* Re: [PATCH v4 2/5] crypto: AF_ALG: add AEAD support
From: Herbert Xu @ 2014-12-05 15:46 UTC (permalink / raw)
To: Stephan Mueller
Cc: Daniel Borkmann, 'Quentin Gouchet', 'LKML',
linux-crypto, linux-api
In-Reply-To: <4875720.jRoMDtjHB4@tachyon.chronox.de>
On Wed, Dec 03, 2014 at 08:57:24PM +0100, Stephan Mueller wrote:
>
> + if (ctx->merge) {
> + sg = sgl->sg + sgl->cur - 1;
> + len = min_t(unsigned long, len,
> + PAGE_SIZE - sg->offset - sg->length);
> +
> + err = memcpy_fromiovec(page_address(sg_page(sg)) +
> + sg->offset + sg->length,
> + msg->msg_iov, len);
> + if (err)
> + goto unlock;
> +
> + sg->length += len;
> + ctx->merge = (sg->offset + sg->length) & (PAGE_SIZE - 1);
> +
> + ctx->used += len;
> + copied += len;
> + size -= len;
> + }
Any reason why you got rid of the outer loop here? This will cause
short writes I think.
> +static struct proto_ops algif_aead_ops = {
> + .family = PF_ALG,
> +
> + .connect = sock_no_connect,
> + .socketpair = sock_no_socketpair,
> + .getname = sock_no_getname,
> + .ioctl = sock_no_ioctl,
> + .listen = sock_no_listen,
> + .shutdown = sock_no_shutdown,
> + .getsockopt = sock_no_getsockopt,
> + .mmap = sock_no_mmap,
> + .bind = sock_no_bind,
> + .accept = sock_no_accept,
> +
> + .release = af_alg_release,
> + .sendmsg = aead_sendmsg,
> + .sendpage = aead_sendpage,
> + .recvmsg = aead_recvmsg,
> + .poll = aead_poll,
> + .setsockopt = aead_setsockopt,
No it should go into the parent setsockopt. Perhaps add a setsockopt
to af_alg_type in order to keep this out of the generic code.
Thanks,
--
Email: Herbert Xu <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
^ permalink raw reply
* Re: [PATCH v4 4/5] crypto: AF_ALG: add random number generator support
From: Herbert Xu @ 2014-12-05 15:53 UTC (permalink / raw)
To: Stephan Mueller
Cc: Daniel Borkmann, 'Quentin Gouchet', 'LKML',
linux-crypto, linux-api
In-Reply-To: <1480776.TeURZV8VLa@tachyon.chronox.de>
On Wed, Dec 03, 2014 at 08:59:01PM +0100, Stephan Mueller wrote:
>
> +static int rng_recvmsg(struct kiocb *unused, struct socket *sock,
> + struct msghdr *msg, size_t len, int flags)
> +{
> + struct sock *sk = sock->sk;
> + struct alg_sock *ask = alg_sk(sk);
> + struct rng_ctx *ctx = ask->private;
> + int err = -EFAULT;
> +
> + if (len == 0)
> + return 0;
> + if (len > MAXSIZE)
> + len = MAXSIZE;
> +
> + lock_sock(sk);
This lock simply protects ctx->result. Since you're using a
tiny buffer why not just put it on the stack?
> + u8 *buf = kmalloc(seedsize, GFP_KERNEL);
> + if (!buf)
> + goto err;
> + get_random_bytes(buf, seedsize);
> + ret = crypto_rng_reset(private, buf, len);
I think you should leave the seeding and the seed to the user.
Perhaps do it through setsockopt (on the parent socket).
Cheers,
--
Email: Herbert Xu <herbert@gondor.apana.org.au>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
^ permalink raw reply
* Re: [PATCH v4 1/5] crypto: AF_ALG: add user space interface for AEAD
From: Herbert Xu @ 2014-12-05 16:01 UTC (permalink / raw)
To: Stephan Mueller
Cc: Daniel Borkmann, 'Quentin Gouchet', 'LKML',
linux-crypto-u79uwXL29TY76Z2rM5mHXA,
linux-api-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <4746255.Vltg6sM4x3-PJstQz4BMNNP20K/wil9xYQuADTiUCJX@public.gmane.org>
On Wed, Dec 03, 2014 at 08:55:42PM +0100, Stephan Mueller wrote:
> AEAD requires the caller to specify the following information separate
> from the data stream. This information allows the AEAD interface handler
> to identify the AAD, ciphertext/plaintext and the authentication tag:
>
> * Associated authentication data of arbitrary length and
> length
>
> * Length of authentication tag for encryption
>
> Signed-off-by: Stephan Mueller <smueller-T9tCv8IpfcWELgA04lAiVw@public.gmane.org>
Patch applied.
--
Email: Herbert Xu <herbert-lOAM2aK0SrRLBo1qDEOMRrpzq4S04n8Q@public.gmane.org>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
^ permalink raw reply
* Re: [tpmdd-devel] [PATCH v9 8/8] tpm: TPM 2.0 FIFO Interface
From: Stefan Berger @ 2014-12-05 20:44 UTC (permalink / raw)
To: Peter Huewe
Cc: christophe.ricard-Re5JQEeQqe8AvxtiuMwx3w,
josh.triplett-ral2JQCrhuEAvxtiuMwx3w,
linux-api-u79uwXL29TY76Z2rM5mHXA, Ashley Lai,
linux-kernel-u79uwXL29TY76Z2rM5mHXA, Will Arthur,
tpmdd-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f,
jason.gunthorpe-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/,
trousers-tech-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f
In-Reply-To: <trinity-0fea8cea-4011-4d04-af4a-e7583a7184b4-1417791661803@3capp-gmx-bs52>
On 12/05/2014 10:01 AM, Peter Huewe wrote:
>>> Am Donnerstag, 4. Dezember 2014, 06:55:18 schrieb Jarkko Sakkinen:
>>>> From: Will Arthur <will.c.arthur-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
>>>>
>>>> Detect TPM 2.0 by using the extended STS (STS3) register. For TPM 2.0,
>>>> instead of calling tpm_get_timeouts(), assign duration and timeout
>>>> values defined in the TPM 2.0 PTP specification.
>>>>
>>>> Signed-off-by: Will Arthur <will.c.arthur-ral2JQCrhuEAvxtiuMwx3w@public.gmane.org>
>>>> Signed-off-by: Jarkko Sakkinen <jarkko.sakkinen-VuQAYsv1563Yd54FQh9/CA@public.gmane.org>
>>>> + sts3 = ioread8(chip->vendor.iobase + TPM_STS3(1));
>>>> + if ((sts3 & TPM_STS3_TPM2_FAM) == TPM_STS3_TPM2_FAM)
>>>> + chip->flags = TPM_CHIP_FLAG_TPM2;
>>>> +
>>>>
>>> When loading tpm_tis force=1 with my tpm1.2 chip on a machine without bios
>>> integration, it fets detected as a TPM2.0 chip :/
>>>
>>> sudo rmmod tpm_tis
>>> # modprobe tpm_tis force=1
>>> modprobe: ERROR: could not insert 'tpm_tis': No such device
>>> # dmesg
>>> [ 263.903828] tpm_tis tpm_tis: 2.0 TPM (device-id 0xB, rev-id 16)
>>> [ 263.948049] tpm_tis tpm_tis: A TPM error (10) occurred continue selftest
>>> [ 263.948120] tpm_tis tpm_tis: TPM self test failed
>>>
>>>
>>> sts3 is reported as 0xff from my TPM1.2
>>>
>>
>> Hmm,
>> my TPM2.0 chip also reports sts3 as 0xff (when loading with force=1 on a
>> machine without bios integration)
>>
>> [ 307.095344] sts3 ff
>> [ 307.095366] tpm_tis tpm_tis: 2.0 TPM (device-id 0x1A, rev-id 16)
>> [ 307.140047] tpm_tis tpm_tis: A TPM error (256) occurred continue selftest
>> [ 307.140056] tpm_tis tpm_tis: TPM self test failed
>
> You are reading "sts3" - before requesting the locality and thus it returns 0xff for a TPM20 chip as well.
> --> You have to have an active locality first.
>
>
> For a TPM2.0 0xFF is not a valid value (if active locality is set), since reading commandCancel and resetEstablishment bit always return 0 on reads (according to spec).
>
> --> 0xFF should be treated as a TPM1.2 (older tpms with TIS 1.2)
> --> 0x04 should be treated as TPM 2.0
> --> 0x08 should be treated as TPM1.2 (newer tpms with TIS1.3 enhanced)
> + sts3 = ioread8(chip->vendor.iobase + TPM_STS3(1));
> + if ((sts3 & TPM_STS3_TPM2_FAM) == TPM_STS3_TPM2_FAM)
> + chip->flags = TPM_CHIP_FLAG_TPM2;
#define TPM_STS3_TPM2_FAM_MASK = 0x0c
then:
if ((sts3 & TPM_STS3_TPM2_FAM_MASK) == TPM_STS3_TPM2_FAM) ...
[see my previous review on this]
Stefan
>
>
>
> Thanks,
> Peter
>
> ------------------------------------------------------------------------------
> Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
> from Actuate! Instantly Supercharge Your Business Reports and Dashboards
> with Interactivity, Sharing, Native Excel Exports, App Integration & more
> Get technology previously reserved for billion-dollar corporations, FREE
> http://pubads.g.doubleclick.net/gampad/clk?id=164703151&iu=/4140/ostg.clktrk
> _______________________________________________
> tpmdd-devel mailing list
> tpmdd-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f@public.gmane.org
> https://lists.sourceforge.net/lists/listinfo/tpmdd-devel
>
^ permalink raw reply
* Re: [PATCH] arch: uapi: asm: mman.h: Let MADV_FREE have same value for all architectures
From: Chen Gang @ 2014-12-05 21:51 UTC (permalink / raw)
To: Darrick J. Wong
Cc: akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org,
rth-hL46jP5Bxq7R7s880joybQ@public.gmane.org,
ink-biIs/Y0ymYJMZLIVYojuPNP0rXTJTi09@public.gmane.org,
mattst88-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org, Ralf Baechle,
jejb-6jwH94ZQLHl74goWV3ctuw@public.gmane.org,
deller-Mmb7MZpHnFY@public.gmane.org,
chris-YvXeqwSYzG2sTnJN9+BGXg@public.gmane.org,
jcmvbkbc-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org, Arnd Bergmann,
minchan-DgEjT+Ai2ygdnm+yROfE0A,
linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
linux-arch-u79uwXL29TY76Z2rM5mHXA,
linux-api-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <20141205001446.GC10064-PTl6brltDGh4DFYR7WNSRA@public.gmane.org>
On 12/05/2014 08:14 AM, Darrick J. Wong wrote:
> On Fri, Dec 05, 2014 at 06:58:29AM +0800, Chen Gang wrote:
>> For uapi, need try to let all macros have same value, and MADV_FREE is
>> added into main branch recently, so need redefine MADV_FREE for it.
>>
>> At present, '8' can be shared with all architectures, so redefine it to
>> '8'.
>
> Hmmm... you might want to talk to the dietlibc people if you do this:
>
> $ grep -rn MADV_FREE /usr/include/
> /usr/include/diet/sys/mman.h:101:#define MADV_FREE 0x5 /* (Solaris) contents can be freed */
>
> Though the sole definition only exists on sparc/powerpc/powerpc64, so who
> knows...
>
Excuse me, I am not quite familiar about dietlibc. If possible, please
help forward this mail to them, and Cc me (if necessary).
Thanks.
--
Chen Gang
Open share and attitude like air water and life which God blessed
^ permalink raw reply
* Re: [PATCH v4 2/5] crypto: AF_ALG: add AEAD support
From: Stephan Mueller @ 2014-12-05 21:51 UTC (permalink / raw)
To: Herbert Xu
Cc: Daniel Borkmann, 'Quentin Gouchet', 'LKML',
linux-crypto, linux-api
In-Reply-To: <20141205154606.GA30180@gondor.apana.org.au>
Am Freitag, 5. Dezember 2014, 23:46:06 schrieb Herbert Xu:
Hi Herbert,
> On Wed, Dec 03, 2014 at 08:57:24PM +0100, Stephan Mueller wrote:
> > + if (ctx->merge) {
> > + sg = sgl->sg + sgl->cur - 1;
> > + len = min_t(unsigned long, len,
> > + PAGE_SIZE - sg->offset - sg->length);
> > +
> > + err = memcpy_fromiovec(page_address(sg_page(sg)) +
> > + sg->offset + sg->length,
> > + msg->msg_iov, len);
> > + if (err)
> > + goto unlock;
> > +
> > + sg->length += len;
> > + ctx->merge = (sg->offset + sg->length) & (PAGE_SIZE - 1);
> > +
> > + ctx->used += len;
> > + copied += len;
> > + size -= len;
> > + }
>
> Any reason why you got rid of the outer loop here? This will cause
> short writes I think.
You are absolutely right. I removed it as I do not have the multiple sgl
entries. But now as you mentioned it, I still need it if size >
aead_sndbuf(sk).
This will be fixed in the next installment.
>
> > +static struct proto_ops algif_aead_ops = {
> > + .family = PF_ALG,
> > +
> > + .connect = sock_no_connect,
> > + .socketpair = sock_no_socketpair,
> > + .getname = sock_no_getname,
> > + .ioctl = sock_no_ioctl,
> > + .listen = sock_no_listen,
> > + .shutdown = sock_no_shutdown,
> > + .getsockopt = sock_no_getsockopt,
> > + .mmap = sock_no_mmap,
> > + .bind = sock_no_bind,
> > + .accept = sock_no_accept,
> > +
> > + .release = af_alg_release,
> > + .sendmsg = aead_sendmsg,
> > + .sendpage = aead_sendpage,
> > + .recvmsg = aead_recvmsg,
> > + .poll = aead_poll,
> > + .setsockopt = aead_setsockopt,
>
> No it should go into the parent setsockopt. Perhaps add a setsockopt
> to af_alg_type in order to keep this out of the generic code.
I was thinking about that for quite a while. My thought for the current
approach was that the actual cipher operation happens in the child FD (i.e.
after accept). AAD is delivered to that FD. Therefore, I thought that the size
of the AAD can be specific to that operational FD.
If we move it to the parent setsockopt, all child FDs have the same AAD size.
If you think that this is the right course of action, I can surely implement
that.
Would you please be so kind and help me understand when some operations are
intended for the parent FD and when for the child FD?
Thanks
--
Ciao
Stephan
^ permalink raw reply
* Re: [PATCH] arch: uapi: asm: mman.h: Let MADV_FREE have same value for all architectures
From: Chen Gang @ 2014-12-05 22:00 UTC (permalink / raw)
To: Minchan Kim
Cc: akpm@linux-foundation.org, rth@twiddle.net,
ink@jurassic.park.msu.ru, mattst88@gmail.com, Ralf Baechle,
jejb@parisc-linux.org, deller@gmx.de, chris@zankel.net,
jcmvbkbc@gmail.com, Arnd Bergmann, linux-kernel@vger.kernel.org,
linux-arch, linux-api, Geert Uytterhoeven, Carlos O'Donell
In-Reply-To: <20141205065417.GA3358@bbox>
On 12/05/2014 02:54 PM, Minchan Kim wrote:
> On Fri, Dec 05, 2014 at 06:58:29AM +0800, Chen Gang wrote:
>> For uapi, need try to let all macros have same value, and MADV_FREE is
>> added into main branch recently, so need redefine MADV_FREE for it.
