* [PATCH 1/5] tune-corei7.inc: fix PACKAGE_EXTRA_ARCHS_tune-corei7-32 typo
2015-10-19 18:59 [PATCH 0/5] tuning support for 1st and 2nd generation Intel Atom CPUs Andre McCurdy
@ 2015-10-19 18:59 ` Andre McCurdy
2015-10-19 18:59 ` [PATCH 2/5] tune-bonnell.inc: support for first generation Intel Atom CPUs Andre McCurdy
` (5 subsequent siblings)
6 siblings, 0 replies; 12+ messages in thread
From: Andre McCurdy @ 2015-10-19 18:59 UTC (permalink / raw)
To: openembedded-core
Signed-off-by: Andre McCurdy <armccurdy@gmail.com>
---
meta/conf/machine/include/tune-corei7.inc | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/meta/conf/machine/include/tune-corei7.inc b/meta/conf/machine/include/tune-corei7.inc
index 42cf932..bea652f 100644
--- a/meta/conf/machine/include/tune-corei7.inc
+++ b/meta/conf/machine/include/tune-corei7.inc
@@ -20,7 +20,7 @@ AVAILTUNES += "corei7-32"
TUNE_FEATURES_tune-corei7-32 = "${TUNE_FEATURES_tune-x86} corei7"
BASE_LIB_tune-corei7-32 = "lib"
TUNE_PKGARCH_tune-corei7-32 = "corei7-32"
-PACKAGE_EXTRA_ARCHS_tune-corei7-32 = "${PACKAGE_EXTRA_ARCHS_tune-core2} corei7-32"
+PACKAGE_EXTRA_ARCHS_tune-corei7-32 = "${PACKAGE_EXTRA_ARCHS_tune-core2-32} corei7-32"
AVAILTUNES += "corei7-64"
TUNE_FEATURES_tune-corei7-64 = "${TUNE_FEATURES_tune-x86-64} corei7"
--
1.9.1
^ permalink raw reply related [flat|nested] 12+ messages in thread* [PATCH 2/5] tune-bonnell.inc: support for first generation Intel Atom CPUs
2015-10-19 18:59 [PATCH 0/5] tuning support for 1st and 2nd generation Intel Atom CPUs Andre McCurdy
2015-10-19 18:59 ` [PATCH 1/5] tune-corei7.inc: fix PACKAGE_EXTRA_ARCHS_tune-corei7-32 typo Andre McCurdy
@ 2015-10-19 18:59 ` Andre McCurdy
2015-10-19 18:59 ` [PATCH 3/5] tune-silvermont.inc: support for second " Andre McCurdy
` (4 subsequent siblings)
6 siblings, 0 replies; 12+ messages in thread
From: Andre McCurdy @ 2015-10-19 18:59 UTC (permalink / raw)
To: openembedded-core
Signed-off-by: Andre McCurdy <armccurdy@gmail.com>
---
meta/conf/machine/include/tune-bonnell.inc | 35 ++++++++++++++++++++++++++++++
1 file changed, 35 insertions(+)
create mode 100644 meta/conf/machine/include/tune-bonnell.inc
diff --git a/meta/conf/machine/include/tune-bonnell.inc b/meta/conf/machine/include/tune-bonnell.inc
new file mode 100644
index 0000000..f6d92dc
--- /dev/null
+++ b/meta/conf/machine/include/tune-bonnell.inc
@@ -0,0 +1,35 @@
+# Settings for the GCC(1) cpu-type "bonnell":
+#
+# Intel Bonnell CPU with 64-bit extensions, MOVBE, MMX, SSE, SSE2, SSE3
+# and SSSE3 instruction set support.
+#
+# This tune is recommended for the processors based on the first generation
+# Bonnell (45nm) and Saltwell (32nm) Intel Atom CPUs.
