The patch below does not apply to the 4.16-stable tree.
If someone wants it applied there, or to any other stable or longterm
tree, then please email the backport, including the original git commit
id to <stable(a)vger.kernel.org>.
thanks,
greg k-h
------------------ original commit in Linus's tree ------------------
>From 1c1a2ee1b53b006754073eefc65d2b2cedb5264b Mon Sep 17 00:00:00 2001
From: Coly Li <colyli(a)suse.de>
Date: Thu, 17 May 2018 23:33:26 +0800
Subject: [PATCH] bcache: return 0 from bch_debug_init() if CONFIG_DEBUG_FS=n
Commit 539d39eb2708 ("bcache: fix wrong return value in bch_debug_init()")
returns the return value of debugfs_create_dir() to bcache_init(). When
CONFIG_DEBUG_FS=n, bch_debug_init() always returns 1 and makes
bcache_init() failedi.
This patch makes bch_debug_init() always returns 0 if CONFIG_DEBUG_FS=n,
so bcache can continue to work for the kernels which don't have debugfs
enanbled.
Changelog:
v4: Add Acked-by from Kent Overstreet.
v3: Use IS_ENABLED(CONFIG_DEBUG_FS) to replace #ifdef DEBUG_FS.
v2: Remove a warning information
v1: Initial version.
Fixes: Commit 539d39eb2708 ("bcache: fix wrong return value in bch_debug_init()")
Cc: stable(a)vger.kernel.org
Signed-off-by: Coly Li <colyli(a)suse.de>
Reported-by: Massimo B. <massimo.b(a)gmx.net>
Reported-by: Kai Krakow <kai(a)kaishome.de>
Tested-by: Kai Krakow <kai(a)kaishome.de>
Acked-by: Kent Overstreet <kent.overstreet(a)gmail.com>
Signed-off-by: Jens Axboe <axboe(a)kernel.dk>
diff --git a/drivers/md/bcache/debug.c b/drivers/md/bcache/debug.c
index 4e63c6f6c04d..d030ce3025a6 100644
--- a/drivers/md/bcache/debug.c
+++ b/drivers/md/bcache/debug.c
@@ -250,7 +250,9 @@ void bch_debug_exit(void)
int __init bch_debug_init(struct kobject *kobj)
{
- bcache_debug = debugfs_create_dir("bcache", NULL);
+ if (!IS_ENABLED(CONFIG_DEBUG_FS))
+ return 0;
+ bcache_debug = debugfs_create_dir("bcache", NULL);
return IS_ERR_OR_NULL(bcache_debug);
}
On Sun, May 20, 2018 at 10:59:32AM +0200, Kai Krakow wrote:
> Hey Greg!
>
> The v1 version applied for me but it shows a compiler warning. I didn't try
> the newer version yet.
>
> I could prepare a back-ported version.
Backported would be good.
Also, the code really is wrong even with this change. No code path
should ever do anything different if debugfs is enabled or not, or based
on the return value of a debugfs call. No need to check anything here
at all, the function should be:
void __init bch_debug_init(void)
{
bcache_debug = debugfs_create_dir("bcache", NULL);
}
That's it, no checking, and all is fine and good. Any result of a
debugfs call can always be fed back into another debugfs call with no
harm or errors happening.
thanks,
greg k-h
The patch below does not apply to the 4.4-stable tree.
If someone wants it applied there, or to any other stable or longterm
tree, then please email the backport, including the original git commit
id to <stable(a)vger.kernel.org>.
thanks,
greg k-h
------------------ original commit in Linus's tree ------------------
>From 23a4d7fd34856da8218c4cfc23dba7a6ec0a423a Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky(a)de.ibm.com>
Date: Wed, 25 Apr 2018 18:35:26 +0200
Subject: [PATCH] s390/ftrace: use expoline for indirect branches
The return from the ftrace_stub, _mcount, ftrace_caller and
return_to_handler functions is done with "br %r14" and "br %r1".
These are indirect branches as well and need to use execute
trampolines for CONFIG_EXPOLINE=y.
The ftrace_caller function is a special case as it returns to the
start of a function and may only use %r0 and %r1. For a pre z10
machine the standard execute trampoline uses a LARL + EX to do
this, but this requires *two* registers in the range %r1..%r15.
To get around this the 'br %r1' located in the lowcore is used,
then the EX instruction does not need an address register.
But the lowcore trick may only be used for pre z14 machines,
with noexec=on the mapping for the first page may not contain
instructions. The solution for that is an ALTERNATIVE in the
expoline THUNK generated by 'GEN_BR_THUNK %r1' to switch to
EXRL, this relies on the fact that a machine that supports
noexec=on has EXRL as well.