>>
>> At present, '8' can be shared with all architectures, so redefine it to
>> '8'.
>>
>> Signed-off-by: Chen Gang <gang.chen.5i5j@gmail.com>
>
>
> Hello Chen,
>
> Thanks for looking at this.
> Feel free to add my sign.
>
> Acked-by: Minchan Kim <minchan@kernel.org>
>
OK, thanks.
Originally I sent the same patch like you sent (but later than yours).
Geert suggested to use same value for MADV_FREE, and Carlos confirmed
it and suggested more things (sorry, I forgot to Cc them in this mail).
Thanks.
--
Chen Gang
Open share and attitude like air water and life which God blessed
^ permalink raw reply
* Re: [PATCH v4 4/5] crypto: AF_ALG: add random number generator support
From: Stephan Mueller @ 2014-12-05 22:02 UTC (permalink / raw)
To: Herbert Xu
Cc: Daniel Borkmann, 'Quentin Gouchet', 'LKML',
linux-crypto-u79uwXL29TY76Z2rM5mHXA,
linux-api-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <20141205155359.GA30224-lOAM2aK0SrRLBo1qDEOMRrpzq4S04n8Q@public.gmane.org>
Am Freitag, 5. Dezember 2014, 23:53:59 schrieb Herbert Xu:
Hi Herbert,
> On Wed, Dec 03, 2014 at 08:59:01PM +0100, Stephan Mueller wrote:
> > +static int rng_recvmsg(struct kiocb *unused, struct socket *sock,
> > + struct msghdr *msg, size_t len, int flags)
> > +{
> > + struct sock *sk = sock->sk;
> > + struct alg_sock *ask = alg_sk(sk);
> > + struct rng_ctx *ctx = ask->private;
> > + int err = -EFAULT;
> > +
> > + if (len == 0)
> > + return 0;
> > + if (len > MAXSIZE)
> > + len = MAXSIZE;
> > +
> > + lock_sock(sk);
>
> This lock simply protects ctx->result. Since you're using a
> tiny buffer why not just put it on the stack?
When I developed the DRBG code, I got comments that 128 byte variables shall
not be on the stack in kernel code.
But if you agree that I can put a 128 byte variable on the stack, I will see
it done.
>
> > + u8 *buf = kmalloc(seedsize, GFP_KERNEL);
> > + if (!buf)
> > + goto err;
> > + get_random_bytes(buf, seedsize);
> > + ret = crypto_rng_reset(private, buf, len);
>
> I think you should leave the seeding and the seed to the user.
> Perhaps do it through setsockopt (on the parent socket).
Sure. But please note that the seeding happens only when seedsize > 0. Such
seeding therefore is not performed for krng, and the DRBG because both seed
automatically.
Therefore, may I propose the following: We offer a setsockopt for (re)seeding.
For all RNGs with seedsize > 0, we return EAGAIN for recvmsg until a
setsockopt for at least seedsize is provided. That would imply that krng and
DRBG would be usable without seeding from user space.
--
Ciao
Stephan
^ permalink raw reply
* Re: [PATCH] uapi: fix to export linux/vm_sockets.h
From: David Miller @ 2014-12-06 5:36 UTC (permalink / raw)
To: yamada.m-NAum8xwdG0+S7A1Ibl2khg
Cc: netdev-u79uwXL29TY76Z2rM5mHXA,
stephen-OTpzqLSitTUnbdJkjeBofR2eb7JE58TQ,
gregkh-hQyY1W1yCW8ekmWlsbkhG0B+6BGkLq7r,
ast-uqk4Ao+rVK5Wk0Htik3J/w, piotr.krol-zILERgsGILsAvxtiuMwx3w,
akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b,
sakari.ailus-VuQAYsv1563Yd54FQh9/CA,
linux-api-u79uwXL29TY76Z2rM5mHXA,
linux-kernel-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1417396577-32499-1-git-send-email-yamada.m-NAum8xwdG0+S7A1Ibl2khg@public.gmane.org>
From: Masahiro Yamada <yamada.m-NAum8xwdG0+S7A1Ibl2khg@public.gmane.org>
Date: Mon, 1 Dec 2014 10:16:17 +0900
> A typo "header=y" was introduced by commit 7071cf7fc435
> (uapi: add missing network related headers to kbuild).
>
> Signed-off-by: Masahiro Yamada <yamada.m-NAum8xwdG0+S7A1Ibl2khg@public.gmane.org>
> Cc: Stephen Hemminger <stephen-OTpzqLSitTUnbdJkjeBofR2eb7JE58TQ@public.gmane.org>
Applied.
^ permalink raw reply
* Re: [PATCH v2 net-next 0/6] allow eBPF programs to be attached to sockets
From: David Miller @ 2014-12-06 5:48 UTC (permalink / raw)
To: ast-uqk4Ao+rVK5Wk0Htik3J/w
Cc: mingo-DgEjT+Ai2ygdnm+yROfE0A, luto-kltTT9wpgjJwATOyAt5JVQ,
dborkman-H+wXaHxf7aLQT0dZR+AlfA,
hannes-tFNcAqjVMyqKXQKiL6tip0B+6BGkLq7r,
edumazet-hpIqsD4AKlfQT0dZR+AlfA, linux-api-u79uwXL29TY76Z2rM5mHXA,
netdev-u79uwXL29TY76Z2rM5mHXA,
linux-kernel-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1417475199-15950-1-git-send-email-ast-uqk4Ao+rVK5Wk0Htik3J/w@public.gmane.org>
From: Alexei Starovoitov <ast-uqk4Ao+rVK5Wk0Htik3J/w@public.gmane.org>
Date: Mon, 1 Dec 2014 15:06:33 -0800
> V1->V2:
>
> fixed comments in sample code to state clearly that packet data is accessed
> with LD_ABS instructions and not internal skb fields.
> Also replaced constants in:
> BPF_LD_ABS(BPF_B, 14 + 9 /* R0 = ip->proto */),
> with:
> BPF_LD_ABS(BPF_B, ETH_HLEN + offsetof(struct iphdr, protocol) /* R0 = ip->proto */),
>
> V1 cover:
>
> Introduce BPF_PROG_TYPE_SOCKET_FILTER type of eBPF programs that can be
> attached to sockets with setsockopt().
> Allow such programs to access maps via lookup/update/delete helpers.
>
> This feature was previewed by bpf manpage in commit b4fc1a460f30("Merge branch 'bpf-next'")
> Now it can actually run.
>
> 1st patch adds LD_ABS/LD_IND instruction verification and
> 2nd patch adds new setsockopt() flag.
> Patches 3-6 are examples in assembler and in C.
>
> Though native eBPF programs are way more powerful than classic filters
> (attachable through similar setsockopt() call), they don't have skb field
> accessors yet. Like skb->pkt_type, skb->dev->ifindex are not accessible.
> There are sevaral ways to achieve that. That will be in the next set of patches.
> So in this set native eBPF programs can only read data from packet and
> access maps.
>
> The most powerful example is sockex2_kern.c from patch 6 where ~200 lines of C
> are compiled into ~300 of eBPF instructions.
> It shows how quite complex packet parsing can be done.
>
> LLVM used to build examples is at https://github.com/iovisor/llvm
> which is fork of llvm trunk that I'm cleaning up for upstreaming.
Series applied, thanks.
^ permalink raw reply
* Re: Re: [PATCH v9 6/8] tpm: TPM 2.0 baseline support
From: Jarkko Sakkinen @ 2014-12-06 11:54 UTC (permalink / raw)
To: Peter Huewe
Cc: Ashley Lai, Marcel Selhorst,
tpmdd-devel-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f,
linux-kernel-u79uwXL29TY76Z2rM5mHXA,
josh.triplett-ral2JQCrhuEAvxtiuMwx3w,
christophe.ricard-Re5JQEeQqe8AvxtiuMwx3w,
jason.gunthorpe-ePGOBjL8dl3ta4EC/59zMFaTQe2KTcn/,
linux-api-u79uwXL29TY76Z2rM5mHXA,
trousers-tech-5NWGOfrQmneRv+LV9MX5uipxlwaOVQ5f, Will Arthur
In-Reply-To: <trinity-78b59db8-b937-4ee0-989b-557fcbf612db-1417790149807@3capp-gmx-bs52>
On Fri, Dec 05, 2014 at 03:35:49PM +0100, Peter Huewe wrote:
> Hi Jarkko,
>
>
> >is it cool if I prepare a separate set of fixes for all issues
> >in v9? I do not see any problem that could not be fixed without
> >major structural changes.
>
>
>
> to be honest - I would prefer a v10 with the fixes included,
> especially since there were quite a few issues/remarks.
> If you already have fixes post them as patches on top of v9 so we
> can test, and then squash them in.
It's as easy for me to just prepare a new patch set.
> I think we are still in schedule if you do fixes over the weekend
> - so we hopefully have a _great_ v10 on monday/tuesday
Yeah, I apologize for the state of v9. I admit that I rushed with it
a bit. I'll prepare a new patch set over the weekend.
> I'll send out the first pull request without TPM2.0 support later
> this evening,
> and then send out another one by next week - this should be fine.
Great!
> The next testing slot is Sunday evening since I'm travelling
> myself (and don't have TPM hardware with me).
>
>
> Thanks,
> Peter
/Jarkko
^ permalink raw reply
* [PATCH v4] media: platform: add VPFE capture driver support for AM437X
From: Lad, Prabhakar @ 2014-12-06 13:07 UTC (permalink / raw)
To: LMML, devicetree, linux-api, Hans Verkuil
Cc: LKML, Hans Verkuil, Lad, Prabhakar
From: Benoit Parrot <bparrot@ti.com>
This patch adds Video Processing Front End (VPFE) driver for
AM437X family of devices
Driver supports the following:
- V4L2 API using MMAP buffer access based on videobuf2 api
- Asynchronous sensor/decoder sub device registration
- DT support
Signed-off-by: Benoit Parrot <bparrot@ti.com>
Signed-off-by: Darren Etheridge <detheridge@ti.com>
Signed-off-by: Lad, Prabhakar <prabhakar.csengg@gmail.com>
---
Changes for v4:
1: Fixed review comments pointed by Hans to check
ycbcr_enc and quantization while comparing formats,
fixed the release function and dropped compose cases
for vpfe_g_selection as compose isn't supported.
.../devicetree/bindings/media/ti-am437x-vpfe.txt | 61 +
MAINTAINERS | 9 +
drivers/media/platform/Kconfig | 1 +
drivers/media/platform/Makefile | 2 +
drivers/media/platform/am437x/Kconfig | 11 +
drivers/media/platform/am437x/Makefile | 3 +
drivers/media/platform/am437x/am437x-vpfe.c | 2778 ++++++++++++++++++++
drivers/media/platform/am437x/am437x-vpfe.h | 283 ++
drivers/media/platform/am437x/am437x-vpfe_regs.h | 140 +
include/uapi/linux/Kbuild | 1 +
include/uapi/linux/am437x-vpfe.h | 122 +
11 files changed, 3411 insertions(+)
create mode 100644 Documentation/devicetree/bindings/media/ti-am437x-vpfe.txt
create mode 100644 drivers/media/platform/am437x/Kconfig
create mode 100644 drivers/media/platform/am437x/Makefile
create mode 100644 drivers/media/platform/am437x/am437x-vpfe.c
create mode 100644 drivers/media/platform/am437x/am437x-vpfe.h
create mode 100644 drivers/media/platform/am437x/am437x-vpfe_regs.h
create mode 100644 include/uapi/linux/am437x-vpfe.h
diff --git a/Documentation/devicetree/bindings/media/ti-am437x-vpfe.txt b/Documentation/devicetree/bindings/media/ti-am437x-vpfe.txt
new file mode 100644
index 0000000..3932e76
--- /dev/null
+++ b/Documentation/devicetree/bindings/media/ti-am437x-vpfe.txt
@@ -0,0 +1,61 @@
+Texas Instruments AM437x CAMERA (VPFE)
+--------------------------------------
+
+The Video Processing Front End (VPFE) is a key component for image capture
+applications. The capture module provides the system interface and the
+processing capability to connect RAW image-sensor modules and video decoders
+to the AM437x device.
+
+Required properties:
+- compatible: must be "ti,am437x-vpfe"
+- reg: physical base address and length of the registers set for the device;
+- interrupts: should contain IRQ line for the VPFE;
+- ti,am437x-vpfe-interface: can be one of the following,
+ 0 - Raw Bayer Interface.
+ 1 - 8 Bit BT656 Interface.
+ 2 - 10 Bit BT656 Interface.
+ 3 - YCbCr 8 Bit Interface.
+ 4 - YCbCr 16 Bit Interface.
+
+VPFE supports a single port node with parallel bus. It should contain one
+'port' child node with child 'endpoint' node. Please refer to the bindings
+defined in Documentation/devicetree/bindings/media/video-interfaces.txt.
+
+Example:
+ vpfe: vpfe@f0034000 {
+ compatible = "ti,am437x-vpfe";
+ reg = <0x48328000 0x2000>;
+ interrupts = <GIC_SPI 50 IRQ_TYPE_LEVEL_HIGH>;
+
+ pinctrl-names = "default", "sleep";
+ pinctrl-0 = <&vpfe_pins_default>;
+ pinctrl-1 = <&vpfe_pins_sleep>;
+
+ port {
+ #address-cells = <1>;
+ #size-cells = <0>;
+
+ vpfe0_ep: endpoint {
+ remote-endpoint = <&ov2659_1>;
+ ti,am437x-vpfe-interface = <0>;
+ bus-width = <8>;
+ hsync-active = <0>;
+ vsync-active = <0>;
+ };
+ };
+ };
+
+ i2c1: i2c@4802a000 {
+
+ ov2659@30 {
+ compatible = "ti,ov2659";
+ reg = <0x30>;
+
+ port {
+ ov2659_1: endpoint {
+ remote-endpoint = <&vpfe0_ep>;
+ bus-width = <8>;
+ mclk-frequency = <12000000>;
+ };
+ };
+ };
diff --git a/MAINTAINERS b/MAINTAINERS
index 9c49eb6..5b1cb1a 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -8543,6 +8543,15 @@ S: Maintained
F: drivers/media/platform/davinci/
F: include/media/davinci/
+TI AM437X VPFE DRIVER
+M: Lad, Prabhakar <prabhakar.csengg@gmail.com>
+L: linux-media@vger.kernel.org
+W: http://linuxtv.org/
+Q: http://patchwork.linuxtv.org/project/linux-media/list/
+T: git git://linuxtv.org/mhadli/v4l-dvb-davinci_devices.git
+S: Maintained
+F: drivers/media/platform/am437x/
+
SIS 190 ETHERNET DRIVER
M: Francois Romieu <romieu@fr.zoreil.com>
L: netdev@vger.kernel.org
diff --git a/drivers/media/platform/Kconfig b/drivers/media/platform/Kconfig
index 0c61155..6d94045 100644
--- a/drivers/media/platform/Kconfig
+++ b/drivers/media/platform/Kconfig
@@ -126,6 +126,7 @@ config VIDEO_S3C_CAMIF
source "drivers/media/platform/soc_camera/Kconfig"
source "drivers/media/platform/exynos4-is/Kconfig"
source "drivers/media/platform/s5p-tv/Kconfig"
+source "drivers/media/platform/am437x/Kconfig"
endif # V4L_PLATFORM_DRIVERS
diff --git a/drivers/media/platform/Makefile b/drivers/media/platform/Makefile
index b818afb..7bb6d46 100644
--- a/drivers/media/platform/Makefile
+++ b/drivers/media/platform/Makefile
@@ -49,4 +49,6 @@ obj-$(CONFIG_VIDEO_RENESAS_VSP1) += vsp1/
obj-y += omap/
+obj-$(CONFIG_VIDEO_AM437X_VPFE) += am437x/
+
ccflags-y += -I$(srctree)/drivers/media/i2c
diff --git a/drivers/media/platform/am437x/Kconfig b/drivers/media/platform/am437x/Kconfig
new file mode 100644
index 0000000..7b023a76
--- /dev/null
+++ b/drivers/media/platform/am437x/Kconfig
@@ -0,0 +1,11 @@
+config VIDEO_AM437X_VPFE
+ tristate "TI AM437x VPFE video capture driver"
+ depends on VIDEO_V4L2 && VIDEO_V4L2_SUBDEV_API
+ depends on SOC_AM43XX || COMPILE_TEST
+ select VIDEOBUF2_DMA_CONTIG
+ help
+ Support for AM437x Video Processing Front End based Video
+ Capture Driver.