+#
+DEFAULTTUNE ?= "bonnell-32"
+
+# Include the previous tune to pull in PACKAGE_EXTRA_ARCHS
+require conf/machine/include/tune-core2.inc
+
+# Extra tune features
+TUNEVALID[bonnell] = "Enable bonnell specific processor optimizations"
+TUNE_CCARGS .= "${@bb.utils.contains("TUNE_FEATURES", "bonnell", " -march=bonnell -mtune=bonnell -msse3 -mfpmath=sse", "", d)}"
+
+# Extra tune selections
+AVAILTUNES += "bonnell-32"
+TUNE_FEATURES_tune-bonnell-32 = "${TUNE_FEATURES_tune-x86} bonnell"
+BASE_LIB_tune-bonnell-32 = "lib"
+TUNE_PKGARCH_tune-bonnell-32 = "bonnell-32"
+PACKAGE_EXTRA_ARCHS_tune-bonnell-32 = "${PACKAGE_EXTRA_ARCHS_tune-core2-32} bonnell-32"
+
+AVAILTUNES += "bonnell-64"
+TUNE_FEATURES_tune-bonnell-64 = "${TUNE_FEATURES_tune-x86-64} bonnell"
+BASE_LIB_tune-bonnell-64 = "lib64"
+TUNE_PKGARCH_tune-bonnell-64 = "bonnell-64"
+PACKAGE_EXTRA_ARCHS_tune-bonnell-64 = "${PACKAGE_EXTRA_ARCHS_tune-core2-64} bonnell-64"
+
+AVAILTUNES += "bonnell-64-x32"
+TUNE_FEATURES_tune-bonnell-64-x32 = "${TUNE_FEATURES_tune-x86-64-x32} bonnell"
+BASE_LIB_tune-bonnell-64-x32 = "libx32"
+TUNE_PKGARCH_tune-bonnell-64-x32 = "bonnell-64-x32"
+PACKAGE_EXTRA_ARCHS_tune-bonnell-64-x32 = "${PACKAGE_EXTRA_ARCHS_tune-core2-64-x32} bonnell-64-x32"
--
1.9.1
^ permalink raw reply related [flat|nested] 12+ messages in thread* [PATCH 3/5] tune-silvermont.inc: support for second generation Intel Atom CPUs
2015-10-19 18:59 [PATCH 0/5] tuning support for 1st and 2nd generation Intel Atom CPUs Andre McCurdy
2015-10-19 18:59 ` [PATCH 1/5] tune-corei7.inc: fix PACKAGE_EXTRA_ARCHS_tune-corei7-32 typo Andre McCurdy
2015-10-19 18:59 ` [PATCH 2/5] tune-bonnell.inc: support for first generation Intel Atom CPUs Andre McCurdy
@ 2015-10-19 18:59 ` Andre McCurdy
2015-10-19 18:59 ` [PATCH 4/5] tune-atom.inc: include tune-bonnell.inc instead of tune-core2.inc Andre McCurdy
` (3 subsequent siblings)
6 siblings, 0 replies; 12+ messages in thread
From: Andre McCurdy @ 2015-10-19 18:59 UTC (permalink / raw)
To: openembedded-core
Signed-off-by: Andre McCurdy <armccurdy@gmail.com>
---
meta/conf/machine/include/tune-silvermont.inc | 36 +++++++++++++++++++++++++++
1 file changed, 36 insertions(+)
create mode 100644 meta/conf/machine/include/tune-silvermont.inc
diff --git a/meta/conf/machine/include/tune-silvermont.inc b/meta/conf/machine/include/tune-silvermont.inc
new file mode 100644
index 0000000..19bff65
--- /dev/null
+++ b/meta/conf/machine/include/tune-silvermont.inc
@@ -0,0 +1,36 @@
+# Settings for the GCC(1) cpu-type "silvermont":
+#
+# Intel Silvermont CPU with 64-bit extensions, MOVBE, MMX, SSE, SSE2, SSE3,
+# SSSE3, SSE4.1, SSE4.2, POPCNT, AES, PCLMUL and RDRND instruction set
+# support.
