All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH] eal/x86: implement x86 specific tsc hz
@ 2017-08-23 15:00 Sergio Gonzalez Monroy
  2017-08-23 15:00 ` [PATCH] eal/x86: use cpuid builtin Sergio Gonzalez Monroy
                   ` (2 more replies)
  0 siblings, 3 replies; 11+ messages in thread
From: Sergio Gonzalez Monroy @ 2017-08-23 15:00 UTC (permalink / raw)
  To: dev; +Cc: konstantin.ananyev, bruce.richardson

First, try to use CPUID Time Stamp Counter and Nominal Core Crystal
Clock Information Leaf to determine the tsc hz on platforms that
supports it (does not require priviledge user).

If the CPUID leaf is not available, then try to determine the tsc hz by
reading the MSR 0xCE (requires priviledge user).

Default to the tsc hz estimation if both methods fail.

Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
---

DEPENDS on Jerin's patch:
http://dpdk.org/dev/patchwork/patch/27526/

 lib/librte_eal/common/arch/x86/rte_cycles.c        | 142 +++++++++++++++++++++
 .../common/include/arch/x86/rte_cycles.h           |   7 +-
 lib/librte_eal/linuxapp/eal/Makefile               |   1 +
 3 files changed, 145 insertions(+), 5 deletions(-)
 create mode 100644 lib/librte_eal/common/arch/x86/rte_cycles.c

diff --git a/lib/librte_eal/common/arch/x86/rte_cycles.c b/lib/librte_eal/common/arch/x86/rte_cycles.c
new file mode 100644
index 0000000..9336947
--- /dev/null
+++ b/lib/librte_eal/common/arch/x86/rte_cycles.c
@@ -0,0 +1,142 @@
+/*-
+ *   BSD LICENSE
+ *
+ *   Copyright(c) 2017 Intel Corporation. All rights reserved.
+ *   All rights reserved.
+ *
+ *   Redistribution and use in source and binary forms, with or without
+ *   modification, are permitted provided that the following conditions
+ *   are met:
+ *
+ *     * Redistributions of source code must retain the above copyright
+ *       notice, this list of conditions and the following disclaimer.
+ *     * Redistributions in binary form must reproduce the above copyright
+ *       notice, this list of conditions and the following disclaimer in
+ *       the documentation and/or other materials provided with the
+ *       distribution.
+ *     * Neither the name of Intel Corporation nor the names of its
+ *       contributors may be used to endorse or promote products derived
+ *       from this software without specific prior written permission.
+ *
+ *   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ *   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ *   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ *   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ *   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ *   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ *   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ *   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ *   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ *   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ *   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include <fcntl.h>
+#include <unistd.h>
+#include <cpuid.h>
+#include <rte_cycles.h>
+
+static unsigned int
+rte_cpu_get_model(uint32_t fam_mod_step)
+{
+	uint32_t family, model, ext_model;
+
+	family = (fam_mod_step >> 8) & 0xf;
+	model = (fam_mod_step >> 4) & 0xf;
+
+	if (family == 6 || family == 15) {
+		ext_model = (fam_mod_step >> 16) & 0xf;
+		model += (ext_model << 4);
+	}
+
+	return model;
+}
+
+static int32_t
+rdmsr(int msr, uint64_t *val)
+{
+	int fd;
+	int ret = 0;
+
+	fd = open("/dev/cpu/0/msr", O_RDONLY);
+	if (fd < 0)
+		return fd;
+
+	ret = pread(fd, val, sizeof(uint64_t), msr);
+
+	close(fd);
+
+	return ret;
+}
+
+static uint32_t
+check_model_wsm_nhm(uint8_t model)
+{
+	switch (model) {
+	/* Westmere */
+	case 0x25:
+	case 0x2C:
+	case 0x2F:
+	/* Nehalem */
+	case 0x1E:
+	case 0x1F:
+	case 0x1A:
+	case 0x2E:
+		return 1;
+	}
+
+	return 0;
+}
+
+static uint32_t
+check_model_gdm_dnv(uint8_t model)
+{
+	switch (model) {
+	/* Goldmont */
+	case 0x5C:
+	/* Denverton */
+	case 0x5F:
+		return 1;
+	}
+
+	return 0;
+}
+
+uint64_t
+rte_rdtsc_arch_hz(void)
+{
+	uint64_t tsc_hz = 0;
+	uint32_t a, b, c, d, maxleaf;
+	uint8_t mult, model;
+	int32_t ret;
+
+	/*
+	 * Time Stamp Counter and Nominal Core Crystal Clock
+	 * Information Leaf
+	 */
+	maxleaf = __get_cpuid_max(0, NULL);
+
+	if (maxleaf >= 0x15) {
+		__cpuid(0x15, a, b, c, d);
+
+		/* EBX : TSC/Crystal ratio, ECX : Crystal Hz */
+		if (b && c)
+			return c * (b / a);
+	}
+
+	__cpuid(0x1, a, b, c, d);
+	model = rte_cpu_get_model(a);
+
+	if (check_model_wsm_nhm(model))
+		mult = 133;
+	else if ((c & bit_AVX) || check_model_gdm_dnv(model))
+		mult = 100;
+	else
+		return 0;
+
+	ret = rdmsr(0xCE, &tsc_hz);
+	if (!(ret < 0))
+		return ((tsc_hz >> 8) & 0xff) * mult * 1E6;
+
+	return 0;
+}
diff --git a/lib/librte_eal/common/include/arch/x86/rte_cycles.h b/lib/librte_eal/common/include/arch/x86/rte_cycles.h
index e2661e2..0db89dc 100644
--- a/lib/librte_eal/common/include/arch/x86/rte_cycles.h
+++ b/lib/librte_eal/common/include/arch/x86/rte_cycles.h
@@ -84,11 +84,8 @@ rte_rdtsc(void)
  *   The number of rdtsc cycles in one second. Return zero if the architecture
  *   support is not available.
  */
-static inline uint64_t
-rte_rdtsc_arch_hz(void)
-{
-	return 0;
-}
+uint64_t
+rte_rdtsc_arch_hz(void);
 
 static inline uint64_t
 rte_rdtsc_precise(void)
diff --git a/lib/librte_eal/linuxapp/eal/Makefile b/lib/librte_eal/linuxapp/eal/Makefile
index 90bca4d..9d44828 100644
--- a/lib/librte_eal/linuxapp/eal/Makefile
+++ b/lib/librte_eal/linuxapp/eal/Makefile
@@ -104,6 +104,7 @@ SRCS-$(CONFIG_RTE_EXEC_ENV_LINUXAPP) += rte_service.c
 # from arch dir
 SRCS-$(CONFIG_RTE_EXEC_ENV_LINUXAPP) += rte_cpuflags.c
 SRCS-$(CONFIG_RTE_ARCH_X86) += rte_spinlock.c
+SRCS-$(CONFIG_RTE_ARCH_X86) += rte_cycles.c
 
 CFLAGS_eal_common_cpuflags.o := $(CPUFLAGS_LIST)
 
-- 
2.9.5

^ permalink raw reply related	[flat|nested] 11+ messages in thread

* [PATCH] eal/x86: use cpuid builtin
  2017-08-23 15:00 [PATCH] eal/x86: implement x86 specific tsc hz Sergio Gonzalez Monroy
@ 2017-08-23 15:00 ` Sergio Gonzalez Monroy
  2017-10-04 23:39   ` Ferruh Yigit
  2017-09-04  9:38 ` [PATCH] eal/x86: implement x86 specific tsc hz Van Haaren, Harry
  2017-10-02 10:09 ` [PATCH v2] " Sergio Gonzalez Monroy
  2 siblings, 1 reply; 11+ messages in thread
From: Sergio Gonzalez Monroy @ 2017-08-23 15:00 UTC (permalink / raw)
  To: dev; +Cc: konstantin.ananyev, bruce.richardson

GCC does have the __get_cpuid_count builtin which checks for maximum
supported leaf, but implementations differ between CLANG and GCC.

