diff mbox

Recent Go patch broke Alpha bootstrap

Message ID CAKOQZ8yP+6RV1s-6crc5-KKVoiGbcAyUCVz066UmG2zN6S4xzQ@mail.gmail.com
State New
Headers show

Commit Message

Ian Lance Taylor Nov. 15, 2013, 5:22 p.m. UTC 
On Fri, Nov 15, 2013 at 4:34 AM, Uros Bizjak <ubizjak@gmail.com> wrote:
> On Fri, Nov 15, 2013 at 12:35 PM, Uros Bizjak <ubizjak@gmail.com> wrote:
>
>>>>>>> panic: runtime error: invalid memory address or nil pointer dereference
>>>>>>> [signal 0xb code=0x1 addr=0x1c]
>>>>>
>>>>>>> FAIL: runtime/pprof
>>>>>>> gmake[2]: *** [runtime/pprof/check] Error 1
>>>>>>>
>>>>>>> This one is new, I have to look into it a bit deeper.
>>>>>>
>>>>>>
>>>>>> I don't know what is happening here.  I can't recreate it.  There was
>>>>>> a different problem that could arise in runtime/pprof, that was fixed
>>>>>> by a patch I submitted on Saturday
>>>>>> (http://gcc.gnu.org/ml/gcc-patches/2013-11/msg01016.html).  So it's
>>>>>> possible that this is fixed now.
>>>>>
>>>>> The failure is specific to !USING_SPLIT_STACK targets:
>>>>
>>>> The same error triggered on CentOS 5.10 x86_64 (another
>>>> !USING_SPLIT_STACK target) for 32bit lib (net, runtime). The panic:
>>>> string is the same, only addr=0x9f. There are also a couple of
>>>> segfaults (database/sql, net/http) and abort in sync/atomic.
>>>
>>> Could you check to see if this patch fixes the problem?  Thanks.
>
> Oh, I was not clear.
>
> The patch fixes the net failure on Alpha and Centos 5.10, but other
> failures on 32bit Centos 5.10 remain.

Thanks for testing it.  I've committed the patch to mainline.

As I noted in an earlier e-mail, the 32-bit failures are due to a
general bug in the middle-end, http://gcc.gnu.org/PR59099 .  That bug
is causing a miscompilation of the libgo runtime code, code that is
written in C.  The miscompilation shows up as test failures.

Ian
diff mbox

Patch

diff -r 9b2a1ae08a21 libgo/runtime/proc.c
--- a/libgo/runtime/proc.c	Thu Nov 14 14:29:49 2013 -0800
+++ b/libgo/runtime/proc.c	Thu Nov 14 18:42:54 2013 -0800
@@ -1983,7 +1983,10 @@ 
 #endif
 	gp->gcnext_sp = nil;
 	runtime_memclr(&gp->gcregs, sizeof gp->gcregs);
-	m->p->syscalltick++;
+
+	// Don't refer to m again, we might be running on a different
+	// thread after returning from runtime_mcall.
+	runtime_m()->p->syscalltick++;
 }
 
 static bool