_cansleep(chip->gpiod_oe, 1);
-
 	spi_set_drvdata(spi, chip);
 
 	chip->gpio_chip.label = spi->modalias;
@@ -153,6 +151,8 @@ static int gen_74x164_probe(struct spi_device *spi)
 		goto exit_destroy;
 	}
 
+	gpiod_set_value_cansleep(chip->gpiod_oe, 1);
+
 	ret = gpiochip_add_data(&chip->gpio_chip, chip);
 	if (!ret)
 		return 0;
Signature · 1 line
2.43.0
author maintainer
#205 lore raw

PATCH 201/215 gpiolib: Fix the error path order in gpiochip_add_data_with_key()

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Andy Shevchenko <[email protected]>

[ Upstream commit e4aec4daa8c009057b5e063db1b7322252c92dc8 ]

After shuffling the code, error path wasn't updated correctly. Fix it here.

Fixes 2f4133bb5f14 ("gpiolib: No need to call gpiochip_remove_pin_ranges() twice")
---
 drivers/gpio/gpiolib.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpio/gpiolib.c +1 -1
index 6d3e3454a6ed6..f646df7f1b41a 100644
@@ -815,11 +815,11 @@ int gpiochip_add_data_with_key(struct gpio_chip *gc, void *data,
 	gpiochip_irqchip_free_valid_mask(gc);
 err_remove_acpi_chip:
 	acpi_gpiochip_remove(gc);
+	gpiochip_remove_pin_ranges(gc);
 err_remove_of_chip:
 	gpiochip_free_hogs(gc);
 	of_gpiochip_remove(gc);
 err_free_gpiochip_mask:
-	gpiochip_remove_pin_ranges(gc);
 	gpiochip_free_valid_mask(gc);
 	if (gdev->dev.release) {
 		/* release() has been registered by gpiochip_setup_dev() */
Signature · 1 line
2.43.0
author maintainer
#206 lore raw

PATCH 202/215 gpio: fix resource unwinding order in error path

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Bartosz Golaszewski <[email protected]>

[ Upstream commit ec5c54a9d3c4f9c15e647b049fea401ee5258696 ]

Hogs are added *after* ACPI so should be removed *before* in error path.

Fixes a411e81e61df ("gpiolib: add hogs support for machine code")
---
 drivers/gpio/gpiolib.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
drivers/gpio/gpiolib.c +4 -4
index f646df7f1b41a..9d8c783124033 100644
@@ -784,11 +784,11 @@ int gpiochip_add_data_with_key(struct gpio_chip *gc, void *data,
 
 	ret = gpiochip_irqchip_init_valid_mask(gc);
 	if (ret)
-		goto err_remove_acpi_chip;
+		goto err_free_hogs;
 
 	ret = gpiochip_irqchip_init_hw(gc);
 	if (ret)
-		goto err_remove_acpi_chip;
+		goto err_remove_irqchip_mask;
 
 	ret = gpiochip_add_irqchip(gc, lock_key, request_key);
 	if (ret)
@@ -813,11 +813,11 @@ int gpiochip_add_data_with_key(struct gpio_chip *gc, void *data,
 	gpiochip_irqchip_remove(gc);
 err_remove_irqchip_mask:
 	gpiochip_irqchip_free_valid_mask(gc);
-err_remove_acpi_chip:
+err_free_hogs:
+	gpiochip_free_hogs(gc);
 	acpi_gpiochip_remove(gc);
 	gpiochip_remove_pin_ranges(gc);
 err_remove_of_chip:
-	gpiochip_free_hogs(gc);
 	of_gpiochip_remove(gc);
 err_free_gpiochip_mask:
 	gpiochip_free_valid_mask(gc);
Signature · 1 line
2.43.0
author maintainer
#207 lore raw

PATCH 203/215 block: define bvec_iter as __packed __aligned(4)

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ming Lei <[email protected]>

[ Upstream commit 7838b4656110d950afdd92a081cc0f33e23e0ea8 ]

In commit 19416123ab3e ("block: define 'struct bvec_iter' as packed"), what we need is to save the 4byte padding, and avoid bio to spread on one extra cache line.

It is enough to define it as '__packed __aligned(4)', as '__packed' alone means byte aligned, and can cause compiler to generate horrible code on architectures that don't support unaligned access in case that bvec_iter is embedded in other structures.

Cc Mikulas Patocka <[email protected]>
Fixes 19416123ab3e ("block: define 'struct bvec_iter' as packed")
---
 include/linux/bvec.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
include/linux/bvec.h +1 -1
index 9e3dac51eb26b..d4dbaae8b5218 100644
@@ -59,7 +59,7 @@ struct bvec_iter {
 
 	unsigned int            bi_bvec_done;	/* number of bytes completed in
 						   current bvec */
-} __packed;
+} __packed __aligned(4);
 
 struct bvec_iter_all {
 	struct bio_vec	bv;
Signature · 1 line
2.43.0
author maintainer
#208 lore raw

PATCH 204/215 Revert "interconnect: Fix locking for runpm vs reclaim"

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Greg Kroah-Hartman <[email protected]>

This reverts commit ee42bfc791aa3cd78e29046f26a09d189beb3efb which is commit af42269c3523492d71ebbe11fefae2653e9cdc78 upstream.

It is reported to cause boot crashes in Android systems, so revert it from the stable trees for now.

Cc Rob Clark <[email protected]>
Cc Georgi Djakov <[email protected]>
Cc Sasha Levin <[email protected]>
---
 drivers/interconnect/core.c |    8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)
drivers/interconnect/core.c +2 -6
@@ -29,7 +29,6 @@ static LIST_HEAD(icc_providers);
 static int providers_count;
 static bool synced_state;
 static DEFINE_MUTEX(icc_lock);
-static DEFINE_MUTEX(icc_bw_lock);
 static struct dentry *icc_debugfs_dir;
 
 static void icc_summary_show_one(struct seq_file *s, struct icc_node *n)
@@ -636,7 +635,7 @@ int icc_set_bw(struct icc_path *path, u3
 	if (WARN_ON(IS_ERR(path) || !path->num_nodes))
 		return -EINVAL;
 
-	mutex_lock(&icc_bw_lock);
+	mutex_lock(&icc_lock);
 
 	old_avg = path->reqs[0].avg_bw;
 	old_peak = path->reqs[0].peak_bw;
@@ -668,7 +667,7 @@ int icc_set_bw(struct icc_path *path, u3
 		apply_constraints(path);
 	}
 
-	mutex_unlock(&icc_bw_lock);
+	mutex_unlock(&icc_lock);
 
 	trace_icc_set_bw_end(path, ret);
 
@@ -971,7 +970,6 @@ void icc_node_add(struct icc_node *node,
 		return;
 
 	mutex_lock(&icc_lock);
-	mutex_lock(&icc_bw_lock);
 
 	node->provider = provider;
 	list_add_tail(&node->node_list, &provider->nodes);
@@ -997,7 +995,6 @@ void icc_node_add(struct icc_node *node,
 	node->avg_bw = 0;
 	node->peak_bw = 0;
 
-	mutex_unlock(&icc_bw_lock);
 	mutex_unlock(&icc_lock);
 }
 EXPORT_SYMBOL_GPL(icc_node_add);
@@ -1137,7 +1134,6 @@ void icc_sync_state(struct device *dev)
 		return;
 
 	mutex_lock(&icc_lock);
-	mutex_lock(&icc_bw_lock);
 	synced_state = true;
 	list_for_each_entry(p, &icc_providers, provider_list) {
 		dev_dbg(p->dev, "interconnect provider is in synced state\n");
author maintainer
#209 lore raw

PATCH 205/215 Revert "interconnect: Teach lockdep about icc_bw_lock order"

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Greg Kroah-Hartman <[email protected]>

This reverts commit 0db211ec0f1d32b93486e8f6565249ad4d1bece5 which is commit 13619170303878e1dae86d9a58b039475c957fcf upstream.

It is reported to cause boot crashes in Android systems, so revert it from the stable trees for now.

Cc Rob Clark <[email protected]>
Cc Georgi Djakov <[email protected]>
Cc Guenter Roeck <[email protected]>
Cc Jon Hunter <[email protected]>
---
 drivers/interconnect/core.c |   10 +---------
 1 file changed, 1 insertion(+), 9 deletions(-)
drivers/interconnect/core.c +1 -9
@@ -1146,21 +1146,13 @@ void icc_sync_state(struct device *dev)
 			}
 		}
 	}
-	mutex_unlock(&icc_bw_lock);
 	mutex_unlock(&icc_lock);
 }
 EXPORT_SYMBOL_GPL(icc_sync_state);
 
 static int __init icc_init(void)
 {
-	struct device_node *root;
-
-	/* Teach lockdep about lock ordering wrt. shrinker: */
-	fs_reclaim_acquire(GFP_KERNEL);
-	might_lock(&icc_bw_lock);
-	fs_reclaim_release(GFP_KERNEL);
-
-	root = of_find_node_by_path("/");
+	struct device_node *root = of_find_node_by_path("/");
 
 	providers_count = of_count_icc_providers(root);
 	of_node_put(root);
author maintainer
#210 lore raw

PATCH 206/215 x86/bugs: Add asm helpers for executing VERW

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Pawan Gupta <[email protected]>

commit baf8361e54550a48a7087b603313ad013cc13386 upstream.