This change provides an implementation compatible with both GCC and
CLANG 3.4+.

Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
---
 lib/librte_eal/common/arch/x86/rte_cpuflags.c | 40 ++++++---------------------
 1 file changed, 8 insertions(+), 32 deletions(-)

diff --git a/lib/librte_eal/common/arch/x86/rte_cpuflags.c b/lib/librte_eal/common/arch/x86/rte_cpuflags.c
index 0138257..7d4a0fe 100644
--- a/lib/librte_eal/common/arch/x86/rte_cpuflags.c
+++ b/lib/librte_eal/common/arch/x86/rte_cpuflags.c
@@ -36,6 +36,7 @@
 #include <stdio.h>
 #include <errno.h>
 #include <stdint.h>
+#include <cpuid.h>
 
 enum cpu_register_t {
 	RTE_REG_EAX = 0,
@@ -156,38 +157,12 @@ const struct feature_entry rte_cpu_feature_table[] = {
 	FEAT_DEF(INVTSC, 0x80000007, 0, RTE_REG_EDX,  8)
 };
 
-/*
- * Execute CPUID instruction and get contents of a specific register
- *
- * This function, when compiled with GCC, will generate architecture-neutral
- * code, as per GCC manual.
- */
-static void
-rte_cpu_get_features(uint32_t leaf, uint32_t subleaf, cpuid_registers_t out)
-{
-#if defined(__i386__) && defined(__PIC__)
-	/* %ebx is a forbidden register if we compile with -fPIC or -fPIE */
-	asm volatile("movl %%ebx,%0 ; cpuid ; xchgl %%ebx,%0"
-		 : "=r" (out[RTE_REG_EBX]),
-		   "=a" (out[RTE_REG_EAX]),
-		   "=c" (out[RTE_REG_ECX]),
-		   "=d" (out[RTE_REG_EDX])
-		 : "a" (leaf), "c" (subleaf));
-#else
-	asm volatile("cpuid"
-		 : "=a" (out[RTE_REG_EAX]),
-		   "=b" (out[RTE_REG_EBX]),
-		   "=c" (out[RTE_REG_ECX]),
-		   "=d" (out[RTE_REG_EDX])
-		 : "a" (leaf), "c" (subleaf));
-#endif
-}
-
 int
 rte_cpu_get_flag_enabled(enum rte_cpu_flag_t feature)
 {
 	const struct feature_entry *feat;
 	cpuid_registers_t regs;
+	unsigned int maxleaf;
 
 	if (feature >= RTE_CPUFLAG_NUMFLAGS)
 		/* Flag does not match anything in the feature tables */
@@ -199,13 +174,14 @@ rte_cpu_get_flag_enabled(enum rte_cpu_flag_t feature)
 		/* This entry in the table wasn't filled out! */
 		return -EFAULT;
 
-	rte_cpu_get_features(feat->leaf & 0xffff0000, 0, regs);
-	if (((regs[RTE_REG_EAX] ^ feat->leaf) & 0xffff0000) ||
-	      regs[RTE_REG_EAX] < feat->leaf)
+	maxleaf = __get_cpuid_max(feat->leaf & 0x80000000, NULL);
+
+	if (maxleaf < feat->leaf)
 		return 0;
 
-	/* get the cpuid leaf containing the desired feature */
-	rte_cpu_get_features(feat->leaf, feat->subleaf, regs);
+	 __cpuid_count(feat->leaf, feat->subleaf,
+			 regs[RTE_REG_EAX], regs[RTE_REG_EBX],
+			 regs[RTE_REG_ECX], regs[RTE_REG_EDX]);
 
 	/* check if the feature is enabled */
 	return (regs[feat->reg] >> feat->bit) & 1;
-- 
2.9.5

^ permalink raw reply related	[flat|nested] 11+ messages in thread

* Re: [PATCH] eal/x86: implement x86 specific tsc hz
  2017-08-23 15:00 [PATCH] eal/x86: implement x86 specific tsc hz Sergio Gonzalez Monroy
  2017-08-23 15:00 ` [PATCH] eal/x86: use cpuid builtin Sergio Gonzalez Monroy
@ 2017-09-04  9:38 ` Van Haaren, Harry
  2017-09-04 10:24   ` Bruce Richardson
  2017-10-02 10:09 ` [PATCH v2] " Sergio Gonzalez Monroy
  2 siblings, 1 reply; 11+ messages in thread
From: Van Haaren, Harry @ 2017-09-04  9:38 UTC (permalink / raw)
  To: Gonzalez Monroy, Sergio, dev; +Cc: Ananyev, Konstantin, Richardson, Bruce

> From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Sergio Gonzalez Monroy
> Sent: Wednesday, August 23, 2017 4:00 PM
> To: dev@dpdk.org
> Cc: Ananyev, Konstantin <konstantin.ananyev@intel.com>; Richardson, Bruce
> <bruce.richardson@intel.com>
> Subject: [dpdk-dev] [PATCH] eal/x86: implement x86 specific tsc hz
> 
> First, try to use CPUID Time Stamp Counter and Nominal Core Crystal
> Clock Information Leaf to determine the tsc hz on platforms that
> supports it (does not require priviledge user).

Checkpatch notifies me that "privilege" is spelled wrong (once above, once below).