+
+ To compile this driver as a module, choose M here. The module
+ will be called am437x-vpfe.
diff --git a/drivers/media/platform/am437x/Makefile b/drivers/media/platform/am437x/Makefile
new file mode 100644
index 0000000..d11fff1
--- /dev/null
+++ b/drivers/media/platform/am437x/Makefile
@@ -0,0 +1,3 @@
+# Makefile for AM437x VPFE driver
+
+obj-$(CONFIG_VIDEO_AM437X_VPFE) += am437x-vpfe.o
diff --git a/drivers/media/platform/am437x/am437x-vpfe.c b/drivers/media/platform/am437x/am437x-vpfe.c
new file mode 100644
index 0000000..ed2c4a1
--- /dev/null
+++ b/drivers/media/platform/am437x/am437x-vpfe.c
@@ -0,0 +1,2778 @@
+/*
+ * TI VPFE capture Driver
+ *
+ * Copyright (C) 2013 - 2014 Texas Instruments, Inc.
+ *
+ * Benoit Parrot <bparrot@ti.com>
+ * Lad, Prabhakar <prabhakar.csengg@gmail.com>
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include <linux/delay.h>
+#include <linux/err.h>
+#include <linux/init.h>
+#include <linux/interrupt.h>
+#include <linux/io.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/pm_runtime.h>
+#include <linux/slab.h>
+#include <linux/uaccess.h>
+#include <linux/videodev2.h>
+
+#include <media/v4l2-common.h>
+#include <media/v4l2-ctrls.h>
+#include <media/v4l2-event.h>
+#include <media/v4l2-of.h>
+
+#include "am437x-vpfe.h"
+
+#define VPFE_MODULE_NAME "vpfe"
+#define VPFE_VERSION "0.1.0"
+
+static int debug;
+module_param(debug, int, 0644);
+MODULE_PARM_DESC(debug, "Debug level 0-8");
+
+#define vpfe_dbg(level, dev, fmt, arg...) \
+ v4l2_dbg(level, debug, &dev->v4l2_dev, fmt, ##arg)
+#define vpfe_info(dev, fmt, arg...) \
+ v4l2_info(&dev->v4l2_dev, fmt, ##arg)
+#define vpfe_err(dev, fmt, arg...) \
+ v4l2_err(&dev->v4l2_dev, fmt, ##arg)
+
+/* standard information */
+struct vpfe_standard {
+ v4l2_std_id std_id;
+ unsigned int width;
+ unsigned int height;
+ struct v4l2_fract pixelaspect;
+ int frame_format;
+};
+
+const struct vpfe_standard vpfe_standards[] = {
+ {V4L2_STD_525_60, 720, 480, {11, 10}, 1},
+ {V4L2_STD_625_50, 720, 576, {54, 59}, 1},
+};
+
+struct bus_format {
+ unsigned int width;
+ unsigned int bpp;
+};
+
+/*
+ * struct vpfe_fmt - VPFE media bus format information
+ * @name: V4L2 format description
+ * @code: V4L2 media bus format code
+ * @shifted: V4L2 media bus format code for the same pixel layout but
+ * shifted to be 8 bits per pixel. =0 if format is not shiftable.
+ * @pixelformat: V4L2 pixel format FCC identifier
+ * @width: Bits per pixel (when transferred over a bus)
+ * @bpp: Bytes per pixel (when stored in memory)
+ * @supported: Indicates format supported by subdev
+ */
+struct vpfe_fmt {
+ const char *name;
+ u32 fourcc;
+ u32 code;
+ struct bus_format l;
+ struct bus_format s;
+ bool supported;
+ u32 index;
+};
+
+static struct vpfe_fmt formats[] = {
+ {
+ .name = "YUV 4:2:2 packed, YCbYCr",
+ .fourcc = V4L2_PIX_FMT_YUYV,
+ .code = MEDIA_BUS_FMT_YUYV8_2X8,
+ .l.width = 10,
+ .l.bpp = 4,
+ .s.width = 8,
+ .s.bpp = 2,
+ .supported = false,
+ }, {
+ .name = "YUV 4:2:2 packed, CbYCrY",
+ .fourcc = V4L2_PIX_FMT_UYVY,
+ .code = MEDIA_BUS_FMT_UYVY8_2X8,
+ .l.width = 10,
+ .l.bpp = 4,
+ .s.width = 8,
+ .s.bpp = 2,
+ .supported = false,
+ }, {
+ .name = "YUV 4:2:2 packed, YCrYCb",
+ .fourcc = V4L2_PIX_FMT_YVYU,
+ .code = MEDIA_BUS_FMT_YVYU8_2X8,
+ .l.width = 10,
+ .l.bpp = 4,
+ .s.width = 8,
+ .s.bpp = 2,
+ .supported = false,
+ }, {
+ .name = "YUV 4:2:2 packed, CrYCbY",
+ .fourcc = V4L2_PIX_FMT_VYUY,
+ .code = MEDIA_BUS_FMT_VYUY8_2X8,
+ .l.width = 10,
+ .l.bpp = 4,
+ .s.width = 8,
+ .s.bpp = 2,
+ .supported = false,
+ }, {
+ .name = "RAW8 BGGR",
+ .fourcc = V4L2_PIX_FMT_SBGGR8,
+ .code = MEDIA_BUS_FMT_SBGGR8_1X8,
+ .l.width = 10,
+ .l.bpp = 2,
+ .s.width = 8,
+ .s.bpp = 1,
+ .supported = false,
+ }, {
+ .name = "RAW8 GBRG",
+ .fourcc = V4L2_PIX_FMT_SGBRG8,
+ .code = MEDIA_BUS_FMT_SGBRG8_1X8,
+ .l.width = 10,
+ .l.bpp = 2,
+ .s.width = 8,
+ .s.bpp = 1,
+ .supported = false,
+ }, {
+ .name = "RAW8 GRBG",
+ .fourcc = V4L2_PIX_FMT_SGRBG8,
+ .code = MEDIA_BUS_FMT_SGRBG8_1X8,
+ .l.width = 10,
+ .l.bpp = 2,
+ .s.width = 8,
+ .s.bpp = 1,
+ .supported = false,
+ }, {
+ .name = "RAW8 RGGB",
+ .fourcc = V4L2_PIX_FMT_SRGGB8,
+ .code = MEDIA_BUS_FMT_SRGGB8_1X8,
+ .l.width = 10,
+ .l.bpp = 2,
+ .s.width = 8,
+ .s.bpp = 1,
+ .supported = false,
+ }, {
+ .name = "RGB565 (LE)",
+ .fourcc = V4L2_PIX_FMT_RGB565,
+ .code = MEDIA_BUS_FMT_RGB565_2X8_LE,
+ .l.width = 10,
+ .l.bpp = 4,
+ .s.width = 8,
+ .s.bpp = 2,
+ .supported = false,
+ }, {
+ .name = "RGB565 (BE)",
+ .fourcc = V4L2_PIX_FMT_RGB565X,
+ .code = MEDIA_BUS_FMT_RGB565_2X8_BE,
+ .l.width = 10,
+ .l.bpp = 4,
+ .s.width = 8,
+ .s.bpp = 2,
+ .supported = false,
+ },
+};
+
+static int
+__vpfe_get_format(struct vpfe_device *vpfe,
+ struct v4l2_format *format, unsigned int *bpp);
+
+static struct vpfe_fmt *find_format_by_code(unsigned int code)
+{
+ struct vpfe_fmt *fmt;
+ unsigned int k;
+
+ for (k = 0; k < ARRAY_SIZE(formats); k++) {
+ fmt = &formats[k];
+ if (fmt->code == code)
+ return fmt;
+ }
+
+ return NULL;
+}
+
+static struct vpfe_fmt *find_format_by_pix(unsigned int pixelformat)
+{
+ struct vpfe_fmt *fmt;
+ unsigned int k;
+
+ for (k = 0; k < ARRAY_SIZE(formats); k++) {
+ fmt = &formats[k];
+ if (fmt->fourcc == pixelformat)
+ return fmt;
+ }
+
+ return NULL;
+}
+
+static void
+mbus_to_pix(struct vpfe_device *vpfe,
+ const struct v4l2_mbus_framefmt *mbus,
+ struct v4l2_pix_format *pix, unsigned int *bpp)
+{
+ struct vpfe_subdev_info *sdinfo = vpfe->current_subdev;
+ unsigned int bus_width = sdinfo->vpfe_param.bus_width;
+ struct vpfe_fmt *fmt;
+
+ fmt = find_format_by_code(mbus->code);
+ if (WARN_ON(fmt == NULL)) {
+ pr_err("Invalid mbus code set\n");
+ *bpp = 1;
+ return;
+ }
+
+ memset(pix, 0, sizeof(*pix));
+ v4l2_fill_pix_format(pix, mbus);
+ pix->pixelformat = fmt->fourcc;
+ *bpp = (bus_width == 10) ? fmt->l.bpp : fmt->s.bpp;
+
+ /* pitch should be 32 bytes aligned */
+ pix->bytesperline = ALIGN(pix->width * *bpp, 32);
+ pix->sizeimage = pix->bytesperline * pix->height;
+}
+
+static void pix_to_mbus(struct vpfe_device *vpfe,
+ struct v4l2_pix_format *pix_fmt,
+ struct v4l2_mbus_framefmt *mbus_fmt)
+{
+ struct vpfe_fmt *fmt;
+
+ fmt = find_format_by_pix(pix_fmt->pixelformat);
+ if (!fmt) {
+ /* default to first entry */
+ vpfe_dbg(3, vpfe, "Invalid pixel code: %x, default used instead\n",
+ pix_fmt->pixelformat);
+ fmt = &formats[0];
+ }
+
+ memset(mbus_fmt, 0, sizeof(*mbus_fmt));
+ v4l2_fill_mbus_format(mbus_fmt, pix_fmt, fmt->code);
+}
+
+/* Print Four-character-code (FOURCC) */
+static char *print_fourcc(u32 fmt)
+{
+ static char code[5];
+
+ code[0] = (unsigned char)(fmt & 0xff);
+ code[1] = (unsigned char)((fmt >> 8) & 0xff);
+ code[2] = (unsigned char)((fmt >> 16) & 0xff);
+ code[3] = (unsigned char)((fmt >> 24) & 0xff);
+ code[4] = '\0';
+
+ return code;
+}
+
+static int
+cmp_v4l2_format(const struct v4l2_format *lhs, const struct v4l2_format *rhs)
+{
+ return lhs->type == rhs->type &&
+ lhs->fmt.pix.width == rhs->fmt.pix.width &&
+ lhs->fmt.pix.height == rhs->fmt.pix.height &&
+ lhs->fmt.pix.pixelformat == rhs->fmt.pix.pixelformat &&
+ lhs->fmt.pix.field == rhs->fmt.pix.field &&
+ lhs->fmt.pix.colorspace == rhs->fmt.pix.colorspace &&
+ lhs->fmt.pix.ycbcr_enc == rhs->fmt.pix.ycbcr_enc &&
+ lhs->fmt.pix.quantization == rhs->fmt.pix.quantization;
+}
+
+static inline u32 vpfe_reg_read(struct vpfe_ccdc *ccdc, u32 offset)
+{
+ return ioread32(ccdc->ccdc_cfg.base_addr + offset);
+}
+
+static inline void vpfe_reg_write(struct vpfe_ccdc *ccdc, u32 val, u32 offset)
+{
+ iowrite32(val, ccdc->ccdc_cfg.base_addr + offset);
+}
+
+static inline struct vpfe_device *to_vpfe(struct vpfe_ccdc *ccdc)
+{
+ return container_of(ccdc, struct vpfe_device, ccdc);
+}
+
+static inline struct vpfe_cap_buffer *to_vpfe_buffer(struct vb2_buffer *vb)
+{
+ return container_of(vb, struct vpfe_cap_buffer, vb);
+}
+
+static inline void vpfe_pcr_enable(struct vpfe_ccdc *ccdc, int flag)
+{
+ vpfe_reg_write(ccdc, !!flag, VPFE_PCR);
+}
+
+static void vpfe_config_enable(struct vpfe_ccdc *ccdc, int flag)
+{
+ unsigned int cfg;
+
+ if (!flag) {
+ cfg = vpfe_reg_read(ccdc, VPFE_CONFIG);
+ cfg &= ~(VPFE_CONFIG_EN_ENABLE << VPFE_CONFIG_EN_SHIFT);
+ } else {
+ cfg = VPFE_CONFIG_EN_ENABLE << VPFE_CONFIG_EN_SHIFT;
+ }
+
+ vpfe_reg_write(ccdc, cfg, VPFE_CONFIG);
+}
+
+static void vpfe_ccdc_setwin(struct vpfe_ccdc *ccdc,
+ struct v4l2_rect *image_win,
+ enum ccdc_frmfmt frm_fmt,
+ int bpp)
+{
+ int horz_start, horz_nr_pixels;
+ int vert_start, vert_nr_lines;
+ int val, mid_img;
+
+ /*
+ * ppc - per pixel count. indicates how many pixels per cell
+ * output to SDRAM. example, for ycbcr, it is one y and one c, so 2.
