[dpdk-dev] ethdev: rename folder to library name
Checks
Commit Message
Library folder name and output library name are same except a few flaws
including librte_ether.
This library is network device abstraction layer, the name "ethdev" fits
better than "ether", and library & header files already named as ethdev.
Also there is a rte_ether.h in the net library which can cause confusion.
Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
---
Not sure if the we are ready for this change J
This is one the issues that the hassle of the change doesn't worth the
benefit and you may prefer to live with the flaw.
Also a concern is this breaks the git history.
---
MAINTAINERS | 8 +++---
doc/api/doxy-api.conf | 2 +-
doc/guides/prog_guide/source_org.rst | 2 +-
.../prog_guide/traffic_metering_and_policing.rst | 2 +-
lib/Makefile | 32 +++++++++++-----------
lib/{librte_ether => librte_ethdev}/Makefile | 0
.../ethdev_profile.c | 0
.../ethdev_profile.h | 0
lib/{librte_ether => librte_ethdev}/meson.build | 0
lib/{librte_ether => librte_ethdev}/rte_dev_info.h | 0
lib/{librte_ether => librte_ethdev}/rte_eth_ctrl.h | 0
lib/{librte_ether => librte_ethdev}/rte_ethdev.c | 0
lib/{librte_ether => librte_ethdev}/rte_ethdev.h | 0
.../rte_ethdev_core.h | 0
.../rte_ethdev_driver.h | 0
.../rte_ethdev_pci.h | 0
.../rte_ethdev_vdev.h | 0
.../rte_ethdev_version.map | 0
lib/{librte_ether => librte_ethdev}/rte_flow.c | 0
lib/{librte_ether => librte_ethdev}/rte_flow.h | 0
.../rte_flow_driver.h | 0
lib/{librte_ether => librte_ethdev}/rte_mtr.c | 0
lib/{librte_ether => librte_ethdev}/rte_mtr.h | 0
.../rte_mtr_driver.h | 0
lib/{librte_ether => librte_ethdev}/rte_tm.c | 0
lib/{librte_ether => librte_ethdev}/rte_tm.h | 0
.../rte_tm_driver.h | 0
lib/meson.build | 2 +-
28 files changed, 24 insertions(+), 24 deletions(-)
rename lib/{librte_ether => librte_ethdev}/Makefile (100%)
rename lib/{librte_ether => librte_ethdev}/ethdev_profile.c (100%)
rename lib/{librte_ether => librte_ethdev}/ethdev_profile.h (100%)
rename lib/{librte_ether => librte_ethdev}/meson.build (100%)
rename lib/{librte_ether => librte_ethdev}/rte_dev_info.h (100%)
rename lib/{librte_ether => librte_ethdev}/rte_eth_ctrl.h (100%)
rename lib/{librte_ether => librte_ethdev}/rte_ethdev.c (100%)
rename lib/{librte_ether => librte_ethdev}/rte_ethdev.h (100%)
rename lib/{librte_ether => librte_ethdev}/rte_ethdev_core.h (100%)
rename lib/{librte_ether => librte_ethdev}/rte_ethdev_driver.h (100%)
rename lib/{librte_ether => librte_ethdev}/rte_ethdev_pci.h (100%)
rename lib/{librte_ether => librte_ethdev}/rte_ethdev_vdev.h (100%)
rename lib/{librte_ether => librte_ethdev}/rte_ethdev_version.map (100%)
rename lib/{librte_ether => librte_ethdev}/rte_flow.c (100%)
rename lib/{librte_ether => librte_ethdev}/rte_flow.h (100%)
rename lib/{librte_ether => librte_ethdev}/rte_flow_driver.h (100%)
rename lib/{librte_ether => librte_ethdev}/rte_mtr.c (100%)
rename lib/{librte_ether => librte_ethdev}/rte_mtr.h (100%)
rename lib/{librte_ether => librte_ethdev}/rte_mtr_driver.h (100%)
rename lib/{librte_ether => librte_ethdev}/rte_tm.c (100%)
rename lib/{librte_ether => librte_ethdev}/rte_tm.h (100%)
rename lib/{librte_ether => librte_ethdev}/rte_tm_driver.h (100%)
Comments
On Tue, Apr 03, 2018 at 10:28:59AM +0100, Ferruh Yigit wrote:
> Library folder name and output library name are same except a few flaws
> including librte_ether.
