[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [PATCH v3 3/9] target/ppc: Fix instruction loading endianness in ali

From: Daniel Henrique Barboza
Subject: Re: [PATCH v3 3/9] target/ppc: Fix instruction loading endianness in alignment interrupt
Date: Tue, 16 May 2023 12:46:55 -0300
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101 Thunderbird/102.10.0

Either this patch or patch 6 broke a gitlab KVM builder (cross-ppc64el-kvm-only)
as follows:

[1441/2019] Compiling C object 
FAILED: libqemu-ppc-softmmu.fa.p/target_ppc_excp_helper.c.o
powerpc64le-linux-gnu-gcc -m64 -mlittle-endian -Ilibqemu-ppc-softmmu.fa.p -I. -I.. -Itarget/ppc 
-I../target/ppc -Iqapi -Itrace -Iui -Iui/shader -I/usr/include/pixman-1 -I/usr/include/capstone 
-I/usr/include/spice-server -I/usr/include/spice-1 -I/usr/include/glib-2.0 
-I/usr/lib/powerpc64le-linux-gnu/glib-2.0/include -fdiagnostics-color=auto -Wall -Winvalid-pch 
-Werror -std=gnu11 -O2 -g -isystem /builds/danielhb/qemu/linux-headers -isystem linux-headers 
-iquote . -iquote /builds/danielhb/qemu -iquote /builds/danielhb/qemu/include -pthread 
-fno-strict-aliasing -fno-common -fwrapv -Wundef -Wwrite-strings -Wmissing-prototypes 
-Wstrict-prototypes -Wredundant-decls -Wold-style-declaration -Wold-style-definition -Wtype-limits 
-Wformat-security -Wformat-y2k -Winit-self -Wignored-qualifiers -Wempty-body -Wnested-externs 
-Wendif-labels -Wexpansion-to-defined -Wimplicit-fallthrough=2 -Wmissing-format-attribute 
-Wno-missing-include-dirs -Wno-shift-negative-value -Wno-psabi -fstack-protector-strong -fPIE 
-isystem../linux-headers -isystemlinux-headers -DNEED_CPU_H 
'-DCONFIG_DEVICES="ppc-softmmu-config-devices.h"' -MD -MQ 
libqemu-ppc-softmmu.fa.p/target_ppc_excp_helper.c.o -MF 
libqemu-ppc-softmmu.fa.p/target_ppc_excp_helper.c.o.d -o 
libqemu-ppc-softmmu.fa.p/target_ppc_excp_helper.c.o -c ../target/ppc/excp_helper.c
../target/ppc/excp_helper.c:143:49: error: unknown type name ‘abi_ptr’; did you 
mean ‘si_ptr’?
  143 | static uint32_t ppc_ldl_code(CPUArchState *env, abi_ptr addr)
      |                                                 ^~~~~~~
      |                                                 si_ptr
[1442/2019] Compiling C object 
ninja: build stopped: subcommand failed.
make: *** [Makefile:165: run-ninja] Error 1

More details here:


I suppose we're missing an ifdef somewhere to gate this code from KVM code. 
is a TCG pointer afaik.



On 5/15/23 06:26, Nicholas Piggin wrote:
powerpc ifetch endianness depends on MSR[LE] so it has to byteswap
after cpu_ldl_code(). This corrects DSISR bits in alignment
interrupts when running in little endian mode.

Reviewed-by: Fabiano Rosas <farosas@suse.de>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Since v2: no change.

  target/ppc/excp_helper.c | 20 +++++++++++++++++++-
  1 file changed, 19 insertions(+), 1 deletion(-)

diff --git a/target/ppc/excp_helper.c b/target/ppc/excp_helper.c
index 199328f4b6..bc2be4a726 100644
--- a/target/ppc/excp_helper.c
+++ b/target/ppc/excp_helper.c
@@ -133,6 +133,24 @@ static void dump_hcall(CPUPPCState *env)
+/* Return true iff byteswap is needed in a scalar memop */
+static inline bool need_byteswap(CPUArchState *env)
+    /* SOFTMMU builds TARGET_BIG_ENDIAN. Need to swap when MSR[LE] is set */
+    return !!(env->msr & ((target_ulong)1 << MSR_LE));
+static uint32_t ppc_ldl_code(CPUArchState *env, abi_ptr addr)
+    uint32_t insn = cpu_ldl_code(env, addr);
+    if (need_byteswap(env)) {
+        insn = bswap32(insn);
+    }
+    return insn;
  static void ppc_excp_debug_sw_tlb(CPUPPCState *env, int excp)
      const char *es;
@@ -3097,7 +3115,7 @@ void ppc_cpu_do_unaligned_access(CPUState *cs, vaddr 
/* Restore state and reload the insn we executed, for filling in DSISR. */
      cpu_restore_state(cs, retaddr);
-    insn = cpu_ldl_code(env, env->nip);
+    insn = ppc_ldl_code(env, env->nip);
switch (env->mmu_model) {
      case POWERPC_MMU_SOFT_4xx:

reply via email to

[Prev in Thread] Current Thread [Next in Thread]