+ * raw capture this is 1
+ */
+ horz_start = image_win->left * bpp;
+ horz_nr_pixels = (image_win->width * bpp) - 1;
+ vpfe_reg_write(ccdc, (horz_start << VPFE_HORZ_INFO_SPH_SHIFT) |
+ horz_nr_pixels, VPFE_HORZ_INFO);
+
+ vert_start = image_win->top;
+
+ if (frm_fmt == CCDC_FRMFMT_INTERLACED) {
+ vert_nr_lines = (image_win->height >> 1) - 1;
+ vert_start >>= 1;
+ /* Since first line doesn't have any data */
+ vert_start += 1;
+ /* configure VDINT0 */
+ val = (vert_start << VPFE_VDINT_VDINT0_SHIFT);
+ } else {
+ /* Since first line doesn't have any data */
+ vert_start += 1;
+ vert_nr_lines = image_win->height - 1;
+ /*
+ * configure VDINT0 and VDINT1. VDINT1 will be at half
+ * of image height
+ */
+ mid_img = vert_start + (image_win->height / 2);
+ val = (vert_start << VPFE_VDINT_VDINT0_SHIFT) |
+ (mid_img & VPFE_VDINT_VDINT1_MASK);
+ }
+
+ vpfe_reg_write(ccdc, val, VPFE_VDINT);
+
+ vpfe_reg_write(ccdc, (vert_start << VPFE_VERT_START_SLV0_SHIFT) |
+ vert_start, VPFE_VERT_START);
+ vpfe_reg_write(ccdc, vert_nr_lines, VPFE_VERT_LINES);
+}
+
+static void vpfe_reg_dump(struct vpfe_ccdc *ccdc)
+{
+ struct vpfe_device *vpfe = to_vpfe(ccdc);
+
+ vpfe_dbg(3, vpfe, "ALAW: 0x%x\n", vpfe_reg_read(ccdc, VPFE_ALAW));
+ vpfe_dbg(3, vpfe, "CLAMP: 0x%x\n", vpfe_reg_read(ccdc, VPFE_CLAMP));
+ vpfe_dbg(3, vpfe, "DCSUB: 0x%x\n", vpfe_reg_read(ccdc, VPFE_DCSUB));
+ vpfe_dbg(3, vpfe, "BLKCMP: 0x%x\n", vpfe_reg_read(ccdc, VPFE_BLKCMP));
+ vpfe_dbg(3, vpfe, "COLPTN: 0x%x\n", vpfe_reg_read(ccdc, VPFE_COLPTN));
+ vpfe_dbg(3, vpfe, "SDOFST: 0x%x\n", vpfe_reg_read(ccdc, VPFE_SDOFST));
+ vpfe_dbg(3, vpfe, "SYN_MODE: 0x%x\n",
+ vpfe_reg_read(ccdc, VPFE_SYNMODE));
+ vpfe_dbg(3, vpfe, "HSIZE_OFF: 0x%x\n",
+ vpfe_reg_read(ccdc, VPFE_HSIZE_OFF));
+ vpfe_dbg(3, vpfe, "HORZ_INFO: 0x%x\n",
+ vpfe_reg_read(ccdc, VPFE_HORZ_INFO));
+ vpfe_dbg(3, vpfe, "VERT_START: 0x%x\n",
+ vpfe_reg_read(ccdc, VPFE_VERT_START));
+ vpfe_dbg(3, vpfe, "VERT_LINES: 0x%x\n",
+ vpfe_reg_read(ccdc, VPFE_VERT_LINES));
+}
+
+static int
+vpfe_ccdc_validate_param(struct vpfe_ccdc *ccdc,
+ struct vpfe_ccdc_config_params_raw *ccdcparam)
+{
+ struct vpfe_device *vpfe = to_vpfe(ccdc);
+ u8 max_gamma, max_data;
+
+ if (!ccdcparam->alaw.enable)
+ return 0;
+
+ max_gamma = ccdc_gamma_width_max_bit(ccdcparam->alaw.gamma_wd);
+ max_data = ccdc_data_size_max_bit(ccdcparam->data_sz);
+
+ if (ccdcparam->alaw.gamma_wd > VPFE_CCDC_GAMMA_BITS_09_0 ||
+ ccdcparam->alaw.gamma_wd < VPFE_CCDC_GAMMA_BITS_15_6 ||
+ max_gamma > max_data) {
+ vpfe_dbg(1, vpfe, "Invalid data line select\n");
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+static void
+vpfe_ccdc_update_raw_params(struct vpfe_ccdc *ccdc,
+ struct vpfe_ccdc_config_params_raw *raw_params)
+{
+ struct vpfe_ccdc_config_params_raw *config_params =
+ &ccdc->ccdc_cfg.bayer.config_params;
+
+ config_params = raw_params;
+}
+
+/*
+ * vpfe_ccdc_restore_defaults()
+ * This function will write defaults to all CCDC registers
+ */
+static void vpfe_ccdc_restore_defaults(struct vpfe_ccdc *ccdc)
+{
+ int i;
+
+ /* Disable CCDC */
+ vpfe_pcr_enable(ccdc, 0);
+
+ /* set all registers to default value */
+ for (i = 4; i <= 0x94; i += 4)
+ vpfe_reg_write(ccdc, 0, i);
+
+ vpfe_reg_write(ccdc, VPFE_NO_CULLING, VPFE_CULLING);
+ vpfe_reg_write(ccdc, VPFE_CCDC_GAMMA_BITS_11_2, VPFE_ALAW);
+}
+
+static int vpfe_ccdc_close(struct vpfe_ccdc *ccdc, struct device *dev)
+{
+ int dma_cntl, i, pcr;
+
+ /* If the CCDC module is still busy wait for it to be done */
+ for (i = 0; i < 10; i++) {
+ usleep_range(5000, 6000);
+ pcr = vpfe_reg_read(ccdc, VPFE_PCR);
+ if (!pcr)
+ break;
+
+ /* make sure it it is disabled */
+ vpfe_pcr_enable(ccdc, 0);
+ }
+
+ /* Disable CCDC by resetting all register to default POR values */
+ vpfe_ccdc_restore_defaults(ccdc);
+
+ /* if DMA_CNTL overflow bit is set. Clear it
+ * It appears to take a while for this to become quiescent ~20ms
+ */
+ for (i = 0; i < 10; i++) {
+ dma_cntl = vpfe_reg_read(ccdc, VPFE_DMA_CNTL);
+ if (!(dma_cntl & VPFE_DMA_CNTL_OVERFLOW))
+ break;
+
+ /* Clear the overflow bit */
+ vpfe_reg_write(ccdc, dma_cntl, VPFE_DMA_CNTL);
+ usleep_range(5000, 6000);
+ }
+
+ /* Disabled the module at the CONFIG level */
+ vpfe_config_enable(ccdc, 0);
+
+ pm_runtime_put_sync(dev);
+
+ return 0;
+}
+
+static int vpfe_ccdc_set_params(struct vpfe_ccdc *ccdc, void __user *params)
+{
+ struct vpfe_device *vpfe = container_of(ccdc, struct vpfe_device, ccdc);
+ struct vpfe_ccdc_config_params_raw raw_params;
+ int x;
+
+ if (ccdc->ccdc_cfg.if_type != VPFE_RAW_BAYER)
+ return -EINVAL;
+
+ x = copy_from_user(&raw_params, params, sizeof(raw_params));
+ if (x) {
+ vpfe_dbg(1, vpfe,
+ "vpfe_ccdc_set_params: error in copying ccdc params, %d\n",
+ x);
+ return -EFAULT;
+ }
+
+ if (!vpfe_ccdc_validate_param(ccdc, &raw_params)) {
+ vpfe_ccdc_update_raw_params(ccdc, &raw_params);
+ return 0;
+ }
+
+ return -EINVAL;
+}
+
+/*
+ * vpfe_ccdc_config_ycbcr()
+ * This function will configure CCDC for YCbCr video capture
+ */
+static void vpfe_ccdc_config_ycbcr(struct vpfe_ccdc *ccdc)
+{
+ struct vpfe_device *vpfe = container_of(ccdc, struct vpfe_device, ccdc);
+ struct ccdc_params_ycbcr *params = &ccdc->ccdc_cfg.ycbcr;
+ u32 syn_mode;
+
+ vpfe_dbg(3, vpfe, "vpfe_ccdc_config_ycbcr:\n");
+ /*
+ * first restore the CCDC registers to default values
+ * This is important since we assume default values to be set in
+ * a lot of registers that we didn't touch
+ */
+ vpfe_ccdc_restore_defaults(ccdc);
+
+ /*
+ * configure pixel format, frame format, configure video frame
+ * format, enable output to SDRAM, enable internal timing generator
+ * and 8bit pack mode
+ */
+ syn_mode = (((params->pix_fmt & VPFE_SYN_MODE_INPMOD_MASK) <<
+ VPFE_SYN_MODE_INPMOD_SHIFT) |
+ ((params->frm_fmt & VPFE_SYN_FLDMODE_MASK) <<
+ VPFE_SYN_FLDMODE_SHIFT) | VPFE_VDHDEN_ENABLE |
+ VPFE_WEN_ENABLE | VPFE_DATA_PACK_ENABLE);
+
+ /* setup BT.656 sync mode */
+ if (params->bt656_enable) {
+ vpfe_reg_write(ccdc, VPFE_REC656IF_BT656_EN, VPFE_REC656IF);
+
+ /*
+ * configure the FID, VD, HD pin polarity,
+ * fld,hd pol positive, vd negative, 8-bit data
+ */
+ syn_mode |= VPFE_SYN_MODE_VD_POL_NEGATIVE;
+ if (ccdc->ccdc_cfg.if_type == VPFE_BT656_10BIT)
+ syn_mode |= VPFE_SYN_MODE_10BITS;
+ else
+ syn_mode |= VPFE_SYN_MODE_8BITS;
+ } else {
+ /* y/c external sync mode */
+ syn_mode |= (((params->fid_pol & VPFE_FID_POL_MASK) <<
+ VPFE_FID_POL_SHIFT) |
+ ((params->hd_pol & VPFE_HD_POL_MASK) <<
+ VPFE_HD_POL_SHIFT) |
+ ((params->vd_pol & VPFE_VD_POL_MASK) <<
+ VPFE_VD_POL_SHIFT));
+ }
+ vpfe_reg_write(ccdc, syn_mode, VPFE_SYNMODE);
+
+ /* configure video window */
+ vpfe_ccdc_setwin(ccdc, ¶ms->win,
+ params->frm_fmt, params->bytesperpixel);
+
+ /*
+ * configure the order of y cb cr in SDRAM, and disable latch
+ * internal register on vsync
+ */
+ if (ccdc->ccdc_cfg.if_type == VPFE_BT656_10BIT)
+ vpfe_reg_write(ccdc,
+ (params->pix_order << VPFE_CCDCFG_Y8POS_SHIFT) |
+ VPFE_LATCH_ON_VSYNC_DISABLE |
+ VPFE_CCDCFG_BW656_10BIT, VPFE_CCDCFG);
+ else
+ vpfe_reg_write(ccdc,
+ (params->pix_order << VPFE_CCDCFG_Y8POS_SHIFT) |
+ VPFE_LATCH_ON_VSYNC_DISABLE, VPFE_CCDCFG);
+
+ /*
+ * configure the horizontal line offset. This should be a
+ * on 32 byte boundary. So clear LSB 5 bits
+ */
+ vpfe_reg_write(ccdc, params->bytesperline, VPFE_HSIZE_OFF);
+
+ /* configure the memory line offset */
+ if (params->buf_type == CCDC_BUFTYPE_FLD_INTERLEAVED)
+ /* two fields are interleaved in memory */
+ vpfe_reg_write(ccdc, VPFE_SDOFST_FIELD_INTERLEAVED,
+ VPFE_SDOFST);
+}
+
+static void
+vpfe_ccdc_config_black_clamp(struct vpfe_ccdc *ccdc,
+ struct vpfe_ccdc_black_clamp *bclamp)
+{
+ u32 val;
+
+ if (!bclamp->enable) {
+ /* configure DCSub */
+ val = (bclamp->dc_sub) & VPFE_BLK_DC_SUB_MASK;
+ vpfe_reg_write(ccdc, val, VPFE_DCSUB);
+ vpfe_reg_write(ccdc, VPFE_CLAMP_DEFAULT_VAL, VPFE_CLAMP);
+ return;
+ }
+ /*
+ * Configure gain, Start pixel, No of line to be avg,
+ * No of pixel/line to be avg, & Enable the Black clamping
+ */
+ val = ((bclamp->sgain & VPFE_BLK_SGAIN_MASK) |
+ ((bclamp->start_pixel & VPFE_BLK_ST_PXL_MASK) <<
+ VPFE_BLK_ST_PXL_SHIFT) |
+ ((bclamp->sample_ln & VPFE_BLK_SAMPLE_LINE_MASK) <<
+ VPFE_BLK_SAMPLE_LINE_SHIFT) |
+ ((bclamp->sample_pixel & VPFE_BLK_SAMPLE_LN_MASK) <<
+ VPFE_BLK_SAMPLE_LN_SHIFT) | VPFE_BLK_CLAMP_ENABLE);
+ vpfe_reg_write(ccdc, val, VPFE_CLAMP);
+ /* If Black clamping is enable then make dcsub 0 */
+ vpfe_reg_write(ccdc, VPFE_DCSUB_DEFAULT_VAL, VPFE_DCSUB);
+}
+
+static void
+vpfe_ccdc_config_black_compense(struct vpfe_ccdc *ccdc,
+ struct vpfe_ccdc_black_compensation *bcomp)
+{
+ u32 val;
+
+ val = ((bcomp->b & VPFE_BLK_COMP_MASK) |
+ ((bcomp->gb & VPFE_BLK_COMP_MASK) <<
+ VPFE_BLK_COMP_GB_COMP_SHIFT) |
+ ((bcomp->gr & VPFE_BLK_COMP_MASK) <<
+ VPFE_BLK_COMP_GR_COMP_SHIFT) |
+ ((bcomp->r & VPFE_BLK_COMP_MASK) <<
+ VPFE_BLK_COMP_R_COMP_SHIFT));
+ vpfe_reg_write(ccdc, val, VPFE_BLKCMP);
+}
+
+/*
+ * vpfe_ccdc_config_raw()
+ * This function will configure CCDC for Raw capture mode
+ */
+static void vpfe_ccdc_config_raw(struct vpfe_ccdc *ccdc)
+{
+ struct vpfe_device *vpfe = container_of(ccdc, struct vpfe_device, ccdc);
+ struct vpfe_ccdc_config_params_raw *config_params =
+ &ccdc->ccdc_cfg.bayer.config_params;
+ struct ccdc_params_raw *params = &ccdc->ccdc_cfg.bayer;
+ unsigned int syn_mode;
+ unsigned int val;
+
+ vpfe_dbg(3, vpfe, "vpfe_ccdc_config_raw:\n");
+
+ /* Reset CCDC */
+ vpfe_ccdc_restore_defaults(ccdc);
+
+ /* Disable latching function registers on VSYNC */
+ vpfe_reg_write(ccdc, VPFE_LATCH_ON_VSYNC_DISABLE, VPFE_CCDCFG);
+
+ /*
+ * Configure the vertical sync polarity(SYN_MODE.VDPOL),
+ * horizontal sync polarity (SYN_MODE.HDPOL), frame id polarity
+ * (SYN_MODE.FLDPOL), frame format(progressive or interlace),
+ * data size(SYNMODE.DATSIZ), &pixel format (Input mode), output
+ * SDRAM, enable internal timing generator
+ */
+ syn_mode = (((params->vd_pol & VPFE_VD_POL_MASK) << VPFE_VD_POL_SHIFT) |
+ ((params->hd_pol & VPFE_HD_POL_MASK) << VPFE_HD_POL_SHIFT) |
+ ((params->fid_pol & VPFE_FID_POL_MASK) <<
+ VPFE_FID_POL_SHIFT) | ((params->frm_fmt &
+ VPFE_FRM_FMT_MASK) << VPFE_FRM_FMT_SHIFT) |
+ ((config_params->data_sz & VPFE_DATA_SZ_MASK) <<
+ VPFE_DATA_SZ_SHIFT) | ((params->pix_fmt &
+ VPFE_PIX_FMT_MASK) << VPFE_PIX_FMT_SHIFT) |
+ VPFE_WEN_ENABLE | VPFE_VDHDEN_ENABLE);
+
+ /* Enable and configure aLaw register if needed */
+ if (config_params->alaw.enable) {
+ val = ((config_params->alaw.gamma_wd &
+ VPFE_ALAW_GAMMA_WD_MASK) | VPFE_ALAW_ENABLE);
+ vpfe_reg_write(ccdc, val, VPFE_ALAW);
+ vpfe_dbg(3, vpfe, "\nWriting 0x%x to ALAW...\n", val);
+ }
+
+ /* Configure video window */
+ vpfe_ccdc_setwin(ccdc, ¶ms->win, params->frm_fmt,
+ params->bytesperpixel);
+
+ /* Configure Black Clamp */
+ vpfe_ccdc_config_black_clamp(ccdc, &config_params->blk_clamp);
+
+ /* Configure Black level compensation */
+ vpfe_ccdc_config_black_compense(ccdc, &config_params->blk_comp);
+
+ /* If data size is 8 bit then pack the data */
+ if ((config_params->data_sz == VPFE_CCDC_DATA_8BITS) ||
+ config_params->alaw.enable)
+ syn_mode |= VPFE_DATA_PACK_ENABLE;
+
+ /*
+ * Configure Horizontal offset register. If pack 8 is enabled then
+ * 1 pixel will take 1 byte
+ */
+ vpfe_reg_write(ccdc, params->bytesperline, VPFE_HSIZE_OFF);
+
+ vpfe_dbg(3, vpfe, "Writing %d (%x) to HSIZE_OFF\n",
+ params->bytesperline, params->bytesperline);
+
+ /* Set value for SDOFST */
+ if (params->frm_fmt == CCDC_FRMFMT_INTERLACED) {
+ if (params->image_invert_enable) {
+ /* For interlace inverse mode */
+ vpfe_reg_write(ccdc, VPFE_INTERLACED_IMAGE_INVERT,
+ VPFE_SDOFST);
+ } else {
+ /* For interlace non inverse mode */
+ vpfe_reg_write(ccdc, VPFE_INTERLACED_NO_IMAGE_INVERT,
+ VPFE_SDOFST);
+ }
+ } else if (params->frm_fmt == CCDC_FRMFMT_PROGRESSIVE) {
+ vpfe_reg_write(ccdc, VPFE_PROGRESSIVE_NO_IMAGE_INVERT,
+ VPFE_SDOFST);
+ }
+
+ vpfe_reg_write(ccdc, syn_mode, VPFE_SYNMODE);
+
+ vpfe_reg_dump(ccdc);
+}
+
+static inline int
+vpfe_ccdc_set_buftype(struct vpfe_ccdc *ccdc,
+ enum ccdc_buftype buf_type)
+{
+ if (ccdc->ccdc_cfg.if_type == VPFE_RAW_BAYER)
+ ccdc->ccdc_cfg.bayer.buf_type = buf_type;
+ else
+ ccdc->ccdc_cfg.ycbcr.buf_type = buf_type;
+
+ return 0;
+}
+
+static inline enum ccdc_buftype vpfe_ccdc_get_buftype(struct vpfe_ccdc *ccdc)
+{
+ if (ccdc->ccdc_cfg.if_type == VPFE_RAW_BAYER)
+ return ccdc->ccdc_cfg.bayer.buf_type;
+
+ return ccdc->ccdc_cfg.ycbcr.buf_type;
+}
+
+static int vpfe_ccdc_set_pixel_format(struct vpfe_ccdc *ccdc, u32 pixfmt)
+{
+ struct vpfe_device *vpfe = container_of(ccdc, struct vpfe_device, ccdc);
+
+ vpfe_dbg(1, vpfe, "vpfe_ccdc_set_pixel_format: if_type: %d, pixfmt:%s\n",
+ ccdc->ccdc_cfg.if_type, print_fourcc(pixfmt));
+
+ if (ccdc->ccdc_cfg.if_type == VPFE_RAW_BAYER) {
+ ccdc->ccdc_cfg.bayer.pix_fmt = CCDC_PIXFMT_RAW;
+ /*
+ * Need to clear it in case it was left on
+ * after the last capture.
