[dpdk-dev] [PATCH 18/22] timer: option --vmware-tsc-map for VMware guest

[Thomas Monjalon]

From: Damien Millescamps <damien.millescamps at 6wind.com>

The VMware TSC mapping uses a hook to RDPMC to read the physical TSC
in the case of VMware ESXi.

Introduce new option --vmware-tsc-map, ignored if
CONFIG_RTE_LIBRTE_EAL_VMWARE_TSC_MAP_SUPPORT is not set.

Default is CONFIG_RTE_LIBRTE_EAL_VMWARE_TSC_MAP_SUPPORT=y.

if CONFIG_RTE_LIBRTE_EAL_VMWARE_TSC_MAP_SUPPORT is set:
	rte_rdtsc() selects at runtime between Vmware mapping of
	TSC or native TSC
else
	rte_rdtsc() always uses native rdtsc.

When running DPDK on VMware guest, enable --vmware-tsc-map to
read the physical TSC.
Caution: ESXi should pass monitor_control.pseudo_perfctr = TRUE
otherwise it results in general protection fault.

Acked-by: Vincent Jardin <vincent.jardin at 6wind.com>
Acked-by: Olivier Matz <olivier.matz at 6wind.com>
Acked-by: Thomas Monjalon <thomas.monjalon at 6wind.com>
Signed-off-by: Damien Millescamps <damien.millescamps at 6wind.com>
Signed-off-by: Jean-Mickael Guerin <jean-mickael.guerin at 6wind.com>
---
 config/defconfig_i686-default-linuxapp-gcc         |    5 ++++
 config/defconfig_i686-default-linuxapp-icc         |    5 ++++
 config/defconfig_x86_64-default-linuxapp-gcc       |    5 ++++
 config/defconfig_x86_64-default-linuxapp-icc       |    5 ++++
 lib/librte_eal/common/include/rte_cycles.h         |   19 ++++++++++++++++
 lib/librte_eal/linuxapp/eal/eal.c                  |   23 ++++++++++++++++++++
 .../linuxapp/eal/include/eal_internal_cfg.h        |    1 +
 7 files changed, 63 insertions(+), 0 deletions(-)

diff --git a/config/defconfig_i686-default-linuxapp-gcc b/config/defconfig_i686-default-linuxapp-gcc
index ed54434..e64bb86 100644
--- a/config/defconfig_i686-default-linuxapp-gcc
+++ b/config/defconfig_i686-default-linuxapp-gcc
@@ -121,6 +121,11 @@ CONFIG_RTE_LIBRTE_EAL_LINUXAPP=y
 CONFIG_RTE_LIBRTE_EAL_BAREMETAL=n
 
 #
+# Compile Environment Abstraction Layer to support VMware TSC map
+#
+CONFIG_RTE_LIBRTE_EAL_VMWARE_TSC_MAP_SUPPORT=y
+
+#
 # Compile generic ethernet library
 #
 CONFIG_RTE_LIBRTE_ETHER=y
diff --git a/config/defconfig_i686-default-linuxapp-icc b/config/defconfig_i686-default-linuxapp-icc
index cb0d017..3f15a53 100644
--- a/config/defconfig_i686-default-linuxapp-icc
+++ b/config/defconfig_i686-default-linuxapp-icc
@@ -121,6 +121,11 @@ CONFIG_RTE_LIBRTE_EAL_LINUXAPP=y
 CONFIG_RTE_LIBRTE_EAL_BAREMETAL=n
 
 #
+# Compile Environment Abstraction Layer to support VMware TSC map
+#
+CONFIG_RTE_LIBRTE_EAL_VMWARE_TSC_MAP_SUPPORT=y
+
+#
 # Compile generic ethernet library
 #
 CONFIG_RTE_LIBRTE_ETHER=y
diff --git a/config/defconfig_x86_64-default-linuxapp-gcc b/config/defconfig_x86_64-default-linuxapp-gcc
index 3555187..f5213d5 100644
--- a/config/defconfig_x86_64-default-linuxapp-gcc
+++ b/config/defconfig_x86_64-default-linuxapp-gcc
@@ -121,6 +121,11 @@ CONFIG_RTE_LIBRTE_EAL_LINUXAPP=y
 CONFIG_RTE_LIBRTE_EAL_BAREMETAL=n
 