MDS mitigation requires clearing the CPU buffers before returning to user. This needs to be done late in the exit-to-user path. Current location of VERW leaves a possibility of kernel data ending up in CPU buffers for memory accesses done after VERW such as:

1. Kernel data accessed by an NMI between VERW and return-to-user can remain in CPU buffers since NMI returning to kernel does not execute VERW to clear CPU buffers.
2. Alyssa reported that after VERW is executed, CONFIG_GCC_PLUGIN_STACKLEAK=y scrubs the stack used by a system call. Memory accesses during stack scrubbing can move kernel stack contents into CPU buffers.
3. When caller saved registers are restored after a return from function executing VERW, the kernel stack accesses can remain in CPU buffers(since they occur after VERW).

To fix this VERW needs to be moved very late in exit-to-user path.

In preparation for moving VERW to entry/exit asm code, create macros that can be used in asm. Also make VERW patching depend on a new feature flag X86_FEATURE_CLEAR_CPU_BUF.

[pawan: - Runtime patch jmp instead of verw in macro CLEAR_CPU_BUFFERS

	    due to lack of relative addressing support for relocations
	    in kernels < v6.5.
	  - Add UNWIND_HINT_EMPTY to avoid warning:
	    arch/x86/entry/entry.o: warning: objtool: mds_verw_sel+0x0: unreachable instruction]
---
 arch/x86/entry/entry.S               |   23 +++++++++++++++++++++++
 arch/x86/include/asm/cpufeatures.h   |    2 +-
 arch/x86/include/asm/nospec-branch.h |   15 +++++++++++++++
 3 files changed, 39 insertions(+), 1 deletion(-)
arch/x86/entry/entry.S +23 -0
@@ -6,6 +6,9 @@
 #include <linux/linkage.h>
 #include <asm/export.h>
 #include <asm/msr-index.h>
+#include <asm/unwind_hints.h>
+#include <asm/segment.h>
+#include <asm/cache.h>
 
 .pushsection .noinstr.text, "ax"
 
@@ -20,3 +23,23 @@ SYM_FUNC_END(entry_ibpb)
 EXPORT_SYMBOL_GPL(entry_ibpb);
 
 .popsection
+
+/*
+ * Define the VERW operand that is disguised as entry code so that
+ * it can be referenced with KPTI enabled. This ensure VERW can be
+ * used late in exit-to-user path after page tables are switched.
+ */
+.pushsection .entry.text, "ax"
+
+.align L1_CACHE_BYTES, 0xcc
+SYM_CODE_START_NOALIGN(mds_verw_sel)
+	UNWIND_HINT_EMPTY
+	ANNOTATE_NOENDBR
+	.word __KERNEL_DS
+.align L1_CACHE_BYTES, 0xcc
+SYM_CODE_END(mds_verw_sel);
+/* For KVM */
+EXPORT_SYMBOL_GPL(mds_verw_sel);
+
+.popsection
+
arch/x86/include/asm/cpufeatures.h +1 -1
@@ -304,7 +304,7 @@
 #define X86_FEATURE_UNRET		(11*32+15) /* "" AMD BTB untrain return */
 #define X86_FEATURE_USE_IBPB_FW		(11*32+16) /* "" Use IBPB during runtime firmware calls */
 #define X86_FEATURE_RSB_VMEXIT_LITE	(11*32+17) /* "" Fill RSB on VM exit when EIBRS is enabled */
-
+#define X86_FEATURE_CLEAR_CPU_BUF	(11*32+18) /* "" Clear CPU buffers using VERW */
 
 #define X86_FEATURE_MSR_TSX_CTRL	(11*32+20) /* "" MSR IA32_TSX_CTRL (Intel) implemented */
 
arch/x86/include/asm/nospec-branch.h +15 -0
@@ -194,6 +194,19 @@
 #endif
 .endm
 
+/*
+ * Macro to execute VERW instruction that mitigate transient data sampling
+ * attacks such as MDS. On affected systems a microcode update overloaded VERW
+ * instruction to also clear the CPU buffers. VERW clobbers CFLAGS.ZF.
+ *
+ * Note: Only the memory operand variant of VERW clears the CPU buffers.
+ */
+.macro CLEAR_CPU_BUFFERS
+	ALTERNATIVE "jmp .Lskip_verw_\@", "", X86_FEATURE_CLEAR_CPU_BUF
+	verw _ASM_RIP(mds_verw_sel)
+.Lskip_verw_\@:
+.endm
+
 #else /* __ASSEMBLY__ */
 
 #define ANNOTATE_RETPOLINE_SAFE					\
@@ -375,6 +388,8 @@ DECLARE_STATIC_KEY_FALSE(switch_mm_cond_
 
 DECLARE_STATIC_KEY_FALSE(mmio_stale_data_clear);
 
+extern u16 mds_verw_sel;
+
 #include <asm/segment.h>
 
 /**
author maintainer
#211 lore raw

PATCH 207/215 x86/entry_64: Add VERW just before userspace transition

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Pawan Gupta <[email protected]>

commit 3c7501722e6b31a6e56edd23cea5e77dbb9ffd1a upstream.

Mitigation for MDS is to use VERW instruction to clear any secrets in CPU Buffers. Any memory accesses after VERW execution can still remain in CPU buffers. It is safer to execute VERW late in return to user path to minimize the window in which kernel data can end up in CPU buffers. There are not many kernel secrets to be had after SWITCH_TO_USER_CR3.

Add support for deploying VERW mitigation after user register state is restored. This helps minimize the chances of kernel data ending up into CPU buffers after executing VERW.

Note that the mitigation at the new location is not yet enabled.

Corner case not handled
=======================
Interrupts returning to kernel don't clear CPUs buffers since the exit-to-user path is expected to do that anyways. But, there could be a case when an NMI is generated in kernel after the exit-to-user path has cleared the buffers. This case is not handled and NMI returning to kernel don't clear CPU buffers because:

1. It is rare to get an NMI after VERW, but before returning to user.
2. For an unprivileged user, there is no known way to make that NMI less rare or target it.
3. It would take a large number of these precisely-timed NMIs to mount an actual attack. There's presumably not enough bandwidth.
4. The NMI in question occurs after a VERW, i.e. when user state is restored and most interesting data is already scrubbed. Whats left is only the data that NMI touches, and that may or may not be of any interest.

[ pawan: resolved conflict for hunk swapgs_restore_regs_and_return_to_usermode in backport ]

---
 arch/x86/entry/entry_64.S        |   11 +++++++++++
 arch/x86/entry/entry_64_compat.S |    1 +
 2 files changed, 12 insertions(+)
arch/x86/entry/entry_64.S +11 -0
@@ -223,6 +223,7 @@ syscall_return_via_sysret:
 SYM_INNER_LABEL(entry_SYSRETQ_unsafe_stack, SYM_L_GLOBAL)
 	ANNOTATE_NOENDBR
 	swapgs
+	CLEAR_CPU_BUFFERS
 	sysretq
 SYM_INNER_LABEL(entry_SYSRETQ_end, SYM_L_GLOBAL)
 	ANNOTATE_NOENDBR
@@ -656,6 +657,7 @@ SYM_INNER_LABEL(swapgs_restore_regs_and_
 	/* Restore RDI. */
 	popq	%rdi
 	swapgs
+	CLEAR_CPU_BUFFERS
 	jmp	.Lnative_iret
 
 
@@ -767,6 +769,8 @@ native_irq_return_ldt:
 	 */
 	popq	%rax				/* Restore user RAX */
 
+	CLEAR_CPU_BUFFERS
+
 	/*
 	 * RSP now points to an ordinary IRET frame, except that the page
 	 * is read-only and RSP[31:16] are preloaded with the userspace
@@ -1494,6 +1498,12 @@ nmi_restore:
 	movq	$0, 5*8(%rsp)		/* clear "NMI executing" */
 
 	/*
+	 * Skip CLEAR_CPU_BUFFERS here, since it only helps in rare cases like
+	 * NMI in kernel after user state is restored. For an unprivileged user
+	 * these conditions are hard to meet.
+	 */
+
+	/*
 	 * iretq reads the "iret" frame and exits the NMI stack in a
 	 * single instruction.  We are returning to kernel mode, so this
 	 * cannot result in a fault.  Similarly, we don't need to worry
@@ -1511,6 +1521,7 @@ SYM_CODE_START(ignore_sysret)
 	UNWIND_HINT_EMPTY
 	ENDBR
 	mov	$-ENOSYS, %eax
+	CLEAR_CPU_BUFFERS
 	sysretl
 SYM_CODE_END(ignore_sysret)
 #endif
arch/x86/entry/entry_64_compat.S +1 -0
@@ -272,6 +272,7 @@ SYM_INNER_LABEL(entry_SYSRETL_compat_uns
 	xorl	%r9d, %r9d
 	xorl	%r10d, %r10d
 	swapgs
+	CLEAR_CPU_BUFFERS
 	sysretl
 SYM_INNER_LABEL(entry_SYSRETL_compat_end, SYM_L_GLOBAL)
 	ANNOTATE_NOENDBR
author maintainer
#212 lore raw

PATCH 208/215 x86/entry_32: Add VERW just before userspace transition

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Pawan Gupta <[email protected]>

commit a0e2dab44d22b913b4c228c8b52b2a104434b0b3 upstream.

As done for entry_64, add support for executing VERW late in exit to user path for 32-bit mode.

