Re: [PATCH v2 2/3] arm64: fpsimd: Fix bad si_code for undiagnosed SIGFPE

On Fri, Mar 09, 2018 at 01:10:18PM +0000, Will Deacon wrote:

Hi Eric,

On Thu, Mar 08, 2018 at 04:40:12PM -0600, Eric W. Biederman wrote:

quoted

Will Deacon [off-list ref] writes:

quoted

On Thu, Mar 01, 2018 at 05:44:07PM +0000, Dave Martin wrote:

quoted

Currently a SIGFPE delivered in response to a floating-point
exception trap may have si_code set to 0 on arm64.  As reported by
Eric, this is a bad idea since this is the value of SI_USER -- yet
this signal is definitely not the result of kill(2), tgkill(2) etc.
and si_uid and si_pid make limited sense whereas we do want to
yield a value for si_addr (which doesn't exist for SI_USER).

[...]

quoted

diff --git a/arch/arm64/kernel/fpsimd.c b/arch/arm64/kernel/fpsimd.c
index e7226c4..9040038 100644
--- a/arch/arm64/kernel/fpsimd.c
+++ b/arch/arm64/kernel/fpsimd.c

@@ -39,6 +39,7 @@
 #include <linux/slab.h>
 #include <linux/sysctl.h>
 
+#include <asm/esr.h>
 #include <asm/fpsimd.h>
 #include <asm/cputype.h>
 #include <asm/simd.h>

@@ -867,18 +868,20 @@ asmlinkage void do_fpsimd_acc(unsigned int esr, struct pt_regs *regs)
 asmlinkage void do_fpsimd_exc(unsigned int esr, struct pt_regs *regs)
 {
 	siginfo_t info;
-	unsigned int si_code = FPE_FIXME;
-
-	if (esr & FPEXC_IOF)
-		si_code = FPE_FLTINV;
-	else if (esr & FPEXC_DZF)
-		si_code = FPE_FLTDIV;
-	else if (esr & FPEXC_OFF)
-		si_code = FPE_FLTOVF;
-	else if (esr & FPEXC_UFF)
-		si_code = FPE_FLTUND;
-	else if (esr & FPEXC_IXF)
-		si_code = FPE_FLTRES;
+	unsigned int si_code = FPE_FLTUNK;

Happy to take this patch once the dependency on FPE_FLTUNK in core code is
resolved.

Would it help for me to take the FPE_FLTUNK patch into my siginfo-next
branch?   So that there is a common branch with the code so we don't
need to worry about conflicts.  If so I will look at that on Monday.

Yes please, that would be helpful actually. I can then pull that into
arm64 if you give me a stable branch or tag. Alternatively, I can define
FPE_FLTUNK locally and remove it at -rc1.

OK Eric, let me know if you need me to rebase anything for the first 2
patches.

Cheers
---Dave