From: "Ulrich Weigand" <uweigand@de.ibm.com>
To: michael.hope@linaro.org (Michael Hope)
Cc: gcc-patches@gcc.gnu.org, rearnsha@arm.com,
ramana.radhakrishnan@linaro.org, patches@linaro.org
Subject: Re: [patch, arm] Fix PR target/50305 (arm_legitimize_reload_address problem)
Date: Fri, 23 Sep 2011 17:28:00 -0000 [thread overview]
Message-ID: <201109232303.p8NN3Gjt018587@d06av02.portsmouth.uk.ibm.com> (raw)
In-Reply-To: <CANLjY-nad=ndSvOAGURuRkDDrOhfn=fG-mL_BH3YMVedh_UJ5w@mail.gmail.com> from "Michael Hope" at Sep 12, 2011 09:50:45 AM
Michael Hope wrote:
> On Sat, Sep 10, 2011 at 5:04 AM, Ulrich Weigand <uweigand@de.ibm.com> wrote:
> > the problem in PR 50305 turned out to be caused by the ARM back-end
> > LEGITIMIZE_RELOAD_ADDRESS implementation.
>
> Interesting the fault goes away with -mfpu=neon, perhaps due to the DI
> mode operations getting pushed out into NEON registers. You might
> want to be explicit about the FPU in dg-options.
Hmm, reload problems tend to be very difficult to reproduce in general,
everything need to line up just so ... For some reason, I didn't see
the -mfpu=neon effect you describe; I'm seeing the reload failure with
that setting as well.
Nevertheless, I've updated the dg-options as suggested. I've also added
a dg-skip-if line to prevent problems when using a -march multilib (as
currently discussed in other threads ...).
Richard/Ramana, any thoughts on this? Is this OK?
Thanks,
Ulrich
ChangeLog:
gcc/
PR target/50305
* config/arm/arm.c (arm_legitimize_reload_address): Recognize
output of a previous pass through legitimize_reload_address.
Do not attempt to optimize addresses if the base register is
equivalent to a constant.
gcc/testsuite/
PR target/50305
* gcc.target/arm/pr50305.c: New test.
Index: gcc/testsuite/gcc.target/arm/pr50305.c
===================================================================
--- gcc/testsuite/gcc.target/arm/pr50305.c (revision 0)
+++ gcc/testsuite/gcc.target/arm/pr50305.c (revision 0)
@@ -0,0 +1,60 @@
+/* { dg-do compile } */
+/* { dg-skip-if "incompatible options" { arm*-*-* } { "-march=*" } { "-march=armv7-a" } } */
+/* { dg-options "-O2 -fno-omit-frame-pointer -marm -march=armv7-a -mfpu=vfp3" } */
+
+struct event {
+ unsigned long long id;
+ unsigned int flag;
+};
+
+void dummy(void)
+{
+ /* This is here to ensure that the offset of perf_event_id below
+ relative to the LANCHOR symbol exceeds the allowed displacement. */
+ static int __warned[300];
+ __warned[0] = 1;
+}
+
+extern void *kmem_cache_alloc_trace (void *cachep);
+extern void *cs_cachep;
+extern int nr_cpu_ids;
+
+struct event *
+event_alloc (int cpu)
+{
+ static unsigned long long __attribute__((aligned(8))) perf_event_id;
+ struct event *event;
+ unsigned long long result;
+ unsigned long tmp;
+
+ if (cpu >= nr_cpu_ids)
+ return 0;
+
+ event = kmem_cache_alloc_trace (cs_cachep);
+
+ __asm__ __volatile__ ("dmb" : : : "memory");
+
+ __asm__ __volatile__("@ atomic64_add_return\n"
+"1: ldrexd %0, %H0, [%3]\n"
+" adds %0, %0, %4\n"
+" adc %H0, %H0, %H4\n"
+" strexd %1, %0, %H0, [%3]\n"
+" teq %1, #0\n"
+" bne 1b"
+ : "=&r" (result), "=&r" (tmp), "+Qo" (perf_event_id)
+ : "r" (&perf_event_id), "r" (1LL)
+ : "cc");
+
+ __asm__ __volatile__ ("dmb" : : : "memory");
+
+ event->id = result;
+
+ if (cpu)
+ event->flag = 1;
+
+ for (cpu = 0; cpu < nr_cpu_ids; cpu++)
+ kmem_cache_alloc_trace (cs_cachep);
+
+ return event;
+}
+
Index: gcc/config/arm/arm.c
===================================================================
--- gcc/config/arm/arm.c (revision 179082)
+++ gcc/config/arm/arm.c (working copy)
@@ -6550,9 +6550,26 @@
int opnum, int type,
int ind_levels ATTRIBUTE_UNUSED)
{
+ /* We must recognize output that we have already generated ourselves. */
if (GET_CODE (*p) == PLUS
+ && GET_CODE (XEXP (*p, 0)) == PLUS
+ && GET_CODE (XEXP (XEXP (*p, 0), 0)) == REG
+ && GET_CODE (XEXP (XEXP (*p, 0), 1)) == CONST_INT
+ && GET_CODE (XEXP (*p, 1)) == CONST_INT)
+ {
+ push_reload (XEXP (*p, 0), NULL_RTX, &XEXP (*p, 0), NULL,
+ MODE_BASE_REG_CLASS (mode), GET_MODE (*p),
+ VOIDmode, 0, 0, opnum, (enum reload_type) type);
+ return true;
+ }
+
+ if (GET_CODE (*p) == PLUS
&& GET_CODE (XEXP (*p, 0)) == REG
&& ARM_REGNO_OK_FOR_BASE_P (REGNO (XEXP (*p, 0)))
+ /* If the base register is equivalent to a constant, let the generic
+ code handle it. Otherwise we will run into problems if a future
+ reload pass decides to rematerialize the constant. */
+ && !reg_equiv_constant (ORIGINAL_REGNO (XEXP (*p, 0)))
&& GET_CODE (XEXP (*p, 1)) == CONST_INT)
{
HOST_WIDE_INT val = INTVAL (XEXP (*p, 1));
--
Dr. Ulrich Weigand
GNU Toolchain for Linux on System z and Cell BE
Ulrich.Weigand@de.ibm.com
next prev parent reply other threads:[~2011-09-23 16:03 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-09-09 18:14 Ulrich Weigand
2011-09-11 23:26 ` Michael Hope
2011-09-23 17:28 ` Ulrich Weigand [this message]
2011-09-26 10:26 ` Ramana Radhakrishnan
2011-09-26 14:55 ` Ulrich Weigand
2011-09-26 18:02 ` Ramana Radhakrishnan
2011-10-04 15:22 ` Ulrich Weigand
2011-10-06 9:28 ` Ramana Radhakrishnan
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=201109232303.p8NN3Gjt018587@d06av02.portsmouth.uk.ibm.com \
--to=uweigand@de.ibm.com \
--cc=gcc-patches@gcc.gnu.org \
--cc=michael.hope@linaro.org \
--cc=patches@linaro.org \
--cc=ramana.radhakrishnan@linaro.org \
--cc=rearnsha@arm.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).