---
 arch/x86/entry/entry_32.S |    3 +++
 1 file changed, 3 insertions(+)
arch/x86/entry/entry_32.S +3 -0
@@ -912,6 +912,7 @@ SYM_FUNC_START(entry_SYSENTER_32)
 	BUG_IF_WRONG_CR3 no_user_check=1
 	popfl
 	popl	%eax
+	CLEAR_CPU_BUFFERS
 
 	/*
 	 * Return back to the vDSO, which will pop ecx and edx.
@@ -981,6 +982,7 @@ restore_all_switch_stack:
 
 	/* Restore user state */
 	RESTORE_REGS pop=4			# skip orig_eax/error_code
+	CLEAR_CPU_BUFFERS
 .Lirq_return:
 	/*
 	 * ARCH_HAS_MEMBARRIER_SYNC_CORE rely on IRET core serialization
@@ -1173,6 +1175,7 @@ SYM_CODE_START(asm_exc_nmi)
 
 	/* Not on SYSENTER stack. */
 	call	exc_nmi
+	CLEAR_CPU_BUFFERS
 	jmp	.Lnmi_return
 
 .Lnmi_from_sysenter_stack:
author maintainer
#213 lore raw

PATCH 209/215 x86/bugs: Use ALTERNATIVE() instead of mds_user_clear static key

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Pawan Gupta <[email protected]>

commit 6613d82e617dd7eb8b0c40b2fe3acea655b1d611 upstream.

The VERW mitigation at exit-to-user is enabled via a static branch mds_user_clear. This static branch is never toggled after boot, and can be safely replaced with an ALTERNATIVE() which is convenient to use in asm.

Switch to ALTERNATIVE() to use the VERW mitigation late in exit-to-user path. Also remove the now redundant VERW in exc_nmi() and arch_exit_to_user_mode().

---
 Documentation/x86/mds.rst            |   36 +++++++++++++++++++++++++----------
 arch/x86/include/asm/entry-common.h  |    1 
 arch/x86/include/asm/nospec-branch.h |   12 -----------
 arch/x86/kernel/cpu/bugs.c           |   15 +++++---------
 arch/x86/kernel/nmi.c                |    3 --
 arch/x86/kvm/vmx/vmx.c               |    2 -
 6 files changed, 33 insertions(+), 36 deletions(-)
Documentation/x86/mds.rst +26 -10
@@ -95,6 +95,9 @@ The kernel provides a function to invoke
 
     mds_clear_cpu_buffers()
 
+Also macro CLEAR_CPU_BUFFERS can be used in ASM late in exit-to-user path.
+Other than CFLAGS.ZF, this macro doesn't clobber any registers.
+
 The mitigation is invoked on kernel/userspace, hypervisor/guest and C-state
 (idle) transitions.
 
@@ -138,17 +141,30 @@ Mitigation points
 
    When transitioning from kernel to user space the CPU buffers are flushed
    on affected CPUs when the mitigation is not disabled on the kernel
-   command line. The migitation is enabled through the static key
-   mds_user_clear.
+   command line. The mitigation is enabled through the feature flag
+   X86_FEATURE_CLEAR_CPU_BUF.
 
-   The mitigation is invoked in prepare_exit_to_usermode() which covers
-   all but one of the kernel to user space transitions.  The exception
-   is when we return from a Non Maskable Interrupt (NMI), which is
-   handled directly in do_nmi().
-
-   (The reason that NMI is special is that prepare_exit_to_usermode() can
-    enable IRQs.  In NMI context, NMIs are blocked, and we don't want to
-    enable IRQs with NMIs blocked.)
+   The mitigation is invoked just before transitioning to userspace after
+   user registers are restored. This is done to minimize the window in
+   which kernel data could be accessed after VERW e.g. via an NMI after
+   VERW.
+
+   **Corner case not handled**
+   Interrupts returning to kernel don't clear CPUs buffers since the
+   exit-to-user path is expected to do that anyways. But, there could be
+   a case when an NMI is generated in kernel after the exit-to-user path
+   has cleared the buffers. This case is not handled and NMI returning to
+   kernel don't clear CPU buffers because:
+
+   1. It is rare to get an NMI after VERW, but before returning to userspace.
+   2. For an unprivileged user, there is no known way to make that NMI
+      less rare or target it.
+   3. It would take a large number of these precisely-timed NMIs to mount
+      an actual attack.  There's presumably not enough bandwidth.
+   4. The NMI in question occurs after a VERW, i.e. when user state is
+      restored and most interesting data is already scrubbed. Whats left
+      is only the data that NMI touches, and that may or may not be of
+      any interest.
 
 
 2. C-State transition
arch/x86/include/asm/entry-common.h +0 -1
@@ -91,7 +91,6 @@ static inline void arch_exit_to_user_mod
 
 static __always_inline void arch_exit_to_user_mode(void)
 {
-	mds_user_clear_cpu_buffers();
 	amd_clear_divider();
 }
 #define arch_exit_to_user_mode arch_exit_to_user_mode
arch/x86/include/asm/nospec-branch.h +0 -12
@@ -381,7 +381,6 @@ DECLARE_STATIC_KEY_FALSE(switch_to_cond_
 DECLARE_STATIC_KEY_FALSE(switch_mm_cond_ibpb);
 DECLARE_STATIC_KEY_FALSE(switch_mm_always_ibpb);
 
-DECLARE_STATIC_KEY_FALSE(mds_user_clear);
 DECLARE_STATIC_KEY_FALSE(mds_idle_clear);
 
 DECLARE_STATIC_KEY_FALSE(switch_mm_cond_l1d_flush);
@@ -416,17 +415,6 @@ static __always_inline void mds_clear_cp
 }
 
 /**
- * mds_user_clear_cpu_buffers - Mitigation for MDS and TAA vulnerability
- *
- * Clear CPU buffers if the corresponding static key is enabled
- */
-static __always_inline void mds_user_clear_cpu_buffers(void)
-{
-	if (static_branch_likely(&mds_user_clear))
-		mds_clear_cpu_buffers();
-}
-
-/**
  * mds_idle_clear_cpu_buffers - Mitigation for MDS vulnerability
  *
  * Clear CPU buffers if the corresponding static key is enabled
arch/x86/kernel/cpu/bugs.c +6 -9
@@ -110,9 +110,6 @@ DEFINE_STATIC_KEY_FALSE(switch_mm_cond_i
 /* Control unconditional IBPB in switch_mm() */
 DEFINE_STATIC_KEY_FALSE(switch_mm_always_ibpb);
 
-/* Control MDS CPU buffer clear before returning to user space */
-DEFINE_STATIC_KEY_FALSE(mds_user_clear);
-EXPORT_SYMBOL_GPL(mds_user_clear);
 /* Control MDS CPU buffer clear before idling (halt, mwait) */
 DEFINE_STATIC_KEY_FALSE(mds_idle_clear);
 EXPORT_SYMBOL_GPL(mds_idle_clear);
@@ -251,7 +248,7 @@ static void __init mds_select_mitigation
 		if (!boot_cpu_has(X86_FEATURE_MD_CLEAR))
 			mds_mitigation = MDS_MITIGATION_VMWERV;
 
-		static_branch_enable(&mds_user_clear);
+		setup_force_cpu_cap(X86_FEATURE_CLEAR_CPU_BUF);
 
 		if (!boot_cpu_has(X86_BUG_MSBDS_ONLY) &&
 		    (mds_nosmt || cpu_mitigations_auto_nosmt()))
@@ -355,7 +352,7 @@ static void __init taa_select_mitigation
 	 * For guests that can't determine whether the correct microcode is
 	 * present on host, enable the mitigation for UCODE_NEEDED as well.
 	 */
-	static_branch_enable(&mds_user_clear);
+	setup_force_cpu_cap(X86_FEATURE_CLEAR_CPU_BUF);
 
 	if (taa_nosmt || cpu_mitigations_auto_nosmt())
 		cpu_smt_disable(false);
@@ -423,7 +420,7 @@ static void __init mmio_select_mitigatio
 	 */
 	if (boot_cpu_has_bug(X86_BUG_MDS) || (boot_cpu_has_bug(X86_BUG_TAA) &&
 					      boot_cpu_has(X86_FEATURE_RTM)))
-		static_branch_enable(&mds_user_clear);
+		setup_force_cpu_cap(X86_FEATURE_CLEAR_CPU_BUF);
 	else
 		static_branch_enable(&mmio_stale_data_clear);
 
@@ -483,12 +480,12 @@ static void __init md_clear_update_mitig
 	if (cpu_mitigations_off())
 		return;
 
-	if (!static_key_enabled(&mds_user_clear))
+	if (!boot_cpu_has(X86_FEATURE_CLEAR_CPU_BUF))
 		goto out;
 
 	/*
-	 * mds_user_clear is now enabled. Update MDS, TAA and MMIO Stale Data
-	 * mitigation, if necessary.
+	 * X86_FEATURE_CLEAR_CPU_BUF is now enabled. Update MDS, TAA and MMIO
+	 * Stale Data mitigation, if necessary.
 	 */
 	if (mds_mitigation == MDS_MITIGATION_OFF &&
 	    boot_cpu_has_bug(X86_BUG_MDS)) {
arch/x86/kernel/nmi.c +0 -3
@@ -522,9 +522,6 @@ nmi_restart:
 		write_cr2(this_cpu_read(nmi_cr2));
 	if (this_cpu_dec_return(nmi_state))
 		goto nmi_restart;
-
-	if (user_mode(regs))
-		mds_user_clear_cpu_buffers();
 }
 
 #if defined(CONFIG_X86_64) && IS_ENABLED(CONFIG_KVM_INTEL)
arch/x86/kvm/vmx/vmx.c +1 -1
@@ -7123,7 +7123,7 @@ static noinstr void vmx_vcpu_enter_exit(
 	/* L1D Flush includes CPU buffer clear to mitigate MDS */
 	if (static_branch_unlikely(&vmx_l1d_should_flush))
 		vmx_l1d_flush(vcpu);
-	else if (static_branch_unlikely(&mds_user_clear))
+	else if (cpu_feature_enabled(X86_FEATURE_CLEAR_CPU_BUF))
 		mds_clear_cpu_buffers();
 	else if (static_branch_unlikely(&mmio_stale_data_clear) &&
 		 kvm_arch_has_assigned_device(vcpu->kvm))
author maintainer
#214 lore raw

PATCH 210/215 KVM/VMX: Use BT+JNC, i.e. EFLAGS.CF to select VMRESUME vs. VMLAUNCH

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Pawan Gupta <[email protected]>

From: Sean Christopherson <[email protected]>

commit 706a189dcf74d3b3f955e9384785e726ed6c7c80 upstream.

