[PATCH AUTOSEL 5.0 09/34] ARC: PAE40: don't panic and instead turn off hw ioc

[PATCH AUTOSEL 5.0 09/34] ARC: PAE40: don't panic and instead turn off hw ioc

[Sasha Levin]

From: Vineet Gupta <vgupta@synopsys.com>

[ Upstream commit 99bd5fcc505d65ea9c60619202f0b2d926eabbe9 ]

HSDK currently panics when built for HIGHMEM/ARC_HAS_PAE40 because ioc
is enabled with default which doesn't work for the 2 non contiguous
memory nodes. So get PAE working by disabling ioc instead.

Tested with !PAE40 by forcing @ioc_enable=0 and running the glibc
testsuite over ssh

Signed-off-by: Vineet Gupta <vgupta at synopsys.com>
Signed-off-by: Sasha Levin <sashal at kernel.org>
---
 arch/arc/mm/cache.c | 31 ++++++++++++++++---------------
 1 file changed, 16 insertions(+), 15 deletions(-)

diff --git a/arch/arc/mm/cache.c b/arch/arc/mm/cache.c
index 4135abec3fb09..63e6e65046992 100644
--- a/arch/arc/mm/cache.c
+++ b/arch/arc/mm/cache.c
@@ -113,10 +113,24 @@ static void read_decode_cache_bcr_arcv2(int cpu)
 	}
 
 	READ_BCR(ARC_REG_CLUSTER_BCR, cbcr);
-	if (cbcr.c)
+	if (cbcr.c) {
 		ioc_exists = 1;
-	else
+
+		/*
+		 * As for today we don't support both IOC and ZONE_HIGHMEM enabled
+		 * simultaneously. This happens because as of today IOC aperture covers
+		 * only ZONE_NORMAL (low mem) and any dma transactions outside this
+		 * region won't be HW coherent.
+		 * If we want to use both IOC and ZONE_HIGHMEM we can use
+		 * bounce_buffer to handle dma transactions to HIGHMEM.
+		 * Also it is possible to modify dma_direct cache ops or increase IOC
+		 * aperture size if we are planning to use HIGHMEM without PAE.
+		 */
+		if (IS_ENABLED(CONFIG_HIGHMEM) || is_pae40_enabled())
+			ioc_enable = 0;
+	} else {
 		ioc_enable = 0;
+	}
 
 	/* HS 2.0 didn't have AUX_VOL */
 	if (cpuinfo_arc700[cpu].core.family > 0x51) {
@@ -1158,19 +1172,6 @@ noinline void __init arc_ioc_setup(void)
 	if (!ioc_enable)
 		return;
 
-	/*
-	 * As for today we don't support both IOC and ZONE_HIGHMEM enabled
-	 * simultaneously. This happens because as of today IOC aperture covers
-	 * only ZONE_NORMAL (low mem) and any dma transactions outside this
-	 * region won't be HW coherent.
-	 * If we want to use both IOC and ZONE_HIGHMEM we can use
-	 * bounce_buffer to handle dma transactions to HIGHMEM.
-	 * Also it is possible to modify dma_direct cache ops or increase IOC
-	 * aperture size if we are planning to use HIGHMEM without PAE.
-	 */
-	if (IS_ENABLED(CONFIG_HIGHMEM))
-		panic("IOC and HIGHMEM can't be used simultaneously");
-
 	/* Flush + invalidate + disable L1 dcache */
 	__dc_disable();
 
-- 
2.20.1

[PATCH AUTOSEL 5.0 32/34] perf bench numa: Add define for RUSAGE_THREAD if not present

[Sasha Levin]

From: Arnaldo Carvalho de Melo <acme@redhat.com>

[ Upstream commit bf561d3c13423fc54daa19b5d49dc15fafdb7acc ]

While cross building perf to the ARC architecture on a fedora 30 host,
we were failing with:

      CC       /tmp/build/perf/bench/numa.o
  bench/numa.c: In function ?worker_thread?:
  bench/numa.c:1261:12: error: ?RUSAGE_THREAD? undeclared (first use in this function); did you mean ?SIGEV_THREAD??
    getrusage(RUSAGE_THREAD, &rusage);
              ^~~~~~~~~~~~~
              SIGEV_THREAD
  bench/numa.c:1261:12: note: each undeclared identifier is reported only once for each function it appears in

[perfbuilder at 60d5802468f6 perf]$ /arc_gnu_2019.03-rc1_prebuilt_uclibc_le_archs_linux_install/bin/arc-linux-gcc --version | head -1
arc-linux-gcc (ARCv2 ISA Linux uClibc toolchain 2019.03-rc1) 8.3.1 20190225
[perfbuilder at 60d5802468f6 perf]$

Trying to reproduce a report by Vineet, I noticed that, with just
cross-built zlib and numactl libraries, I ended up with the above
failure.

So, since RUSAGE_THREAD is available as a define, check for that and
numactl libraries, I ended up with the above failure.

So, since RUSAGE_THREAD is available as a define in the system headers,
check if it is defined in the 'perf bench numa' sources and define it if
not.

Now it builds and I have to figure out if the problem reported by Vineet
only takes place if we have libelf or some other library available.

Cc: Arnd Bergmann <arnd at arndb.de>
Cc: Jiri Olsa <jolsa at kernel.org>
Cc: linux-snps-arc at lists.infradead.org
Cc: Namhyung Kim <namhyung at kernel.org>
Cc: Vineet Gupta <Vineet.Gupta1 at synopsys.com>
Link: https://lkml.kernel.org/n/tip-2wb4r1gir9xrevbpq7qp0amk at git.kernel.org
Signed-off-by: Arnaldo Carvalho de Melo <acme at redhat.com>
Signed-off-by: Sasha Levin <sashal at kernel.org>
---
 tools/perf/bench/numa.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/tools/perf/bench/numa.c b/tools/perf/bench/numa.c
index 44195514b19e6..fa56fde6e8d80 100644
--- a/tools/perf/bench/numa.c
+++ b/tools/perf/bench/numa.c
@@ -38,6 +38,10 @@
 #include <numa.h>
 #include <numaif.h>
 
+#ifndef RUSAGE_THREAD
+# define RUSAGE_THREAD 1
+#endif
+
 /*
  * Regular printout to the terminal, supressed if -q is specified:
  */
-- 
2.20.1