+#
+# This tune is recommended for the processors based on the second generation
+# Silvermont (22nm) Intel Atom CPUs.
+#
+DEFAULTTUNE ?= "silvermont-32"
+
+# Include the previous tune to pull in PACKAGE_EXTRA_ARCHS
+require conf/machine/include/tune-bonnell.inc
+
+# Extra tune features
+TUNEVALID[silvermont] = "Enable silvermont specific processor optimizations"
+TUNE_CCARGS .= "${@bb.utils.contains("TUNE_FEATURES", "silvermont", " -march=silvermont -mtune=silvermont -msse4.2 -mfpmath=sse", "", d)}"
+
+# Extra tune selections
+AVAILTUNES += "silvermont-32"
+TUNE_FEATURES_tune-silvermont-32 = "${TUNE_FEATURES_tune-x86} silvermont"
+BASE_LIB_tune-silvermont-32 = "lib"
+TUNE_PKGARCH_tune-silvermont-32 = "silvermont-32"
+PACKAGE_EXTRA_ARCHS_tune-silvermont-32 = "${PACKAGE_EXTRA_ARCHS_tune-bonnell-32} silvermont-32"
+
+AVAILTUNES += "silvermont-64"
+TUNE_FEATURES_tune-silvermont-64 = "${TUNE_FEATURES_tune-x86-64} silvermont"
+BASE_LIB_tune-silvermont-64 = "lib64"
+TUNE_PKGARCH_tune-silvermont-64 = "silvermont-64"
+PACKAGE_EXTRA_ARCHS_tune-silvermont-64 = "${PACKAGE_EXTRA_ARCHS_tune-bonnell-64} silvermont-64"
+
+AVAILTUNES += "silvermont-64-x32"
+TUNE_FEATURES_tune-silvermont-64-x32 = "${TUNE_FEATURES_tune-x86-64-x32} silvermont"
+BASE_LIB_tune-silvermont-64-x32 = "libx32"
+TUNE_PKGARCH_tune-silvermont-64-x32 = "silvermont-64-x32"
+PACKAGE_EXTRA_ARCHS_tune-silvermont-64-x32 = "${PACKAGE_EXTRA_ARCHS_tune-bonnell-64-x32} silvermont-64-x32"
--
1.9.1
^ permalink raw reply related [flat|nested] 12+ messages in thread* [PATCH 4/5] tune-atom.inc: include tune-bonnell.inc instead of tune-core2.inc
2015-10-19 18:59 [PATCH 0/5] tuning support for 1st and 2nd generation Intel Atom CPUs Andre McCurdy
` (2 preceding siblings ...)
2015-10-19 18:59 ` [PATCH 3/5] tune-silvermont.inc: support for second " Andre McCurdy
@ 2015-10-19 18:59 ` Andre McCurdy
2015-10-19 19:37 ` Khem Raj
2015-10-20 9:16 ` Richard Purdie
2015-10-19 18:59 ` [PATCH 5/5] tune-corei7.inc: update comments regarding Silvermont support Andre McCurdy
` (2 subsequent siblings)
6 siblings, 2 replies; 12+ messages in thread
From: Andre McCurdy @ 2015-10-19 18:59 UTC (permalink / raw)
To: openembedded-core
Use 'atom' as an alias for the first generation Intel Atom CPUs.
---
meta/conf/machine/include/tune-atom.inc | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/meta/conf/machine/include/tune-atom.inc b/meta/conf/machine/include/tune-atom.inc
index 5e1bb74..24cd676 100644
--- a/meta/conf/machine/include/tune-atom.inc
+++ b/meta/conf/machine/include/tune-atom.inc
@@ -1,2 +1,2 @@
-# Atom tunings are the same as core2 for now...
-require conf/machine/include/tune-core2.inc
+# Alias for the first generation of Intel Atom CPUs.