Cc: stable(a)vger.kernel.org # 4.16
Fixes: f19fbd5ed6 ("s390: introduce execute-trampolines for branches")
Signed-off-by: Martin Schwidefsky <schwidefsky(a)de.ibm.com>
diff --git a/arch/s390/include/asm/nospec-insn.h b/arch/s390/include/asm/nospec-insn.h
index 440689cbcf51..7d7640e1cf90 100644
--- a/arch/s390/include/asm/nospec-insn.h
+++ b/arch/s390/include/asm/nospec-insn.h
@@ -2,12 +2,16 @@
#ifndef _ASM_S390_NOSPEC_ASM_H
#define _ASM_S390_NOSPEC_ASM_H
+#include <asm/alternative-asm.h>
+#include <asm/asm-offsets.h>
#include <asm/dwarf.h>
#ifdef __ASSEMBLY__
#ifdef CONFIG_EXPOLINE
+_LC_BR_R1 = __LC_BR_R1
+
/*
* The expoline macros are used to create thunks in the same format
* as gcc generates them. The 'comdat' section flag makes sure that
@@ -78,13 +82,21 @@
.endm
.macro __THUNK_EX_BR reg,ruse
+ # Be very careful when adding instructions to this macro!
+ # The ALTERNATIVE replacement code has a .+10 which targets
+ # the "br \reg" after the code has been patched.
#ifdef CONFIG_HAVE_MARCH_Z10_FEATURES
exrl 0,555f
j .
#else
+ .ifc \reg,%r1
+ ALTERNATIVE "ex %r0,_LC_BR_R1", ".insn ril,0xc60000000000,0,.+10", 35
+ j .
+ .else
larl \ruse,555f
ex 0,0(\ruse)
j .
+ .endif
#endif
555: br \reg
.endm
diff --git a/arch/s390/kernel/asm-offsets.c b/arch/s390/kernel/asm-offsets.c
index eb2a5c0443cd..11aea745a2a6 100644
--- a/arch/s390/kernel/asm-offsets.c
+++ b/arch/s390/kernel/asm-offsets.c
@@ -181,6 +181,7 @@ int main(void)
OFFSET(__LC_MACHINE_FLAGS, lowcore, machine_flags);
OFFSET(__LC_PREEMPT_COUNT, lowcore, preempt_count);
OFFSET(__LC_GMAP, lowcore, gmap);
+ OFFSET(__LC_BR_R1, lowcore, br_r1_trampoline);
/* software defined ABI-relevant lowcore locations 0xe00 - 0xe20 */
OFFSET(__LC_DUMP_REIPL, lowcore, ipib);
/* hardware defined lowcore locations 0x1000 - 0x18ff */
diff --git a/arch/s390/kernel/mcount.S b/arch/s390/kernel/mcount.S
index 82df7d80fab2..27110f3294ed 100644
--- a/arch/s390/kernel/mcount.S
+++ b/arch/s390/kernel/mcount.S
@@ -9,13 +9,17 @@
#include <linux/linkage.h>
#include <asm/asm-offsets.h>
#include <asm/ftrace.h>
+#include <asm/nospec-insn.h>
#include <asm/ptrace.h>
#include <asm/export.h>
+ GEN_BR_THUNK %r1
+ GEN_BR_THUNK %r14
+
.section .kprobes.text, "ax"
ENTRY(ftrace_stub)
- br %r14
+ BR_EX %r14
#define STACK_FRAME_SIZE (STACK_FRAME_OVERHEAD + __PT_SIZE)
#define STACK_PTREGS (STACK_FRAME_OVERHEAD)
@@ -23,7 +27,7 @@ ENTRY(ftrace_stub)
#define STACK_PTREGS_PSW (STACK_PTREGS + __PT_PSW)
ENTRY(_mcount)
- br %r14
+ BR_EX %r14
EXPORT_SYMBOL(_mcount)
@@ -53,7 +57,7 @@ ENTRY(ftrace_caller)
#endif
lgr %r3,%r14
la %r5,STACK_PTREGS(%r15)
- basr %r14,%r1
+ BASR_EX %r14,%r1
#ifdef CONFIG_FUNCTION_GRAPH_TRACER
# The j instruction gets runtime patched to a nop instruction.
# See ftrace_enable_ftrace_graph_caller.
@@ -68,7 +72,7 @@ ftrace_graph_caller_end:
#endif
lg %r1,(STACK_PTREGS_PSW+8)(%r15)
lmg %r2,%r15,(STACK_PTREGS_GPRS+2*8)(%r15)
- br %r1
+ BR_EX %r1
#ifdef CONFIG_FUNCTION_GRAPH_TRACER
@@ -81,6 +85,6 @@ ENTRY(return_to_handler)
aghi %r15,STACK_FRAME_OVERHEAD
lgr %r14,%r2
lmg %r2,%r5,32(%r15)
- br %r14
+ BR_EX %r14
#endif
The patch below does not apply to the 4.9-stable tree.