+ */
+ ccdc->ccdc_cfg.bayer.config_params.alaw.enable = 0;
+
+ switch (pixfmt) {
+ case V4L2_PIX_FMT_SBGGR8:
+ ccdc->ccdc_cfg.bayer.config_params.alaw.enable = 1;
+ break;
+
+ case V4L2_PIX_FMT_YUYV:
+ case V4L2_PIX_FMT_UYVY:
+ case V4L2_PIX_FMT_YUV420:
+ case V4L2_PIX_FMT_NV12:
+ case V4L2_PIX_FMT_RGB565X:
+ break;
+
+ case V4L2_PIX_FMT_SBGGR16:
+ default:
+ return -EINVAL;
+ }
+ } else {
+ switch (pixfmt) {
+ case V4L2_PIX_FMT_YUYV:
+ ccdc->ccdc_cfg.ycbcr.pix_order = CCDC_PIXORDER_YCBYCR;
+ break;
+
+ case V4L2_PIX_FMT_UYVY:
+ ccdc->ccdc_cfg.ycbcr.pix_order = CCDC_PIXORDER_CBYCRY;
+ break;
+
+ default:
+ return -EINVAL;
+ }
+ }
+
+ return 0;
+}
+
+static u32 vpfe_ccdc_get_pixel_format(struct vpfe_ccdc *ccdc)
+{
+ u32 pixfmt;
+
+ if (ccdc->ccdc_cfg.if_type == VPFE_RAW_BAYER) {
+ pixfmt = V4L2_PIX_FMT_YUYV;
+ } else {
+ if (ccdc->ccdc_cfg.ycbcr.pix_order == CCDC_PIXORDER_YCBYCR)
+ pixfmt = V4L2_PIX_FMT_YUYV;
+ else
+ pixfmt = V4L2_PIX_FMT_UYVY;
+ }
+
+ return pixfmt;
+}
+
+static int
+vpfe_ccdc_set_image_window(struct vpfe_ccdc *ccdc,
+ struct v4l2_rect *win, unsigned int bpp)
+{
+ if (ccdc->ccdc_cfg.if_type == VPFE_RAW_BAYER) {
+ ccdc->ccdc_cfg.bayer.win = *win;
+ ccdc->ccdc_cfg.bayer.bytesperpixel = bpp;
+ ccdc->ccdc_cfg.bayer.bytesperline = ALIGN(win->width * bpp, 32);
+ } else {
+ ccdc->ccdc_cfg.ycbcr.win = *win;
+ ccdc->ccdc_cfg.ycbcr.bytesperpixel = bpp;
+ ccdc->ccdc_cfg.ycbcr.bytesperline = ALIGN(win->width * bpp, 32);
+ }
+
+ return 0;
+}
+
+static inline void
+vpfe_ccdc_get_image_window(struct vpfe_ccdc *ccdc,
+ struct v4l2_rect *win)
+{
+ if (ccdc->ccdc_cfg.if_type == VPFE_RAW_BAYER)
+ *win = ccdc->ccdc_cfg.bayer.win;
+ else
+ *win = ccdc->ccdc_cfg.ycbcr.win;
+}
+
+static inline unsigned int vpfe_ccdc_get_line_length(struct vpfe_ccdc *ccdc)
+{
+ if (ccdc->ccdc_cfg.if_type == VPFE_RAW_BAYER)
+ return ccdc->ccdc_cfg.bayer.bytesperline;
+
+ return ccdc->ccdc_cfg.ycbcr.bytesperline;
+}
+
+static inline int
+vpfe_ccdc_set_frame_format(struct vpfe_ccdc *ccdc,
+ enum ccdc_frmfmt frm_fmt)
+{
+ if (ccdc->ccdc_cfg.if_type == VPFE_RAW_BAYER)
+ ccdc->ccdc_cfg.bayer.frm_fmt = frm_fmt;
+ else
+ ccdc->ccdc_cfg.ycbcr.frm_fmt = frm_fmt;
+
+ return 0;
+}
+
+static inline enum ccdc_frmfmt
+vpfe_ccdc_get_frame_format(struct vpfe_ccdc *ccdc)
+{
+ if (ccdc->ccdc_cfg.if_type == VPFE_RAW_BAYER)
+ return ccdc->ccdc_cfg.bayer.frm_fmt;
+
+ return ccdc->ccdc_cfg.ycbcr.frm_fmt;
+}
+
+static inline int vpfe_ccdc_getfid(struct vpfe_ccdc *ccdc)
+{
+ return (vpfe_reg_read(ccdc, VPFE_SYNMODE) >> 15) & 1;
+}
+
+static inline void vpfe_set_sdr_addr(struct vpfe_ccdc *ccdc, unsigned long addr)
+{
+ vpfe_reg_write(ccdc, addr & 0xffffffe0, VPFE_SDR_ADDR);
+}
+
+static int vpfe_ccdc_set_hw_if_params(struct vpfe_ccdc *ccdc,
+ struct vpfe_hw_if_param *params)
+{
+ struct vpfe_device *vpfe = container_of(ccdc, struct vpfe_device, ccdc);
+
+ ccdc->ccdc_cfg.if_type = params->if_type;
+
+ switch (params->if_type) {
+ case VPFE_BT656:
+ case VPFE_YCBCR_SYNC_16:
+ case VPFE_YCBCR_SYNC_8:
+ case VPFE_BT656_10BIT:
+ ccdc->ccdc_cfg.ycbcr.vd_pol = params->vdpol;
+ ccdc->ccdc_cfg.ycbcr.hd_pol = params->hdpol;
+ break;
+
+ case VPFE_RAW_BAYER:
+ ccdc->ccdc_cfg.bayer.vd_pol = params->vdpol;
+ ccdc->ccdc_cfg.bayer.hd_pol = params->hdpol;
+ if (params->bus_width == 10)
+ ccdc->ccdc_cfg.bayer.config_params.data_sz =
+ VPFE_CCDC_DATA_10BITS;
+ else
+ ccdc->ccdc_cfg.bayer.config_params.data_sz =
+ VPFE_CCDC_DATA_8BITS;
+ vpfe_dbg(1, vpfe, "params.bus_width: %d\n",
+ params->bus_width);
+ vpfe_dbg(1, vpfe, "config_params.data_sz: %d\n",
+ ccdc->ccdc_cfg.bayer.config_params.data_sz);
+ break;
+
+ default:
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+static void vpfe_clear_intr(struct vpfe_ccdc *ccdc, int vdint)
+{
+ unsigned int vpfe_int_status;
+
+ vpfe_int_status = vpfe_reg_read(ccdc, VPFE_IRQ_STS);
+
+ switch (vdint) {
+ /* VD0 interrupt */
+ case VPFE_VDINT0:
+ vpfe_int_status &= ~VPFE_VDINT0;
+ vpfe_int_status |= VPFE_VDINT0;
+ break;
+
+ /* VD1 interrupt */
+ case VPFE_VDINT1:
+ vpfe_int_status &= ~VPFE_VDINT1;
+ vpfe_int_status |= VPFE_VDINT1;
+ break;
+
+ /* VD2 interrupt */
+ case VPFE_VDINT2:
+ vpfe_int_status &= ~VPFE_VDINT2;
+ vpfe_int_status |= VPFE_VDINT2;
+ break;
+
+ /* Clear all interrupts */
+ default:
+ vpfe_int_status &= ~(VPFE_VDINT0 |
+ VPFE_VDINT1 |
+ VPFE_VDINT2);
+ vpfe_int_status |= (VPFE_VDINT0 |
+ VPFE_VDINT1 |
+ VPFE_VDINT2);
+ break;
+ }
+ /* Clear specific VDINT from the status register */
+ vpfe_reg_write(ccdc, vpfe_int_status, VPFE_IRQ_STS);
+
+ vpfe_int_status = vpfe_reg_read(ccdc, VPFE_IRQ_STS);
+
+ /* Acknowledge that we are done with all interrupts */
+ vpfe_reg_write(ccdc, 1, VPFE_IRQ_EOI);
+}
+
+static void vpfe_ccdc_config_defaults(struct vpfe_ccdc *ccdc)
+{
+ ccdc->ccdc_cfg.if_type = VPFE_RAW_BAYER;
+
+ ccdc->ccdc_cfg.ycbcr.pix_fmt = CCDC_PIXFMT_YCBCR_8BIT;
+ ccdc->ccdc_cfg.ycbcr.frm_fmt = CCDC_FRMFMT_INTERLACED;
+ ccdc->ccdc_cfg.ycbcr.fid_pol = VPFE_PINPOL_POSITIVE;
+ ccdc->ccdc_cfg.ycbcr.vd_pol = VPFE_PINPOL_POSITIVE;
+ ccdc->ccdc_cfg.ycbcr.hd_pol = VPFE_PINPOL_POSITIVE;
+ ccdc->ccdc_cfg.ycbcr.pix_order = CCDC_PIXORDER_CBYCRY;
+ ccdc->ccdc_cfg.ycbcr.buf_type = CCDC_BUFTYPE_FLD_INTERLEAVED;
+
+ ccdc->ccdc_cfg.ycbcr.win.left = 0;
+ ccdc->ccdc_cfg.ycbcr.win.top = 0;
+ ccdc->ccdc_cfg.ycbcr.win.width = 720;
+ ccdc->ccdc_cfg.ycbcr.win.height = 576;
+ ccdc->ccdc_cfg.ycbcr.bt656_enable = 1;
+
+ ccdc->ccdc_cfg.bayer.pix_fmt = CCDC_PIXFMT_RAW;
+ ccdc->ccdc_cfg.bayer.frm_fmt = CCDC_FRMFMT_PROGRESSIVE;
+ ccdc->ccdc_cfg.bayer.fid_pol = VPFE_PINPOL_POSITIVE;
+ ccdc->ccdc_cfg.bayer.vd_pol = VPFE_PINPOL_POSITIVE;
+ ccdc->ccdc_cfg.bayer.hd_pol = VPFE_PINPOL_POSITIVE;
+
+ ccdc->ccdc_cfg.bayer.win.left = 0;
+ ccdc->ccdc_cfg.bayer.win.top = 0;
+ ccdc->ccdc_cfg.bayer.win.width = 800;
+ ccdc->ccdc_cfg.bayer.win.height = 600;
+ ccdc->ccdc_cfg.bayer.config_params.data_sz = VPFE_CCDC_DATA_8BITS;
+ ccdc->ccdc_cfg.bayer.config_params.alaw.gamma_wd =
+ VPFE_CCDC_GAMMA_BITS_09_0;
+}
+
+/*
+ * vpfe_get_ccdc_image_format - Get image parameters based on CCDC settings
+ */
+static int vpfe_get_ccdc_image_format(struct vpfe_device *vpfe,
+ struct v4l2_format *f)
+{
+ struct v4l2_rect image_win;
+ enum ccdc_buftype buf_type;
+ enum ccdc_frmfmt frm_fmt;
+
+ memset(f, 0, sizeof(*f));
+ f->type = V4L2_BUF_TYPE_VIDEO_CAPTURE;
+ vpfe_ccdc_get_image_window(&vpfe->ccdc, &image_win);
+ f->fmt.pix.width = image_win.width;
+ f->fmt.pix.height = image_win.height;
+ f->fmt.pix.bytesperline = vpfe_ccdc_get_line_length(&vpfe->ccdc);
+ f->fmt.pix.sizeimage = f->fmt.pix.bytesperline *
+ f->fmt.pix.height;
+ buf_type = vpfe_ccdc_get_buftype(&vpfe->ccdc);
+ f->fmt.pix.pixelformat = vpfe_ccdc_get_pixel_format(&vpfe->ccdc);
+ frm_fmt = vpfe_ccdc_get_frame_format(&vpfe->ccdc);
+
+ if (frm_fmt == CCDC_FRMFMT_PROGRESSIVE) {
+ f->fmt.pix.field = V4L2_FIELD_NONE;
+ } else if (frm_fmt == CCDC_FRMFMT_INTERLACED) {
+ if (buf_type == CCDC_BUFTYPE_FLD_INTERLEAVED) {
+ f->fmt.pix.field = V4L2_FIELD_INTERLACED;
+ } else if (buf_type == CCDC_BUFTYPE_FLD_SEPARATED) {
+ f->fmt.pix.field = V4L2_FIELD_SEQ_TB;
+ } else {
+ vpfe_err(vpfe, "Invalid buf_type\n");
+ return -EINVAL;
+ }
+ } else {
+ vpfe_err(vpfe, "Invalid frm_fmt\n");
+ return -EINVAL;
+ }
+ return 0;
+}
+
+static int vpfe_config_ccdc_image_format(struct vpfe_device *vpfe)
+{
+ enum ccdc_frmfmt frm_fmt = CCDC_FRMFMT_INTERLACED;
+ int ret;
+
+ vpfe_dbg(2, vpfe, "vpfe_config_ccdc_image_format\n");
+
+ vpfe_dbg(1, vpfe, "pixelformat: %s\n",
+ print_fourcc(vpfe->fmt.fmt.pix.pixelformat));
+
+ if (vpfe_ccdc_set_pixel_format(&vpfe->ccdc,
+ vpfe->fmt.fmt.pix.pixelformat) < 0) {
+ vpfe_err(vpfe, "couldn't set pix format in ccdc\n");
+ return -EINVAL;
+ }
+
+ /* configure the image window */
+ vpfe_ccdc_set_image_window(&vpfe->ccdc, &vpfe->crop, vpfe->bpp);
+
+ switch (vpfe->fmt.fmt.pix.field) {
+ case V4L2_FIELD_INTERLACED:
+ /* do nothing, since it is default */
+ ret = vpfe_ccdc_set_buftype(
+ &vpfe->ccdc,
+ CCDC_BUFTYPE_FLD_INTERLEAVED);
+ break;
+
+ case V4L2_FIELD_NONE:
+ frm_fmt = CCDC_FRMFMT_PROGRESSIVE;
+ /* buffer type only applicable for interlaced scan */
+ break;
+
+ case V4L2_FIELD_SEQ_TB:
+ ret = vpfe_ccdc_set_buftype(
+ &vpfe->ccdc,
+ CCDC_BUFTYPE_FLD_SEPARATED);
+ break;
+
+ default:
+ return -EINVAL;
+ }
+
+ if (ret)
+ return ret;
+
+ return vpfe_ccdc_set_frame_format(&vpfe->ccdc, frm_fmt);
+}
+
+/*
+ * vpfe_config_image_format()
+ * For a given standard, this functions sets up the default
+ * pix format & crop values in the vpfe device and ccdc. It first
+ * starts with defaults based values from the standard table.