>
> This library is network device abstraction layer, the name "ethdev" fits
> better than "ether", and library & header files already named as ethdev.
>
> Also there is a rte_ether.h in the net library which can cause confusion.
>
> Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
> ---
> Not sure if the we are ready for this change J
>
> This is one the issues that the hassle of the change doesn't worth the
> benefit and you may prefer to live with the flaw.
>
> Also a concern is this breaks the git history.
> ---
I strongly support this change. The impacts should all be internal, since
the installed headers and final library name are all unchanged, so I see
little reason not to do this.
Acked-by: Bruce Richardson <bruce.richardson@intel.com>
03/04/2018 11:28, Ferruh Yigit:
> Library folder name and output library name are same except a few flaws
> including librte_ether.
>
> This library is network device abstraction layer, the name "ethdev" fits
> better than "ether", and library & header files already named as ethdev.
>
> Also there is a rte_ether.h in the net library which can cause confusion.
>
> Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
> ---
> Not sure if the we are ready for this change J
>
> This is one the issues that the hassle of the change doesn't worth the
> benefit and you may prefer to live with the flaw.
>
> Also a concern is this breaks the git history.
Yes it breaks the git history.
I am a big user of gitk. Do you know how to follow the history after rename
with gitk?
I think it is a good change but I would like to be sure it will not
bother us every day.
On Tue, Apr 03, 2018 at 12:24:15PM +0200, Thomas Monjalon wrote:
> 03/04/2018 11:28, Ferruh Yigit:
> > Library folder name and output library name are same except a few flaws
> > including librte_ether.
> >
> > This library is network device abstraction layer, the name "ethdev" fits
> > better than "ether", and library & header files already named as ethdev.
> >
> > Also there is a rte_ether.h in the net library which can cause confusion.
> >
> > Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
> > ---
> > Not sure if the we are ready for this change J
> >
> > This is one the issues that the hassle of the change doesn't worth the
> > benefit and you may prefer to live with the flaw.
> >
> > Also a concern is this breaks the git history.
>
> Yes it breaks the git history.
> I am a big user of gitk. Do you know how to follow the history after rename
> with gitk?
>
> I think it is a good change but I would like to be sure it will not
> bother us every day.
>
For those of use with even slight OCD, having it the way it is bothers us
every day! :-)
I view this as something that we really should do some day - it's just
broken the way the way it is right now - and so the sooner we take the hit
and fix it, the better.
If you do need to find the history of the file, would git blame work well
enough for you?
/Bruce
03/04/2018 12:57, Bruce Richardson:
> On Tue, Apr 03, 2018 at 12:24:15PM +0200, Thomas Monjalon wrote:
> > 03/04/2018 11:28, Ferruh Yigit:
> > > Library folder name and output library name are same except a few flaws
> > > including librte_ether.
> > >
> > > This library is network device abstraction layer, the name "ethdev" fits
> > > better than "ether", and library & header files already named as ethdev.
> > >
> > > Also there is a rte_ether.h in the net library which can cause confusion.
> > >
> > > Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
> > > ---
> > > Not sure if the we are ready for this change J
> > >
> > > This is one the issues that the hassle of the change doesn't worth the
> > > benefit and you may prefer to live with the flaw.
> > >
> > > Also a concern is this breaks the git history.
> >
> > Yes it breaks the git history.
> > I am a big user of gitk. Do you know how to follow the history after rename
> > with gitk?
> >
> > I think it is a good change but I would like to be sure it will not
> > bother us every day.
> >
> For those of use with even slight OCD, having it the way it is bothers us
> every day! :-)
> I view this as something that we really should do some day - it's just
> broken the way the way it is right now - and so the sooner we take the hit
> and fix it, the better.
Yes
> If you do need to find the history of the file, would git blame work well
> enough for you?
No, blame is not enough, but it does not matter.
When should we do this rename? 18.05-rc1?
