Skip to contentSkip to conversation

signal/m68k: Use force_sigsegv(SIGSEGV) in fpsp040_die

PATCH In review Finn Thain 3 participants 4 messages Feb 13, 2025 – Feb 25, 2025 m68k architecture

lore
Share & tools
lore

Finn Thain

author
#0 lore raw

PATCH 1/1 signal/m68k: Use force_sigsegv(SIGSEGV) in fpsp040_die

From: "Eric W. Biederman" <[email protected]>

[ Upstream commit a3616a3c02722d1edb95acc7fceade242f6553ba ]

In the fpsp040 code when copyin or copyout fails call force_sigsegv(SIGSEGV) instead of do_exit(SIGSEGV).

This solves a couple of problems. Because do_exit embeds the ptrace stop PTRACE_EVENT_EXIT a complete stack frame needs to be present for that to work correctly. There is always the information needed for a ptrace stop where get_signal is called. So exiting with a signal solves the ptrace issue.

Further exiting with a signal ensures that all of the threads in a process are killed not just the thread that malfunctioned. Which avoids confusing userspace.

To make force_sigsegv(SIGSEGV) work in fpsp040_die modify the code to save all of the registers and jump to ret_from_exception (which ultimately calls get_signal) after fpsp040_die returns.

v2: Updated the branches to use gas's pseudo ops that automatically
calculate the best branch instruction to use for the purpose.

v1: https://lkml.kernel.org/r/87a6m8kgtx.fsf_-_@disp2133

---
 arch/m68k/fpsp040/skeleton.S | 3 ++-
 arch/m68k/kernel/traps.c     | 2 +-
 2 files changed, 3 insertions(+), 2 deletions(-)
arch/m68k/fpsp040/skeleton.S +2 -1
index 31a9c634c81e..081922c72daa 100644
@@ -502,7 +502,8 @@ in_ea:
 	.section .fixup,"ax"
 	.even
 1:
-	jbra	fpsp040_die
+	jbsr	fpsp040_die
+	jbra	.Lnotkern
 
 	.section __ex_table,"a"
 	.align	4
arch/m68k/kernel/traps.c +1 -1
index 35f706d836c5..c6f18dc5884b 100644
@@ -1155,7 +1155,7 @@ asmlinkage void set_esp0(unsigned long ssp)
  */
 asmlinkage void fpsp040_die(void)
 {
-	do_exit(SIGSEGV);
+	force_sigsegv(SIGSEGV, current);
 }
 
 #ifdef CONFIG_M68KFPU_EMU
#1 lore raw

Sorry, I saw these five minutes too late. Applied for the next release.

CU
Uli

Quoted 65 lines · CET Finn Thain

From: "Eric W. Biederman" <[email protected]>

[ Upstream commit a3616a3c02722d1edb95acc7fceade242f6553ba ]

In the fpsp040 code when copyin or copyout fails call force_sigsegv(SIGSEGV) instead of do_exit(SIGSEGV).

This solves a couple of problems. Because do_exit embeds the ptrace stop PTRACE_EVENT_EXIT a complete stack frame needs to be present for that to work correctly. There is always the information needed for a ptrace stop where get_signal is called. So exiting with a signal solves the ptrace issue.

Further exiting with a signal ensures that all of the threads in a process are killed not just the thread that malfunctioned. Which avoids confusing userspace.

To make force_sigsegv(SIGSEGV) work in fpsp040_die modify the code to save all of the registers and jump to ret_from_exception (which ultimately calls get_signal) after fpsp040_die returns.

v2: Updated the branches to use gas's pseudo ops that automatically
calculate the best branch instruction to use for the purpose.

v1: https://lkml.kernel.org/r/87a6m8kgtx.fsf_-_@disp2133

---
 arch/m68k/fpsp040/skeleton.S | 3 ++-
 arch/m68k/kernel/traps.c     | 2 +-
 2 files changed, 3 insertions(+), 2 deletions(-)
arch/m68k/fpsp040/skeleton.S +2 -1
index 31a9c634c81e..081922c72daa 100644
@@ -502,7 +502,8 @@ in_ea:
 	.section .fixup,"ax"
 	.even
 1:
-	jbra	fpsp040_die
+	jbsr	fpsp040_die
+	jbra	.Lnotkern
 