 #
+# Compile Environment Abstraction Layer to support VMware TSC map
+#
+CONFIG_RTE_LIBRTE_EAL_VMWARE_TSC_MAP_SUPPORT=y
+
+#
 # Compile generic ethernet library
 #
 CONFIG_RTE_LIBRTE_ETHER=y
diff --git a/config/defconfig_x86_64-default-linuxapp-icc b/config/defconfig_x86_64-default-linuxapp-icc
index f527f53..f28943a 100644
--- a/config/defconfig_x86_64-default-linuxapp-icc
+++ b/config/defconfig_x86_64-default-linuxapp-icc
@@ -121,6 +121,11 @@ CONFIG_RTE_LIBRTE_EAL_LINUXAPP=y
 CONFIG_RTE_LIBRTE_EAL_BAREMETAL=n
 
 #
+# Compile Environment Abstraction Layer to support VMware TSC map
+#
+CONFIG_RTE_LIBRTE_EAL_VMWARE_TSC_MAP_SUPPORT=y
+
+#
 # Compile generic ethernet library
 #
 CONFIG_RTE_LIBRTE_ETHER=y
diff --git a/lib/librte_eal/common/include/rte_cycles.h b/lib/librte_eal/common/include/rte_cycles.h
index a1eca6c..fa0b408 100644
--- a/lib/librte_eal/common/include/rte_cycles.h
+++ b/lib/librte_eal/common/include/rte_cycles.h
@@ -2,6 +2,7 @@
  *   BSD LICENSE
  * 
  *   Copyright(c) 2010-2012 Intel Corporation. All rights reserved.
+ *   Copyright(c) 2012-2013 6WIND.
  *   All rights reserved.
  * 
  *   Redistribution and use in source and binary forms, with or without 
@@ -48,6 +49,13 @@ extern "C" {
 
 #include <stdint.h>
 
+#ifdef RTE_LIBRTE_EAL_VMWARE_TSC_MAP_SUPPORT
+/** Global switch to use VMware mapping of TSC instead of RDTSC */
+extern int rte_cycles_vmware_tsc_map;
+#include <rte_branch_prediction.h>
+#endif
+
+
 /**
  * Read the TSC register.
  *
@@ -65,9 +73,20 @@ rte_rdtsc(void)
 		};
 	} tsc;
 
+#ifdef RTE_LIBRTE_EAL_VMWARE_TSC_MAP_SUPPORT
+	if (unlikely(rte_cycles_vmware_tsc_map)) {
+		/* ecx = 0x10000 corresponds to the Physical TSC for VMware */
+		asm volatile("rdpmc" :
+		             "=a" (tsc.lo_32),
+		             "=d" (tsc.hi_32) :
+		             "c"(0x10000));
+		return tsc.tsc_64;
+	}
+#endif
 	asm volatile("rdtsc" :
 		     "=a" (tsc.lo_32),
 		     "=d" (tsc.hi_32));
+
 	return tsc.tsc_64;
 }
 
diff --git a/lib/librte_eal/linuxapp/eal/eal.c b/lib/librte_eal/linuxapp/eal/eal.c
index 8d82cc3..a63881c 100644
--- a/lib/librte_eal/linuxapp/eal/eal.c
+++ b/lib/librte_eal/linuxapp/eal/eal.c
@@ -78,6 +78,7 @@
 #define OPT_PROC_TYPE   "proc-type"
 #define OPT_NO_SHCONF   "no-shconf"
 #define OPT_NO_HPET     "no-hpet"
+#define OPT_VMWARE_TSC_MAP   "vmware-tsc-map"
 #define OPT_NO_PCI      "no-pci"
 #define OPT_NO_HUGE     "no-huge"
 #define OPT_FILE_PREFIX "file-prefix"
@@ -125,6 +126,9 @@ struct lcore_config lcore_config[RTE_MAX_LCORE];
 /* internal configuration */
 struct internal_config internal_config;
 