If someone wants it applied there, or to any other stable or longterm
tree, then please email the backport, including the original git commit
id to <stable(a)vger.kernel.org>.
thanks,
greg k-h
------------------ original commit in Linus's tree ------------------
>From 23a4d7fd34856da8218c4cfc23dba7a6ec0a423a Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky(a)de.ibm.com>
Date: Wed, 25 Apr 2018 18:35:26 +0200
Subject: [PATCH] s390/ftrace: use expoline for indirect branches
The return from the ftrace_stub, _mcount, ftrace_caller and
return_to_handler functions is done with "br %r14" and "br %r1".
These are indirect branches as well and need to use execute
trampolines for CONFIG_EXPOLINE=y.
The ftrace_caller function is a special case as it returns to the
start of a function and may only use %r0 and %r1. For a pre z10
machine the standard execute trampoline uses a LARL + EX to do
this, but this requires *two* registers in the range %r1..%r15.
To get around this the 'br %r1' located in the lowcore is used,
then the EX instruction does not need an address register.
But the lowcore trick may only be used for pre z14 machines,
with noexec=on the mapping for the first page may not contain
instructions. The solution for that is an ALTERNATIVE in the
expoline THUNK generated by 'GEN_BR_THUNK %r1' to switch to
EXRL, this relies on the fact that a machine that supports
noexec=on has EXRL as well.
Cc: stable(a)vger.kernel.org # 4.16
Fixes: f19fbd5ed6 ("s390: introduce execute-trampolines for branches")
Signed-off-by: Martin Schwidefsky <schwidefsky(a)de.ibm.com>
diff --git a/arch/s390/include/asm/nospec-insn.h b/arch/s390/include/asm/nospec-insn.h
index 440689cbcf51..7d7640e1cf90 100644
--- a/arch/s390/include/asm/nospec-insn.h
+++ b/arch/s390/include/asm/nospec-insn.h
@@ -2,12 +2,16 @@
#ifndef _ASM_S390_NOSPEC_ASM_H
#define _ASM_S390_NOSPEC_ASM_H
+#include <asm/alternative-asm.h>
+#include <asm/asm-offsets.h>
#include <asm/dwarf.h>
#ifdef __ASSEMBLY__
#ifdef CONFIG_EXPOLINE
+_LC_BR_R1 = __LC_BR_R1
+
/*
* The expoline macros are used to create thunks in the same format
* as gcc generates them. The 'comdat' section flag makes sure that
@@ -78,13 +82,21 @@
.endm
.macro __THUNK_EX_BR reg,ruse
+ # Be very careful when adding instructions to this macro!
+ # The ALTERNATIVE replacement code has a .+10 which targets
+ # the "br \reg" after the code has been patched.
#ifdef CONFIG_HAVE_MARCH_Z10_FEATURES
exrl 0,555f
j .
#else
+ .ifc \reg,%r1
+ ALTERNATIVE "ex %r0,_LC_BR_R1", ".insn ril,0xc60000000000,0,.+10", 35
+ j .
+ .else
larl \ruse,555f
ex 0,0(\ruse)
j .