+require conf/machine/include/tune-bonnell.inc
--
1.9.1
^ permalink raw reply related [flat|nested] 12+ messages in thread* Re: [PATCH 4/5] tune-atom.inc: include tune-bonnell.inc instead of tune-core2.inc
2015-10-19 18:59 ` [PATCH 4/5] tune-atom.inc: include tune-bonnell.inc instead of tune-core2.inc Andre McCurdy
@ 2015-10-19 19:37 ` Khem Raj
2015-10-19 19:51 ` Andre McCurdy
2015-10-20 9:16 ` Richard Purdie
1 sibling, 1 reply; 12+ messages in thread
From: Khem Raj @ 2015-10-19 19:37 UTC (permalink / raw)
To: Andre McCurdy; +Cc: Patches and discussions about the oe-core layer
[-- Attachment #1: Type: text/plain, Size: 972 bytes --]
What do we get with this change
On Oct 19, 2015 12:00 PM, "Andre McCurdy" <armccurdy@gmail.com> wrote:
> Use 'atom' as an alias for the first generation Intel Atom CPUs.
> ---
> meta/conf/machine/include/tune-atom.inc | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/meta/conf/machine/include/tune-atom.inc
> b/meta/conf/machine/include/tune-atom.inc
> index 5e1bb74..24cd676 100644
> --- a/meta/conf/machine/include/tune-atom.inc
> +++ b/meta/conf/machine/include/tune-atom.inc
> @@ -1,2 +1,2 @@
> -# Atom tunings are the same as core2 for now...
> -require conf/machine/include/tune-core2.inc
> +# Alias for the first generation of Intel Atom CPUs.
> +require conf/machine/include/tune-bonnell.inc
> --
> 1.9.1
>
> --
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.openembedded.org/mailman/listinfo/openembedded-core
>
[-- Attachment #2: Type: text/html, Size: 1437 bytes --]
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH 4/5] tune-atom.inc: include tune-bonnell.inc instead of tune-core2.inc
2015-10-19 19:37 ` Khem Raj
@ 2015-10-19 19:51 ` Andre McCurdy
0 siblings, 0 replies; 12+ messages in thread
From: Andre McCurdy @ 2015-10-19 19:51 UTC (permalink / raw)
To: Khem Raj; +Cc: Patches and discussions about the oe-core layer
On Mon, Oct 19, 2015 at 12:37 PM, Khem Raj <raj.khem@gmail.com> wrote:
> What do we get with this change
It's mainly just a convenience for anyone with an existing machine
config using "tune-atom.inc". They would get atom specific tuning
without needing to update their machine configs.
> On Oct 19, 2015 12:00 PM, "Andre McCurdy" <armccurdy@gmail.com> wrote:
>>
>> Use 'atom' as an alias for the first generation Intel Atom CPUs.
>> ---
>> meta/conf/machine/include/tune-atom.inc | 4 ++--
>> 1 file changed, 2 insertions(+), 2 deletions(-)
>>
>> diff --git a/meta/conf/machine/include/tune-atom.inc
>> b/meta/conf/machine/include/tune-atom.inc
>> index 5e1bb74..24cd676 100644
>> --- a/meta/conf/machine/include/tune-atom.inc
>> +++ b/meta/conf/machine/include/tune-atom.inc
>> @@ -1,2 +1,2 @@
>> -# Atom tunings are the same as core2 for now...
>> -require conf/machine/include/tune-core2.inc
>> +# Alias for the first generation of Intel Atom CPUs.