+/* used by rte_rdtsc() */
+int rte_cycles_vmware_tsc_map;
+
 /* Return a pointer to the configuration structure */
 struct rte_config *
 rte_eal_get_configuration(void)
@@ -264,6 +268,7 @@ eal_usage(const char *prgname)
 	       "  --"OPT_HUGE_DIR" : directory where hugetlbfs is mounted\n"
 	       "  --"OPT_PROC_TYPE": type of this process\n"
 	       "  --"OPT_FILE_PREFIX": prefix for hugepage filenames\n"
+	       "  --"OPT_VMWARE_TSC_MAP"  : use VMware TSC map instead of native RDTSC\n\n"
 	       "\nEAL options for DEBUG use only:\n"
 	       "  --"OPT_NO_HUGE"  : use malloc instead of hugetlbfs\n"
 	       "  --"OPT_NO_PCI"   : disable pci\n"
@@ -379,6 +384,7 @@ eal_parse_args(int argc, char **argv)
 		{OPT_NO_HUGE, 0, 0, 0},
 		{OPT_NO_PCI, 0, 0, 0},
 		{OPT_NO_HPET, 0, 0, 0},
+		{OPT_VMWARE_TSC_MAP, 0, 0, 0},
 		{OPT_HUGE_DIR, 1, 0, 0},
 		{OPT_NO_SHCONF, 0, 0, 0},
 		{OPT_PROC_TYPE, 1, 0, 0},
@@ -399,6 +405,8 @@ eal_parse_args(int argc, char **argv)
 	internal_config.no_hpet = 1;
 #endif
 
+	internal_config.vmware_tsc_map = 0;
+
 	while ((opt = getopt_long(argc, argvopt, "b:c:m:n:r:v",
 				  lgopts, &option_index)) != EOF) {
 
@@ -464,6 +472,9 @@ eal_parse_args(int argc, char **argv)
 			else if (!strcmp(lgopts[option_index].name, OPT_NO_HPET)) {
 				internal_config.no_hpet = 1;
 			}
+			else if (!strcmp(lgopts[option_index].name, OPT_VMWARE_TSC_MAP)) {
+				internal_config.vmware_tsc_map = 1;
+			}
 			else if (!strcmp(lgopts[option_index].name, OPT_NO_SHCONF)) {
 				internal_config.no_shconf = 1;
 			}
@@ -547,6 +558,18 @@ rte_eal_init(int argc, char **argv)
 			internal_config.memory = eal_get_hugepage_mem_size();
 	}
 
+	if (internal_config.vmware_tsc_map == 1) {
+#ifdef RTE_LIBRTE_EAL_VMWARE_TSC_MAP_SUPPORT
+		rte_cycles_vmware_tsc_map = 1;
+		RTE_LOG (DEBUG, EAL, "Using VMware TSC map, "
+				"you must have monitor_control.pseudo_perfctr = TRUE\n");
+#else
+		RTE_LOG (WARNING, EAL, "Ignoring --vmware-tsc-map because "
+				"RTE_LIBRTE_EAL_VMWARE_TSC_MAP_SUPPORT is not set\n");
+#endif
+	} else {
+		RTE_LOG (DEBUG, EAL, "Using native RDTSC\n");
+	}
 	rte_srand(rte_rdtsc());
 	rte_config_init();
 
diff --git a/lib/librte_eal/linuxapp/eal/include/eal_internal_cfg.h b/lib/librte_eal/linuxapp/eal/include/eal_internal_cfg.h
index 70d5afb..470b746 100644
--- a/lib/librte_eal/linuxapp/eal/include/eal_internal_cfg.h
+++ b/lib/librte_eal/linuxapp/eal/include/eal_internal_cfg.h
@@ -63,6 +63,7 @@ struct internal_config {
 	volatile unsigned no_hugetlbfs;   /* true to disable hugetlbfs */
 	volatile unsigned no_pci;         /* true to disable PCI */
 	volatile unsigned no_hpet;        /* true to disable HPET */
+	volatile unsigned vmware_tsc_map; /* true to use VMware TSC mapping instead of native TSC */
 	volatile unsigned no_shconf;      /* true if there is no shared config */
 	volatile enum rte_proc_type_t process_type; /* multi-process proc type */
 	const char *hugefile_prefix;      /* the base filename of hugetlbfs files */
-- 
1.7.2.5