From: Ferruh Yigit <ferruh.yigit@intel.com>
To: David Marchand <david.marchand@redhat.com>
Cc: Gagandeep Singh <g.singh@nxp.com>,
Hemant Agrawal <hemant.agrawal@nxp.com>,
Akhil Goyal <akhil.goyal@nxp.com>, dev <dev@dpdk.org>,
dpdk stable <stable@dpdk.org>
Subject: Re: [dpdk-dev] [dpdk-stable] [PATCH 05/10] crypto/dpaa2_sec: fix global variable multiple definitions
Date: Fri, 25 Oct 2019 11:25:25 +0100 [thread overview]
Message-ID: <3fafc4a9-4514-60aa-b4dd-a1c5c6053112@intel.com> (raw)
In-Reply-To: <CAJFAV8xomOod6sbadiBDsoagQCGEoZ1iMsi9k5vcmTR37FjcMw@mail.gmail.com>
On 10/24/2019 5:56 PM, David Marchand wrote:
> On Thu, Oct 24, 2019 at 4:55 PM Ferruh Yigit <ferruh.yigit@intel.com> wrote:
>>
>> On 10/24/2019 3:53 PM, David Marchand wrote:
>>> On Thu, Sep 5, 2019 at 4:54 PM Ferruh Yigit <ferruh.yigit@intel.com> wrote:
>>>>
>>>> 'rta_sec_era' global variable is defined in 'crypto/dpaa2_sec',
>>>> 'crypto/caam_jr' & 'crypto/dpaa_sec' PMDs with the same name. This means
>>>> they share same storage when application linked with static DPDK
>>>> library, which is not the intention. Three differing drivers sharing
>>>> same global variable is a defect.
>>>>
>>>> Variable has been used by multiple header files in static inline
>>>> functions and these header files are shared by all three PMDs, this
>>>> forces using same variable name in three PMDs.
>>>>
>>>> Since the variable is used only single .c file in 'crypto/dpaa2_sec' &
>>>> 'crypto/dpaa_sec' converting global variable to 'static', this requires
>>>> removing 'extern' definition from header files, and this requires moving
>>>> the global variable definition before including those headers in .c
>>>> files.
>>>>
>>>> For 'crypto/caam_jr' multiple .c files exists and includes these
>>>> headers which prevent making variable static, so only one file has the
>>>> global variable and others has 'extern' keyword on .c files.
>>>>
>>>> This change should let all three drivers have their own storage for the
>>>> 'rta_sec_era' global variable without multiple definitions.
>>>>
>>>> Fixes: 1d678de329ab ("crypto/caam_jr: add basic job ring routines")
>>>> Fixes: c3e85bdcc6e6 ("crypto/dpaa_sec: add crypto driver for NXP DPAA platform")
>>>> Cc: stable@dpdk.org
>>>
>>> Hit a build issue, with gcc, static build:
>>>
>>> ../drivers/crypto/dpaa_sec/dpaa_sec.c:33:25: error: static declaration
>>> of ‘rta_sec_era’ follows non-static declaration
>>> 33 | static enum rta_sec_era rta_sec_era;
>>> | ^~~~~~~~~~~
>>> In file included from
>>> ../drivers/crypto/dpaa_sec/../dpaa2_sec/hw/rta/sec_run_time_asm.h:10,
>>> from ../drivers/crypto/dpaa_sec/dpaa_sec.c:32:
>>> ../drivers/crypto/dpaa_sec/../dpaa2_sec/hw/desc.h:21:25: note:
>>> previous declaration of ‘rta_sec_era’ was here
>>> 21 | extern enum rta_sec_era rta_sec_era;
>>> | ^~~~~~~~~~~
>>>
>>> And
>>>
>>> ../drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c:33:25: error: static
>>> declaration of ‘rta_sec_era’ follows non-static declaration
>>> 33 | static enum rta_sec_era rta_sec_era = RTA_SEC_ERA_8;
>>> | ^~~~~~~~~~~
>>> In file included from ../drivers/crypto/dpaa2_sec/hw/rta/sec_run_time_asm.h:10,
>>> from ../drivers/crypto/dpaa2_sec/dpaa2_sec_dpseci.c:32:
>>> ../drivers/crypto/dpaa2_sec/hw/desc.h:21:25: note: previous
>>> declaration of ‘rta_sec_era’ was here
>>> 21 | extern enum rta_sec_era rta_sec_era;
>>> | ^~~~~~~~~~~
>>> [5/86] Compiling C object
>>> 'drivers/a715181@@tmp_rte_pmd_caam_jr@sta/crypto_caam_jr_caam_jr.c.o'.
>>>
>>>
>> Would you be OK to defer the set to the next release?
>
> Either that or I only take the other 7 patches that are fine (dropped
> patch 4 and 9, as discussed some time ago).
>
If you are OK to pick the patches in the series, +1 from my end.
next prev parent reply other threads:[~2019-10-25 10:25 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-09-05 14:53 [dpdk-dev] [PATCH 00/10] " Ferruh Yigit
2019-09-05 14:53 ` [dpdk-dev] [PATCH 01/10] bus/fslmc: " Ferruh Yigit
2019-09-10 16:36 ` Sachin Saxena
2019-09-05 14:53 ` [dpdk-dev] [PATCH 02/10] net/igb: " Ferruh Yigit
2019-09-05 14:53 ` [dpdk-dev] [PATCH 03/10] crypto/null: " Ferruh Yigit
2019-09-05 14:53 ` [dpdk-dev] [PATCH 04/10] crypto/octeontx: " Ferruh Yigit
2019-09-26 11:20 ` [dpdk-dev] [EXT] " Anoob Joseph
2019-09-26 18:03 ` Ferruh Yigit
2019-09-05 14:53 ` [dpdk-dev] [PATCH 05/10] crypto/dpaa2_sec: " Ferruh Yigit
2019-09-10 16:53 ` Sachin Saxena
2019-10-24 14:53 ` [dpdk-dev] [dpdk-stable] " David Marchand
2019-10-24 14:55 ` Ferruh Yigit
2019-10-24 16:56 ` David Marchand
2019-10-25 10:25 ` Ferruh Yigit [this message]
2019-09-05 14:53 ` [dpdk-dev] [PATCH 06/10] crypto/virtio: " Ferruh Yigit
2019-09-05 14:53 ` [dpdk-dev] [PATCH 07/10] compress/octeontx: " Ferruh Yigit
2019-09-05 16:00 ` [dpdk-dev] [EXT] " Ashish Gupta
2019-09-05 14:53 ` [dpdk-dev] [PATCH 08/10] app/testpmd: " Ferruh Yigit
2019-10-12 12:36 ` [dpdk-dev] [dpdk-stable] " Thomas Monjalon
2019-09-05 14:53 ` [dpdk-dev] [PATCH 09/10] app/test-pipeline: " Ferruh Yigit
2019-09-05 15:01 ` Dumitrescu, Cristian
2019-09-05 15:19 ` Ferruh Yigit
2019-09-05 14:53 ` [dpdk-dev] [PATCH 10/10] test: " Ferruh Yigit
2019-09-05 15:45 ` Honnappa Nagarahalli
2019-10-25 12:53 ` [dpdk-dev] [PATCH 00/10] " David Marchand
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=3fafc4a9-4514-60aa-b4dd-a1c5c6053112@intel.com \
--to=ferruh.yigit@intel.com \
--cc=akhil.goyal@nxp.com \
--cc=david.marchand@redhat.com \
--cc=dev@dpdk.org \
--cc=g.singh@nxp.com \
--cc=hemant.agrawal@nxp.com \
--cc=stable@dpdk.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).