 	.section __ex_table,"a"
 	.align	4
arch/m68k/kernel/traps.c +1 -1
index 35f706d836c5..c6f18dc5884b 100644
@@ -1155,7 +1155,7 @@ asmlinkage void set_esp0(unsigned long ssp)
  */
 asmlinkage void fpsp040_die(void)
 {
-	do_exit(SIGSEGV);
+	force_sigsegv(SIGSEGV, current);
 }
 
 #ifdef CONFIG_M68KFPU_EMU
#2 lore raw

Hi!

Quoted 1 line

Sorry, I saw these five minutes too late. Applied for the next release.

I guess these should go to 4.4-st, but I don't see them there. Did you forget to push?

Best regards,
Pavel

Quoted 72 lines
Quoted 54 lines

From: "Eric W. Biederman" <[email protected]>

[ Upstream commit a3616a3c02722d1edb95acc7fceade242f6553ba ]

In the fpsp040 code when copyin or copyout fails call force_sigsegv(SIGSEGV) instead of do_exit(SIGSEGV).

This solves a couple of problems. Because do_exit embeds the ptrace stop PTRACE_EVENT_EXIT a complete stack frame needs to be present for that to work correctly. There is always the information needed for a ptrace stop where get_signal is called. So exiting with a signal solves the ptrace issue.

Further exiting with a signal ensures that all of the threads in a process are killed not just the thread that malfunctioned. Which avoids confusing userspace.

To make force_sigsegv(SIGSEGV) work in fpsp040_die modify the code to save all of the registers and jump to ret_from_exception (which ultimately calls get_signal) after fpsp040_die returns.

v2: Updated the branches to use gas's pseudo ops that automatically
calculate the best branch instruction to use for the purpose.

v1: https://lkml.kernel.org/r/87a6m8kgtx.fsf_-_@disp2133

---
 arch/m68k/fpsp040/skeleton.S | 3 ++-
 arch/m68k/kernel/traps.c     | 2 +-
 2 files changed, 3 insertions(+), 2 deletions(-)
arch/m68k/fpsp040/skeleton.S +2 -1
index 31a9c634c81e..081922c72daa 100644
@@ -502,7 +502,8 @@ in_ea:
 	.section .fixup,"ax"
 	.even
 1:
-	jbra	fpsp040_die
+	jbsr	fpsp040_die
+	jbra	.Lnotkern
 
 	.section __ex_table,"a"
 	.align	4
arch/m68k/kernel/traps.c +0 -0
index 35f706d836c5..c6f18dc5884b 100644

arch/m68k/kernel/traps.c asmlinkage void set_esp0(unsigned long ssp) · line 1161

Quoted 8 lines
arch/m68k/kernel/traps.c +1 -1
@@ -1155,7 +1155,7 @@ asmlinkage void set_esp0(unsigned long ssp)
  */
 asmlinkage void fpsp040_die(void)
 {
-	do_exit(SIGSEGV);
+	force_sigsegv(SIGSEGV, current);
 }
 
 #ifdef CONFIG_M68KFPU_EMU

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#17828): https://lists.cip-project.org/g/cip-dev/message/17828
Mute This Topic: https://lists.cip-project.org/mt/111159546/4520429
Group Owner: [email protected]
Unsubscribe: https://lists.cip-project.org/g/cip-dev/leave/8129118/4520429/629823354/xyzzy [[email protected]]
-=-=-=-=-=-=-=-=-=-=-=-

Signature · 2 lines
DENX Software Engineering GmbH,        Managing Director: Erika Unter
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
#3 lore raw
Quoted 6 lines · CET Pavel Machek via lists.cip-project.org
Quoted 1 line

Sorry, I saw these five minutes too late. Applied for the next release.

I guess these should go to 4.4-st, but I don't see them there. Did you forget to push?

I don't push the tree unless there is a release or an rc. I don't think we get enough external input ATM to warrant queue branches or something like that.

CU
Uli