Use EFLAGS.CF instead of EFLAGS.ZF to track whether to use VMRESUME versus VMLAUNCH. Freeing up EFLAGS.ZF will allow doing VERW, which clobbers ZF, for MDS mitigations as late as possible without needing to duplicate VERW for both paths.

[ pawan: resolved merge conflict in __vmx_vcpu_run in backport. ]

---
 arch/x86/kvm/vmx/run_flags.h |    7 +++++--
 arch/x86/kvm/vmx/vmenter.S   |    6 +++---
 2 files changed, 8 insertions(+), 5 deletions(-)
arch/x86/kvm/vmx/run_flags.h +5 -2
@@ -2,7 +2,10 @@
 #ifndef __KVM_X86_VMX_RUN_FLAGS_H
 #define __KVM_X86_VMX_RUN_FLAGS_H
 
-#define VMX_RUN_VMRESUME	(1 << 0)
-#define VMX_RUN_SAVE_SPEC_CTRL	(1 << 1)
+#define VMX_RUN_VMRESUME_SHIFT		0
+#define VMX_RUN_SAVE_SPEC_CTRL_SHIFT	1
+
+#define VMX_RUN_VMRESUME		BIT(VMX_RUN_VMRESUME_SHIFT)
+#define VMX_RUN_SAVE_SPEC_CTRL		BIT(VMX_RUN_SAVE_SPEC_CTRL_SHIFT)
 
 #endif /* __KVM_X86_VMX_RUN_FLAGS_H */
arch/x86/kvm/vmx/vmenter.S +3 -3
@@ -106,7 +106,7 @@ SYM_FUNC_START(__vmx_vcpu_run)
 	mov (%_ASM_SP), %_ASM_AX
 
 	/* Check if vmlaunch or vmresume is needed */
-	testb $VMX_RUN_VMRESUME, %bl
+	bt   $VMX_RUN_VMRESUME_SHIFT, %bx
 
 	/* Load guest registers.  Don't clobber flags. */
 	mov VCPU_RCX(%_ASM_AX), %_ASM_CX
@@ -128,8 +128,8 @@ SYM_FUNC_START(__vmx_vcpu_run)
 	/* Load guest RAX.  This kills the @regs pointer! */
 	mov VCPU_RAX(%_ASM_AX), %_ASM_AX
 
-	/* Check EFLAGS.ZF from 'testb' above */
-	jz .Lvmlaunch
+	/* Check EFLAGS.CF from the VMX_RUN_VMRESUME bit test above. */
+	jnc .Lvmlaunch
 
 	/*
 	 * After a successful VMRESUME/VMLAUNCH, control flow "magically"
author maintainer
#215 lore raw

PATCH 211/215 KVM/VMX: Move VERW closer to VMentry for MDS mitigation

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Pawan Gupta <[email protected]>

commit 43fb862de8f628c5db5e96831c915b9aebf62d33 upstream.

During VMentry VERW is executed to mitigate MDS. After VERW, any memory access like register push onto stack may put host data in MDS affected CPU buffers. A guest can then use MDS to sample host data.

Although likelihood of secrets surviving in registers at current VERW callsite is less, but it can't be ruled out. Harden the MDS mitigation by moving the VERW mitigation late in VMentry path.

Note that VERW for MMIO Stale Data mitigation is unchanged because of the complexity of per-guest conditional VERW which is not easy to handle that late in asm with no GPRs available. If the CPU is also affected by MDS, VERW is unconditionally executed late in asm regardless of guest having MMIO access.

[ pawan: conflict resolved in backport ]

---
 arch/x86/kvm/vmx/vmenter.S |    3 +++
 arch/x86/kvm/vmx/vmx.c     |   12 ++++++++----
 2 files changed, 11 insertions(+), 4 deletions(-)
arch/x86/kvm/vmx/vmenter.S +3 -0
@@ -128,6 +128,9 @@ SYM_FUNC_START(__vmx_vcpu_run)
 	/* Load guest RAX.  This kills the @regs pointer! */
 	mov VCPU_RAX(%_ASM_AX), %_ASM_AX
 
+	/* Clobbers EFLAGS.ZF */
+	CLEAR_CPU_BUFFERS
+
 	/* Check EFLAGS.CF from the VMX_RUN_VMRESUME bit test above. */
 	jnc .Lvmlaunch
 
arch/x86/kvm/vmx/vmx.c +8 -4
@@ -407,7 +407,8 @@ static __always_inline void vmx_enable_f
 
 static void vmx_update_fb_clear_dis(struct kvm_vcpu *vcpu, struct vcpu_vmx *vmx)
 {
-	vmx->disable_fb_clear = vmx_fb_clear_ctrl_available;
+	vmx->disable_fb_clear = !cpu_feature_enabled(X86_FEATURE_CLEAR_CPU_BUF) &&
+		vmx_fb_clear_ctrl_available;
 
 	/*
 	 * If guest will not execute VERW, there is no need to set FB_CLEAR_DIS
@@ -7120,11 +7121,14 @@ static noinstr void vmx_vcpu_enter_exit(
 {
 	guest_state_enter_irqoff();
 
-	/* L1D Flush includes CPU buffer clear to mitigate MDS */
+	/*
+	 * L1D Flush includes CPU buffer clear to mitigate MDS, but VERW
+	 * mitigation for MDS is done late in VMentry and is still
+	 * executed in spite of L1D Flush. This is because an extra VERW
+	 * should not matter much after the big hammer L1D Flush.
+	 */
 	if (static_branch_unlikely(&vmx_l1d_should_flush))
 		vmx_l1d_flush(vcpu);
-	else if (cpu_feature_enabled(X86_FEATURE_CLEAR_CPU_BUF))
-		mds_clear_cpu_buffers();
 	else if (static_branch_unlikely(&mmio_stale_data_clear) &&
 		 kvm_arch_has_assigned_device(vcpu->kvm))
 		mds_clear_cpu_buffers();
author maintainer
#216 lore raw

PATCH 212/215 bpf: Add table ID to bpf_fib_lookup BPF helper

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Louis DeLosSantos <[email protected]>

commit 8ad77e72caae22a1ddcfd0c03f2884929e93b7a4 upstream.

Add ability to specify routing table ID to the bpf_fib_lookup BPF helper.

A new field tbid is added to struct bpf_fib_lookup used as parameters to the bpf_fib_lookup BPF helper.

When the helper is called with the BPF_FIB_LOOKUP_DIRECT and BPF_FIB_LOOKUP_TBID flags the tbid field in struct bpf_fib_lookup will be used as the table ID for the fib lookup.

If the tbid does not exist the fib lookup will fail with BPF_FIB_LKUP_RET_NOT_FWDED.

The tbid field becomes a union over the vlan related output fields in struct bpf_fib_lookup and will be zeroed immediately after usage.

This functionality is useful in containerized environments.

For instance, if a CNI wants to dictate the next-hop for traffic leaving a container it can create a container-specific routing table and perform a fib lookup against this table in a "host-net-namespace-side" TC program.

This functionality also allows ip rule like functionality at the TC layer, allowing an eBPF program to pick a routing table based on some aspect of the sk_buff.

As a concrete use case, this feature will be used in Cilium's SRv6 L3VPN datapath.

When egress traffic leaves a Pod an eBPF program attached by Cilium will determine which VRF the egress traffic should target, and then perform a FIB lookup in a specific table representing this VRF's FIB.