> If the CPUID leaf is not available, then try to determine the tsc hz by
> reading the MSR 0xCE (requires priviledge user).
> 
> Default to the tsc hz estimation if both methods fail.
> 
> Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
> ---
> 
> DEPENDS on Jerin's patch:
> http://dpdk.org/dev/patchwork/patch/27526/
> 
>  lib/librte_eal/common/arch/x86/rte_cycles.c        | 142 +++++++++++++++++++++
>  .../common/include/arch/x86/rte_cycles.h           |   7 +-
>  lib/librte_eal/linuxapp/eal/Makefile               |   1 +
>  3 files changed, 145 insertions(+), 5 deletions(-)
>  create mode 100644 lib/librte_eal/common/arch/x86/rte_cycles.c
> 
> diff --git a/lib/librte_eal/common/arch/x86/rte_cycles.c
> b/lib/librte_eal/common/arch/x86/rte_cycles.c
> new file mode 100644
> index 0000000..9336947
> --- /dev/null
> +++ b/lib/librte_eal/common/arch/x86/rte_cycles.c
> @@ -0,0 +1,142 @@
> +/*-
> + *   BSD LICENSE
> + *
> + *   Copyright(c) 2017 Intel Corporation. All rights reserved.
> + *   All rights reserved.
> + *
> + *   Redistribution and use in source and binary forms, with or without
> + *   modification, are permitted provided that the following conditions
> + *   are met:
> + *
> + *     * Redistributions of source code must retain the above copyright
> + *       notice, this list of conditions and the following disclaimer.
> + *     * Redistributions in binary form must reproduce the above copyright
> + *       notice, this list of conditions and the following disclaimer in
> + *       the documentation and/or other materials provided with the
> + *       distribution.
> + *     * Neither the name of Intel Corporation nor the names of its
> + *       contributors may be used to endorse or promote products derived
> + *       from this software without specific prior written permission.
> + *
> + *   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
> + *   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
> + *   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
> + *   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
> + *   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
> + *   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
> + *   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
> + *   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
> + *   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
> + *   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
> + *   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
> + */
> +
> +#include <fcntl.h>
> +#include <unistd.h>
> +#include <cpuid.h>
> +#include <rte_cycles.h>
> +
> +static unsigned int
> +rte_cpu_get_model(uint32_t fam_mod_step)
> +{
> +	uint32_t family, model, ext_model;
> +
> +	family = (fam_mod_step >> 8) & 0xf;
> +	model = (fam_mod_step >> 4) & 0xf;
> +
> +	if (family == 6 || family == 15) {
> +		ext_model = (fam_mod_step >> 16) & 0xf;
> +		model += (ext_model << 4);
> +	}
> +
> +	return model;
> +}
> +
> +static int32_t
> +rdmsr(int msr, uint64_t *val)
> +{
> +	int fd;
> +	int ret = 0;

Initialization of ret will always be overwritten by pread() below, so no need to initialize.

> +
> +	fd = open("/dev/cpu/0/msr", O_RDONLY);
> +	if (fd < 0)
> +		return fd;
> +
> +	ret = pread(fd, val, sizeof(uint64_t), msr);
> +
> +	close(fd);
> +
> +	return ret;
> +}
> +
> +static uint32_t
> +check_model_wsm_nhm(uint8_t model)
> +{
> +	switch (model) {
> +	/* Westmere */
> +	case 0x25:
> +	case 0x2C:
> +	case 0x2F:
> +	/* Nehalem */
> +	case 0x1E:
> +	case 0x1F:
> +	case 0x1A:
> +	case 0x2E:
> +		return 1;
> +	}

DPDK coding standards say /* fallthrough */ comments required when falling through cases.
In this case I feel it would reduce readability, more than it improves it, but I recall
some recent gcc/clang prints warnings if no /* fallthrough */ comments exist.. opinions?

Same for switch() below.

> +
> +	return 0;
> +}
> +
> +static uint32_t
> +check_model_gdm_dnv(uint8_t model)
> +{
> +	switch (model) {
> +	/* Goldmont */
> +	case 0x5C:
> +	/* Denverton */
> +	case 0x5F:
> +		return 1;
> +	}
> +
> +	return 0;
> +}
> +
> +uint64_t
> +rte_rdtsc_arch_hz(void)
> +{
> +	uint64_t tsc_hz = 0;
> +	uint32_t a, b, c, d, maxleaf;
> +	uint8_t mult, model;
> +	int32_t ret;
> +
> +	/*
> +	 * Time Stamp Counter and Nominal Core Crystal Clock
> +	 * Information Leaf
> +	 */
> +	maxleaf = __get_cpuid_max(0, NULL);
> +
> +	if (maxleaf >= 0x15) {
> +		__cpuid(0x15, a, b, c, d);
> +
> +		/* EBX : TSC/Crystal ratio, ECX : Crystal Hz */
> +		if (b && c)
> +			return c * (b / a);
> +	}
> +
> +	__cpuid(0x1, a, b, c, d);
> +	model = rte_cpu_get_model(a);
> +
> +	if (check_model_wsm_nhm(model))
> +		mult = 133;
> +	else if ((c & bit_AVX) || check_model_gdm_dnv(model))
> +		mult = 100;
> +	else
> +		return 0;
> +
> +	ret = rdmsr(0xCE, &tsc_hz);
> +	if (!(ret < 0))
> +		return ((tsc_hz >> 8) & 0xff) * mult * 1E6;
> +
> +	return 0;


if (!(ret < 0))   feels a little awkward, end of the function could be reworked to, which reads a little cleaner to me?

+     if (ret < 0)
+		return 0;
+	return ((tsc_hz >> 8) & 0xff) * mult * 1E6;


> +}
> diff --git a/lib/librte_eal/common/include/arch/x86/rte_cycles.h
> b/lib/librte_eal/common/include/arch/x86/rte_cycles.h
> index e2661e2..0db89dc 100644
> --- a/lib/librte_eal/common/include/arch/x86/rte_cycles.h
> +++ b/lib/librte_eal/common/include/arch/x86/rte_cycles.h
> @@ -84,11 +84,8 @@ rte_rdtsc(void)
>   *   The number of rdtsc cycles in one second. Return zero if the architecture
>   *   support is not available.
>   */
> -static inline uint64_t
> -rte_rdtsc_arch_hz(void)
> -{
> -	return 0;
> -}
> +uint64_t
> +rte_rdtsc_arch_hz(void);
> 
>  static inline uint64_t
>  rte_rdtsc_precise(void)
> diff --git a/lib/librte_eal/linuxapp/eal/Makefile b/lib/librte_eal/linuxapp/eal/Makefile
> index 90bca4d..9d44828 100644
> --- a/lib/librte_eal/linuxapp/eal/Makefile
> +++ b/lib/librte_eal/linuxapp/eal/Makefile
> @@ -104,6 +104,7 @@ SRCS-$(CONFIG_RTE_EXEC_ENV_LINUXAPP) += rte_service.c
>  # from arch dir
>  SRCS-$(CONFIG_RTE_EXEC_ENV_LINUXAPP) += rte_cpuflags.c
>  SRCS-$(CONFIG_RTE_ARCH_X86) += rte_spinlock.c
> +SRCS-$(CONFIG_RTE_ARCH_X86) += rte_cycles.c
> 
>  CFLAGS_eal_common_cpuflags.o := $(CPUFLAGS_LIST)
> 
> --
> 2.9.5

With above changes, and optionally /* fallthrough */ comments;

Acked-by: Harry van Haaren <harry.van.haaren@intel.com>

^ permalink raw reply	[flat|nested] 11+ messages in thread

* Re: [PATCH] eal/x86: implement x86 specific tsc hz
  2017-09-04  9:38 ` [PATCH] eal/x86: implement x86 specific tsc hz Van Haaren, Harry
@ 2017-09-04 10:24   ` Bruce Richardson
  2017-09-04 10:32     ` Bruce Richardson
  0 siblings, 1 reply; 11+ messages in thread
From: Bruce Richardson @ 2017-09-04 10:24 UTC (permalink / raw)
  To: Van Haaren, Harry; +Cc: Gonzalez Monroy, Sergio, dev, Ananyev, Konstantin