+ .endif
#endif
555: br \reg
.endm
diff --git a/arch/s390/kernel/asm-offsets.c b/arch/s390/kernel/asm-offsets.c
index eb2a5c0443cd..11aea745a2a6 100644
--- a/arch/s390/kernel/asm-offsets.c
+++ b/arch/s390/kernel/asm-offsets.c
@@ -181,6 +181,7 @@ int main(void)
OFFSET(__LC_MACHINE_FLAGS, lowcore, machine_flags);
OFFSET(__LC_PREEMPT_COUNT, lowcore, preempt_count);
OFFSET(__LC_GMAP, lowcore, gmap);
+ OFFSET(__LC_BR_R1, lowcore, br_r1_trampoline);
/* software defined ABI-relevant lowcore locations 0xe00 - 0xe20 */
OFFSET(__LC_DUMP_REIPL, lowcore, ipib);
/* hardware defined lowcore locations 0x1000 - 0x18ff */
diff --git a/arch/s390/kernel/mcount.S b/arch/s390/kernel/mcount.S
index 82df7d80fab2..27110f3294ed 100644
--- a/arch/s390/kernel/mcount.S
+++ b/arch/s390/kernel/mcount.S
@@ -9,13 +9,17 @@
#include <linux/linkage.h>
#include <asm/asm-offsets.h>
#include <asm/ftrace.h>
+#include <asm/nospec-insn.h>
#include <asm/ptrace.h>
#include <asm/export.h>
+ GEN_BR_THUNK %r1
+ GEN_BR_THUNK %r14
+
.section .kprobes.text, "ax"
ENTRY(ftrace_stub)
- br %r14
+ BR_EX %r14
#define STACK_FRAME_SIZE (STACK_FRAME_OVERHEAD + __PT_SIZE)
#define STACK_PTREGS (STACK_FRAME_OVERHEAD)
@@ -23,7 +27,7 @@ ENTRY(ftrace_stub)
#define STACK_PTREGS_PSW (STACK_PTREGS + __PT_PSW)
ENTRY(_mcount)
- br %r14
+ BR_EX %r14
EXPORT_SYMBOL(_mcount)
@@ -53,7 +57,7 @@ ENTRY(ftrace_caller)
#endif
lgr %r3,%r14
la %r5,STACK_PTREGS(%r15)
- basr %r14,%r1
+ BASR_EX %r14,%r1
#ifdef CONFIG_FUNCTION_GRAPH_TRACER
# The j instruction gets runtime patched to a nop instruction.
# See ftrace_enable_ftrace_graph_caller.
@@ -68,7 +72,7 @@ ftrace_graph_caller_end:
#endif
lg %r1,(STACK_PTREGS_PSW+8)(%r15)
lmg %r2,%r15,(STACK_PTREGS_GPRS+2*8)(%r15)
- br %r1
+ BR_EX %r1
#ifdef CONFIG_FUNCTION_GRAPH_TRACER
@@ -81,6 +85,6 @@ ENTRY(return_to_handler)
aghi %r15,STACK_FRAME_OVERHEAD
lgr %r14,%r2
lmg %r2,%r5,32(%r15)
- br %r14
+ BR_EX %r14
#endif
The patch below does not apply to the 4.14-stable tree.
If someone wants it applied there, or to any other stable or longterm
tree, then please email the backport, including the original git commit
id to <stable(a)vger.kernel.org>.
thanks,
greg k-h
------------------ original commit in Linus's tree ------------------
>From 23a4d7fd34856da8218c4cfc23dba7a6ec0a423a Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky(a)de.ibm.com>
Date: Wed, 25 Apr 2018 18:35:26 +0200
Subject: [PATCH] s390/ftrace: use expoline for indirect branches
The return from the ftrace_stub, _mcount, ftrace_caller and
return_to_handler functions is done with "br %r14" and "br %r1".
These are indirect branches as well and need to use execute
trampolines for CONFIG_EXPOLINE=y.
The ftrace_caller function is a special case as it returns to the
start of a function and may only use %r0 and %r1. For a pre z10
machine the standard execute trampoline uses a LARL + EX to do
this, but this requires *two* registers in the range %r1..%r15.
To get around this the 'br %r1' located in the lowcore is used,
then the EX instruction does not need an address register.
But the lowcore trick may only be used for pre z14 machines,
with noexec=on the mapping for the first page may not contain
instructions. The solution for that is an ALTERNATIVE in the
expoline THUNK generated by 'GEN_BR_THUNK %r1' to switch to
EXRL, this relies on the fact that a machine that supports
noexec=on has EXRL as well.
Cc: stable(a)vger.kernel.org # 4.16
Fixes: f19fbd5ed6 ("s390: introduce execute-trampolines for branches")
Signed-off-by: Martin Schwidefsky <schwidefsky(a)de.ibm.com>
diff --git a/arch/s390/include/asm/nospec-insn.h b/arch/s390/include/asm/nospec-insn.h
index 440689cbcf51..7d7640e1cf90 100644
--- a/arch/s390/include/asm/nospec-insn.h
+++ b/arch/s390/include/asm/nospec-insn.h
@@ -2,12 +2,16 @@
#ifndef _ASM_S390_NOSPEC_ASM_H
#define _ASM_S390_NOSPEC_ASM_H
+#include <asm/alternative-asm.h>
+#include <asm/asm-offsets.h>
#include <asm/dwarf.h>
#ifdef __ASSEMBLY__
#ifdef CONFIG_EXPOLINE
+_LC_BR_R1 = __LC_BR_R1
+
/*
* The expoline macros are used to create thunks in the same format
* as gcc generates them. The 'comdat' section flag makes sure that
@@ -78,13 +82,21 @@
.endm
.macro __THUNK_EX_BR reg,ruse
+ # Be very careful when adding instructions to this macro!
