Skip to contentSkip to conversation

target/ppc: don't byte swap ELFv2 signal handler

PATCH Superseded Vincent Fazio 3 participants 3 messages Mar 18, 2020 – Mar 19, 2020

v2 of 3 Superseded Latest is v3 →

lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 target/ppc: don't byte swap ELFv2 signal handler

From: Vincent Fazio <[email protected]>

Previously, the signal handler would be byte swapped if the target and host CPU used different endianness. This would cause a SIGSEGV when attempting to translate the opcode pointed to by the swapped address.

Thread 1 "qemu-ppc64" received signal SIGSEGV, Segmentation fault. 0x00000000600a9257 in ldl_he_p (ptr=0x4c2c061000000000) at qemu/include/qemu/bswap.h:351
351 __builtin_memcpy(&r, ptr, sizeof(r));

 #0  0x00000000600a9257 in ldl_he_p (ptr=0x4c2c061000000000) at qemu/include/qemu/bswap.h:351
 #1  0x00000000600a92fe in ldl_be_p (ptr=0x4c2c061000000000) at qemu/include/qemu/bswap.h:449
 #2  0x00000000600c0790 in translator_ldl_swap at qemu/include/exec/translator.h:201
 #3  0x000000006011c1ab in ppc_tr_translate_insn at qemu/target/ppc/translate.c:7856
 #4  0x000000006005ae70 in translator_loop at qemu/accel/tcg/translator.c:102

Now, no swap is performed and execution continues properly.

---
Changes since v1:
- Drop host/target endianness callouts
- Drop unnecessary pointer cast
- Clarify commit message

 linux-user/ppc/signal.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)
linux-user/ppc/signal.c +2 -4
index 5b82af6cb6..b8613c5e1b 100644
@@ -567,10 +567,8 @@ void setup_rt_frame(int sig, struct target_sigaction *ka,
         env->nip = tswapl(handler->entry);
         env->gpr[2] = tswapl(handler->toc);
     } else {
-        /* ELFv2 PPC64 function pointers are entry points, but R12
-         * must also be set */
-        env->nip = tswapl((target_ulong) ka->_sa_handler);
-        env->gpr[12] = env->nip;
+        /* ELFv2 PPC64 function pointers are entry points. R12 must also be set. */
+        env->gpr[12] = env->nip = ka->_sa_handler;
     }
 #else
     env->nip = (target_ulong) ka->_sa_handler;
Signature · 1 line
2.25.1
Reviewed-by on v2 1/1
#1 lore raw
Quoted 18 lines · Vincent Fazio

From: Vincent Fazio <[email protected]>

Previously, the signal handler would be byte swapped if the target and host CPU used different endianness. This would cause a SIGSEGV when attempting to translate the opcode pointed to by the swapped address.

Thread 1 "qemu-ppc64" received signal SIGSEGV, Segmentation fault. 0x00000000600a9257 in ldl_he_p (ptr=0x4c2c061000000000) at qemu/include/qemu/bswap.h:351
351 __builtin_memcpy(&r, ptr, sizeof(r));

 #0  0x00000000600a9257 in ldl_he_p (ptr=0x4c2c061000000000) at qemu/include/qemu/bswap.h:351
 #1  0x00000000600a92fe in ldl_be_p (ptr=0x4c2c061000000000) at qemu/include/qemu/bswap.h:449
 #2  0x00000000600c0790 in translator_ldl_swap at qemu/include/exec/translator.h:201
 #3  0x000000006011c1ab in ppc_tr_translate_insn at qemu/target/ppc/translate.c:7856
 #4  0x000000006005ae70 in translator_loop at qemu/accel/tcg/translator.c:102

Now, no swap is performed and execution continues properly.

Quoted 7 lines

---
Changes since v1:
- Drop host/target endianness callouts
- Drop unnecessary pointer cast
- Clarify commit message

r~

#2 lore raw
In reply to Vincent Fazio
Quoted 10 lines · Vincent Fazio

From: Vincent Fazio <[email protected]>

Previously, the signal handler would be byte swapped if the target and host CPU used different endianness. This would cause a SIGSEGV when attempting to translate the opcode pointed to by the swapped address.

Thread 1 "qemu-ppc64" received signal SIGSEGV, Segmentation fault. 0x00000000600a9257 in ldl_he_p (ptr=0x4c2c061000000000) at qemu/include/qemu/bswap.h:351
351 __builtin_memcpy(&r, ptr, sizeof(r));

Quoted 7 lines
 #0  0x00000000600a9257 in ldl_he_p (ptr=0x4c2c061000000000) at qemu/include/qemu/bswap.h:351
 #1  0x00000000600a92fe in ldl_be_p (ptr=0x4c2c061000000000) at qemu/include/qemu/bswap.h:449
 #2  0x00000000600c0790 in translator_ldl_swap at qemu/include/exec/translator.h:201
 #3  0x000000006011c1ab in ppc_tr_translate_insn at qemu/target/ppc/translate.c:7856
 #4  0x000000006005ae70 in translator_loop at qemu/accel/tcg/translator.c:102

Now, no swap is performed and execution continues properly.

Patch is good, but the message still needs some work. It should point out that the necessary swap is already done at sigaction() time by the __get_user().

Quoted 29 lines

---
Changes since v1:
- Drop host/target endianness callouts
- Drop unnecessary pointer cast
- Clarify commit message

 linux-user/ppc/signal.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)
linux-user/ppc/signal.c +2 -4
index 5b82af6cb6..b8613c5e1b 100644
@@ -567,10 +567,8 @@ void setup_rt_frame(int sig, struct target_sigaction *ka,
         env->nip = tswapl(handler->entry);
         env->gpr[2] = tswapl(handler->toc);
     } else {
-        /* ELFv2 PPC64 function pointers are entry points, but R12
-         * must also be set */
-        env->nip = tswapl((target_ulong) ka->_sa_handler);
-        env->gpr[12] = env->nip;
+        /* ELFv2 PPC64 function pointers are entry points. R12 must also be set. */
+        env->gpr[12] = env->nip = ka->_sa_handler;
     }
 #else
     env->nip = (target_ulong) ka->_sa_handler;
Signature · 4 lines
David Gibson			| I'll have my music baroque, and my code
david AT gibson.dropbear.id.au	| minimalist, thank you.  NOT _the_ _other_
				| _way_ _around_!
http://www.ozlabs.org/~dgibson