On Mon, Sep 04, 2017 at 10:38:08AM +0100, Van Haaren, Harry wrote:
> > From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Sergio Gonzalez Monroy
> > Sent: Wednesday, August 23, 2017 4:00 PM
> > To: dev@dpdk.org
> > Cc: Ananyev, Konstantin <konstantin.ananyev@intel.com>; Richardson, Bruce
> > <bruce.richardson@intel.com>
> > Subject: [dpdk-dev] [PATCH] eal/x86: implement x86 specific tsc hz
> > 
> > First, try to use CPUID Time Stamp Counter and Nominal Core Crystal
> > Clock Information Leaf to determine the tsc hz on platforms that
> > supports it (does not require priviledge user).
> 
> Checkpatch notifies me that "privilege" is spelled wrong (once above, once below).
> 
> > If the CPUID leaf is not available, then try to determine the tsc hz by
> > reading the MSR 0xCE (requires priviledge user).
> > 
> > Default to the tsc hz estimation if both methods fail.
> > 
> > Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
> > ---
> > 
> > DEPENDS on Jerin's patch:
> > http://dpdk.org/dev/patchwork/patch/27526/
> > 
> >  lib/librte_eal/common/arch/x86/rte_cycles.c        | 142 +++++++++++++++++++++
> >  .../common/include/arch/x86/rte_cycles.h           |   7 +-
> >  lib/librte_eal/linuxapp/eal/Makefile               |   1 +
> >  3 files changed, 145 insertions(+), 5 deletions(-)
> >  create mode 100644 lib/librte_eal/common/arch/x86/rte_cycles.c
> > 
> > diff --git a/lib/librte_eal/common/arch/x86/rte_cycles.c
> > b/lib/librte_eal/common/arch/x86/rte_cycles.c
> > new file mode 100644
> > index 0000000..9336947
> > --- /dev/null
> > +++ b/lib/librte_eal/common/arch/x86/rte_cycles.c
> > @@ -0,0 +1,142 @@
> > +/*-
> > + *   BSD LICENSE
> > + *
> > + *   Copyright(c) 2017 Intel Corporation. All rights reserved.
> > + *   All rights reserved.
> > + *
> > + *   Redistribution and use in source and binary forms, with or without
> > + *   modification, are permitted provided that the following conditions
> > + *   are met:
> > + *
> > + *     * Redistributions of source code must retain the above copyright
> > + *       notice, this list of conditions and the following disclaimer.
> > + *     * Redistributions in binary form must reproduce the above copyright
> > + *       notice, this list of conditions and the following disclaimer in
> > + *       the documentation and/or other materials provided with the
> > + *       distribution.
> > + *     * Neither the name of Intel Corporation nor the names of its
> > + *       contributors may be used to endorse or promote products derived
> > + *       from this software without specific prior written permission.
> > + *
> > + *   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
> > + *   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
> > + *   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
> > + *   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
> > + *   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
> > + *   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
> > + *   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
> > + *   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
> > + *   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
> > + *   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
> > + *   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
> > + */
> > +
> > +#include <fcntl.h>
> > +#include <unistd.h>
> > +#include <cpuid.h>
> > +#include <rte_cycles.h>
> > +
> > +static unsigned int
> > +rte_cpu_get_model(uint32_t fam_mod_step)
> > +{
> > +	uint32_t family, model, ext_model;
> > +
> > +	family = (fam_mod_step >> 8) & 0xf;
> > +	model = (fam_mod_step >> 4) & 0xf;
> > +
> > +	if (family == 6 || family == 15) {
> > +		ext_model = (fam_mod_step >> 16) & 0xf;
> > +		model += (ext_model << 4);
> > +	}
> > +
> > +	return model;
> > +}
> > +
> > +static int32_t
> > +rdmsr(int msr, uint64_t *val)
> > +{
> > +	int fd;
> > +	int ret = 0;
> 
> Initialization of ret will always be overwritten by pread() below, so no need to initialize.
> 
> > +
> > +	fd = open("/dev/cpu/0/msr", O_RDONLY);
> > +	if (fd < 0)
> > +		return fd;
> > +
> > +	ret = pread(fd, val, sizeof(uint64_t), msr);
> > +
> > +	close(fd);
> > +
> > +	return ret;
> > +}
> > +
> > +static uint32_t
> > +check_model_wsm_nhm(uint8_t model)
> > +{
> > +	switch (model) {
> > +	/* Westmere */
> > +	case 0x25:
> > +	case 0x2C:
> > +	case 0x2F:
> > +	/* Nehalem */
> > +	case 0x1E:
> > +	case 0x1F:
> > +	case 0x1A:
> > +	case 0x2E:
> > +		return 1;
> > +	}
> 
> DPDK coding standards say /* fallthrough */ comments required when falling through cases.
> In this case I feel it would reduce readability, more than it improves it, but I recall
> some recent gcc/clang prints warnings if no /* fallthrough */ comments exist.. opinions?
> 
> Same for switch() below.
>

I see no warnings in this case with gcc 7.x. I don't think it counts as
a fallthrough unless there is code after the label - i.e. multiple
labels though technically fallthrough are treated as such by compiler.

> > +
> > +	return 0;
> > +}
> > +
> > +static uint32_t
> > +check_model_gdm_dnv(uint8_t model)
> > +{
> > +	switch (model) {
> > +	/* Goldmont */
> > +	case 0x5C:
> > +	/* Denverton */
> > +	case 0x5F:
> > +		return 1;
> > +	}
> > +
> > +	return 0;
> > +}
> > +
> > +uint64_t
> > +rte_rdtsc_arch_hz(void)
> > +{
> > +	uint64_t tsc_hz = 0;
> > +	uint32_t a, b, c, d, maxleaf;
> > +	uint8_t mult, model;
> > +	int32_t ret;
> > +
> > +	/*
> > +	 * Time Stamp Counter and Nominal Core Crystal Clock
> > +	 * Information Leaf
> > +	 */
> > +	maxleaf = __get_cpuid_max(0, NULL);
> > +
> > +	if (maxleaf >= 0x15) {
> > +		__cpuid(0x15, a, b, c, d);
> > +
> > +		/* EBX : TSC/Crystal ratio, ECX : Crystal Hz */
> > +		if (b && c)
> > +			return c * (b / a);
> > +	}
> > +
> > +	__cpuid(0x1, a, b, c, d);
> > +	model = rte_cpu_get_model(a);
> > +
> > +	if (check_model_wsm_nhm(model))
> > +		mult = 133;
> > +	else if ((c & bit_AVX) || check_model_gdm_dnv(model))
> > +		mult = 100;
> > +	else
> > +		return 0;
> > +
> > +	ret = rdmsr(0xCE, &tsc_hz);
> > +	if (!(ret < 0))
> > +		return ((tsc_hz >> 8) & 0xff) * mult * 1E6;
> > +
> > +	return 0;
> 
> 
> if (!(ret < 0))   feels a little awkward, end of the function could be reworked to, which reads a little cleaner to me?
> 
> +     if (ret < 0)
> +		return 0;
> +	return ((tsc_hz >> 8) & 0xff) * mult * 1E6;
> 
> 
> > +}
> > diff --git a/lib/librte_eal/common/include/arch/x86/rte_cycles.h
> > b/lib/librte_eal/common/include/arch/x86/rte_cycles.h
> > index e2661e2..0db89dc 100644
> > --- a/lib/librte_eal/common/include/arch/x86/rte_cycles.h
> > +++ b/lib/librte_eal/common/include/arch/x86/rte_cycles.h
> > @@ -84,11 +84,8 @@ rte_rdtsc(void)
> >   *   The number of rdtsc cycles in one second. Return zero if the architecture
> >   *   support is not available.
> >   */
> > -static inline uint64_t
> > -rte_rdtsc_arch_hz(void)
> > -{
> > -	return 0;
> > -}
> > +uint64_t
> > +rte_rdtsc_arch_hz(void);
> > 
> >  static inline uint64_t
> >  rte_rdtsc_precise(void)
> > diff --git a/lib/librte_eal/linuxapp/eal/Makefile b/lib/librte_eal/linuxapp/eal/Makefile
> > index 90bca4d..9d44828 100644
> > --- a/lib/librte_eal/linuxapp/eal/Makefile
> > +++ b/lib/librte_eal/linuxapp/eal/Makefile
> > @@ -104,6 +104,7 @@ SRCS-$(CONFIG_RTE_EXEC_ENV_LINUXAPP) += rte_service.c
> >  # from arch dir
> >  SRCS-$(CONFIG_RTE_EXEC_ENV_LINUXAPP) += rte_cpuflags.c
> >  SRCS-$(CONFIG_RTE_ARCH_X86) += rte_spinlock.c
> > +SRCS-$(CONFIG_RTE_ARCH_X86) += rte_cycles.c
> > 
> >  CFLAGS_eal_common_cpuflags.o := $(CPUFLAGS_LIST)
> > 
> > --
> > 2.9.5
> 
> With above changes, and optionally /* fallthrough */ comments;
> 
> Acked-by: Harry van Haaren <harry.van.haaren@intel.com>
> 
No need for fallthrough comments.
On my system with "Intel(R) Xeon(R) Gold 6154 CPU @ 3.00GHz" the TSC
frequency is now reported as 3,000,000 KHz, rather than 2,999,998 KHz as
before.

