All of lore.kernel.org
 help / color / mirror / Atom feed
From: Denys Dmytriyenko <denys@ti.com>
To: "Maupin, Chase" <chase.maupin@ti.com>
Cc: "meta-ti@yoctoproject.org" <meta-ti@yoctoproject.org>
Subject: Re: [PATCH] mmap-lld; Add new recipe for mmap lld
Date: Thu, 20 Nov 2014 14:53:00 -0500	[thread overview]
Message-ID: <20141120195300.GD31396@edge> (raw)
In-Reply-To: <7D46E86EC0A8354091174257B2FED1015D375D5C@DLEE11.ent.ti.com>

Chase,

I had pretty much all the same questions before. Please see the discussion we 
had previously on hyplnk-lld patch:

http://thread.gmane.org/gmane.linux.embedded.yocto.meta-ti/4651

-- 
Denys


On Wed, Nov 19, 2014 at 05:05:51PM +0000, Maupin, Chase wrote:
> 
> diff --git a/recipes-bsp/mmap-lld/mmap-lld-test_git.bb b/recipes-bsp/mmap-lld/mmap-lld-test_git.bb
> new file mode 100755
> index 0000000..41e46f7
> --- /dev/null
> +++ b/recipes-bsp/mmap-lld/mmap-lld-test_git.bb
> @@ -0,0 +1,40 @@
> +DESCRIPTION = "TI KEYSTONE MMAP driver test binaries "
> +include mmap-lld.inc
> +
> +DEPENDS="common-csl-ip mmap-lld"
> +RDEPENDS_${PN}="uio-module-drv"
> +
> +DEVICELIST = "k2h \
> +              k2k \
> +              k2e \
> +"
> 
> I don't have a fundamental issue with a single  recipe for all devices, but would it be easier to make this per device and then keep a generic binary name?  i.e. a machine specific recipe that installs the binary build for that device into the file system?
> 
> +
> +CHOICELIST = " yes \
> +               no \
> +"
> +
> +do_compile () {
> +	for device in ${DEVICELIST}
> +	do
> +		make -f makefile_armv7 clean \
> +			PDK_INSTALL_PATH="${STAGING_INCDIR}" \
> +			DEVICE="$device" MPAX_SRC_DIR="${S}"
> +		for choice in ${CHOICELIST}
> +		do
> +			make -f makefile_armv7 tests examples \
> +				PDK_INSTALL_PATH=${STAGING_INCDIR} \
> +				DEVICE="$device" MPAX_SRC_DIR="${S}" \
> +				USEDYNAMIC_LIB="$choice"
> +		done
> 
> What is the purpose for both choice options?  Are you expecting users to use both?
> 
> +	done
> +}
> +
> +do_install () {
> +	for device in ${DEVICELIST}
> +	do
> +		make -f makefile_armv7 installbin \
> +		PDK_INSTALL_PATH=${STAGING_INCDIR} \
> +		DEVICE="$device" MPAX_SRC_DIR="${S}" \
> +		INSTALL_BIN_BASE_DIR="${D}/${bindir}"
> +	done
> +}
> diff --git a/recipes-bsp/mmap-lld/mmap-lld.inc b/recipes-bsp/mmap-lld/mmap-lld.inc
> new file mode 100644
> index 0000000..b8a5b32
> --- /dev/null
> +++ b/recipes-bsp/mmap-lld/mmap-lld.inc
> @@ -0,0 +1,16 @@
> +LICENSE = "BSD-3-Clause"
> +
> +COMPATIBLE_MACHINE = "keystone"
> +
> +LIC_FILES_CHKSUM = "file://COPYING.txt;md5=b3315b8190036fe1c6a8263d02e15606"
> 
> Nit pick here but can this checksum go after the LICENSE setting?
> 
> +
> +
> +BRANCH="master"
> +SRC_URI = "git://git.ti.com/keystone-rtos/mmap-lld.git;destsuffix=git/ti/runtime/mmap;protocol=git;branch=${BRANCH}"
> +# Commit correspond to DEV.MMAP.02.00.00.03 SRCREV = 
> +"601f9284c392f67bb7f69ed1a5df5231a43c3137"
> 
> I would put the SRCREV in the _git file, not in the .inc since you would expect the .inc to be common across all versions.  I guess you are using the .inc to handle the different test vs driver recipes.  Maybe a better idea would be a single recipe that builds both but package the driver as the default package and make a new PACKAGE for the test applications? 
> 
> +
> +PV="2.0.0"
> +PR = "r0"
> +
> +S = "${WORKDIR}/git/ti/runtime/mmap"
> \ No newline at end of file
> diff --git a/recipes-bsp/mmap-lld/mmap-lld_git.bb b/recipes-bsp/mmap-lld/mmap-lld_git.bb
> new file mode 100755
> index 0000000..3a56a62
> --- /dev/null
> +++ b/recipes-bsp/mmap-lld/mmap-lld_git.bb
> @@ -0,0 +1,12 @@
> +DESCRIPTION = "TI KEYSTONE mmap driver using Keystone MPAX Programming"
> +include mmap-lld.inc
> +
> +DEPENDS="common-csl-ip"
> +
> +do_compile () {
> +	make -f makefile_armv7 clean lib PDK_INSTALL_PATH=${STAGING_INCDIR}
> +}
> +
> +do_install () {
> +	make -f makefile_armv7 install PDK_INSTALL_PATH=${STAGING_INCDIR} 
> +INSTALL_INC_BASE_DIR=${D}/${includedir} 
> +INSTALL_LIB_BASE_DIR=${D}${libdir}
> +}
> --
> 1.7.9.5
> 
> --
> _______________________________________________
> meta-ti mailing list
> meta-ti@yoctoproject.org
> https://lists.yoctoproject.org/listinfo/meta-ti
> -- 
> _______________________________________________
> meta-ti mailing list
> meta-ti@yoctoproject.org
> https://lists.yoctoproject.org/listinfo/meta-ti


  parent reply	other threads:[~2014-11-20 19:53 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-11-19 11:47 [PATCH] mmap-lld; Add new recipe for mmap lld Sam Nelson
2014-11-19 17:05 ` Maupin, Chase
2014-11-20 17:42   ` Nelson, Sam
2014-11-20 19:53   ` Denys Dmytriyenko [this message]
2014-12-02 15:07     ` Nelson, Sam
2014-12-09 20:04       ` Nelson, Sam

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=20141120195300.GD31396@edge \
    --to=denys@ti.com \
    --cc=chase.maupin@ti.com \
    --cc=meta-ti@yoctoproject.org \
    /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 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.