---
 include/uapi/linux/bpf.h       |   21 ++++++++++++++++++---
 net/core/filter.c              |   14 +++++++++++++-
 tools/include/uapi/linux/bpf.h |   21 ++++++++++++++++++---
 3 files changed, 49 insertions(+), 7 deletions(-)
include/uapi/linux/bpf.h +18 -3
@@ -3109,6 +3109,10 @@ union bpf_attr {
  *		**BPF_FIB_LOOKUP_DIRECT**
  *			Do a direct table lookup vs full lookup using FIB
  *			rules.
+ *		**BPF_FIB_LOOKUP_TBID**
+ *			Used with BPF_FIB_LOOKUP_DIRECT.
+ *			Use the routing table ID present in *params*->tbid
+ *			for the fib lookup.
  *		**BPF_FIB_LOOKUP_OUTPUT**
  *			Perform lookup from an egress perspective (default is
  *			ingress).
@@ -6687,6 +6691,7 @@ enum {
 	BPF_FIB_LOOKUP_DIRECT  = (1U << 0),
 	BPF_FIB_LOOKUP_OUTPUT  = (1U << 1),
 	BPF_FIB_LOOKUP_SKIP_NEIGH = (1U << 2),
+	BPF_FIB_LOOKUP_TBID    = (1U << 3),
 };
 
 enum {
@@ -6747,9 +6752,19 @@ struct bpf_fib_lookup {
 		__u32		ipv6_dst[4];  /* in6_addr; network order */
 	};
 
-	/* output */
-	__be16	h_vlan_proto;
-	__be16	h_vlan_TCI;
+	union {
+		struct {
+			/* output */
+			__be16	h_vlan_proto;
+			__be16	h_vlan_TCI;
+		};
+		/* input: when accompanied with the
+		 * 'BPF_FIB_LOOKUP_DIRECT | BPF_FIB_LOOKUP_TBID` flags, a
+		 * specific routing table to use for the fib lookup.
+		 */
+		__u32	tbid;
+	};
+
 	__u8	smac[6];     /* ETH_ALEN */
 	__u8	dmac[6];     /* ETH_ALEN */
 };
net/core/filter.c +13 -1
@@ -5752,6 +5752,12 @@ static int bpf_ipv4_fib_lookup(struct ne
 		u32 tbid = l3mdev_fib_table_rcu(dev) ? : RT_TABLE_MAIN;
 		struct fib_table *tb;
 
+		if (flags & BPF_FIB_LOOKUP_TBID) {
+			tbid = params->tbid;
+			/* zero out for vlan output */
+			params->tbid = 0;
+		}
+
 		tb = fib_get_table(net, tbid);
 		if (unlikely(!tb))
 			return BPF_FIB_LKUP_RET_NOT_FWDED;
@@ -5885,6 +5891,12 @@ static int bpf_ipv6_fib_lookup(struct ne
 		u32 tbid = l3mdev_fib_table_rcu(dev) ? : RT_TABLE_MAIN;
 		struct fib6_table *tb;
 
+		if (flags & BPF_FIB_LOOKUP_TBID) {
+			tbid = params->tbid;
+			/* zero out for vlan output */
+			params->tbid = 0;
+		}
+
 		tb = ipv6_stub->fib6_get_table(net, tbid);
 		if (unlikely(!tb))
 			return BPF_FIB_LKUP_RET_NOT_FWDED;
@@ -5957,7 +5969,7 @@ set_fwd_params:
 #endif
 
 #define BPF_FIB_LOOKUP_MASK (BPF_FIB_LOOKUP_DIRECT | BPF_FIB_LOOKUP_OUTPUT | \
-			     BPF_FIB_LOOKUP_SKIP_NEIGH)
+			     BPF_FIB_LOOKUP_SKIP_NEIGH | BPF_FIB_LOOKUP_TBID)
 
 BPF_CALL_4(bpf_xdp_fib_lookup, struct xdp_buff *, ctx,
 	   struct bpf_fib_lookup *, params, int, plen, u32, flags)
tools/include/uapi/linux/bpf.h +18 -3
@@ -3109,6 +3109,10 @@ union bpf_attr {
  *		**BPF_FIB_LOOKUP_DIRECT**
  *			Do a direct table lookup vs full lookup using FIB
  *			rules.
+ *		**BPF_FIB_LOOKUP_TBID**
+ *			Used with BPF_FIB_LOOKUP_DIRECT.
+ *			Use the routing table ID present in *params*->tbid
+ *			for the fib lookup.
  *		**BPF_FIB_LOOKUP_OUTPUT**
  *			Perform lookup from an egress perspective (default is
  *			ingress).
@@ -6687,6 +6691,7 @@ enum {
 	BPF_FIB_LOOKUP_DIRECT  = (1U << 0),
 	BPF_FIB_LOOKUP_OUTPUT  = (1U << 1),
 	BPF_FIB_LOOKUP_SKIP_NEIGH = (1U << 2),
+	BPF_FIB_LOOKUP_TBID    = (1U << 3),
 };
 
 enum {
@@ -6747,9 +6752,19 @@ struct bpf_fib_lookup {
 		__u32		ipv6_dst[4];  /* in6_addr; network order */
 	};
 
-	/* output */
-	__be16	h_vlan_proto;
-	__be16	h_vlan_TCI;
+	union {
+		struct {
+			/* output */
+			__be16	h_vlan_proto;
+			__be16	h_vlan_TCI;
+		};
+		/* input: when accompanied with the
+		 * 'BPF_FIB_LOOKUP_DIRECT | BPF_FIB_LOOKUP_TBID` flags, a
+		 * specific routing table to use for the fib lookup.
+		 */
+		__u32	tbid;
+	};
+
 	__u8	smac[6];     /* ETH_ALEN */
 	__u8	dmac[6];     /* ETH_ALEN */
 };
author maintainer
#217 lore raw

PATCH 213/215 bpf: Derive source IP addr via bpf_*_fib_lookup()

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Martynas Pumputis <[email protected]>

commit dab4e1f06cabb6834de14264394ccab197007302 upstream.

Extend the bpf_fib_lookup() helper by making it to return the source IPv4/IPv6 address if the BPF_FIB_LOOKUP_SRC flag is set.

For example, the following snippet can be used to derive the desired source IP address:

    struct bpf_fib_lookup p = { .ipv4_dst = ip4->daddr };

    ret = bpf_skb_fib_lookup(skb, p, sizeof(p),
            BPF_FIB_LOOKUP_SRC | BPF_FIB_LOOKUP_SKIP_NEIGH);
    if (ret != BPF_FIB_LKUP_RET_SUCCESS)
        return TC_ACT_SHOT;

    /* the p.ipv4_src now contains the source address */

The inability to derive the proper source address may cause malfunctions in BPF-based dataplanes for hosts containing netdevs with more than one routable IP address or for multi-homed hosts.

For example, Cilium implements packet masquerading in BPF. If an egressing netdev to which the Cilium's BPF prog is attached has multiple IP addresses, then only one [hardcoded] IP address can be used for masquerading. This breaks connectivity if any other IP address should have been selected instead, for example, when a public and private addresses are attached to the same egress interface.

The change was tested with Cilium [1].

Nikolay Aleksandrov helped to figure out the IPv6 addr selection.

[1]: https://github.com/cilium/cilium/pull/28283

---
 include/net/ipv6_stubs.h       |    5 +++++
 include/uapi/linux/bpf.h       |   10 ++++++++++
 net/core/filter.c              |   18 +++++++++++++++++-
 net/ipv6/af_inet6.c            |    1 +
 tools/include/uapi/linux/bpf.h |   10 ++++++++++
 5 files changed, 43 insertions(+), 1 deletion(-)
include/net/ipv6_stubs.h +5 -0
@@ -85,6 +85,11 @@ struct ipv6_bpf_stub {
 			       sockptr_t optval, unsigned int optlen);
 	int (*ipv6_getsockopt)(struct sock *sk, int level, int optname,
 			       sockptr_t optval, sockptr_t optlen);
+	int (*ipv6_dev_get_saddr)(struct net *net,
+				  const struct net_device *dst_dev,
+				  const struct in6_addr *daddr,
+				  unsigned int prefs,
+				  struct in6_addr *saddr);
 };
 extern const struct ipv6_bpf_stub *ipv6_bpf_stub __read_mostly;
 
include/uapi/linux/bpf.h +10 -0
@@ -3121,6 +3121,11 @@ union bpf_attr {
  *			and *params*->smac will not be set as output. A common
  *			use case is to call **bpf_redirect_neigh**\ () after
  *			doing **bpf_fib_lookup**\ ().
+ *		**BPF_FIB_LOOKUP_SRC**
+ *			Derive and set source IP addr in *params*->ipv{4,6}_src
+ *			for the nexthop. If the src addr cannot be derived,
+ *			**BPF_FIB_LKUP_RET_NO_SRC_ADDR** is returned. In this
+ *			case, *params*->dmac and *params*->smac are not set either.
  *
  *		*ctx* is either **struct xdp_md** for XDP programs or
  *		**struct sk_buff** tc cls_act programs.
@@ -6692,6 +6697,7 @@ enum {
 	BPF_FIB_LOOKUP_OUTPUT  = (1U << 1),
 	BPF_FIB_LOOKUP_SKIP_NEIGH = (1U << 2),
 	BPF_FIB_LOOKUP_TBID    = (1U << 3),
+	BPF_FIB_LOOKUP_SRC     = (1U << 4),
 };
 
 enum {
@@ -6704,6 +6710,7 @@ enum {
 	BPF_FIB_LKUP_RET_UNSUPP_LWT,   /* fwd requires encapsulation */
 	BPF_FIB_LKUP_RET_NO_NEIGH,     /* no neighbor entry for nh */
 	BPF_FIB_LKUP_RET_FRAG_NEEDED,  /* fragmentation required to fwd */
+	BPF_FIB_LKUP_RET_NO_SRC_ADDR,  /* failed to derive IP src addr */
 };
 
 struct bpf_fib_lookup {
@@ -6738,6 +6745,9 @@ struct bpf_fib_lookup {
 		__u32	rt_metric;
 	};
 
+	/* input: source address to consider for lookup
+	 * output: source address result from lookup
+	 */
 	union {
 		__be32		ipv4_src;
 		__u32		ipv6_src[4];  /* in6_addr; network order */
net/core/filter.c +17 -1
@@ -5809,6 +5809,9 @@ static int bpf_ipv4_fib_lookup(struct ne
 	params->rt_metric = res.fi->fib_priority;
 	params->ifindex = dev->ifindex;
 