Tested-by: Bruce Richardson <bruce.richardson@intel.com>

^ permalink raw reply	[flat|nested] 11+ messages in thread

* Re: [PATCH] eal/x86: implement x86 specific tsc hz
  2017-09-04 10:24   ` Bruce Richardson
@ 2017-09-04 10:32     ` Bruce Richardson
  0 siblings, 0 replies; 11+ messages in thread
From: Bruce Richardson @ 2017-09-04 10:32 UTC (permalink / raw)
  To: Van Haaren, Harry; +Cc: Gonzalez Monroy, Sergio, dev, Ananyev, Konstantin

On Mon, Sep 04, 2017 at 11:24:07AM +0100, Bruce Richardson wrote:
> On Mon, Sep 04, 2017 at 10:38:08AM +0100, Van Haaren, Harry wrote:
> > > From: dev [mailto:dev-bounces@dpdk.org] On Behalf Of Sergio Gonzalez Monroy
> > > Sent: Wednesday, August 23, 2017 4:00 PM
> > > To: dev@dpdk.org
> > > Cc: Ananyev, Konstantin <konstantin.ananyev@intel.com>; Richardson, Bruce
> > > <bruce.richardson@intel.com>
> > > Subject: [dpdk-dev] [PATCH] eal/x86: implement x86 specific tsc hz
> > > 
<snip>
> > +
> > > +static uint32_t
> > > +check_model_wsm_nhm(uint8_t model)
> > > +{
> > > +	switch (model) {
> > > +	/* Westmere */
> > > +	case 0x25:
> > > +	case 0x2C:
> > > +	case 0x2F:
> > > +	/* Nehalem */
> > > +	case 0x1E:
> > > +	case 0x1F:
> > > +	case 0x1A:
> > > +	case 0x2E:
> > > +		return 1;
> > > +	}
> > 
> > DPDK coding standards say /* fallthrough */ comments required when falling through cases.
> > In this case I feel it would reduce readability, more than it improves it, but I recall
> > some recent gcc/clang prints warnings if no /* fallthrough */ comments exist.. opinions?
> > 
> > Same for switch() below.
> >
> 
> I see no warnings in this case with gcc 7.x. I don't think it counts as
> a fallthrough unless there is code after the label - i.e. multiple
> labels though technically fallthrough are treated as such by compiler.
> 
apologies, typo: ... are NOT treated as such ...

^ permalink raw reply	[flat|nested] 11+ messages in thread

* [PATCH v2] eal/x86: implement x86 specific tsc hz
  2017-08-23 15:00 [PATCH] eal/x86: implement x86 specific tsc hz Sergio Gonzalez Monroy
  2017-08-23 15:00 ` [PATCH] eal/x86: use cpuid builtin Sergio Gonzalez Monroy
  2017-09-04  9:38 ` [PATCH] eal/x86: implement x86 specific tsc hz Van Haaren, Harry
@ 2017-10-02 10:09 ` Sergio Gonzalez Monroy
  2017-10-02 11:17   ` [PATCH v3] " Sergio Gonzalez Monroy
  2 siblings, 1 reply; 11+ messages in thread
From: Sergio Gonzalez Monroy @ 2017-10-02 10:09 UTC (permalink / raw)
  To: dev; +Cc: harry.van.haaren, bruce.richardson

First, try to use CPUID Time Stamp Counter and Nominal Core Crystal
Clock Information Leaf to determine the tsc hz on platforms that
supports it (does not require privileged user).

If the CPUID leaf is not available, then try to determine the tsc hz by
reading the MSR 0xCE (requires privileged user).

Default to the tsc hz estimation if both methods fail.

Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
---
DEPENDS on:
http://dpdk.org/dev/patchwork/patch/29086/

v2:
 - fix misspelled word in commit message
 - address comment for more clear code

 lib/librte_eal/common/arch/x86/rte_cycles.c        | 142 +++++++++++++++++++++
 .../common/include/arch/x86/rte_cycles.h           |   7 +-
 lib/librte_eal/linuxapp/eal/Makefile               |   1 +
 3 files changed, 145 insertions(+), 5 deletions(-)
 create mode 100644 lib/librte_eal/common/arch/x86/rte_cycles.c