>> +require conf/machine/include/tune-bonnell.inc
>> --
>> 1.9.1
>>
>> --
>> _______________________________________________
>> Openembedded-core mailing list
>> Openembedded-core@lists.openembedded.org
>> http://lists.openembedded.org/mailman/listinfo/openembedded-core
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH 4/5] tune-atom.inc: include tune-bonnell.inc instead of tune-core2.inc
2015-10-19 18:59 ` [PATCH 4/5] tune-atom.inc: include tune-bonnell.inc instead of tune-core2.inc Andre McCurdy
2015-10-19 19:37 ` Khem Raj
@ 2015-10-20 9:16 ` Richard Purdie
2015-10-20 20:55 ` Andre McCurdy
1 sibling, 1 reply; 12+ messages in thread
From: Richard Purdie @ 2015-10-20 9:16 UTC (permalink / raw)
To: Andre McCurdy; +Cc: openembedded-core
On Mon, 2015-10-19 at 11:59 -0700, Andre McCurdy wrote:
> Use 'atom' as an alias for the first generation Intel Atom CPUs.
> ---
> meta/conf/machine/include/tune-atom.inc | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/meta/conf/machine/include/tune-atom.inc b/meta/conf/machine/include/tune-atom.inc
> index 5e1bb74..24cd676 100644
> --- a/meta/conf/machine/include/tune-atom.inc
> +++ b/meta/conf/machine/include/tune-atom.inc
> @@ -1,2 +1,2 @@
> -# Atom tunings are the same as core2 for now...
> -require conf/machine/include/tune-core2.inc
> +# Alias for the first generation of Intel Atom CPUs.
> +require conf/machine/include/tune-bonnell.inc
This is actually pretty nasty to anyone who uses package feeds or
packages since all of a sudden, the system rebuilds with a completely
different package architecture. Not sure we can take this change for
that reason.
I'd also like to understand how much of a difference these specific
tunes actually give. I know the Intel people have been trying to focus
on a smaller number of tunes that work well on the majority of platforms
rather than micro optimising the tunes.
Are there some benchmark numbers or other analysis which shows these
tunes as being more effective?
Cheers,
Richard
^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: [PATCH 4/5] tune-atom.inc: include tune-bonnell.inc instead of tune-core2.inc
2015-10-20 9:16 ` Richard Purdie
@ 2015-10-20 20:55 ` Andre McCurdy
0 siblings, 0 replies; 12+ messages in thread
From: Andre McCurdy @ 2015-10-20 20:55 UTC (permalink / raw)
To: Richard Purdie; +Cc: OE Core mailing list
On Tue, Oct 20, 2015 at 2:16 AM, Richard Purdie
<richard.purdie@linuxfoundation.org> wrote:
> On Mon, 2015-10-19 at 11:59 -0700, Andre McCurdy wrote:
>> Use 'atom' as an alias for the first generation Intel Atom CPUs.
>> ---
>> meta/conf/machine/include/tune-atom.inc | 4 ++--
>> 1 file changed, 2 insertions(+), 2 deletions(-)
>>
>> diff --git a/meta/conf/machine/include/tune-atom.inc b/meta/conf/machine/include/tune-atom.inc
>> index 5e1bb74..24cd676 100644
>> --- a/meta/conf/machine/include/tune-atom.inc
>> +++ b/meta/conf/machine/include/tune-atom.inc
>> @@ -1,2 +1,2 @@
>> -# Atom tunings are the same as core2 for now...
>> -require conf/machine/include/tune-core2.inc
>> +# Alias for the first generation of Intel Atom CPUs.
>> +require conf/machine/include/tune-bonnell.inc
>
> This is actually pretty nasty to anyone who uses package feeds or
> packages since all of a sudden, the system rebuilds with a completely
> different package architecture. Not sure we can take this change for
> that reason.
OK. I was thinking the comment in tune-atom.inc "Atom tunings are the
same as core2 FOR NOW..." gave people fair warning that tune-atom.inc
might not remain an alias for tune-core2.inc forever, but no problem
to drop this patch.