+ # The ALTERNATIVE replacement code has a .+10 which targets
+ # the "br \reg" after the code has been patched.
#ifdef CONFIG_HAVE_MARCH_Z10_FEATURES
exrl 0,555f
j .
#else
+ .ifc \reg,%r1
+ ALTERNATIVE "ex %r0,_LC_BR_R1", ".insn ril,0xc60000000000,0,.+10", 35
+ j .
+ .else
larl \ruse,555f
ex 0,0(\ruse)
j .
+ .endif
#endif
555: br \reg
.endm
diff --git a/arch/s390/kernel/asm-offsets.c b/arch/s390/kernel/asm-offsets.c
index eb2a5c0443cd..11aea745a2a6 100644
--- a/arch/s390/kernel/asm-offsets.c
+++ b/arch/s390/kernel/asm-offsets.c
@@ -181,6 +181,7 @@ int main(void)
OFFSET(__LC_MACHINE_FLAGS, lowcore, machine_flags);
OFFSET(__LC_PREEMPT_COUNT, lowcore, preempt_count);
OFFSET(__LC_GMAP, lowcore, gmap);
+ OFFSET(__LC_BR_R1, lowcore, br_r1_trampoline);
/* software defined ABI-relevant lowcore locations 0xe00 - 0xe20 */
OFFSET(__LC_DUMP_REIPL, lowcore, ipib);
/* hardware defined lowcore locations 0x1000 - 0x18ff */
diff --git a/arch/s390/kernel/mcount.S b/arch/s390/kernel/mcount.S
index 82df7d80fab2..27110f3294ed 100644
--- a/arch/s390/kernel/mcount.S
+++ b/arch/s390/kernel/mcount.S
@@ -9,13 +9,17 @@
#include <linux/linkage.h>
#include <asm/asm-offsets.h>
#include <asm/ftrace.h>
+#include <asm/nospec-insn.h>
#include <asm/ptrace.h>
#include <asm/export.h>
+ GEN_BR_THUNK %r1
+ GEN_BR_THUNK %r14
+
.section .kprobes.text, "ax"
ENTRY(ftrace_stub)
- br %r14
+ BR_EX %r14
#define STACK_FRAME_SIZE (STACK_FRAME_OVERHEAD + __PT_SIZE)
#define STACK_PTREGS (STACK_FRAME_OVERHEAD)
@@ -23,7 +27,7 @@ ENTRY(ftrace_stub)
#define STACK_PTREGS_PSW (STACK_PTREGS + __PT_PSW)
ENTRY(_mcount)
- br %r14
+ BR_EX %r14
EXPORT_SYMBOL(_mcount)
@@ -53,7 +57,7 @@ ENTRY(ftrace_caller)
#endif
lgr %r3,%r14
la %r5,STACK_PTREGS(%r15)
- basr %r14,%r1
+ BASR_EX %r14,%r1
#ifdef CONFIG_FUNCTION_GRAPH_TRACER
# The j instruction gets runtime patched to a nop instruction.
# See ftrace_enable_ftrace_graph_caller.
@@ -68,7 +72,7 @@ ftrace_graph_caller_end:
#endif
lg %r1,(STACK_PTREGS_PSW+8)(%r15)
lmg %r2,%r15,(STACK_PTREGS_GPRS+2*8)(%r15)
- br %r1
+ BR_EX %r1
#ifdef CONFIG_FUNCTION_GRAPH_TRACER
@@ -81,6 +85,6 @@ ENTRY(return_to_handler)
aghi %r15,STACK_FRAME_OVERHEAD
lgr %r14,%r2
lmg %r2,%r5,32(%r15)
- br %r14
+ BR_EX %r14
#endif
The patch below does not apply to the 4.16-stable tree.
If someone wants it applied there, or to any other stable or longterm
tree, then please email the backport, including the original git commit
id to <stable(a)vger.kernel.org>.
thanks,
greg k-h
------------------ original commit in Linus's tree ------------------
>From 23a4d7fd34856da8218c4cfc23dba7a6ec0a423a Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky(a)de.ibm.com>
Date: Wed, 25 Apr 2018 18:35:26 +0200
Subject: [PATCH] s390/ftrace: use expoline for indirect branches
The return from the ftrace_stub, _mcount, ftrace_caller and
return_to_handler functions is done with "br %r14" and "br %r1".
These are indirect branches as well and need to use execute
trampolines for CONFIG_EXPOLINE=y.