diff --git a/lib/librte_eal/common/arch/x86/rte_cycles.c b/lib/librte_eal/common/arch/x86/rte_cycles.c
new file mode 100644
index 0000000..7cf6093
--- /dev/null
+++ b/lib/librte_eal/common/arch/x86/rte_cycles.c
@@ -0,0 +1,142 @@
+/*-
+ *   BSD LICENSE
+ *
+ *   Copyright(c) 2017 Intel Corporation. All rights reserved.
+ *   All rights reserved.
+ *
+ *   Redistribution and use in source and binary forms, with or without
+ *   modification, are permitted provided that the following conditions
+ *   are met:
+ *
+ *     * Redistributions of source code must retain the above copyright
+ *       notice, this list of conditions and the following disclaimer.
+ *     * Redistributions in binary form must reproduce the above copyright
+ *       notice, this list of conditions and the following disclaimer in
+ *       the documentation and/or other materials provided with the
+ *       distribution.
+ *     * Neither the name of Intel Corporation nor the names of its
+ *       contributors may be used to endorse or promote products derived
+ *       from this software without specific prior written permission.
+ *
+ *   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ *   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ *   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ *   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ *   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ *   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ *   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ *   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ *   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ *   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ *   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include <fcntl.h>
+#include <unistd.h>
+#include <cpuid.h>
+#include <rte_cycles.h>
+
+static unsigned int
+rte_cpu_get_model(uint32_t fam_mod_step)
+{
+	uint32_t family, model, ext_model;
+
+	family = (fam_mod_step >> 8) & 0xf;
+	model = (fam_mod_step >> 4) & 0xf;
+
+	if (family == 6 || family == 15) {
+		ext_model = (fam_mod_step >> 16) & 0xf;
+		model += (ext_model << 4);
+	}
+
+	return model;
+}
+
+static int32_t
+rdmsr(int msr, uint64_t *val)
+{
+	int fd;
+	int ret;
+
+	fd = open("/dev/cpu/0/msr", O_RDONLY);
+	if (fd < 0)
+		return fd;
+
+	ret = pread(fd, val, sizeof(uint64_t), msr);
+
+	close(fd);
+
+	return ret;
+}
+
+static uint32_t
+check_model_wsm_nhm(uint8_t model)
+{
+	switch (model) {
+	/* Westmere */
+	case 0x25:
+	case 0x2C:
+	case 0x2F:
+	/* Nehalem */
+	case 0x1E:
+	case 0x1F:
+	case 0x1A:
+	case 0x2E:
+		return 1;
+	}
+
+	return 0;
+}
+
+static uint32_t
+check_model_gdm_dnv(uint8_t model)
+{
+	switch (model) {
+	/* Goldmont */
+	case 0x5C:
+	/* Denverton */
+	case 0x5F:
+		return 1;
+	}
+
+	return 0;
+}
+
+uint64_t
+rte_rdtsc_arch_hz(void)
+{
+	uint64_t tsc_hz = 0;
+	uint32_t a, b, c, d, maxleaf;
+	uint8_t mult, model;
+	int32_t ret;
+
+	/*
+	 * Time Stamp Counter and Nominal Core Crystal Clock
+	 * Information Leaf
+	 */
+	maxleaf = __get_cpuid_max(0, NULL);
+
+	if (maxleaf >= 0x15) {
+		__cpuid(0x15, a, b, c, d);
+
+		/* EBX : TSC/Crystal ratio, ECX : Crystal Hz */
+		if (b && c)
+			return c * (b / a);
+	}
+
+	__cpuid(0x1, a, b, c, d);
+	model = rte_cpu_get_model(a);
+
+	if (check_model_wsm_nhm(model))
+		mult = 133;
+	else if ((c & bit_AVX) || check_model_gdm_dnv(model))
+		mult = 100;
+	else
+		return 0;
+
+	ret = rdmsr(0xCE, &tsc_hz);
+	if (ret < 0)
+		return 0;
+
+	return ((tsc_hz >> 8) & 0xff) * mult * 1E6;
+}
diff --git a/lib/librte_eal/common/include/arch/x86/rte_cycles.h b/lib/librte_eal/common/include/arch/x86/rte_cycles.h
index e2661e2..0db89dc 100644
--- a/lib/librte_eal/common/include/arch/x86/rte_cycles.h
+++ b/lib/librte_eal/common/include/arch/x86/rte_cycles.h
@@ -84,11 +84,8 @@ rte_rdtsc(void)
  *   The number of rdtsc cycles in one second. Return zero if the architecture
  *   support is not available.
  */
-static inline uint64_t
-rte_rdtsc_arch_hz(void)
-{
-	return 0;
-}
+uint64_t
+rte_rdtsc_arch_hz(void);
 
 static inline uint64_t
 rte_rdtsc_precise(void)
diff --git a/lib/librte_eal/linuxapp/eal/Makefile b/lib/librte_eal/linuxapp/eal/Makefile
index 90bca4d..9d44828 100644
--- a/lib/librte_eal/linuxapp/eal/Makefile
+++ b/lib/librte_eal/linuxapp/eal/Makefile
@@ -104,6 +104,7 @@ SRCS-$(CONFIG_RTE_EXEC_ENV_LINUXAPP) += rte_service.c
 # from arch dir
 SRCS-$(CONFIG_RTE_EXEC_ENV_LINUXAPP) += rte_cpuflags.c
 SRCS-$(CONFIG_RTE_ARCH_X86) += rte_spinlock.c
+SRCS-$(CONFIG_RTE_ARCH_X86) += rte_cycles.c
 
 CFLAGS_eal_common_cpuflags.o := $(CPUFLAGS_LIST)
 
-- 
2.9.5

^ permalink raw reply related	[flat|nested] 11+ messages in thread

* [PATCH v3] eal/x86: implement x86 specific tsc hz
  2017-10-02 10:09 ` [PATCH v2] " Sergio Gonzalez Monroy
@ 2017-10-02 11:17   ` Sergio Gonzalez Monroy
  2017-10-02 11:24     ` Jerin Jacob
  0 siblings, 1 reply; 11+ messages in thread
From: Sergio Gonzalez Monroy @ 2017-10-02 11:17 UTC (permalink / raw)
  To: dev; +Cc: harry.van.haaren, bruce.richardson

First, try to use CPUID Time Stamp Counter and Nominal Core Crystal
Clock Information Leaf to determine the tsc hz on platforms that
supports it (does not require privileged user).

If the CPUID leaf is not available, then try to determine the tsc hz by
reading the MSR 0xCE (requires privileged user).

Default to the tsc hz estimation if both methods fail.

Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
Acked-by: Harry van Haaren <harry.van.haaren@intel.com>
Tested-by: Bruce Richardson <bruce.richardson@intel.com>
---
DEPENDS on:
http://dpdk.org/dev/patchwork/patch/29086/

v3:
 - acked-by and tested-by tags

v2:
 - fix misspelled word in commit message
 - address comment for more clear code

 lib/librte_eal/common/arch/x86/rte_cycles.c        | 142 +++++++++++++++++++++
 .../common/include/arch/x86/rte_cycles.h           |   7 +-
 lib/librte_eal/linuxapp/eal/Makefile               |   1 +
 3 files changed, 145 insertions(+), 5 deletions(-)
 create mode 100644 lib/librte_eal/common/arch/x86/rte_cycles.c