Another approach might be to drop tune-atom.inc altogether. Atom is
now ambiguous in terms of defining a CPU architecture and it hasn't
been a documented gcc x86 tuning option since gcc 4.8
https://gcc.gnu.org/onlinedocs/gcc-4.8.5/gcc/i386-and-x86-64-Options.html#i386-and-x86-64-Options
https://gcc.gnu.org/onlinedocs/gcc-4.9.3/gcc/i386-and-x86-64-Options.html#i386-and-x86-64-Options
https://gcc.gnu.org/onlinedocs/gcc-5.2.0/gcc/x86-Options.html#x86-Options
> I'd also like to understand how much of a difference these specific
> tunes actually give. I know the Intel people have been trying to focus
> on a smaller number of tunes that work well on the majority of platforms
> rather than micro optimising the tunes.
>
> Are there some benchmark numbers or other analysis which shows these
> tunes as being more effective?
Not that I'm aware of. My own tests suggest that the benefits of
bonnell -vs- core2 tuning on a first generation Atom are pretty
unspectacular (maybe ~0 to 5 % performance increase and ~5% increase
in code size). I don't have a Silvermont board. This patch set was
more about enabling further experiments though, so if solid benchmark
improvements are a prerequisite then this patch set isn't ready to
merge.
>
> Cheers,
>
> Richard
>
>
>
^ permalink raw reply [flat|nested] 12+ messages in thread
* [PATCH 5/5] tune-corei7.inc: update comments regarding Silvermont support
2015-10-19 18:59 [PATCH 0/5] tuning support for 1st and 2nd generation Intel Atom CPUs Andre McCurdy
` (3 preceding siblings ...)
2015-10-19 18:59 ` [PATCH 4/5] tune-atom.inc: include tune-bonnell.inc instead of tune-core2.inc Andre McCurdy
@ 2015-10-19 18:59 ` Andre McCurdy
2015-10-22 18:57 ` [PATCH 0/5] tuning support for 1st and 2nd generation Intel Atom CPUs Saul Wold
2015-10-23 15:59 ` Saul Wold
6 siblings, 0 replies; 12+ messages in thread
From: Andre McCurdy @ 2015-10-19 18:59 UTC (permalink / raw)
To: openembedded-core
Although still compatible, tune-corei7.inc is no longer the recommended
way to tune for Silvermont (which now has its own dedicated .inc file).
Signed-off-by: Andre McCurdy <armccurdy@gmail.com>
---
meta/conf/machine/include/tune-corei7.inc | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/meta/conf/machine/include/tune-corei7.inc b/meta/conf/machine/include/tune-corei7.inc
index bea652f..13da46b 100644
--- a/meta/conf/machine/include/tune-corei7.inc
+++ b/meta/conf/machine/include/tune-corei7.inc
@@ -3,8 +3,8 @@
# Intel Core i7 CPU with 64-bit extensions, MMX, SSE, SSE2, SSE3, SSSE3, SSE4.1
# and SSE4.2 instruction set support.
#
-# This tune is recommended for Intel Nehalem and Silvermont (e.g. Bay Trail) CPUs
-# (and beyond).
+# This tune is recommended for Intel Nehalem CPUs and beyond.
+# It is also compatible with second generation Silvermont Intel Atom CPUs.
#
DEFAULTTUNE ?= "corei7-64"
--
1.9.1
^ permalink raw reply related [flat|nested] 12+ messages in thread* Re: [PATCH 0/5] tuning support for 1st and 2nd generation Intel Atom CPUs
2015-10-19 18:59 [PATCH 0/5] tuning support for 1st and 2nd generation Intel Atom CPUs Andre McCurdy
` (4 preceding siblings ...)
2015-10-19 18:59 ` [PATCH 5/5] tune-corei7.inc: update comments regarding Silvermont support Andre McCurdy
@ 2015-10-22 18:57 ` Saul Wold
2015-10-23 15:59 ` Saul Wold
6 siblings, 0 replies; 12+ messages in thread
From: Saul Wold @ 2015-10-22 18:57 UTC (permalink / raw)
To: Andre McCurdy, openembedded-core; +Cc: dvhart@linux.intel.com
Hi Andre,
On 10/19/2015 11:59 AM, Andre McCurdy wrote:
> Andre McCurdy (5):
> tune-corei7.inc: fix PACKAGE_EXTRA_ARCHS_tune-corei7-32 typo
This change makes sense and addresses an issue.