On Tue, Apr 03, 2018 at 02:20:30PM +0200, Thomas Monjalon wrote:
> 03/04/2018 12:57, Bruce Richardson:
> > On Tue, Apr 03, 2018 at 12:24:15PM +0200, Thomas Monjalon wrote:
> > > 03/04/2018 11:28, Ferruh Yigit:
> > > > Library folder name and output library name are same except a few flaws
> > > > including librte_ether.
> > > >
> > > > This library is network device abstraction layer, the name "ethdev" fits
> > > > better than "ether", and library & header files already named as ethdev.
> > > >
> > > > Also there is a rte_ether.h in the net library which can cause confusion.
> > > >
> > > > Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
> > > > ---
> > > > Not sure if the we are ready for this change J
> > > >
> > > > This is one the issues that the hassle of the change doesn't worth the
> > > > benefit and you may prefer to live with the flaw.
> > > >
> > > > Also a concern is this breaks the git history.
> > >
> > > Yes it breaks the git history.
> > > I am a big user of gitk. Do you know how to follow the history after rename
> > > with gitk?
> > >
> > > I think it is a good change but I would like to be sure it will not
> > > bother us every day.
> > >
> > For those of use with even slight OCD, having it the way it is bothers us
> > every day! :-)
> > I view this as something that we really should do some day - it's just
> > broken the way the way it is right now - and so the sooner we take the hit
> > and fix it, the better.
>
> Yes
>
> > If you do need to find the history of the file, would git blame work well
> > enough for you?
>
> No, blame is not enough, but it does not matter.
>
> When should we do this rename? 18.05-rc1?
>
If we are going to do this rename, I don't see why we should wait.
/Bruce
> On Apr 3, 2018, at 4:28 AM, Ferruh Yigit <ferruh.yigit@intel.com> wrote:
>
> Library folder name and output library name are same except a few flaws
> including librte_ether.
>
> This library is network device abstraction layer, the name "ethdev" fits
> better than "ether", and library & header files already named as ethdev.
>
> Also there is a rte_ether.h in the net library which can cause confusion.
>
> Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
> ---
> Not sure if the we are ready for this change J
>
> This is one the issues that the hassle of the change doesn't worth the
> benefit and you may prefer to live with the flaw.
>
> Also a concern is this breaks the git history.
I believe as long as you used ‘git mv’ the git history should not be broken, correct?
> ---
> On Apr 3, 2018, at 7:33 AM, Bruce Richardson <bruce.richardson@intel.com> wrote:
>
> On Tue, Apr 03, 2018 at 02:20:30PM +0200, Thomas Monjalon wrote:
>> 03/04/2018 12:57, Bruce Richardson:
>>> On Tue, Apr 03, 2018 at 12:24:15PM +0200, Thomas Monjalon wrote:
>>>> 03/04/2018 11:28, Ferruh Yigit:
>>>>> Library folder name and output library name are same except a few flaws
>>>>> including librte_ether.
>>>>>
>>>>> This library is network device abstraction layer, the name "ethdev" fits
>>>>> better than "ether", and library & header files already named as ethdev.
>>>>>
>>>>> Also there is a rte_ether.h in the net library which can cause confusion.
>>>>>
>>>>> Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
>>>>> ---
>>>>> Not sure if the we are ready for this change J
>>>>>
>>>>> This is one the issues that the hassle of the change doesn't worth the
>>>>> benefit and you may prefer to live with the flaw.
>>>>>
>>>>> Also a concern is this breaks the git history.
>>>>
>>>> Yes it breaks the git history.
>>>> I am a big user of gitk. Do you know how to follow the history after rename
>>>> with gitk?
>>>>
>>>> I think it is a good change but I would like to be sure it will not
>>>> bother us every day.
>>>>
>>> For those of use with even slight OCD, having it the way it is bothers us
>>> every day! :-)
>>> I view this as something that we really should do some day - it's just
>>> broken the way the way it is right now - and so the sooner we take the hit
>>> and fix it, the better.
>>
>> Yes
>>
>>> If you do need to find the history of the file, would git blame work well
>>> enough for you?
>>
>> No, blame is not enough, but it does not matter.
>>
>> When should we do this rename? 18.05-rc1?
>>
> If we are going to do this rename, I don't see why we should wait.
