From mboxrd@z Thu Jan 1 00:00:00 1970 From: jon.medhurst@linaro.org (Jon Medhurst (Tixy)) Date: Thu, 22 Sep 2011 14:19:37 +0100 Subject: [PATCH] arm: Add unwinding annotations for 64bit division functions In-Reply-To: <20110922130031.GK12025@e102109-lin.cambridge.arm.com> References: <1316470297-5063-1-git-send-email-lauraa@codeaurora.org> <2285dff3fee56758b6279062a5a30dc7.squirrel@www.codeaurora.org> <20110921113906.GB2872@arm.com> <20110921115553.GF17169@n2100.arm.linux.org.uk> <1316676488.2053.9.camel@linaro1> <1316689606.2053.29.camel@linaro1> <20110922115713.GJ12025@e102109-lin.cambridge.arm.com> <1316693581.2053.38.camel@linaro1> <20110922130031.GK12025@e102109-lin.cambridge.arm.com> Message-ID: <1316697577.2053.48.camel@linaro1> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On Thu, 2011-09-22 at 14:00 +0100, Catalin Marinas wrote: > The unwinding fix should be simple (I haven't tested it yet): > > 8<----------------------------- > ARM: Ignore the unwinding information for the first instruction in a function > > From: Catalin Marinas > > When backtracing from the first instruction of a function, the prologue > has not been executed and the unwinding information is not valid. This > patch checks for this case and just assumes that the return address is > in LR. > > Signed-off-by: Catalin Marinas > --- > arch/arm/kernel/unwind.c | 10 ++++++++++ > 1 files changed, 10 insertions(+), 0 deletions(-) > > diff --git a/arch/arm/kernel/unwind.c b/arch/arm/kernel/unwind.c > index d2cb0b3..946face 100644 > --- a/arch/arm/kernel/unwind.c > +++ b/arch/arm/kernel/unwind.c > @@ -293,6 +293,16 @@ int unwind_frame(struct stackframe *frame) > return -URC_FAILURE; > } > > + /* > + * Check for backtrace on the first instruction of a function. The > + * prologue has not been executed yet and the unwinding information is > + * not valid. Assume that the return address is in LR. > + */ > + if (idx.addr == frame->pc) { > + frame->pc = frame->lr; > + return URC_OK; > + } > + > ctrl.vrs[FP] = frame->fp; > ctrl.vrs[SP] = frame->sp; > ctrl.vrs[LR] = frame->lr; > I've never looked at the unwinding code before but the one comment I would make is: does the patch work with Thumb code? I.e. does bit zero of idx.addr, frame->pc or frame->lr ever get set to indicate Thumb state? And if so, they had better all get set otherwise it won't work :-) -- Tixy