The ftrace_caller function is a special case as it returns to the
start of a function and may only use %r0 and %r1. For a pre z10
machine the standard execute trampoline uses a LARL + EX to do
this, but this requires *two* registers in the range %r1..%r15.
To get around this the 'br %r1' located in the lowcore is used,
then the EX instruction does not need an address register.
But the lowcore trick may only be used for pre z14 machines,
with noexec=on the mapping for the first page may not contain
instructions. The solution for that is an ALTERNATIVE in the
expoline THUNK generated by 'GEN_BR_THUNK %r1' to switch to
EXRL, this relies on the fact that a machine that supports
noexec=on has EXRL as well.
Cc: stable(a)vger.kernel.org # 4.16
Fixes: f19fbd5ed6 ("s390: introduce execute-trampolines for branches")
Signed-off-by: Martin Schwidefsky <schwidefsky(a)de.ibm.com>
diff --git a/arch/s390/include/asm/nospec-insn.h b/arch/s390/include/asm/nospec-insn.h
index 440689cbcf51..7d7640e1cf90 100644
--- a/arch/s390/include/asm/nospec-insn.h
+++ b/arch/s390/include/asm/nospec-insn.h
@@ -2,12 +2,16 @@
#ifndef _ASM_S390_NOSPEC_ASM_H
#define _ASM_S390_NOSPEC_ASM_H
+#include <asm/alternative-asm.h>
+#include <asm/asm-offsets.h>
#include <asm/dwarf.h>
#ifdef __ASSEMBLY__
#ifdef CONFIG_EXPOLINE
+_LC_BR_R1 = __LC_BR_R1
+
/*
* The expoline macros are used to create thunks in the same format
* as gcc generates them. The 'comdat' section flag makes sure that
@@ -78,13 +82,21 @@
.endm
.macro __THUNK_EX_BR reg,ruse
+ # Be very careful when adding instructions to this macro!
+ # The ALTERNATIVE replacement code has a .+10 which targets
+ # the "br \reg" after the code has been patched.
#ifdef CONFIG_HAVE_MARCH_Z10_FEATURES
exrl 0,555f
j .
#else
+ .ifc \reg,%r1
+ ALTERNATIVE "ex %r0,_LC_BR_R1", ".insn ril,0xc60000000000,0,.+10", 35
+ j .
+ .else
larl \ruse,555f
ex 0,0(\ruse)
j .
+ .endif
#endif
555: br \reg
.endm
diff --git a/arch/s390/kernel/asm-offsets.c b/arch/s390/kernel/asm-offsets.c
index eb2a5c0443cd..11aea745a2a6 100644
--- a/arch/s390/kernel/asm-offsets.c
+++ b/arch/s390/kernel/asm-offsets.c
@@ -181,6 +181,7 @@ int main(void)
OFFSET(__LC_MACHINE_FLAGS, lowcore, machine_flags);
OFFSET(__LC_PREEMPT_COUNT, lowcore, preempt_count);
OFFSET(__LC_GMAP, lowcore, gmap);
+ OFFSET(__LC_BR_R1, lowcore, br_r1_trampoline);
/* software defined ABI-relevant lowcore locations 0xe00 - 0xe20 */
OFFSET(__LC_DUMP_REIPL, lowcore, ipib);
/* hardware defined lowcore locations 0x1000 - 0x18ff */
diff --git a/arch/s390/kernel/mcount.S b/arch/s390/kernel/mcount.S
index 82df7d80fab2..27110f3294ed 100644
--- a/arch/s390/kernel/mcount.S
+++ b/arch/s390/kernel/mcount.S
@@ -9,13 +9,17 @@
#include <linux/linkage.h>
#include <asm/asm-offsets.h>
#include <asm/ftrace.h>
+#include <asm/nospec-insn.h>
#include <asm/ptrace.h>
#include <asm/export.h>
+ GEN_BR_THUNK %r1
+ GEN_BR_THUNK %r14
+
.section .kprobes.text, "ax"
ENTRY(ftrace_stub)
- br %r14
+ BR_EX %r14
#define STACK_FRAME_SIZE (STACK_FRAME_OVERHEAD + __PT_SIZE)
#define STACK_PTREGS (STACK_FRAME_OVERHEAD)
@@ -23,7 +27,7 @@ ENTRY(ftrace_stub)
#define STACK_PTREGS_PSW (STACK_PTREGS + __PT_PSW)
ENTRY(_mcount)
- br %r14
+ BR_EX %r14
EXPORT_SYMBOL(_mcount)
@@ -53,7 +57,7 @@ ENTRY(ftrace_caller)
#endif
lgr %r3,%r14
la %r5,STACK_PTREGS(%r15)
