On Wed, Jan 15, 2014 at 06:21:30PM +0200, Horia Geantă wrote: > On 1/15/2014 5:46 PM, Phil Sutter wrote: > > Hi, > > > > On Wed, Jan 15, 2014 at 05:22:18PM +0200, Horia Geanta wrote: > >> From: Cosmin Paraschiv <cosmin.parasc...@freescale.com> > >> > >> In the 3.12 Linux kernel, the INIT_COMPLETION macro has been replaced > >> with an inline function, reinit_completion [1][2]. We are currently > >> using the 3.13-rc3 Linux kernel, which leads to the following error: > >> > >> cryptlib.c:220:2: error: implicit declaration of function > >> 'INIT_COMPLETION' [-Werror=implicit-function-declaration] > >> INIT_COMPLETION(cdata->async.result->completion); > >> > >> [1] > >> https://github.com/torvalds/linux/commit/c32f74ab2872994bc8336ed367313da3139350ca > >> [2] > >> https://github.com/torvalds/linux/commit/62026aedaacedbe1ffe94a3599ad4acd8ecdf587 > >> > >> Signed-off-by: Cosmin Paraschiv <cosmin.parasc...@freescale.com> > >> Reviewed-by: Cristian Stoica <cristian.sto...@freescale.com> > >> Tested-by: Cristian Stoica <cristian.sto...@freescale.com> > >> Signed-off-by: Horia Geanta <horia.gea...@freescale.com> > >> --- > >> cryptlib.c | 8 ++++---- > >> 1 file changed, 4 insertions(+), 4 deletions(-) > >> > >> diff --git a/cryptlib.c b/cryptlib.c > >> index 54d5d41..a923c14 100644 > >> --- a/cryptlib.c > >> +++ b/cryptlib.c > >> @@ -217,7 +217,7 @@ ssize_t cryptodev_cipher_encrypt(struct cipher_data > >> *cdata, > >> { > >> int ret; > >> > >> - INIT_COMPLETION(cdata->async.result->completion); > >> + reinit_completion(&cdata->async.result->completion); > > > > Wouldn't it be better to conditionally define INIT_COMPLETION to > > reinit_completion for newer kernels? Doesn't this break downwards > > compatibility? > > That's a good question. > Before posting v2, I would like to find out what is the goal here - to > be compatible with kernel 2.6.x onwards or...?
To not break existing compatibility. I have no idea which is the earliest version of Linux cryptodev-linux runs on, but the proposed change smells like breaking cryptodev-linux for everything before 3.12, no? > I am sure this is not the first time changes in the kernel affect > out-of-tree modules. Yes, that's why most of them are more and more cluttered with those #IF checks over time. Probably one of the main reasons why (sane) people try to get their kernel modules mainline sooner or later. Greetings, Phil _______________________________________________ Cryptodev-linux-devel mailing list Cryptodev-linux-devel@gna.org https://mail.gna.org/listinfo/cryptodev-linux-devel