+ * It then checks if sub device support g_mbus_fmt and then override the
+ * values based on that.Sets crop values to match with scan resolution
+ * starting at 0,0. It calls vpfe_config_ccdc_image_format() set the
+ * values in ccdc
+ */
+static int vpfe_config_image_format(struct vpfe_device *vpfe,
+ v4l2_std_id std_id)
+{
+ struct v4l2_pix_format *pix = &vpfe->fmt.fmt.pix;
+ int i, ret;
+
+ for (i = 0; i < ARRAY_SIZE(vpfe_standards); i++) {
+ if (vpfe_standards[i].std_id & std_id) {
+ vpfe->std_info.active_pixels =
+ vpfe_standards[i].width;
+ vpfe->std_info.active_lines =
+ vpfe_standards[i].height;
+ vpfe->std_info.frame_format =
+ vpfe_standards[i].frame_format;
+ vpfe->std_index = i;
+
+ break;
+ }
+ }
+
+ if (i == ARRAY_SIZE(vpfe_standards)) {
+ vpfe_err(vpfe, "standard not supported\n");
+ return -EINVAL;
+ }
+
+ vpfe->crop.top = vpfe->crop.left = 0;
+ vpfe->crop.width = vpfe->std_info.active_pixels;
+ vpfe->crop.height = vpfe->std_info.active_lines;
+ pix->width = vpfe->crop.width;
+ pix->height = vpfe->crop.height;
+ pix->pixelformat = V4L2_PIX_FMT_YUYV;
+
+ /* first field and frame format based on standard frame format */
+ if (vpfe->std_info.frame_format)
+ pix->field = V4L2_FIELD_INTERLACED;
+ else
+ pix->field = V4L2_FIELD_NONE;
+
+ ret = __vpfe_get_format(vpfe, &vpfe->fmt, &vpfe->bpp);
+ if (ret)
+ return ret;
+
+ /* Update the crop window based on found values */
+ vpfe->crop.width = pix->width;
+ vpfe->crop.height = pix->height;
+
+ return vpfe_config_ccdc_image_format(vpfe);
+}
+
+static int vpfe_initialize_device(struct vpfe_device *vpfe)
+{
+ struct vpfe_subdev_info *sdinfo;
+ int ret;
+
+ sdinfo = &vpfe->cfg->sub_devs[0];
+ sdinfo->sd = vpfe->sd[0];
+ vpfe->current_input = 0;
+ vpfe->std_index = 0;
+ /* Configure the default format information */
+ ret = vpfe_config_image_format(vpfe,
+ vpfe_standards[vpfe->std_index].std_id);
+ if (ret)
+ return ret;
+
+ pm_runtime_get_sync(vpfe->pdev);
+
+ vpfe_config_enable(&vpfe->ccdc, 1);
+
+ vpfe_ccdc_restore_defaults(&vpfe->ccdc);
+
+ /* Clear all VPFE interrupts */
+ vpfe_clear_intr(&vpfe->ccdc, -1);
+
+ return ret;
+}
+
+/*
+ * vpfe_release : This function is based on the vb2_fop_release
+ * helper function.
+ * It has been augmented to handle module power management,
+ * by disabling/enabling h/w module fcntl clock when necessary.
+ */
+static int vpfe_release(struct file *file)
+{
+ struct vpfe_device *vpfe = video_drvdata(file);
+ int ret;
+
+ vpfe_dbg(2, vpfe, "vpfe_release\n");
+
+ if (!v4l2_fh_is_singular_file(file))
+ return vb2_fop_release(file);
+
+ mutex_lock(&vpfe->lock);
+ ret = _vb2_fop_release(file, NULL);
+ vpfe_ccdc_close(&vpfe->ccdc, vpfe->pdev);
+ mutex_unlock(&vpfe->lock);
+
+ return ret;
+}
+
+/*
+ * vpfe_open : This function is based on the v4l2_fh_open helper function.
+ * It has been augmented to handle module power management,
+ * by disabling/enabling h/w module fcntl clock when necessary.
+ */
+static int vpfe_open(struct file *file)
+{
+ struct vpfe_device *vpfe = video_drvdata(file);
+ int ret;
+
+ ret = v4l2_fh_open(file);
+ if (ret) {
+ vpfe_err(vpfe, "v4l2_fh_open failed\n");
+ return ret;
+ }
+
+ if (!v4l2_fh_is_singular_file(file))
+ return 0;
+
+ mutex_lock(&vpfe->lock);
+ if (vpfe_initialize_device(vpfe)) {
+ mutex_unlock(&vpfe->lock);
+ v4l2_fh_release(file);
+ return -ENODEV;
+ }
+ mutex_unlock(&vpfe->lock);
+
+ return 0;
+}
+
+/**
+ * vpfe_schedule_next_buffer: set next buffer address for capture
+ * @vpfe : ptr to vpfe device
+ *
+ * This function will get next buffer from the dma queue and
+ * set the buffer address in the vpfe register for capture.
+ * the buffer is marked active
+ *
+ * Assumes caller is holding vpfe->dma_queue_lock already
+ */
+static inline void vpfe_schedule_next_buffer(struct vpfe_device *vpfe)
+{
+ vpfe->next_frm = list_entry(vpfe->dma_queue.next,
+ struct vpfe_cap_buffer, list);
+ list_del(&vpfe->next_frm->list);
+
+ vpfe_set_sdr_addr(&vpfe->ccdc,
+ vb2_dma_contig_plane_dma_addr(&vpfe->next_frm->vb, 0));
+}
+
+static inline void vpfe_schedule_bottom_field(struct vpfe_device *vpfe)
+{
+ unsigned long addr;
+
+ addr = vb2_dma_contig_plane_dma_addr(&vpfe->next_frm->vb, 0) +
+ vpfe->field_off;
+
+ vpfe_set_sdr_addr(&vpfe->ccdc, addr);
+}
+
+/*
+ * vpfe_process_buffer_complete: process a completed buffer
+ * @vpfe : ptr to vpfe device
+ *
+ * This function time stamp the buffer and mark it as DONE. It also
+ * wake up any process waiting on the QUEUE and set the next buffer
+ * as current
+ */
+static inline void vpfe_process_buffer_complete(struct vpfe_device *vpfe)
+{
+ v4l2_get_timestamp(&vpfe->cur_frm->vb.v4l2_buf.timestamp);
+ vpfe->cur_frm->vb.v4l2_buf.field = vpfe->fmt.fmt.pix.field;
+ vpfe->cur_frm->vb.v4l2_buf.sequence = vpfe->sequence++;
+ vb2_buffer_done(&vpfe->cur_frm->vb, VB2_BUF_STATE_DONE);
+ vpfe->cur_frm = vpfe->next_frm;
+}
+
+/*
+ * vpfe_isr : ISR handler for vpfe capture (VINT0)
+ * @irq: irq number
+ * @dev_id: dev_id ptr
+ *
+ * It changes status of the captured buffer, takes next buffer from the queue
+ * and sets its address in VPFE registers
+ */
+static irqreturn_t vpfe_isr(int irq, void *dev)
+{
+ struct vpfe_device *vpfe = (struct vpfe_device *)dev;
+ enum v4l2_field field;
+ int intr_status;
+ int fid;
+
+ intr_status = vpfe_reg_read(&vpfe->ccdc, VPFE_IRQ_STS);
+
+ if (intr_status & VPFE_VDINT0) {
+ field = vpfe->fmt.fmt.pix.field;
+
+ if (field == V4L2_FIELD_NONE) {
+ /* handle progressive frame capture */
+ if (vpfe->cur_frm != vpfe->next_frm)
+ vpfe_process_buffer_complete(vpfe);
+ goto next_intr;
+ }
+
+ /* interlaced or TB capture check which field
+ we are in hardware */
+ fid = vpfe_ccdc_getfid(&vpfe->ccdc);
+
+ /* switch the software maintained field id */
+ vpfe->field ^= 1;
+ if (fid == vpfe->field) {
+ /* we are in-sync here,continue */
+ if (fid == 0) {
+ /*
+ * One frame is just being captured. If the
+ * next frame is available, release the
+ * current frame and move on
+ */
+ if (vpfe->cur_frm != vpfe->next_frm)
+ vpfe_process_buffer_complete(vpfe);
+ /*
+ * based on whether the two fields are stored
+ * interleave or separately in memory,
+ * reconfigure the CCDC memory address
+ */
+ if (field == V4L2_FIELD_SEQ_TB)
+ vpfe_schedule_bottom_field(vpfe);
+
+ goto next_intr;
+ }
+ /*
+ * if one field is just being captured configure
+ * the next frame get the next frame from the empty
+ * queue if no frame is available hold on to the
+ * current buffer
+ */
+ spin_lock(&vpfe->dma_queue_lock);
+ if (!list_empty(&vpfe->dma_queue) &&
+ vpfe->cur_frm == vpfe->next_frm)
+ vpfe_schedule_next_buffer(vpfe);
+ spin_unlock(&vpfe->dma_queue_lock);
+ } else if (fid == 0) {
+ /*
+ * out of sync. Recover from any hardware out-of-sync.