- basr %r14,%r1
+ BASR_EX %r14,%r1
#ifdef CONFIG_FUNCTION_GRAPH_TRACER
# The j instruction gets runtime patched to a nop instruction.
# See ftrace_enable_ftrace_graph_caller.
@@ -68,7 +72,7 @@ ftrace_graph_caller_end:
#endif
lg %r1,(STACK_PTREGS_PSW+8)(%r15)
lmg %r2,%r15,(STACK_PTREGS_GPRS+2*8)(%r15)
- br %r1
+ BR_EX %r1
#ifdef CONFIG_FUNCTION_GRAPH_TRACER
@@ -81,6 +85,6 @@ ENTRY(return_to_handler)
aghi %r15,STACK_FRAME_OVERHEAD
lgr %r14,%r2
lmg %r2,%r5,32(%r15)
- br %r14
+ BR_EX %r14
#endif
The patch below does not apply to the 4.4-stable tree.
If someone wants it applied there, or to any other stable or longterm
tree, then please email the backport, including the original git commit
id to <stable(a)vger.kernel.org>.
thanks,
greg k-h
------------------ original commit in Linus's tree ------------------
>From 467a3bf219cee12259182c5cb4821f88fd518a51 Mon Sep 17 00:00:00 2001
From: Martin Schwidefsky <schwidefsky(a)de.ibm.com>
Date: Mon, 23 Apr 2018 14:31:36 +0200
Subject: [PATCH] s390/crc32-vx: use expoline for indirect branches
The return from the crc32_le_vgfm_16/crc32c_le_vgfm_16 and the
crc32_be_vgfm_16 functions are done with "br %r14". These are indirect
branches as well and need to use execute trampolines for CONFIG_EXPOLINE=y.
Cc: stable(a)vger.kernel.org # 4.16
Fixes: f19fbd5ed6 ("s390: introduce execute-trampolines for branches")
Reviewed-by: Hendrik Brueckner <brueckner(a)linux.vnet.ibm.com>
Signed-off-by: Martin Schwidefsky <schwidefsky(a)de.ibm.com>
diff --git a/arch/s390/crypto/crc32be-vx.S b/arch/s390/crypto/crc32be-vx.S
index e8077f0971f8..2bf01ba44107 100644
--- a/arch/s390/crypto/crc32be-vx.S
+++ b/arch/s390/crypto/crc32be-vx.S
@@ -13,6 +13,7 @@
*/
#include <linux/linkage.h>
+#include <asm/nospec-insn.h>
#include <asm/vx-insn.h>
/* Vector register range containing CRC-32 constants */
@@ -67,6 +68,8 @@
.previous
+ GEN_BR_THUNK %r14
+
.text
/*
* The CRC-32 function(s) use these calling conventions:
@@ -203,6 +206,6 @@ ENTRY(crc32_be_vgfm_16)
.Ldone:
VLGVF %r2,%v2,3
- br %r14
+ BR_EX %r14
.previous
diff --git a/arch/s390/crypto/crc32le-vx.S b/arch/s390/crypto/crc32le-vx.S
index d8c67a58c0c5..7d6f568bd3ad 100644
--- a/arch/s390/crypto/crc32le-vx.S
+++ b/arch/s390/crypto/crc32le-vx.S
@@ -14,6 +14,7 @@
*/
#include <linux/linkage.h>
+#include <asm/nospec-insn.h>
#include <asm/vx-insn.h>
/* Vector register range containing CRC-32 constants */
@@ -76,6 +77,7 @@
.previous
+ GEN_BR_THUNK %r14
.text
@@ -264,6 +266,6 @@ crc32_le_vgfm_generic:
.Ldone:
VLGVF %r2,%v2,2
- br %r14
+ BR_EX %r14
.previous
The patch below does not apply to the 4.14-stable tree.
If someone wants it applied there, or to any other stable or longterm
tree, then please email the backport, including the original git commit
id to <stable(a)vger.kernel.org>.
thanks,
greg k-h
------------------ original commit in Linus's tree ------------------
>From 91ba9f28a3de97761c2b5fd5df5d88421268e507 Mon Sep 17 00:00:00 2001
From: Deepak Rawat <drawat(a)vmware.com>
Date: Tue, 15 May 2018 15:39:09 +0200
Subject: [PATCH] drm/vmwgfx: Set dmabuf_size when vmw_dmabuf_init is
successful
SOU primary plane prepare_fb hook depends upon dmabuf_size to pin up BO
(and not call a new vmw_dmabuf_init) when a new fb size is same as
current fb. This was changed in a recent commit which is causing
page_flip to fail on VM with low display memory and multi-mon failure
when cycle monitors from secondary display.