+	if (flags & BPF_FIB_LOOKUP_SRC)
+		params->ipv4_src = fib_result_prefsrc(net, &res);
+
 	/* xdp and cls_bpf programs are run in RCU-bh so
 	 * rcu_read_lock_bh is not needed here
 	 */
@@ -5951,6 +5954,18 @@ static int bpf_ipv6_fib_lookup(struct ne
 	params->rt_metric = res.f6i->fib6_metric;
 	params->ifindex = dev->ifindex;
 
+	if (flags & BPF_FIB_LOOKUP_SRC) {
+		if (res.f6i->fib6_prefsrc.plen) {
+			*src = res.f6i->fib6_prefsrc.addr;
+		} else {
+			err = ipv6_bpf_stub->ipv6_dev_get_saddr(net, dev,
+								&fl6.daddr, 0,
+								src);
+			if (err)
+				return BPF_FIB_LKUP_RET_NO_SRC_ADDR;
+		}
+	}
+
 	if (flags & BPF_FIB_LOOKUP_SKIP_NEIGH)
 		goto set_fwd_params;
 
@@ -5969,7 +5984,8 @@ set_fwd_params:
 #endif
 
 #define BPF_FIB_LOOKUP_MASK (BPF_FIB_LOOKUP_DIRECT | BPF_FIB_LOOKUP_OUTPUT | \
-			     BPF_FIB_LOOKUP_SKIP_NEIGH | BPF_FIB_LOOKUP_TBID)
+			     BPF_FIB_LOOKUP_SKIP_NEIGH | BPF_FIB_LOOKUP_TBID | \
+			     BPF_FIB_LOOKUP_SRC)
 
 BPF_CALL_4(bpf_xdp_fib_lookup, struct xdp_buff *, ctx,
 	   struct bpf_fib_lookup *, params, int, plen, u32, flags)
net/ipv6/af_inet6.c +1 -0
@@ -1077,6 +1077,7 @@ static const struct ipv6_bpf_stub ipv6_b
 	.udp6_lib_lookup = __udp6_lib_lookup,
 	.ipv6_setsockopt = do_ipv6_setsockopt,
 	.ipv6_getsockopt = do_ipv6_getsockopt,
+	.ipv6_dev_get_saddr = ipv6_dev_get_saddr,
 };
 
 static int __init inet6_init(void)
tools/include/uapi/linux/bpf.h +10 -0
@@ -3121,6 +3121,11 @@ union bpf_attr {
  *			and *params*->smac will not be set as output. A common
  *			use case is to call **bpf_redirect_neigh**\ () after
  *			doing **bpf_fib_lookup**\ ().
+ *		**BPF_FIB_LOOKUP_SRC**
+ *			Derive and set source IP addr in *params*->ipv{4,6}_src
+ *			for the nexthop. If the src addr cannot be derived,
+ *			**BPF_FIB_LKUP_RET_NO_SRC_ADDR** is returned. In this
+ *			case, *params*->dmac and *params*->smac are not set either.
  *
  *		*ctx* is either **struct xdp_md** for XDP programs or
  *		**struct sk_buff** tc cls_act programs.
@@ -6692,6 +6697,7 @@ enum {
 	BPF_FIB_LOOKUP_OUTPUT  = (1U << 1),
 	BPF_FIB_LOOKUP_SKIP_NEIGH = (1U << 2),
 	BPF_FIB_LOOKUP_TBID    = (1U << 3),
+	BPF_FIB_LOOKUP_SRC     = (1U << 4),
 };
 
 enum {
@@ -6704,6 +6710,7 @@ enum {
 	BPF_FIB_LKUP_RET_UNSUPP_LWT,   /* fwd requires encapsulation */
 	BPF_FIB_LKUP_RET_NO_NEIGH,     /* no neighbor entry for nh */
 	BPF_FIB_LKUP_RET_FRAG_NEEDED,  /* fragmentation required to fwd */
+	BPF_FIB_LKUP_RET_NO_SRC_ADDR,  /* failed to derive IP src addr */
 };
 
 struct bpf_fib_lookup {
@@ -6738,6 +6745,9 @@ struct bpf_fib_lookup {
 		__u32	rt_metric;
 	};
 
+	/* input: source address to consider for lookup
+	 * output: source address result from lookup
+	 */
 	union {
 		__be32		ipv4_src;
 		__u32		ipv6_src[4];  /* in6_addr; network order */
author maintainer
#218 lore raw

PATCH 214/215 x86/efistub: Give up if memory attribute protocol returns an error

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ard Biesheuvel <[email protected]>

commit a7a6a01f88e87dec4bf2365571dd2dc7403d52d0 upstream.

The recently introduced EFI memory attributes protocol should be used if it exists to ensure that the memory allocation created for the kernel permits execution. This is needed for compatibility with tightened requirements related to Windows logo certification for x86 PCs.

Currently, we simply strip the execute protect (XP) attribute from the entire range, but this might be rejected under some firmware security policies, and so in a subsequent patch, this will be changed to only strip XP from the executable region that runs early, and make it read-only (RO) as well.

In order to catch any issues early, ensure that the memory attribute protocol works as intended, and give up if it produces spurious errors.

Note that the DXE services based fallback was always based on best effort, so don't propagate any errors returned by that API.

Fixes a1b87d54f4e4 ("x86/efistub: Avoid legacy decompressor when doing EFI boot")
---
 drivers/firmware/efi/libstub/x86-stub.c |   24 ++++++++++++++----------
 drivers/firmware/efi/libstub/x86-stub.h |    4 ++--
 2 files changed, 16 insertions(+), 12 deletions(-)
drivers/firmware/efi/libstub/x86-stub.c +14 -10
@@ -212,8 +212,8 @@ static void retrieve_apple_device_proper
 	}
 }
 
-void efi_adjust_memory_range_protection(unsigned long start,
-					unsigned long size)
+efi_status_t efi_adjust_memory_range_protection(unsigned long start,
+						unsigned long size)
 {
 	efi_status_t status;
 	efi_gcd_memory_space_desc_t desc;
@@ -225,13 +225,17 @@ void efi_adjust_memory_range_protection(
 	rounded_end = roundup(start + size, EFI_PAGE_SIZE);
 
 	if (memattr != NULL) {
-		efi_call_proto(memattr, clear_memory_attributes, rounded_start,
-			       rounded_end - rounded_start, EFI_MEMORY_XP);
-		return;
+		status = efi_call_proto(memattr, clear_memory_attributes,
+					rounded_start,
+					rounded_end - rounded_start,
+					EFI_MEMORY_XP);
+		if (status != EFI_SUCCESS)
+			efi_warn("Failed to clear EFI_MEMORY_XP attribute\n");
+		return status;
 	}
 
 	if (efi_dxe_table == NULL)
-		return;
+		return EFI_SUCCESS;
 
 	/*
 	 * Don't modify memory region attributes, they are
@@ -244,7 +248,7 @@ void efi_adjust_memory_range_protection(
 		status = efi_dxe_call(get_memory_space_descriptor, start, &desc);
 
 		if (status != EFI_SUCCESS)
-			return;
+			break;
 
 		next = desc.base_address + desc.length;
 
@@ -269,8 +273,10 @@ void efi_adjust_memory_range_protection(
 				 unprotect_start,
 				 unprotect_start + unprotect_size,
 				 status);
+			break;
 		}
 	}
+	return EFI_SUCCESS;
 }
 
 static efi_char16_t *efistub_fw_vendor(void)
@@ -800,9 +806,7 @@ static efi_status_t efi_decompress_kerne
 
 	*kernel_entry = addr + entry;
 
-	efi_adjust_memory_range_protection(addr, kernel_total_size);
-
-	return EFI_SUCCESS;
+	return efi_adjust_memory_range_protection(addr, kernel_total_size);
 }
 
 static void __noreturn enter_kernel(unsigned long kernel_addr,
drivers/firmware/efi/libstub/x86-stub.h +2 -2
@@ -5,8 +5,8 @@
 extern void trampoline_32bit_src(void *, bool);
 extern const u16 trampoline_ljmp_imm_offset;
 
-void efi_adjust_memory_range_protection(unsigned long start,
-					unsigned long size);
+efi_status_t efi_adjust_memory_range_protection(unsigned long start,
+						unsigned long size);
 
 #ifdef CONFIG_X86_64
 efi_status_t efi_setup_5level_paging(void);
author maintainer
#219 lore raw

PATCH 215/215 xen/events: close evtchn after mapping cleanup

In reply to Greg Kroah-Hartman

6.1-stable review patch. If anyone has any objections, please let me know.

------------------

From: Maximilian Heyne <[email protected]>

commit fa765c4b4aed2d64266b694520ecb025c862c5a9 upstream.

shutdown_pirq and startup_pirq are not taking the irq_mapping_update_lock because they can't due to lock inversion. Both are called with the irq_desc->lock being taking. The lock order, however, is first irq_mapping_update_lock and then irq_desc->lock.

This opens multiple races:
- shutdown_pirq can be interrupted by a function that allocates an event channel:

  CPU0                        CPU1
  shutdown_pirq {
    xen_evtchn_close(e)
                              __startup_pirq {
                                EVTCHNOP_bind_pirq
                                  -> returns just freed evtchn e
                                set_evtchn_to_irq(e, irq)
                              }
    xen_irq_info_cleanup() {
      set_evtchn_to_irq(e, -1)
    }
  }

Assume here event channel e refers here to the same event channel number.
After this race the evtchn_to_irq mapping for e is invalid (-1).