+ * May loose one frame
+ */
+ vpfe->field = fid;
+ }
+ }
+
+next_intr:
+ if (intr_status & VPFE_VDINT1) {
+ spin_lock(&vpfe->dma_queue_lock);
+ if (vpfe->fmt.fmt.pix.field == V4L2_FIELD_NONE &&
+ !list_empty(&vpfe->dma_queue) &&
+ vpfe->cur_frm == vpfe->next_frm)
+ vpfe_schedule_next_buffer(vpfe);
+ spin_unlock(&vpfe->dma_queue_lock);
+ }
+
+ vpfe_clear_intr(&vpfe->ccdc, intr_status);
+
+ return IRQ_HANDLED;
+}
+
+static inline void vpfe_detach_irq(struct vpfe_device *vpfe)
+{
+ unsigned int intr = VPFE_VDINT0;
+ enum ccdc_frmfmt frame_format;
+
+ frame_format = vpfe_ccdc_get_frame_format(&vpfe->ccdc);
+ if (frame_format == CCDC_FRMFMT_PROGRESSIVE)
+ intr |= VPFE_VDINT1;
+
+ vpfe_reg_write(&vpfe->ccdc, intr, VPFE_IRQ_EN_CLR);
+}
+
+static inline void vpfe_attach_irq(struct vpfe_device *vpfe)
+{
+ unsigned int intr = VPFE_VDINT0;
+ enum ccdc_frmfmt frame_format;
+
+ frame_format = vpfe_ccdc_get_frame_format(&vpfe->ccdc);
+ if (frame_format == CCDC_FRMFMT_PROGRESSIVE)
+ intr |= VPFE_VDINT1;
+
+ vpfe_reg_write(&vpfe->ccdc, intr, VPFE_IRQ_EN_SET);
+}
+
+static int vpfe_querycap(struct file *file, void *priv,
+ struct v4l2_capability *cap)
+{
+ struct vpfe_device *vpfe = video_drvdata(file);
+
+ vpfe_dbg(2, vpfe, "vpfe_querycap\n");
+
+ strlcpy(cap->driver, VPFE_MODULE_NAME, sizeof(cap->driver));
+ strlcpy(cap->card, "TI AM437x VPFE", sizeof(cap->card));
+ snprintf(cap->bus_info, sizeof(cap->bus_info),
+ "platform:%s", vpfe->v4l2_dev.name);
+ cap->device_caps = V4L2_CAP_VIDEO_CAPTURE | V4L2_CAP_STREAMING |
+ V4L2_CAP_READWRITE;
+ cap->capabilities = cap->device_caps | V4L2_CAP_DEVICE_CAPS;
+
+ return 0;
+}
+
+/* get the format set at output pad of the adjacent subdev */
+static int __vpfe_get_format(struct vpfe_device *vpfe,
+ struct v4l2_format *format, unsigned int *bpp)
+{
+ struct v4l2_mbus_framefmt mbus_fmt;
+ struct vpfe_subdev_info *sdinfo;
+ struct v4l2_subdev_format fmt;
+ int ret;
+
+ sdinfo = vpfe->current_subdev;
+ if (!sdinfo->sd)
+ return -EINVAL;
+
+ fmt.which = V4L2_SUBDEV_FORMAT_ACTIVE;
+ fmt.pad = 0;
+
+ ret = v4l2_subdev_call(sdinfo->sd, pad, get_fmt, NULL, &fmt);
+ if (ret && ret != -ENOIOCTLCMD && ret != -ENODEV)
+ return ret;
+
+ if (!ret) {
+ v4l2_fill_pix_format(&format->fmt.pix, &fmt.format);
+ mbus_to_pix(vpfe, &fmt.format, &format->fmt.pix, bpp);
+ } else {
+ ret = v4l2_device_call_until_err(&vpfe->v4l2_dev,
+ sdinfo->grp_id,
+ video, g_mbus_fmt,
+ &mbus_fmt);
+ if (ret && ret != -ENOIOCTLCMD && ret != -ENODEV)
+ return ret;
+ v4l2_fill_pix_format(&format->fmt.pix, &mbus_fmt);
+ mbus_to_pix(vpfe, &mbus_fmt, &format->fmt.pix, bpp);
+ }
+
+ format->type = vpfe->fmt.type;
+
+ vpfe_dbg(1, vpfe,
+ "%s size %dx%d (%s) bytesperline = %d, size = %d, bpp = %d\n",
+ __func__, format->fmt.pix.width, format->fmt.pix.height,
+ print_fourcc(format->fmt.pix.pixelformat),
+ format->fmt.pix.bytesperline, format->fmt.pix.sizeimage, *bpp);
+
+ return 0;
+}
+
+/* set the format at output pad of the adjacent subdev */
+static int __vpfe_set_format(struct vpfe_device *vpfe,
+ struct v4l2_format *format, unsigned int *bpp)
+{
+ struct v4l2_mbus_framefmt mbus_fmt;
+ struct vpfe_subdev_info *sdinfo;
+ struct v4l2_subdev_format fmt;
+ int ret;
+
+ vpfe_dbg(2, vpfe, "__vpfe_set_format\n");
+
+ sdinfo = vpfe->current_subdev;
+ if (!sdinfo->sd)
+ return -EINVAL;
+
+ fmt.which = V4L2_SUBDEV_FORMAT_ACTIVE;
+ fmt.pad = 0;
+
+ pix_to_mbus(vpfe, &format->fmt.pix, &fmt.format);
+
+ ret = v4l2_subdev_call(sdinfo->sd, pad, set_fmt, NULL, &fmt);
+ if (ret && ret != -ENOIOCTLCMD && ret != -ENODEV)
+ return ret;
+
+ if (!ret) {
+ v4l2_fill_pix_format(&format->fmt.pix, &fmt.format);
+ mbus_to_pix(vpfe, &fmt.format, &format->fmt.pix, bpp);
+ } else {
+ ret = v4l2_device_call_until_err(&vpfe->v4l2_dev,
+ sdinfo->grp_id,
+ video, s_mbus_fmt,
+ &mbus_fmt);
+ if (ret && ret != -ENOIOCTLCMD && ret != -ENODEV)
+ return ret;
+
+ v4l2_fill_pix_format(&format->fmt.pix, &mbus_fmt);
+ mbus_to_pix(vpfe, &mbus_fmt, &format->fmt.pix, bpp);
+ }
+
+ format->type = vpfe->fmt.type;
+
+ vpfe_dbg(1, vpfe,
+ "%s size %dx%d (%s) bytesperline = %d, size = %d, bpp = %d\n",
+ __func__, format->fmt.pix.width, format->fmt.pix.height,
+ print_fourcc(format->fmt.pix.pixelformat),
+ format->fmt.pix.bytesperline, format->fmt.pix.sizeimage, *bpp);
+
+ return 0;
+}
+
+static int vpfe_g_fmt(struct file *file, void *priv,
+ struct v4l2_format *fmt)
+{
+ struct vpfe_device *vpfe = video_drvdata(file);
+
+ vpfe_dbg(2, vpfe, "vpfe_g_fmt\n");
+
+ *fmt = vpfe->fmt;
+
+ return 0;
+}
+
+static int vpfe_enum_fmt(struct file *file, void *priv,
+ struct v4l2_fmtdesc *f)
+{
+ struct vpfe_device *vpfe = video_drvdata(file);
+ struct vpfe_subdev_info *sdinfo;
+ struct vpfe_fmt *fmt = NULL;
+ unsigned int k;
+
+ vpfe_dbg(2, vpfe, "vpfe_enum_format index:%d\n",
+ f->index);
+
+ sdinfo = vpfe->current_subdev;
+ if (!sdinfo->sd)
+ return -EINVAL;
+
+ if (f->index > ARRAY_SIZE(formats))
+ return -EINVAL;
+
+ for (k = 0; k < ARRAY_SIZE(formats); k++) {
+ if (formats[k].index == f->index) {
+ fmt = &formats[k];
+ break;
+ }
+ }
+ if (!fmt)
+ return -EINVAL;
+
+ strncpy(f->description, fmt->name, sizeof(f->description) - 1);
+ f->pixelformat = fmt->fourcc;
+ f->type = vpfe->fmt.type;
+
+ vpfe_dbg(1, vpfe, "vpfe_enum_format: mbus index: %d code: %x pixelformat: %s [%s]\n",
+ f->index, fmt->code, print_fourcc(fmt->fourcc), fmt->name);
+
+ return 0;
+}
+
+static int vpfe_try_fmt(struct file *file, void *priv,
+ struct v4l2_format *fmt)
+{
+ struct vpfe_device *vpfe = video_drvdata(file);
+ unsigned int bpp;
+
+ vpfe_dbg(2, vpfe, "vpfe_try_fmt\n");
+
+ return __vpfe_get_format(vpfe, fmt, &bpp);
+}
+
+static int vpfe_s_fmt(struct file *file, void *priv,
+ struct v4l2_format *fmt)
+{
+ struct vpfe_device *vpfe = video_drvdata(file);
+ struct v4l2_format format;
+ unsigned int bpp;
+ int ret;
+
+ vpfe_dbg(2, vpfe, "vpfe_s_fmt\n");
+
+ /* If streaming is started, return error */
+ if (vb2_is_busy(&vpfe->buffer_queue)) {
+ vpfe_err(vpfe, "%s device busy\n", __func__);
+ return -EBUSY;
+ }
+
+ ret = vpfe_try_fmt(file, priv, fmt);
+ if (ret)
+ return ret;
+
+
+ if (!cmp_v4l2_format(fmt, &format)) {
+ /* Sensor format is different from the requested format
+ * so we need to change it
+ */
+ ret = __vpfe_set_format(vpfe, fmt, &bpp);
+ if (ret)
+ return ret;
+ } else /* Just make sure all of the fields are consistent */
+ *fmt = format;
+
+ /* First detach any IRQ if currently attached */
+ vpfe_detach_irq(vpfe);
+ vpfe->fmt = *fmt;
+ vpfe->bpp = bpp;
+
+ /* Update the crop window based on found values */
+ vpfe->crop.width = fmt->fmt.pix.width;
+ vpfe->crop.height = fmt->fmt.pix.height;
+
+ /* set image capture parameters in the ccdc */
+ return vpfe_config_ccdc_image_format(vpfe);
+}
+
+static int vpfe_enum_size(struct file *file, void *priv,
+ struct v4l2_frmsizeenum *fsize)
+{
+ struct vpfe_device *vpfe = video_drvdata(file);
+ struct v4l2_subdev_frame_size_enum fse;
+ struct vpfe_subdev_info *sdinfo;
+ struct v4l2_mbus_framefmt mbus;
+ struct v4l2_pix_format pix;
+ struct vpfe_fmt *fmt;
+ int ret;
+
+ vpfe_dbg(2, vpfe, "vpfe_enum_size\n");
+
+ /* check for valid format */
+ fmt = find_format_by_pix(fsize->pixel_format);
+ if (!fmt) {
+ vpfe_dbg(3, vpfe, "Invalid pixel code: %x, default used instead\n",
+ fsize->pixel_format);
+ return -EINVAL;
+ }
+
+ memset(fsize->reserved, 0x0, sizeof(fsize->reserved));
+
+ sdinfo = vpfe->current_subdev;
+ if (!sdinfo->sd)
+ return -EINVAL;
+
+ memset(&pix, 0x0, sizeof(pix));
+ /* Construct pix from parameter and use default for the rest */
+ pix.pixelformat = fsize->pixel_format;
+ pix.width = 640;
+ pix.height = 480;
+ pix.colorspace = V4L2_COLORSPACE_SRGB;
+ pix.field = V4L2_FIELD_NONE;
+ pix_to_mbus(vpfe, &pix, &mbus);
+
+ memset(&fse, 0x0, sizeof(fse));
+ fse.index = fsize->index;
+ fse.pad = 0;
+ fse.code = mbus.code;
+ ret = v4l2_subdev_call(sdinfo->sd, pad, enum_frame_size, NULL, &fse);
+ if (ret)
+ return -EINVAL;
+
+ vpfe_dbg(1, vpfe, "vpfe_enum_size: index: %d code: %x W:[%d,%d] H:[%d,%d]\n",
+ fse.index, fse.code, fse.min_width, fse.max_width,
+ fse.min_height, fse.max_height);
+
+ fsize->type = V4L2_FRMSIZE_TYPE_DISCRETE;
+ fsize->discrete.width = fse.max_width;
+ fsize->discrete.height = fse.max_height;
+
+ vpfe_dbg(1, vpfe, "vpfe_enum_size: index: %d pixformat: %s size: %dx%d\n",
+ fsize->index, print_fourcc(fsize->pixel_format),
+ fsize->discrete.width, fsize->discrete.height);
+
+ return 0;
+}
+
+/*
+ * vpfe_get_subdev_input_index - Get subdev index and subdev input index for a
+ * given app input index
+ */
+static int
+vpfe_get_subdev_input_index(struct vpfe_device *vpfe,
+ int *subdev_index,
+ int *subdev_input_index,
+ int app_input_index)
+{
+ struct vpfe_config *cfg = vpfe->cfg;
+ struct vpfe_subdev_info *sdinfo;
+ int i, j = 0;
+
+ for (i = 0; i < ARRAY_SIZE(vpfe->cfg->asd); i++) {
+ sdinfo = &cfg->sub_devs[i];
+ if (app_input_index < (j + 1)) {
+ *subdev_index = i;
+ *subdev_input_index = app_input_index - j;
+ return 0;
+ }
+ j++;
+ }
+ return -EINVAL;
+}
+
+/*
+ * vpfe_get_app_input - Get app input index for a given subdev input index
+ * driver stores the input index of the current sub device and translate it
+ * when application request the current input
+ */
+static int vpfe_get_app_input_index(struct vpfe_device *vpfe,
+ int *app_input_index)
+{
+ struct vpfe_config *cfg = vpfe->cfg;
+ struct vpfe_subdev_info *sdinfo;
+ int i, j = 0;
+
+ for (i = 0; i < ARRAY_SIZE(vpfe->cfg->asd); i++) {
+ sdinfo = &cfg->sub_devs[i];
+ if (!strcmp(sdinfo->name, vpfe->current_subdev->name)) {
+ if (vpfe->current_input >= 1)
+ return -1;
+ *app_input_index = j + vpfe->current_input;
+ return 0;
+ }
+ j++;
+ }
+ return -EINVAL;
+}
+
+static int vpfe_enum_input(struct file *file, void *priv,
+ struct v4l2_input *inp)
+{
+ struct vpfe_device *vpfe = video_drvdata(file);
+ struct vpfe_subdev_info *sdinfo;
+ int subdev, index;
+
+ vpfe_dbg(2, vpfe, "vpfe_enum_input\n");
+
+ if (vpfe_get_subdev_input_index(vpfe, &subdev, &index,
+ inp->index) < 0) {
+ vpfe_dbg(1, vpfe,
+ "input information not found for the subdev\n");
+ return -EINVAL;
+ }
+ sdinfo = &vpfe->cfg->sub_devs[subdev];
+ *inp = sdinfo->inputs[index];
+
+ return 0;
+}
+
+static int vpfe_g_input(struct file *file, void *priv, unsigned int *index)
+{
+ struct vpfe_device *vpfe = video_drvdata(file);
+
+ vpfe_dbg(2, vpfe, "vpfe_g_input\n");
+
+ return vpfe_get_app_input_index(vpfe, index);
+}
+
+/* Assumes caller is holding vpfe_dev->lock */
+static int vpfe_set_input(struct vpfe_device *vpfe, unsigned int index)
+{
+ int subdev_index = 0, inp_index = 0;
+ struct vpfe_subdev_info *sdinfo;
+ struct vpfe_route *route;
+ u32 input, output;
+ int ret;
+
+ vpfe_dbg(2, vpfe, "vpfe_set_input: index: %d\n", index);
+
+ /* If streaming is started, return error */
+ if (vb2_is_busy(&vpfe->buffer_queue)) {
+ vpfe_err(vpfe, "%s device busy\n", __func__);
+ return -EBUSY;
+ }
+ ret = vpfe_get_subdev_input_index(vpfe,
+ &subdev_index,
+ &inp_index,
+ index);
+ if (ret < 0) {
+ vpfe_err(vpfe, "invalid input index: %d\n", index);
+ goto get_out;
+ }
+
+ sdinfo = &vpfe->cfg->sub_devs[subdev_index];
+ sdinfo->sd = vpfe->sd[subdev_index];
+ route = &sdinfo->routes[inp_index];
+ if (route && sdinfo->can_route) {
+ input = route->input;
+ output = route->output;
+ if (sdinfo->sd) {
+ ret = v4l2_subdev_call(sdinfo->sd, video,
+ s_routing, input, output, 0);
+ if (ret) {
+ vpfe_err(vpfe, "s_routing failed\n");
+ ret = -EINVAL;
+ goto get_out;
+ }
+ }
+
+ }
+
+ vpfe->current_subdev = sdinfo;
+ if (sdinfo->sd)
+ vpfe->v4l2_dev.ctrl_handler = sdinfo->sd->ctrl_handler;
+ vpfe->current_input = index;
+ vpfe->std_index = 0;
+
+ /* set the bus/interface parameter for the sub device in ccdc */
+ ret = vpfe_ccdc_set_hw_if_params(&vpfe->ccdc, &sdinfo->vpfe_param);
+ if (ret)
+ return ret;
+
+ /* set the default image parameters in the device */
+ return vpfe_config_image_format(vpfe,
+ vpfe_standards[vpfe->std_index].std_id);
+
+get_out:
+ return ret;
+}
+
+static int vpfe_s_input(struct file *file, void *priv, unsigned int index)
+{
+ struct vpfe_device *vpfe = video_drvdata(file);
+
+ vpfe_dbg(2, vpfe,
+ "vpfe_s_input: index: %d\n", index);
+
+ return vpfe_set_input(vpfe, index);
+}
+
+static int vpfe_querystd(struct file *file, void *priv, v4l2_std_id *std_id)
+{
+ struct vpfe_device *vpfe = video_drvdata(file);
+ struct vpfe_subdev_info *sdinfo;
+
+ vpfe_dbg(2, vpfe, "vpfe_querystd\n");
+
+ sdinfo = vpfe->current_subdev;
+ if (!(sdinfo->inputs[0].capabilities & V4L2_IN_CAP_STD))
+ return -ENODATA;
+
+ /* Call querystd function of decoder device */
+ return v4l2_device_call_until_err(&vpfe->v4l2_dev, sdinfo->grp_id,
+ video, querystd, std_id);
+}
+
+static int vpfe_s_std(struct file *file, void *priv, v4l2_std_id std_id)
+{
+ struct vpfe_device *vpfe = video_drvdata(file);
+ struct vpfe_subdev_info *sdinfo;
+ int ret;
+
+ vpfe_dbg(2, vpfe, "vpfe_s_std\n");
+
+ sdinfo = vpfe->current_subdev;
+ if (!(sdinfo->inputs[0].capabilities & V4L2_IN_CAP_STD))
+ return -ENODATA;
+
+ /* If streaming is started, return error */
+ if (vb2_is_busy(&vpfe->buffer_queue)) {
+ vpfe_err(vpfe, "%s device busy\n", __func__);
+ ret = -EBUSY;
+ return ret;
+ }
+
+ ret = v4l2_device_call_until_err(&vpfe->v4l2_dev, sdinfo->grp_id,
+ video, s_std, std_id);
+ if (ret < 0) {
+ vpfe_err(vpfe, "Failed to set standard\n");
+ return ret;
+ }
+ ret = vpfe_config_image_format(vpfe, std_id);
+
+ return ret;
+}
+
+static int vpfe_g_std(struct file *file, void *priv, v4l2_std_id *std_id)
+{
+ struct vpfe_device *vpfe = video_drvdata(file);
+ struct vpfe_subdev_info *sdinfo;
+
+ vpfe_dbg(2, vpfe, "vpfe_g_std\n");
+
+ sdinfo = vpfe->current_subdev;
+ if (sdinfo->inputs[0].capabilities != V4L2_IN_CAP_STD)
+ return -ENODATA;
+
+ *std_id = vpfe_standards[vpfe->std_index].std_id;
+
+ return 0;
+}
+
+/*
+ * vpfe_calculate_offsets : This function calculates buffers offset
+ * for top and bottom field
+ */
+static void vpfe_calculate_offsets(struct vpfe_device *vpfe)
+{
+ struct v4l2_rect image_win;
+
+ vpfe_dbg(2, vpfe, "vpfe_calculate_offsets\n");
+
+ vpfe_ccdc_get_image_window(&vpfe->ccdc, &image_win);
+ vpfe->field_off = image_win.height * image_win.width;
+}
+
+/*
+ * vpfe_queue_setup - Callback function for buffer setup.