+1 make the change ASAP IMO.
>
> /Bruce
Regards,
Keith
On 4/3/2018 2:17 PM, Wiles, Keith wrote:
>
>
>> On Apr 3, 2018, at 4:28 AM, Ferruh Yigit <ferruh.yigit@intel.com> wrote:
>>
>> Library folder name and output library name are same except a few flaws
>> including librte_ether.
>>
>> This library is network device abstraction layer, the name "ethdev" fits
>> better than "ether", and library & header files already named as ethdev.
>>
>> Also there is a rte_ether.h in the net library which can cause confusion.
>>
>> Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
>> ---
>> Not sure if the we are ready for this change J
>>
>> This is one the issues that the hassle of the change doesn't worth the
>> benefit and you may prefer to live with the flaw.
>>
>> Also a concern is this breaks the git history.
>
> I believe as long as you used ‘git mv’ the git history should not be broken, correct?
I think this will help in some conflict resolving or git blame, but if you check
the history directly with path, it won't show the past:
"git log --oneline lib/librte_ethdev/": Will start from this commit
03/04/2018 15:29, Ferruh Yigit:
> On 4/3/2018 2:17 PM, Wiles, Keith wrote:
> >> On Apr 3, 2018, at 4:28 AM, Ferruh Yigit <ferruh.yigit@intel.com> wrote:
> >>
> >> Library folder name and output library name are same except a few flaws
> >> including librte_ether.
> >>
> >> This library is network device abstraction layer, the name "ethdev" fits
> >> better than "ether", and library & header files already named as ethdev.
> >>
> >> Also there is a rte_ether.h in the net library which can cause confusion.
> >>
> >> Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
> >> ---
> >> Not sure if the we are ready for this change J
> >>
> >> This is one the issues that the hassle of the change doesn't worth the
> >> benefit and you may prefer to live with the flaw.
> >>
> >> Also a concern is this breaks the git history.
> >
> > I believe as long as you used ‘git mv’ the git history should not be broken, correct?
I think "git mv" does nothing special. Git tracks the content, so it is the
same as using "mv + git add + git rm".
> I think this will help in some conflict resolving or git blame, but if you check
> the history directly with path, it won't show the past:
>
> "git log --oneline lib/librte_ethdev/": Will start from this commit
You can use --follow.
Unfortunately, it does not work well with gitk.
You need to use a trick like https://stackoverflow.com/a/37375502
But there is no good solution when tracking a directory, like ethdev,
which had a file split recently, and a directory rename.
-----Original Message-----
> Date: Tue, 3 Apr 2018 10:28:59 +0100
> From: Ferruh Yigit <ferruh.yigit@intel.com>
> To: Thomas Monjalon <thomas@monjalon.net>, John McNamara
> <john.mcnamara@intel.com>, Marko Kovacevic <marko.kovacevic@intel.com>
> CC: dev@dpdk.org, Ferruh Yigit <ferruh.yigit@intel.com>, Bruce Richardson
> <bruce.richardson@intel.com>
> Subject: [dpdk-dev] [PATCH] ethdev: rename folder to library name
> X-Mailer: git-send-email 2.14.3
>
> Library folder name and output library name are same except a few flaws
> including librte_ether.
>
> This library is network device abstraction layer, the name "ethdev" fits
> better than "ether", and library & header files already named as ethdev.
>
> Also there is a rte_ether.h in the net library which can cause confusion.
>
> Signed-off-by: Ferruh Yigit <ferruh.yigit@intel.com>
I strongly support this change.