Cc: <stable(a)vger.kernel.org> # 4.14, 4.16
Fixes: 20fb5a635a0c ("drm/vmwgfx: Unpin the screen object backup buffer when not used")
Signed-off-by: Deepak Rawat <drawat(a)vmware.com>
Reviewed-by: Sinclair Yeh <syeh(a)vmware.com>
Signed-off-by: Thomas Hellstrom <thellstrom(a)vmware.com>
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c b/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c
index 648f8127f65a..3d667e903beb 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c
@@ -482,6 +482,8 @@ vmw_sou_primary_plane_prepare_fb(struct drm_plane *plane,
return ret;
}
+ vps->dmabuf_size = size;
+
/*
* TTM already thinks the buffer is pinned, but make sure the
* pin_count is upped.
The patch below does not apply to the 4.16-stable tree.
If someone wants it applied there, or to any other stable or longterm
tree, then please email the backport, including the original git commit
id to <stable(a)vger.kernel.org>.
thanks,
greg k-h
------------------ original commit in Linus's tree ------------------
>From 91ba9f28a3de97761c2b5fd5df5d88421268e507 Mon Sep 17 00:00:00 2001
From: Deepak Rawat <drawat(a)vmware.com>
Date: Tue, 15 May 2018 15:39:09 +0200
Subject: [PATCH] drm/vmwgfx: Set dmabuf_size when vmw_dmabuf_init is
successful
SOU primary plane prepare_fb hook depends upon dmabuf_size to pin up BO
(and not call a new vmw_dmabuf_init) when a new fb size is same as
current fb. This was changed in a recent commit which is causing
page_flip to fail on VM with low display memory and multi-mon failure
when cycle monitors from secondary display.
Cc: <stable(a)vger.kernel.org> # 4.14, 4.16
Fixes: 20fb5a635a0c ("drm/vmwgfx: Unpin the screen object backup buffer when not used")
Signed-off-by: Deepak Rawat <drawat(a)vmware.com>
Reviewed-by: Sinclair Yeh <syeh(a)vmware.com>
Signed-off-by: Thomas Hellstrom <thellstrom(a)vmware.com>
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c b/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c
index 648f8127f65a..3d667e903beb 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_scrn.c
@@ -482,6 +482,8 @@ vmw_sou_primary_plane_prepare_fb(struct drm_plane *plane,
return ret;
}
+ vps->dmabuf_size = size;
+
/*
* TTM already thinks the buffer is pinned, but make sure the
* pin_count is upped.
The patch below does not apply to the 4.14-stable tree.
If someone wants it applied there, or to any other stable or longterm
tree, then please email the backport, including the original git commit
id to <stable(a)vger.kernel.org>.
thanks,
greg k-h
------------------ original commit in Linus's tree ------------------
>From a057344806d035cb9ac991619fa07854e807562d Mon Sep 17 00:00:00 2001
From: Maxime Chevallier <maxime.chevallier(a)bootlin.com>
Date: Wed, 25 Apr 2018 13:07:31 +0200
Subject: [PATCH] ARM64: dts: marvell: armada-cp110: Add clocks for the xmdio
node
The Marvell XSMI controller needs 3 clocks to operate correctly :
- The MG clock (clk 5)
- The MG Core clock (clk 6)
- The GOP clock (clk 18)
This commit adds them, to avoid system hangs when using these
interfaces.
[gregory.clement: use the real first commit to fix and add the cc:stable
flag]
Fixes: f66b2aff46ea ("arm64: dts: marvell: add xmdio nodes for 7k/8k")
Cc: <stable(a)vger.kernel.org>
Signed-off-by: Maxime Chevallier <maxime.chevallier(a)bootlin.com>
Signed-off-by: Gregory CLEMENT <gregory.clement(a)bootlin.com>
diff --git a/arch/arm64/boot/dts/marvell/armada-cp110.dtsi b/arch/arm64/boot/dts/marvell/armada-cp110.dtsi
index 48cad7919efa..ca22f9d100f5 100644
--- a/arch/arm64/boot/dts/marvell/armada-cp110.dtsi
+++ b/arch/arm64/boot/dts/marvell/armada-cp110.dtsi
@@ -141,6 +141,8 @@
#size-cells = <0>;
compatible = "marvell,xmdio";
reg = <0x12a600 0x10>;
+ clocks = <&CP110_LABEL(clk) 1 5>,
+ <&CP110_LABEL(clk) 1 6>, <&CP110_LABEL(clk) 1 18>;
status = "disabled";
};