- __startup_pirq races with __unbind_from_irq in a similar way. Because __startup_pirq doesn't take irq_mapping_update_lock it can grab the evtchn that __unbind_from_irq is currently freeing and cleaning up. In this case even though the event channel is allocated, its mapping can be unset in evtchn_to_irq.

The fix is to first cleanup the mappings and then close the event channel. In this way, when an event channel gets allocated it's potential previous evtchn_to_irq mappings are guaranteed to be unset already. This is also the reverse order of the allocation where first the event channel is allocated and then the mappings are setup.

On a 5.10 kernel prior to commit 3fcdaf3d7634 ("xen/events: modify internal [un]bind interfaces"), we hit a BUG like the following during probing of NVMe devices. The issue is that during nvme_setup_io_queues, pci_free_irq is called for every device which results in a call to shutdown_pirq. With many nvme devices it's therefore likely to hit this race during boot because there will be multiple calls to shutdown_pirq and startup_pirq are running potentially in parallel.

  ------------[ cut here ]------------
  blkfront: xvda: barrier or flush: disabled; persistent grants: enabled; indirect descriptors: enabled; bounce buffer: enabled
  kernel BUG at drivers/xen/events/events_base.c:499!
  invalid opcode: 0000 [#1] SMP PTI
  CPU: 44 PID: 375 Comm: kworker/u257:23 Not tainted 5.10.201-191.748.amzn2.x86_64 #1
  Hardware name: Xen HVM domU, BIOS 4.11.amazon 08/24/2006
  Workqueue: nvme-reset-wq nvme_reset_work
  RIP: 0010:bind_evtchn_to_cpu+0xdf/0xf0
  Code: 5d 41 5e c3 cc cc cc cc 44 89 f7 e8 2b 55 ad ff 49 89 c5 48 85 c0 0f 84 64 ff ff ff 4c 8b 68 30 41 83 fe ff 0f 85 60 ff ff ff <0f> 0b 66 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 00 0f 1f 44 00 00
  RSP: 0000:ffffc9000d533b08 EFLAGS: 00010046
  RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000006
  RDX: 0000000000000028 RSI: 00000000ffffffff RDI: 00000000ffffffff
  RBP: ffff888107419680 R08: 0000000000000000 R09: ffffffff82d72b00
  R10: 0000000000000000 R11: 0000000000000000 R12: 00000000000001ed
  R13: 0000000000000000 R14: 00000000ffffffff R15: 0000000000000002
  FS:  0000000000000000(0000) GS:ffff88bc8b500000(0000) knlGS:0000000000000000
  CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
  CR2: 0000000000000000 CR3: 0000000002610001 CR4: 00000000001706e0
  DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
  DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
  Call Trace:
   ? show_trace_log_lvl+0x1c1/0x2d9
   ? show_trace_log_lvl+0x1c1/0x2d9
   ? set_affinity_irq+0xdc/0x1c0
   ? __die_body.cold+0x8/0xd
   ? die+0x2b/0x50
   ? do_trap+0x90/0x110
   ? bind_evtchn_to_cpu+0xdf/0xf0
   ? do_error_trap+0x65/0x80
   ? bind_evtchn_to_cpu+0xdf/0xf0
   ? exc_invalid_op+0x4e/0x70
   ? bind_evtchn_to_cpu+0xdf/0xf0
   ? asm_exc_invalid_op+0x12/0x20
   ? bind_evtchn_to_cpu+0xdf/0xf0
   ? bind_evtchn_to_cpu+0xc5/0xf0
   set_affinity_irq+0xdc/0x1c0
   irq_do_set_affinity+0x1d7/0x1f0
   irq_setup_affinity+0xd6/0x1a0
   irq_startup+0x8a/0xf0
   __setup_irq+0x639/0x6d0
   ? nvme_suspend+0x150/0x150
   request_threaded_irq+0x10c/0x180
   ? nvme_suspend+0x150/0x150
   pci_request_irq+0xa8/0xf0
   ? __blk_mq_free_request+0x74/0xa0
   queue_request_irq+0x6f/0x80
   nvme_create_queue+0x1af/0x200
   nvme_create_io_queues+0xbd/0xf0
   nvme_setup_io_queues+0x246/0x320
   ? nvme_irq_check+0x30/0x30
   nvme_reset_work+0x1c8/0x400
   process_one_work+0x1b0/0x350
   worker_thread+0x49/0x310
   ? process_one_work+0x350/0x350
   kthread+0x11b/0x140
   ? __kthread_bind_mask+0x60/0x60
   ret_from_fork+0x22/0x30
  Modules linked in:
  ---[ end trace a11715de1eee1873 ]---
---
 drivers/xen/events/events_base.c |    5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)
drivers/xen/events/events_base.c +2 -3
@@ -937,8 +937,8 @@ static void shutdown_pirq(struct irq_dat
 		return;
 
 	do_mask(info, EVT_MASK_REASON_EXPLICIT);
-	xen_evtchn_close(evtchn);
 	xen_irq_info_cleanup(info);
+	xen_evtchn_close(evtchn);
 }
 
 static void enable_pirq(struct irq_data *data)
@@ -982,8 +982,6 @@ static void __unbind_from_irq(unsigned i
 		unsigned int cpu = cpu_from_irq(irq);
 		struct xenbus_device *dev;
 
-		xen_evtchn_close(evtchn);
-
 		switch (type_from_irq(irq)) {
 		case IRQT_VIRQ:
 			per_cpu(virq_to_irq, cpu)[virq_from_irq(irq)] = -1;
@@ -1001,6 +999,7 @@ static void __unbind_from_irq(unsigned i
 		}
 
 		xen_irq_info_cleanup(info);
+		xen_evtchn_close(evtchn);
 	}
 
 	xen_free_irq(irq);
maintainer Tested-by on the v1 series
#220 lore raw
In reply to Greg Kroah-Hartman

Hello,

Quoted 5 lines · On Mon, 4 Mar 2024 21:21:03 +0000 Greg Kroah-Hartman…

This is the start of the stable review cycle for the 6.1.81 release. There are 215 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Quoted 8 lines

Responses should be made by Wed, 06 Mar 2024 21:15:26 +0000. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

	https://www.kernel.org/pub/linux/kernel/v6.x/stable-review/patch-6.1.81-rc1.gz

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-6.1.y
and the diffstat can be found below.

This rc kernel passes DAMON functionality test[1] on my test machine. Attaching the test results summary below. Please note that I retrieved the kernel from linux-stable-rc tree[2].

[1] https://github.com/awslabs/damon-tests/tree/next/corr
[2] cf578ac947cb ("Linux 6.1.81-rc1")

Thanks,
SJ

[...]

---

ok 1 selftests: damon: debugfs_attrs.sh
ok 2 selftests: damon: debugfs_schemes.sh
ok 3 selftests: damon: debugfs_target_ids.sh
ok 4 selftests: damon: debugfs_empty_targets.sh
ok 5 selftests: damon: debugfs_huge_count_read_write.sh
ok 6 selftests: damon: debugfs_duplicate_context_creation.sh
ok 7 selftests: damon: sysfs.sh
ok 1 selftests: damon-tests: kunit.sh
ok 2 selftests: damon-tests: huge_count_read_write.sh
ok 3 selftests: damon-tests: buffer_overflow.sh
ok 4 selftests: damon-tests: rm_contexts.sh
ok 5 selftests: damon-tests: record_null_deref.sh
ok 6 selftests: damon-tests: dbgfs_target_ids_read_before_terminate_race.sh ok 7 selftests: damon-tests: dbgfs_target_ids_pid_leak.sh
ok 8 selftests: damon-tests: damo_tests.sh
ok 9 selftests: damon-tests: masim-record.sh
ok 10 selftests: damon-tests: build_i386.sh
ok 11 selftests: damon-tests: build_arm64.sh
ok 12 selftests: damon-tests: build_m68k.sh
ok 13 selftests: damon-tests: build_i386_idle_flag.sh
ok 14 selftests: damon-tests: build_i386_highpte.sh
ok 15 selftests: damon-tests: build_nomemcg.sh
[33m
[92mPASS [39m

Tested-by on the v1 series
#221 lore raw
In reply to Greg Kroah-Hartman
Quoted 9 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 6.1.81 release. There are 215 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Wed, 06 Mar 2024 21:15:26 +0000. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

Quoted 8 lines
	https://www.kernel.org/pub/linux/kernel/v6.x/stable-review/patch-6.1.81-rc1.gz

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-6.1.y
and the diffstat can be found below.

thanks,

greg k-h

Built and booted successfully on RISC-V RV64 (HiFive Unmatched).

maintainer Tested-by on the v1 series
#222 lore raw
In reply to Greg Kroah-Hartman
Quoted 9 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 6.1.81 release. There are 215 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Wed, 06 Mar 2024 21:15:26 +0000. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

Quoted 8 lines
	https://www.kernel.org/pub/linux/kernel/v6.x/stable-review/patch-6.1.81-rc1.gz

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-6.1.y
and the diffstat can be found below.

thanks,

greg k-h

All tests passing for Tegra ...

Test results for stable-v6.1:

    10 builds:	10 pass, 0 fail
    26 boots:	26 pass, 0 fail
    116 tests:	116 pass, 0 fail

Linux version: 6.1.81-rc1-gcf578ac947cb
Boards tested: tegra124-jetson-tk1, tegra186-p2771-0000,

                tegra194-p2972-0000, tegra194-p3509-0000+p3668-0000,
                tegra20-ventana, tegra210-p2371-2180,
                tegra210-p3450-0000, tegra30-cardhu-a04

Jon

Tested-by on the v1 series
#223 lore raw
In reply to Greg Kroah-Hartman

Hi!

