| Message ID | 1336333113-773-1-git-send-email-andreas.faerber@web.de |
|---|---|
| State | New |
| Headers | show |
On Sun, 6 May 2012, Andreas F?rber wrote: > For targets where TARGET_LONG_BITS != 32, i.e. 64-bit guests, > addr_reg is moved to r4. For hosts with TCG_TARGET_CALL_ALIGN_ARGS > either data_reg2 or data_reg or a masked version thereof would overwrite > r4. Place it in r5 instead, matching TCG_TARGET_CALL_ALIGN_ARGS hosts. > > This might explain crashes observed on Darwin/ppc but not on ppc64. > > Signed-off-by: Andreas F?rber <andreas.faerber@web.de> > --- > tcg/ppc/tcg-target.c | 4 ---- > 1 files changed, 0 insertions(+), 4 deletions(-) > > diff --git a/tcg/ppc/tcg-target.c b/tcg/ppc/tcg-target.c > index dc40716..4b85c89 100644 > --- a/tcg/ppc/tcg-target.c > +++ b/tcg/ppc/tcg-target.c > @@ -816,11 +816,7 @@ static void tcg_out_qemu_st (TCGContext *s, const TCGArg *args, int opc) > #else > tcg_out_mov (s, TCG_TYPE_I32, 3, addr_reg2); > tcg_out_mov (s, TCG_TYPE_I32, 4, addr_reg); > -#ifdef TCG_TARGET_CALL_ALIGN_ARGS > ir = 5; > -#else > - ir = 4; > -#endif > #endif > > switch (opc) { > Right you are.
Am 06.05.2012 21:38, schrieb Andreas Färber: > For targets where TARGET_LONG_BITS != 32, i.e. 64-bit guests, > addr_reg is moved to r4. For hosts with TCG_TARGET_CALL_ALIGN_ARGS Er, obviously I meant "without". :) /-F > either data_reg2 or data_reg or a masked version thereof would overwrite > r4. Place it in r5 instead, matching TCG_TARGET_CALL_ALIGN_ARGS hosts. > > This might explain crashes observed on Darwin/ppc but not on ppc64. > > Signed-off-by: Andreas Färber <andreas.faerber@web.de> > --- > tcg/ppc/tcg-target.c | 4 ---- > 1 files changed, 0 insertions(+), 4 deletions(-) > > diff --git a/tcg/ppc/tcg-target.c b/tcg/ppc/tcg-target.c > index dc40716..4b85c89 100644 > --- a/tcg/ppc/tcg-target.c > +++ b/tcg/ppc/tcg-target.c > @@ -816,11 +816,7 @@ static void tcg_out_qemu_st (TCGContext *s, const TCGArg *args, int opc) > #else > tcg_out_mov (s, TCG_TYPE_I32, 3, addr_reg2); > tcg_out_mov (s, TCG_TYPE_I32, 4, addr_reg); > -#ifdef TCG_TARGET_CALL_ALIGN_ARGS > ir = 5; > -#else > - ir = 4; > -#endif > #endif > > switch (opc) {
Am 06.05.2012 22:01, schrieb malc: > On Sun, 6 May 2012, Andreas F?rber wrote: > >> For targets where TARGET_LONG_BITS != 32, i.e. 64-bit guests, >> addr_reg is moved to r4. For hosts [without] TCG_TARGET_CALL_ALIGN_ARGS >> either data_reg2 or data_reg or a masked version thereof would overwrite >> r4. Place it in r5 instead, matching TCG_TARGET_CALL_ALIGN_ARGS hosts. >> >> This might explain crashes observed on Darwin/ppc but not on ppc64. >> >> Signed-off-by: Andreas F?rber <andreas.faerber@web.de> >> --- >> tcg/ppc/tcg-target.c | 4 ---- >> 1 files changed, 0 insertions(+), 4 deletions(-) >> >> diff --git a/tcg/ppc/tcg-target.c b/tcg/ppc/tcg-target.c >> index dc40716..4b85c89 100644 >> --- a/tcg/ppc/tcg-target.c >> +++ b/tcg/ppc/tcg-target.c >> @@ -816,11 +816,7 @@ static void tcg_out_qemu_st (TCGContext *s, const TCGArg *args, int opc) >> #else >> tcg_out_mov (s, TCG_TYPE_I32, 3, addr_reg2); >> tcg_out_mov (s, TCG_TYPE_I32, 4, addr_reg); >> -#ifdef TCG_TARGET_CALL_ALIGN_ARGS >> ir = 5; >> -#else >> - ir = 4; >> -#endif >> #endif >> >> switch (opc) { >> > > Right you are. Confirmed with qemu-system-x86_64 on Darwin/ppc - hangs without, works with patch. Thanks for making me aware of that line of code. :) I'll have a suggestion how to simplify your updated patch and will resubmit fixed as [PATCH] series after testing. Hoping you can get a non-Umlaut-broken version from Patchwork. ;) Cheers, Andreas
diff --git a/tcg/ppc/tcg-target.c b/tcg/ppc/tcg-target.c index dc40716..4b85c89 100644 --- a/tcg/ppc/tcg-target.c +++ b/tcg/ppc/tcg-target.c @@ -816,11 +816,7 @@ static void tcg_out_qemu_st (TCGContext *s, const TCGArg *args, int opc) #else tcg_out_mov (s, TCG_TYPE_I32, 3, addr_reg2); tcg_out_mov (s, TCG_TYPE_I32, 4, addr_reg); -#ifdef TCG_TARGET_CALL_ALIGN_ARGS ir = 5; -#else - ir = 4; -#endif #endif switch (opc) {
For targets where TARGET_LONG_BITS != 32, i.e. 64-bit guests, addr_reg is moved to r4. For hosts with TCG_TARGET_CALL_ALIGN_ARGS either data_reg2 or data_reg or a masked version thereof would overwrite r4. Place it in r5 instead, matching TCG_TARGET_CALL_ALIGN_ARGS hosts. This might explain crashes observed on Darwin/ppc but not on ppc64. Signed-off-by: Andreas Färber <andreas.faerber@web.de> --- tcg/ppc/tcg-target.c | 4 ---- 1 files changed, 0 insertions(+), 4 deletions(-)