Acked-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
@@ -261,22 +261,22 @@ F: test/test/test_mbuf.c
Ethernet API
M: Thomas Monjalon <thomas@monjalon.net>
T: git://dpdk.org/next/dpdk-next-net
-F: lib/librte_ether/
+F: lib/librte_ethdev/
F: devtools/test-null.sh
Flow API
M: Adrien Mazarguil <adrien.mazarguil@6wind.com>
T: git://dpdk.org/next/dpdk-next-net
-F: lib/librte_ether/rte_flow*
+F: lib/librte_ethdev/rte_flow*
Traffic Management API - EXPERIMENTAL
M: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
T: git://dpdk.org/next/dpdk-next-tm
-F: lib/librte_ether/rte_tm*
+F: lib/librte_ethdev/rte_tm*
Traffic Metering and Policing API - EXPERIMENTAL
M: Cristian Dumitrescu <cristian.dumitrescu@intel.com>
-F: lib/librte_ether/rte_mtr*
+F: lib/librte_ethdev/rte_mtr*
Baseband API - EXPERIMENTAL
M: Amr Mokhtar <amr.mokhtar@intel.com>
@@ -48,7 +48,7 @@ INPUT = doc/api/doxy-api-index.md \
lib/librte_cryptodev \
lib/librte_distributor \
lib/librte_efd \
- lib/librte_ether \
+ lib/librte_ethdev \
lib/librte_eventdev \
lib/librte_flow_classify \
lib/librte_gro \
@@ -40,7 +40,7 @@ The lib directory contains::
+-- librte_cmdline # Command line interface helper
+-- librte_distributor # Packet distributor
+-- librte_eal # Environment abstraction layer
- +-- librte_ether # Generic interface to poll mode driver
+ +-- librte_ethdev # Generic interface to poll mode driver
+-- librte_hash # Hash library
+-- librte_ip_frag # IP fragmentation library
+-- librte_kni # Kernel NIC interface
@@ -28,7 +28,7 @@ The metering and policing stage typically sits on top of flow classification,
which is why the MTR objects are enabled through a special "meter" action.
The MTR objects are created and updated in their own name space (``rte_mtr``)
-within the ``librte_ether`` library. Whether an MTR object is private to a
+within the ``librte_ethdev`` library. Whether an MTR object is private to a
flow or potentially shared by several flows has to be specified at its
creation time.
@@ -18,9 +18,9 @@ DEPDIRS-librte_timer := librte_eal
DIRS-$(CONFIG_RTE_LIBRTE_CFGFILE) += librte_cfgfile
DIRS-$(CONFIG_RTE_LIBRTE_CMDLINE) += librte_cmdline
DEPDIRS-librte_cmdline := librte_eal
-DIRS-$(CONFIG_RTE_LIBRTE_ETHER) += librte_ether
-DEPDIRS-librte_ether := librte_net librte_eal librte_mempool librte_ring
-DEPDIRS-librte_ether += librte_mbuf
+DIRS-$(CONFIG_RTE_LIBRTE_ETHER) += librte_ethdev
+DEPDIRS-librte_ethdev := librte_net librte_eal librte_mempool librte_ring
+DEPDIRS-librte_ethdev += librte_mbuf
DIRS-$(CONFIG_RTE_LIBRTE_BBDEV) += librte_bbdev
DEPDIRS-librte_bbdev := librte_eal librte_mempool librte_mbuf
DIRS-$(CONFIG_RTE_LIBRTE_CRYPTODEV) += librte_cryptodev
@@ -28,14 +28,14 @@ DEPDIRS-librte_cryptodev := librte_eal librte_mempool librte_ring librte_mbuf
DEPDIRS-librte_cryptodev += librte_kvargs
DIRS-$(CONFIG_RTE_LIBRTE_SECURITY) += librte_security
DEPDIRS-librte_security := librte_eal librte_mempool librte_ring librte_mbuf
-DEPDIRS-librte_security += librte_ether
+DEPDIRS-librte_security += librte_ethdev
DEPDIRS-librte_security += librte_cryptodev
DIRS-$(CONFIG_RTE_LIBRTE_EVENTDEV) += librte_eventdev
-DEPDIRS-librte_eventdev := librte_eal librte_ring librte_ether librte_hash