diff --git a/lib/librte_eal/common/arch/x86/rte_cycles.c b/lib/librte_eal/common/arch/x86/rte_cycles.c
new file mode 100644
index 0000000..7cf6093
--- /dev/null
+++ b/lib/librte_eal/common/arch/x86/rte_cycles.c
@@ -0,0 +1,142 @@
+/*-
+ *   BSD LICENSE
+ *
+ *   Copyright(c) 2017 Intel Corporation. All rights reserved.
+ *   All rights reserved.
+ *
+ *   Redistribution and use in source and binary forms, with or without
+ *   modification, are permitted provided that the following conditions
+ *   are met:
+ *
+ *     * Redistributions of source code must retain the above copyright
+ *       notice, this list of conditions and the following disclaimer.
+ *     * Redistributions in binary form must reproduce the above copyright
+ *       notice, this list of conditions and the following disclaimer in
+ *       the documentation and/or other materials provided with the
+ *       distribution.
+ *     * Neither the name of Intel Corporation nor the names of its
+ *       contributors may be used to endorse or promote products derived
+ *       from this software without specific prior written permission.
+ *
+ *   THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
+ *   "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
+ *   LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR
+ *   A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT
+ *   OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
+ *   SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
+ *   LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
+ *   DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
+ *   THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
+ *   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+ *   OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include <fcntl.h>
+#include <unistd.h>
+#include <cpuid.h>
+#include <rte_cycles.h>
+
+static unsigned int
+rte_cpu_get_model(uint32_t fam_mod_step)
+{
+	uint32_t family, model, ext_model;
+
+	family = (fam_mod_step >> 8) & 0xf;
+	model = (fam_mod_step >> 4) & 0xf;
+
+	if (family == 6 || family == 15) {
+		ext_model = (fam_mod_step >> 16) & 0xf;
+		model += (ext_model << 4);
+	}
+
+	return model;
+}
+
+static int32_t
+rdmsr(int msr, uint64_t *val)
+{
+	int fd;
+	int ret;
+
+	fd = open("/dev/cpu/0/msr", O_RDONLY);
+	if (fd < 0)
+		return fd;
+
+	ret = pread(fd, val, sizeof(uint64_t), msr);
+
+	close(fd);
+
+	return ret;
+}
+
+static uint32_t
+check_model_wsm_nhm(uint8_t model)
+{
+	switch (model) {
+	/* Westmere */
+	case 0x25:
+	case 0x2C:
+	case 0x2F:
+	/* Nehalem */
+	case 0x1E:
+	case 0x1F:
+	case 0x1A:
+	case 0x2E:
+		return 1;
+	}
+
+	return 0;
+}
+
+static uint32_t
+check_model_gdm_dnv(uint8_t model)
+{
+	switch (model) {
+	/* Goldmont */
+	case 0x5C:
+	/* Denverton */
+	case 0x5F:
+		return 1;
+	}
+
+	return 0;
+}
+
+uint64_t
+rte_rdtsc_arch_hz(void)
+{
+	uint64_t tsc_hz = 0;
+	uint32_t a, b, c, d, maxleaf;
+	uint8_t mult, model;
+	int32_t ret;
+
+	/*
+	 * Time Stamp Counter and Nominal Core Crystal Clock
+	 * Information Leaf
+	 */
+	maxleaf = __get_cpuid_max(0, NULL);
+
+	if (maxleaf >= 0x15) {
+		__cpuid(0x15, a, b, c, d);
+
+		/* EBX : TSC/Crystal ratio, ECX : Crystal Hz */
+		if (b && c)
+			return c * (b / a);
+	}
+
+	__cpuid(0x1, a, b, c, d);
+	model = rte_cpu_get_model(a);
+
+	if (check_model_wsm_nhm(model))
+		mult = 133;
+	else if ((c & bit_AVX) || check_model_gdm_dnv(model))
+		mult = 100;
+	else
+		return 0;
+
+	ret = rdmsr(0xCE, &tsc_hz);
+	if (ret < 0)
+		return 0;
+
+	return ((tsc_hz >> 8) & 0xff) * mult * 1E6;
+}
diff --git a/lib/librte_eal/common/include/arch/x86/rte_cycles.h b/lib/librte_eal/common/include/arch/x86/rte_cycles.h
index e2661e2..0db89dc 100644
--- a/lib/librte_eal/common/include/arch/x86/rte_cycles.h
+++ b/lib/librte_eal/common/include/arch/x86/rte_cycles.h
@@ -84,11 +84,8 @@ rte_rdtsc(void)
  *   The number of rdtsc cycles in one second. Return zero if the architecture
  *   support is not available.
  */
-static inline uint64_t
-rte_rdtsc_arch_hz(void)
-{
-	return 0;
-}
+uint64_t
+rte_rdtsc_arch_hz(void);
 
 static inline uint64_t
 rte_rdtsc_precise(void)
diff --git a/lib/librte_eal/linuxapp/eal/Makefile b/lib/librte_eal/linuxapp/eal/Makefile
index 90bca4d..9d44828 100644
--- a/lib/librte_eal/linuxapp/eal/Makefile
+++ b/lib/librte_eal/linuxapp/eal/Makefile
@@ -104,6 +104,7 @@ SRCS-$(CONFIG_RTE_EXEC_ENV_LINUXAPP) += rte_service.c
 # from arch dir
 SRCS-$(CONFIG_RTE_EXEC_ENV_LINUXAPP) += rte_cpuflags.c
 SRCS-$(CONFIG_RTE_ARCH_X86) += rte_spinlock.c
+SRCS-$(CONFIG_RTE_ARCH_X86) += rte_cycles.c
 
 CFLAGS_eal_common_cpuflags.o := $(CPUFLAGS_LIST)
 
-- 
2.9.5

^ permalink raw reply related	[flat|nested] 11+ messages in thread

* Re: [PATCH v3] eal/x86: implement x86 specific tsc hz
  2017-10-02 11:17   ` [PATCH v3] " Sergio Gonzalez Monroy
@ 2017-10-02 11:24     ` Jerin Jacob
  2017-10-02 11:27       ` Sergio Gonzalez Monroy
  0 siblings, 1 reply; 11+ messages in thread
From: Jerin Jacob @ 2017-10-02 11:24 UTC (permalink / raw)
  To: Sergio Gonzalez Monroy; +Cc: dev, harry.van.haaren, bruce.richardson

-----Original Message-----
> Date: Mon, 2 Oct 2017 12:17:38 +0100
> From: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
> To: dev@dpdk.org
> CC: harry.van.haaren@intel.com, bruce.richardson@intel.com
> Subject: [dpdk-dev] [PATCH v3] eal/x86: implement x86 specific tsc hz
> X-Mailer: git-send-email 2.9.5
> 
> First, try to use CPUID Time Stamp Counter and Nominal Core Crystal
> Clock Information Leaf to determine the tsc hz on platforms that
> supports it (does not require privileged user).
> 
> If the CPUID leaf is not available, then try to determine the tsc hz by
> reading the MSR 0xCE (requires privileged user).
> 
> Default to the tsc hz estimation if both methods fail.
> 
> Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
> Acked-by: Harry van Haaren <harry.van.haaren@intel.com>
> Tested-by: Bruce Richardson <bruce.richardson@intel.com>
> ---
> DEPENDS on:
> http://dpdk.org/dev/patchwork/patch/29086/
> 
> v3:
>  - acked-by and tested-by tags
> 
> v2:
>  - fix misspelled word in commit message
>  - address comment for more clear code
> 
>  lib/librte_eal/common/arch/x86/rte_cycles.c        | 142 +++++++++++++++++++++
>  .../common/include/arch/x86/rte_cycles.h           |   7 +-
>  lib/librte_eal/linuxapp/eal/Makefile               |   1 +
>  3 files changed, 145 insertions(+), 5 deletions(-)
>  create mode 100644 lib/librte_eal/common/arch/x86/rte_cycles.c
> +
> +static uint32_t
> +check_model_wsm_nhm(uint8_t model)
> +{
> +	switch (model) {
> +	/* Westmere */
> +	case 0x25:
> +	case 0x2C:
> +	case 0x2F:
> +	/* Nehalem */
> +	case 0x1E:
> +	case 0x1F:
> +	case 0x1A:

See next comment.