Quoted 4 lines

This is the start of the stable review cycle for the 6.1.81 release. There are 215 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

CIP testing did not find any problems here:

https://gitlab.com/cip-project/cip-testing/linux-stable-rc-ci/-/tree/linux-6.1.y

Best regards,
Pavel

Signature · 2 lines
DENX Software Engineering GmbH,        Managing Director: Erika Unter
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Tested-by on the v1 series
#224 lore raw
In reply to Greg Kroah-Hartman
Quoted 9 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 6.1.81 release. There are 215 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Wed, 06 Mar 2024 21:15:26 +0000. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

Quoted 9 lines
	https://www.kernel.org/pub/linux/kernel/v6.x/stable-review/patch-6.1.81-rc1.gz

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-6.1.y
and the diffstat can be found below.

thanks,

greg k-h

Compiled and booted on my test system. No dmesg regressions.

thanks,
-- Shuah

Tested-by on the v1 series
#225 lore raw
In reply to Greg Kroah-Hartman

W dniu 4.03.2024 o 22:21, Greg Kroah-Hartman pisze:

Quoted 9 lines

This is the start of the stable review cycle for the 6.1.81 release. There are 215 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Wed, 06 Mar 2024 21:15:26 +0000. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

Quoted 9 lines
	https://www.kernel.org/pub/linux/kernel/v6.x/stable-review/patch-6.1.81-rc1.gz

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-6.1.y
and the diffstat can be found below.

thanks,

greg k-h

Hello,

Tested on a HP 17-by0001nw laptop with an Intel Kaby Lake CPU and Ubuntu 20.04.

Stack:

- amd64,
- ext4 on top of LVM on top of LUKS on top of mdraid on top of NVMe and SATA drives (the SATA drive in a write-mostly mode).

Tested (lightly):
- suspend to RAM,
- suspend to disk,

- GPU (Intel HD Graphics 620, with a Duke Nukem 3D game on Proton (Wine fork) and a Unigine benchmark)
- WiFi (Realtek RTL8822BE),
- Bluetooth (Realtek RTL8822BE),
- PCI soundcard (Intel HD Audio),
- USB soundcard (Logitech Pro X),
    - on 6.1.80-rc1 (did not check other kernels) I had problems with microphone input after suspend-to-RAM, does not happen at the moment,
- webcam,

- filesystems tested very lightly:
    - NFS,
    - NTFS via FUSE
    - exFAT
    - vfat,

Issues found:
    - nothing new

Greetings,

Mateusz

Tested-by on the v1 series
#226 lore raw
In reply to Greg Kroah-Hartman
Quoted 9 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 6.1.81 release. There are 215 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Wed, 06 Mar 2024 21:15:26 +0000. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

Quoted 8 lines
	https://www.kernel.org/pub/linux/kernel/v6.x/stable-review/patch-6.1.81-rc1.gz

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-6.1.y
and the diffstat can be found below.

thanks,

greg k-h

On ARCH_BRCMSTB using 32-bit and 64-bit ARM kernels, build tested on BMIPS_GENERIC:

Signature · 1 line
Florian
Tested-by Linux Kernel Functional Testing on the v1 series
#227 lore raw
In reply to Greg Kroah-Hartman
Quoted 10 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 6.1.81 release. There are 215 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Wed, 06 Mar 2024 21:15:26 +0000. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

Quoted 8 lines
        https://www.kernel.org/pub/linux/kernel/v6.x/stable-review/patch-6.1.81-rc1.gz

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-6.1.y
and the diffstat can be found below.

thanks,

greg k-h

Results from Linaro’s test farm.
No regressions on arm64, arm, x86_64, and i386.

## Build
* kernel: 6.1.81-rc1
* git: https://gitlab.com/Linaro/lkft/mirrors/stable/linux-stable-rc
* git branch: linux-6.1.y
* git commit: cf578ac947cbb5bb72cbaa7a166d0f92318c3126
* git describe: v6.1.79-411-gcf578ac947cb
* test details:
https://qa-reports.linaro.org/lkft/linux-stable-rc-linux-6.1.y/build/v6.1.79-411-gcf578ac947cb

## Test Regressions (compared to v6.1.79)

## Metric Regressions (compared to v6.1.79)

## Test Fixes (compared to v6.1.79)

## Metric Fixes (compared to v6.1.79)

## Test result summary
total: 142733, pass: 122117, fail: 2309, skip: 18166, xfail: 141

## Build Summary
* arc: 5 total, 5 passed, 0 failed
* arm: 139 total, 139 passed, 0 failed
* arm64: 41 total, 41 passed, 0 failed
* i386: 31 total, 31 passed, 0 failed
* mips: 26 total, 26 passed, 0 failed
* parisc: 4 total, 4 passed, 0 failed
* powerpc: 36 total, 34 passed, 2 failed
* riscv: 11 total, 11 passed, 0 failed
* s390: 16 total, 16 passed, 0 failed
* sh: 10 total, 10 passed, 0 failed
* sparc: 8 total, 8 passed, 0 failed
* x86_64: 37 total, 37 passed, 0 failed

## Test suites summary
* boot
* kselftest-android
* kselftest-arm64
* kselftest-breakpoints
* kselftest-capabilities
* kselftest-cgroup
* kselftest-clone3
* kselftest-core
* kselftest-cpu-hotplug
* kselftest-cpufreq
* kselftest-drivers-dma-buf
* kselftest-efivarfs
* kselftest-exec
* kselftest-filesystems
* kselftest-filesystems-binderfs
* kselftest-filesystems-epoll
* kselftest-firmware
* kselftest-fpu
* kselftest-ftrace
* kselftest-futex
* kselftest-gpio
* kselftest-intel_pstate
* kselftest-ipc
* kselftest-ir
* kselftest-kcmp
* kselftest-kexec
* kselftest-kvm
* kselftest-lib
* kselftest-membarrier
* kselftest-memfd
* kselftest-memory-hotplug
* kselftest-mincore
* kselftest-mm
* kselftest-mount
* kselftest-mqueue
* kselftest-net
* kselftest-net-forwarding
* kselftest-net-mptcp
* kselftest-netfilter
* kselftest-nsfs
* kselftest-openat2
* kselftest-pid_namespace
* kselftest-pidfd
* kselftest-proc
* kselftest-pstore
* kselftest-ptrace
* kselftest-rseq
* kselftest-rtc
* kselftest-seccomp
* kselftest-sigaltstack
* kselftest-size
* kselftest-splice
* kselftest-static_keys
* kselftest-sync
* kselftest-sysctl
* kselftest-tc-testing
* kselftest-timens
* kselftest-timers
* kselftest-tmpfs
* kselftest-tpm2
* kselftest-user
* kselftest-user_events
* kselftest-vDSO
* kselftest-watchdog
* kselftest-x86
* kselftest-zram
* kunit
* kvm-unit-tests
* libgpiod
* log-parser-boot
* log-parser-test
* ltp-cap_bounds
* ltp-commands
* ltp-containers
* ltp-controllers
* ltp-cpuhotplug
* ltp-crypto
* ltp-cve
* ltp-dio
* ltp-fcntl-locktests
* ltp-filecaps
* ltp-fs
* ltp-fs_bind
* ltp-fs_perms_simple
* ltp-hugetlb
* ltp-io
* ltp-ipc
* ltp-math
* ltp-mm
* ltp-nptl
* ltp-pty
* ltp-sched
* ltp-securebits
* ltp-smoke
* ltp-smoketest
* ltp-syscalls
* ltp-tracing
* perf
* rcutorture

--
Linaro LKFT
https://lkft.linaro.org

Tested-by on the v1 series
#228 lore raw
In reply to Greg Kroah-Hartman

Hi Greg,

Quoted 8 lines · Greg Kroah-Hartman

This is the start of the stable review cycle for the 6.1.81 release. There are 215 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Wed, 06 Mar 2024 21:15:26 +0000. Anything received after that time might be too late.

Quoted 7 lines

The whole patch series can be found in one patch at:

	https://www.kernel.org/pub/linux/kernel/v6.x/stable-review/patch-6.1.81-rc1.gz

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-6.1.y
and the diffstat can be found below.

thanks,

I tested 6.1.81-rc1 (cf578ac947cb) on Kalray kvx arch (not upstream yet) and everything looks good!

It ran on real hw (k200, k200lp and k300 boards), on qemu as well as on our internal instruction set simulator (ISS).

Tests were run on several interfaces/drivers (usb, qsfp ethernet, eMMC, PCIe endpoint+RC, SPI, remoteproc, uart, iommu). LTP and uClibc-ng testsuites are also run without any regression.

Everything looks fine to us.

Tested-by Allen Pais on the v1 series
#229 lore raw
In reply to Greg Kroah-Hartman
Quoted 9 lines

This is the start of the stable review cycle for the 6.1.81 release. There are 215 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Wed, 06 Mar 2024 21:15:26 +0000. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

Quoted 9 lines
        https://www.kernel.org/pub/linux/kernel/v6.x/stable-review/patch-6.1.81-rc1.gz

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-6.1.y
and the diffstat can be found below.

thanks,

greg k-h

Compiled and booted on my x86_64 and ARM64 test systems. No errors or regressions.

Thanks.