+DEPDIRS-librte_eventdev := librte_eal librte_ring librte_ethdev librte_hash
DIRS-$(CONFIG_RTE_LIBRTE_RAWDEV) += librte_rawdev
-DEPDIRS-librte_rawdev := librte_eal librte_ether
+DEPDIRS-librte_rawdev := librte_eal librte_ethdev
DIRS-$(CONFIG_RTE_LIBRTE_VHOST) += librte_vhost
-DEPDIRS-librte_vhost := librte_eal librte_mempool librte_mbuf librte_ether \
+DEPDIRS-librte_vhost := librte_eal librte_mempool librte_mbuf librte_ethdev \
librte_net
DIRS-$(CONFIG_RTE_LIBRTE_HASH) += librte_hash
DEPDIRS-librte_hash := librte_eal librte_ring
@@ -50,18 +50,18 @@ DEPDIRS-librte_member := librte_eal librte_hash
DIRS-$(CONFIG_RTE_LIBRTE_NET) += librte_net
DEPDIRS-librte_net := librte_mbuf librte_eal
DIRS-$(CONFIG_RTE_LIBRTE_IP_FRAG) += librte_ip_frag
-DEPDIRS-librte_ip_frag := librte_eal librte_mempool librte_mbuf librte_ether
+DEPDIRS-librte_ip_frag := librte_eal librte_mempool librte_mbuf librte_ethdev
DEPDIRS-librte_ip_frag += librte_hash
DIRS-$(CONFIG_RTE_LIBRTE_GRO) += librte_gro
-DEPDIRS-librte_gro := librte_eal librte_mbuf librte_ether librte_net
+DEPDIRS-librte_gro := librte_eal librte_mbuf librte_ethdev librte_net
DIRS-$(CONFIG_RTE_LIBRTE_JOBSTATS) += librte_jobstats
DEPDIRS-librte_jobstats := librte_eal
DIRS-$(CONFIG_RTE_LIBRTE_METRICS) += librte_metrics
DEPDIRS-librte_metrics := librte_eal
DIRS-$(CONFIG_RTE_LIBRTE_BITRATE) += librte_bitratestats
-DEPDIRS-librte_bitratestats := librte_eal librte_metrics librte_ether
+DEPDIRS-librte_bitratestats := librte_eal librte_metrics librte_ethdev
DIRS-$(CONFIG_RTE_LIBRTE_LATENCY_STATS) += librte_latencystats
-DEPDIRS-librte_latencystats := librte_eal librte_metrics librte_ether librte_mbuf
+DEPDIRS-librte_latencystats := librte_eal librte_metrics librte_ethdev librte_mbuf
DIRS-$(CONFIG_RTE_LIBRTE_POWER) += librte_power
DEPDIRS-librte_power := librte_eal
DIRS-$(CONFIG_RTE_LIBRTE_METER) += librte_meter
@@ -74,9 +74,9 @@ DEPDIRS-librte_sched += librte_timer
DIRS-$(CONFIG_RTE_LIBRTE_KVARGS) += librte_kvargs
DEPDIRS-librte_kvargs := librte_eal
DIRS-$(CONFIG_RTE_LIBRTE_DISTRIBUTOR) += librte_distributor
-DEPDIRS-librte_distributor := librte_eal librte_mbuf librte_ether
+DEPDIRS-librte_distributor := librte_eal librte_mbuf librte_ethdev
DIRS-$(CONFIG_RTE_LIBRTE_PORT) += librte_port
-DEPDIRS-librte_port := librte_eal librte_mempool librte_mbuf librte_ether
+DEPDIRS-librte_port := librte_eal librte_mempool librte_mbuf librte_ethdev
DEPDIRS-librte_port += librte_ip_frag librte_sched
ifeq ($(CONFIG_RTE_LIBRTE_KNI),y)
DEPDIRS-librte_port += librte_kni
@@ -93,15 +93,15 @@ DEPDIRS-librte_pipeline += librte_table librte_port
DIRS-$(CONFIG_RTE_LIBRTE_REORDER) += librte_reorder
DEPDIRS-librte_reorder := librte_eal librte_mempool librte_mbuf
DIRS-$(CONFIG_RTE_LIBRTE_PDUMP) += librte_pdump
-DEPDIRS-librte_pdump := librte_eal librte_mempool librte_mbuf librte_ether
+DEPDIRS-librte_pdump := librte_eal librte_mempool librte_mbuf librte_ethdev
DIRS-$(CONFIG_RTE_LIBRTE_GSO) += librte_gso
-DEPDIRS-librte_gso := librte_eal librte_mbuf librte_ether librte_net
+DEPDIRS-librte_gso := librte_eal librte_mbuf librte_ethdev librte_net
DEPDIRS-librte_gso += librte_mempool