> +	case 0x2E:
> +		return 1;
> +	}
> +
> +	return 0;
> +}
> +
> +static uint32_t
> +check_model_gdm_dnv(uint8_t model)
> +{
> +	switch (model) {
> +	/* Goldmont */
> +	case 0x5C:
> +	/* Denverton */

Not adding "/* fall-through */" may break gcc 7 build.

> +	case 0x5F:
> +		return 1;
> +	}
> +
> +	return 0;
> +}
> +

^ permalink raw reply	[flat|nested] 11+ messages in thread

* Re: [PATCH v3] eal/x86: implement x86 specific tsc hz
  2017-10-02 11:24     ` Jerin Jacob
@ 2017-10-02 11:27       ` Sergio Gonzalez Monroy
  0 siblings, 0 replies; 11+ messages in thread
From: Sergio Gonzalez Monroy @ 2017-10-02 11:27 UTC (permalink / raw)
  To: Jerin Jacob; +Cc: dev, harry.van.haaren, bruce.richardson

On 02/10/2017 12:24, Jerin Jacob wrote:
> -----Original Message-----
>> Date: Mon, 2 Oct 2017 12:17:38 +0100
>> From: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
>> To: dev@dpdk.org
>> CC: harry.van.haaren@intel.com, bruce.richardson@intel.com
>> Subject: [dpdk-dev] [PATCH v3] eal/x86: implement x86 specific tsc hz
>> X-Mailer: git-send-email 2.9.5
>>
>> First, try to use CPUID Time Stamp Counter and Nominal Core Crystal
>> Clock Information Leaf to determine the tsc hz on platforms that
>> supports it (does not require privileged user).
>>
>> If the CPUID leaf is not available, then try to determine the tsc hz by
>> reading the MSR 0xCE (requires privileged user).
>>
>> Default to the tsc hz estimation if both methods fail.
>>
>> Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
>> Acked-by: Harry van Haaren <harry.van.haaren@intel.com>
>> Tested-by: Bruce Richardson <bruce.richardson@intel.com>
>> ---
>> DEPENDS on:
>> http://dpdk.org/dev/patchwork/patch/29086/
>>
>> v3:
>>   - acked-by and tested-by tags
>>
>> v2:
>>   - fix misspelled word in commit message
>>   - address comment for more clear code
>>
>>   lib/librte_eal/common/arch/x86/rte_cycles.c        | 142 +++++++++++++++++++++
>>   .../common/include/arch/x86/rte_cycles.h           |   7 +-
>>   lib/librte_eal/linuxapp/eal/Makefile               |   1 +
>>   3 files changed, 145 insertions(+), 5 deletions(-)
>>   create mode 100644 lib/librte_eal/common/arch/x86/rte_cycles.c
>> +
>> +static uint32_t
>> +check_model_wsm_nhm(uint8_t model)
>> +{
>> +	switch (model) {
>> +	/* Westmere */
>> +	case 0x25:
>> +	case 0x2C:
>> +	case 0x2F:
>> +	/* Nehalem */
>> +	case 0x1E:
>> +	case 0x1F:
>> +	case 0x1A:
> See next comment.
>
>> +	case 0x2E:
>> +		return 1;
>> +	}
>> +
>> +	return 0;
>> +}
>> +
>> +static uint32_t
>> +check_model_gdm_dnv(uint8_t model)
>> +{
>> +	switch (model) {
>> +	/* Goldmont */
>> +	case 0x5C:
>> +	/* Denverton */
> Not adding "/* fall-through */" may break gcc 7 build.

See Bruce's comment on:
http://dpdk.org/ml/archives/dev/2017-September/074259.html

Thanks,
Sergio

^ permalink raw reply	[flat|nested] 11+ messages in thread

* Re: [PATCH] eal/x86: use cpuid builtin
  2017-08-23 15:00 ` [PATCH] eal/x86: use cpuid builtin Sergio Gonzalez Monroy
@ 2017-10-04 23:39   ` Ferruh Yigit
  2017-10-11 20:00     ` Thomas Monjalon
  0 siblings, 1 reply; 11+ messages in thread
From: Ferruh Yigit @ 2017-10-04 23:39 UTC (permalink / raw)
  To: Sergio Gonzalez Monroy, dev; +Cc: konstantin.ananyev, bruce.richardson

On 8/23/2017 4:00 PM, Sergio Gonzalez Monroy wrote:
> GCC does have the __get_cpuid_count builtin which checks for maximum
> supported leaf, but implementations differ between CLANG and GCC.
> 
> This change provides an implementation compatible with both GCC and
> CLANG 3.4+.
> 
> Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>

Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

^ permalink raw reply	[flat|nested] 11+ messages in thread

* Re: [PATCH] eal/x86: use cpuid builtin
  2017-10-04 23:39   ` Ferruh Yigit
@ 2017-10-11 20:00     ` Thomas Monjalon
  0 siblings, 0 replies; 11+ messages in thread
From: Thomas Monjalon @ 2017-10-11 20:00 UTC (permalink / raw)
  To: Sergio Gonzalez Monroy
  Cc: dev, Ferruh Yigit, konstantin.ananyev, bruce.richardson

05/10/2017 01:39, Ferruh Yigit:
> On 8/23/2017 4:00 PM, Sergio Gonzalez Monroy wrote:
> > GCC does have the __get_cpuid_count builtin which checks for maximum
> > supported leaf, but implementations differ between CLANG and GCC.
> > 
> > This change provides an implementation compatible with both GCC and
> > CLANG 3.4+.
> > 
> > Signed-off-by: Sergio Gonzalez Monroy <sergio.gonzalez.monroy@intel.com>
> 
> Reviewed-by: Ferruh Yigit <ferruh.yigit@intel.com>

Applied, thanks

^ permalink raw reply	[flat|nested] 11+ messages in thread

end of thread, other threads:[~2017-10-11 20:01 UTC | newest]

Thread overview: 11+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-08-23 15:00 [PATCH] eal/x86: implement x86 specific tsc hz Sergio Gonzalez Monroy
2017-08-23 15:00 ` [PATCH] eal/x86: use cpuid builtin Sergio Gonzalez Monroy
2017-10-04 23:39   ` Ferruh Yigit
2017-10-11 20:00     ` Thomas Monjalon
2017-09-04  9:38 ` [PATCH] eal/x86: implement x86 specific tsc hz Van Haaren, Harry
2017-09-04 10:24   ` Bruce Richardson
2017-09-04 10:32     ` Bruce Richardson
2017-10-02 10:09 ` [PATCH v2] " Sergio Gonzalez Monroy
2017-10-02 11:17   ` [PATCH v3] " Sergio Gonzalez Monroy
2017-10-02 11:24     ` Jerin Jacob
2017-10-02 11:27       ` Sergio Gonzalez Monroy

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.