> tune-bonnell.inc: support for first generation Intel Atom CPUs
> tune-silvermont.inc: support for second generation Intel Atom CPUs
> tune-atom.inc: include tune-bonnell.inc instead of tune-core2.inc
> tune-corei7.inc: update comments regarding Silvermont support
>
I am not sure these make sense, it adds additional complexity to an
already complex testing matrix.
Both the bonnell and silvermont tunes are fully part of their respective
core* tune settings, there is no real need to become more specific. We
know that newer code is starting to use intrinsics that can determine
the correct extensions based on the CPU type.
I know you replied partly to RP's query, but is there something specific
your trying to accomplish with these additional .incs that we don't
already do and if so what?
Sau!
meta-intel maintainer
> meta/conf/machine/include/tune-atom.inc | 4 +--
> meta/conf/machine/include/tune-bonnell.inc | 35 ++++++++++++++++++++++++++
> meta/conf/machine/include/tune-corei7.inc | 6 ++---
> meta/conf/machine/include/tune-silvermont.inc | 36 +++++++++++++++++++++++++++
> 4 files changed, 76 insertions(+), 5 deletions(-)
> create mode 100644 meta/conf/machine/include/tune-bonnell.inc
> create mode 100644 meta/conf/machine/include/tune-silvermont.inc
>
^ permalink raw reply [flat|nested] 12+ messages in thread* Re: [PATCH 0/5] tuning support for 1st and 2nd generation Intel Atom CPUs
2015-10-19 18:59 [PATCH 0/5] tuning support for 1st and 2nd generation Intel Atom CPUs Andre McCurdy
` (5 preceding siblings ...)
2015-10-22 18:57 ` [PATCH 0/5] tuning support for 1st and 2nd generation Intel Atom CPUs Saul Wold
@ 2015-10-23 15:59 ` Saul Wold
6 siblings, 0 replies; 12+ messages in thread
From: Saul Wold @ 2015-10-23 15:59 UTC (permalink / raw)
To: Andre McCurdy, openembedded-core
On 10/19/2015 11:59 AM, Andre McCurdy wrote:
> Andre McCurdy (5):
> tune-corei7.inc: fix PACKAGE_EXTRA_ARCHS_tune-corei7-32 typo
This change makes sense
> tune-bonnell.inc: support for first generation Intel Atom CPUs
> tune-silvermont.inc: support for second generation Intel Atom CPUs
> tune-atom.inc: include tune-bonnell.inc instead of tune-core2.inc
> tune-corei7.inc: update comments regarding Silvermont support
>
I am not sure these make sense, it adds additional complexity to an
already complex testing matrix.
Both the bonnell and silvermont tunes are fully part of their respective
core* tune settings, there is no real need to become more specific. We
know that newer code is starting to use intrinsics that can determine
the correct extensions based on the CPU type.
I know you replied partly to RP's query, but is there something specific
your trying to accomplish with these additional .incs that we don't
already do and if so what?
Sau!
meta-intel maintainer
> meta/conf/machine/include/tune-atom.inc | 4 +--
> meta/conf/machine/include/tune-bonnell.inc | 35 ++++++++++++++++++++++++++
> meta/conf/machine/include/tune-corei7.inc | 6 ++---
> meta/conf/machine/include/tune-silvermont.inc | 36 +++++++++++++++++++++++++++
> 4 files changed, 76 insertions(+), 5 deletions(-)
> create mode 100644 meta/conf/machine/include/tune-bonnell.inc
> create mode 100644 meta/conf/machine/include/tune-silvermont.inc
>
^ permalink raw reply [flat|nested] 12+ messages in thread