ifeq ($(CONFIG_RTE_EXEC_ENV_LINUXAPP),y)
DIRS-$(CONFIG_RTE_LIBRTE_KNI) += librte_kni
endif
-DEPDIRS-librte_kni := librte_eal librte_mempool librte_mbuf librte_ether
+DEPDIRS-librte_kni := librte_eal librte_mempool librte_mbuf librte_ethdev
DEPDIRS-librte_kni += librte_pci
include $(RTE_SDK)/mk/rte.subdir.mk
similarity index 100%
rename from lib/librte_ether/Makefile
rename to lib/librte_ethdev/Makefile
similarity index 100%
rename from lib/librte_ether/ethdev_profile.c
rename to lib/librte_ethdev/ethdev_profile.c
similarity index 100%
rename from lib/librte_ether/ethdev_profile.h
rename to lib/librte_ethdev/ethdev_profile.h
similarity index 100%
rename from lib/librte_ether/meson.build
rename to lib/librte_ethdev/meson.build
similarity index 100%
rename from lib/librte_ether/rte_dev_info.h
rename to lib/librte_ethdev/rte_dev_info.h
similarity index 100%
rename from lib/librte_ether/rte_eth_ctrl.h
rename to lib/librte_ethdev/rte_eth_ctrl.h
similarity index 100%
rename from lib/librte_ether/rte_ethdev.c
rename to lib/librte_ethdev/rte_ethdev.c
similarity index 100%
rename from lib/librte_ether/rte_ethdev.h
rename to lib/librte_ethdev/rte_ethdev.h
similarity index 100%
rename from lib/librte_ether/rte_ethdev_core.h
rename to lib/librte_ethdev/rte_ethdev_core.h
similarity index 100%
rename from lib/librte_ether/rte_ethdev_driver.h
rename to lib/librte_ethdev/rte_ethdev_driver.h
similarity index 100%
rename from lib/librte_ether/rte_ethdev_pci.h
rename to lib/librte_ethdev/rte_ethdev_pci.h
similarity index 100%
rename from lib/librte_ether/rte_ethdev_vdev.h
rename to lib/librte_ethdev/rte_ethdev_vdev.h
similarity index 100%
rename from lib/librte_ether/rte_ethdev_version.map
rename to lib/librte_ethdev/rte_ethdev_version.map
similarity index 100%
rename from lib/librte_ether/rte_flow.c
rename to lib/librte_ethdev/rte_flow.c
similarity index 100%
rename from lib/librte_ether/rte_flow.h
rename to lib/librte_ethdev/rte_flow.h
similarity index 100%
rename from lib/librte_ether/rte_flow_driver.h
rename to lib/librte_ethdev/rte_flow_driver.h
similarity index 100%
rename from lib/librte_ether/rte_mtr.c
rename to lib/librte_ethdev/rte_mtr.c
similarity index 100%
rename from lib/librte_ether/rte_mtr.h
rename to lib/librte_ethdev/rte_mtr.h
similarity index 100%
rename from lib/librte_ether/rte_mtr_driver.h
rename to lib/librte_ethdev/rte_mtr_driver.h
similarity index 100%
rename from lib/librte_ether/rte_tm.c
rename to lib/librte_ethdev/rte_tm.c
similarity index 100%
rename from lib/librte_ether/rte_tm.h
rename to lib/librte_ethdev/rte_tm.h
similarity index 100%
rename from lib/librte_ether/rte_tm_driver.h
rename to lib/librte_ethdev/rte_tm_driver.h
@@ -9,7 +9,7 @@
# given as a dep, no need to mention ring. This is especially true for the
# core libs which are widely reused, so their deps are kept to a minimum.
libraries = [ 'compat', # just a header, used for versioning
- 'eal', 'ring', 'mempool', 'mbuf', 'net', 'ether', 'pci', # core
+ 'eal', 'ring', 'mempool', 'mbuf', 'net', 'ethdev', 'pci', # core
'metrics', # bitrate/latency stats depends on this
'hash', # efd depends on this
'kvargs', # cryptodev depends on this