+ * @vq: vb2_queue ptr
+ * @fmt: v4l2 format
+ * @nbuffers: ptr to number of buffers requested by application
+ * @nplanes:: contains number of distinct video planes needed to hold a frame
+ * @sizes[]: contains the size (in bytes) of each plane.
+ * @alloc_ctxs: ptr to allocation context
+ *
+ * This callback function is called when reqbuf() is called to adjust
+ * the buffer count and buffer size
+ */
+static int vpfe_queue_setup(struct vb2_queue *vq,
+ const struct v4l2_format *fmt,
+ unsigned int *nbuffers, unsigned int *nplanes,
+ unsigned int sizes[], void *alloc_ctxs[])
+{
+ struct vpfe_device *vpfe = vb2_get_drv_priv(vq);
+
+ if (fmt && fmt->fmt.pix.sizeimage < vpfe->fmt.fmt.pix.sizeimage)
+ return -EINVAL;
+
+ if (vq->num_buffers + *nbuffers < 3)
+ *nbuffers = 3 - vq->num_buffers;
+
+ *nplanes = 1;
+ sizes[0] = fmt ? fmt->fmt.pix.sizeimage : vpfe->fmt.fmt.pix.sizeimage;
+ alloc_ctxs[0] = vpfe->alloc_ctx;
+
+ vpfe_dbg(1, vpfe,
+ "nbuffers=%d, size=%u\n", *nbuffers, sizes[0]);
+
+ /* Calculate field offset */
+ vpfe_calculate_offsets(vpfe);
+
+ return 0;
+}
+
+/*
+ * vpfe_buffer_prepare : callback function for buffer prepare
+ * @vb: ptr to vb2_buffer
+ *
+ * This is the callback function for buffer prepare when vb2_qbuf()
+ * function is called. The buffer is prepared and user space virtual address
+ * or user address is converted into physical address
+ */
+static int vpfe_buffer_prepare(struct vb2_buffer *vb)
+{
+ struct vpfe_device *vpfe = vb2_get_drv_priv(vb->vb2_queue);
+
+ vb2_set_plane_payload(vb, 0, vpfe->fmt.fmt.pix.sizeimage);
+
+ if (vb2_get_plane_payload(vb, 0) > vb2_plane_size(vb, 0))
+ return -EINVAL;
+
+ vb->v4l2_buf.field = vpfe->fmt.fmt.pix.field;
+
+ return 0;
+}
+
+/*
+ * vpfe_buffer_queue : Callback function to add buffer to DMA queue
+ * @vb: ptr to vb2_buffer
+ */
+static void vpfe_buffer_queue(struct vb2_buffer *vb)
+{
+ struct vpfe_device *vpfe = vb2_get_drv_priv(vb->vb2_queue);
+ struct vpfe_cap_buffer *buf = to_vpfe_buffer(vb);
+ unsigned long flags = 0;
+
+ /* add the buffer to the DMA queue */
+ spin_lock_irqsave(&vpfe->dma_queue_lock, flags);
+ list_add_tail(&buf->list, &vpfe->dma_queue);
+ spin_unlock_irqrestore(&vpfe->dma_queue_lock, flags);
+}
+
+/*
+ * vpfe_start_streaming : Starts the DMA engine for streaming
+ * @vb: ptr to vb2_buffer
+ * @count: number of buffers
+ */
+static int vpfe_start_streaming(struct vb2_queue *vq, unsigned int count)
+{
+ struct vpfe_device *vpfe = vb2_get_drv_priv(vq);
+ struct vpfe_cap_buffer *buf, *tmp;
+ struct vpfe_subdev_info *sdinfo;
+ unsigned long flags;
+ unsigned long addr;
+ int ret;
+
+ spin_lock_irqsave(&vpfe->dma_queue_lock, flags);
+
+ vpfe->field = 0;
+ vpfe->sequence = 0;
+
+ sdinfo = vpfe->current_subdev;
+
+ vpfe_attach_irq(vpfe);
+
+ if (vpfe->ccdc.ccdc_cfg.if_type == VPFE_RAW_BAYER)
+ vpfe_ccdc_config_raw(&vpfe->ccdc);
+ else
+ vpfe_ccdc_config_ycbcr(&vpfe->ccdc);
+
+ /* Get the next frame from the buffer queue */
+ vpfe->next_frm = list_entry(vpfe->dma_queue.next,
+ struct vpfe_cap_buffer, list);
+ vpfe->cur_frm = vpfe->next_frm;
+ /* Remove buffer from the buffer queue */
+ list_del(&vpfe->cur_frm->list);
+ spin_unlock_irqrestore(&vpfe->dma_queue_lock, flags);
+
+ addr = vb2_dma_contig_plane_dma_addr(&vpfe->cur_frm->vb, 0);
+
+ vpfe_set_sdr_addr(&vpfe->ccdc, (unsigned long)(addr));
+
+ vpfe_pcr_enable(&vpfe->ccdc, 1);
+
+ ret = v4l2_subdev_call(sdinfo->sd, video, s_stream, 1);
+ if (ret < 0) {
+ vpfe_err(vpfe, "Error in attaching interrupt handle\n");
+ goto err;
+ }
+
+ return 0;
+
+err:
+ list_for_each_entry_safe(buf, tmp, &vpfe->dma_queue, list) {
+ list_del(&buf->list);
+ vb2_buffer_done(&buf->vb, VB2_BUF_STATE_QUEUED);
+ }
+ spin_unlock_irqrestore(&vpfe->dma_queue_lock, flags);
+
+ return ret;
+}
+
+/*
+ * vpfe_stop_streaming : Stop the DMA engine
+ * @vq: ptr to vb2_queue
+ *
+ * This callback stops the DMA engine and any remaining buffers
+ * in the DMA queue are released.
+ */
+static void vpfe_stop_streaming(struct vb2_queue *vq)
+{
+ struct vpfe_device *vpfe = vb2_get_drv_priv(vq);
+ struct vpfe_subdev_info *sdinfo;
+ unsigned long flags;
+ int ret;
+
+ vpfe_pcr_enable(&vpfe->ccdc, 0);
+
+ vpfe_detach_irq(vpfe);
+
+ sdinfo = vpfe->current_subdev;
+ ret = v4l2_subdev_call(sdinfo->sd, video, s_stream, 0);
+ if (ret && ret != -ENOIOCTLCMD && ret != -ENODEV)
+ vpfe_dbg(1, vpfe, "stream off failed in subdev\n");
+
+ /* release all active buffers */
+ spin_lock_irqsave(&vpfe->dma_queue_lock, flags);
+ if (vpfe->cur_frm == vpfe->next_frm) {
+ vb2_buffer_done(&vpfe->cur_frm->vb, VB2_BUF_STATE_ERROR);
+ } else {
+ if (vpfe->cur_frm != NULL)
+ vb2_buffer_done(&vpfe->cur_frm->vb,
+ VB2_BUF_STATE_ERROR);
+ if (vpfe->next_frm != NULL)
+ vb2_buffer_done(&vpfe->next_frm->vb,
+ VB2_BUF_STATE_ERROR);
+ }
+
+ while (!list_empty(&vpfe->dma_queue)) {
+ vpfe->next_frm = list_entry(vpfe->dma_queue.next,
+ struct vpfe_cap_buffer, list);
+ list_del(&vpfe->next_frm->list);
+ vb2_buffer_done(&vpfe->next_frm->vb, VB2_BUF_STATE_ERROR);
+ }
+ spin_unlock_irqrestore(&vpfe->dma_queue_lock, flags);
+}
+
+static int vpfe_cropcap(struct file *file, void *priv,
+ struct v4l2_cropcap *crop)
+{
+ struct vpfe_device *vpfe = video_drvdata(file);
+
+ vpfe_dbg(2, vpfe, "vpfe_cropcap\n");
+
+ if (vpfe->std_index >= ARRAY_SIZE(vpfe_standards))
+ return -EINVAL;
+
+ memset(crop, 0, sizeof(struct v4l2_cropcap));
+
+ crop->type = V4L2_BUF_TYPE_VIDEO_CAPTURE;
+ crop->defrect.width = vpfe_standards[vpfe->std_index].width;
+ crop->bounds.width = crop->defrect.width;
+ crop->defrect.height = vpfe_standards[vpfe->std_index].height;
+ crop->bounds.height = crop->defrect.height;
+ crop->pixelaspect = vpfe_standards[vpfe->std_index].pixelaspect;
+
+ return 0;
+}
+
+static int
+vpfe_g_selection(struct file *file, void *fh, struct v4l2_selection *s)
+{
+ struct vpfe_device *vpfe = video_drvdata(file);
+
+ switch (s->target) {
+ case V4L2_SEL_TGT_CROP_BOUNDS:
+ case V4L2_SEL_TGT_CROP_DEFAULT:
+ s->r.left = s->r.top = 0;
+ s->r.width = vpfe->crop.width;
+ s->r.height = vpfe->crop.height;
+ break;
+
+ case V4L2_SEL_TGT_CROP:
+ s->r = vpfe->crop;
+ break;
+
+ default:
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
+static int enclosed_rectangle(struct v4l2_rect *a, struct v4l2_rect *b)
+{
+ if (a->left < b->left || a->top < b->top)
+ return 0;
+
+ if (a->left + a->width > b->left + b->width)
+ return 0;
+
+ if (a->top + a->height > b->top + b->height)
+ return 0;
+
+ return 1;
+}
+
+static int
+vpfe_s_selection(struct file *file, void *fh, struct v4l2_selection *s)
+{
+ struct vpfe_device *vpfe = video_drvdata(file);
+ struct v4l2_rect cr = vpfe->crop;
+ struct v4l2_rect r = s->r;
+
+ /* If streaming is started, return error */
+ if (vb2_is_busy(&vpfe->buffer_queue)) {
+ vpfe_err(vpfe, "%s device busy\n", __func__);
+ return -EBUSY;
+ }
+
+ if (s->type != V4L2_BUF_TYPE_VIDEO_CAPTURE ||
+ s->target != V4L2_SEL_TGT_CROP)
+ return -EINVAL;
+
+ v4l_bound_align_image(&r.width, 0, cr.width, 0,
+ &r.height, 0, cr.height, 0, 0);
+
+ r.left = clamp_t(unsigned int, r.left, 0, cr.width - r.width);
+ r.top = clamp_t(unsigned int, r.top, 0, cr.height - r.height);
+
+ if (s->flags & V4L2_SEL_FLAG_LE && !enclosed_rectangle(&r, &s->r))
+ return -ERANGE;
+
+ if (s->flags & V4L2_SEL_FLAG_GE && !enclosed_rectangle(&s->r, &r))
+ return -ERANGE;
+
+ s->r = vpfe->crop = r;
+
+ vpfe_ccdc_set_image_window(&vpfe->ccdc, &r, vpfe->bpp);
+ vpfe->fmt.fmt.pix.width = r.width;
+ vpfe->fmt.fmt.pix.height = r.height;
+ vpfe->fmt.fmt.pix.bytesperline = vpfe_ccdc_get_line_length(&vpfe->ccdc);
+ vpfe->fmt.fmt.pix.sizeimage = vpfe->fmt.fmt.pix.bytesperline *
+ vpfe->fmt.fmt.pix.height;
+
+ vpfe_dbg(1, vpfe, "cropped (%d,%d)/%dx%d of %dx%d\n",
+ r.left, r.top, r.width, r.height, cr.width, cr.height);
+
+ return 0;
+}
+
+static long vpfe_ioctl_default(struct file *file, void *priv,
+ bool valid_prio, unsigned int cmd, void *param)
+{
+ struct vpfe_device *vpfe = video_drvdata(file);
+ int ret;
+
+ vpfe_dbg(2, vpfe, "vpfe_ioctl_default\n");
+
+ if (!valid_prio) {
+ vpfe_err(vpfe, "%s device busy\n", __func__);
+ return -EBUSY;
+ }
+
+ /* If streaming is started, return error */
+ if (vb2_is_busy(&vpfe->buffer_queue)) {
+ vpfe_err(vpfe, "%s device busy\n", __func__);
+ return -EBUSY;
+ }
+
+ switch (cmd) {
+ case VIDIOC_AM437X_CCDC_CFG:
+ ret = vpfe_ccdc_set_params(&vpfe->ccdc, param);
+ if (ret) {
+ vpfe_dbg(2, vpfe,
+ "Error setting parameters in CCDC\n");
+ return ret;
+ }
+ ret = vpfe_get_ccdc_image_format(vpfe,
+ &vpfe->fmt);
+ if (ret < 0) {
+ vpfe_dbg(2, vpfe,
+ "Invalid image format at CCDC\n");
+ return ret;
+ }
+ break;
+
+ default:
+ ret = -ENOTTY;
+ break;
+ }
+
+ return ret;
+}
+
+static const struct vb2_ops vpfe_video_qops = {
+ .wait_prepare = vb2_ops_wait_prepare,
+ .wait_finish = vb2_ops_wait_finish,
+ .queue_setup = vpfe_queue_setup,
+ .buf_prepare = vpfe_buffer_prepare,
+ .buf_queue = vpfe_buffer_queue,
+ .start_streaming = vpfe_start_streaming,
+ .stop_streaming = vpfe_stop_streaming,
+};
+
+/* vpfe capture driver file operations */
+static const struct v4l2_file_operations vpfe_fops = {
+ .owner = THIS_MODULE,
+ .open = vpfe_open,
+ .release = vpfe_release,
+ .read = vb2_fop_read,
+ .poll = vb2_fop_poll,
+ .unlocked_ioctl = video_ioctl2,
+ .mmap = vb2_fop_mmap,
+};
+
+/* vpfe capture ioctl operations */
+static const struct v4l2_ioctl_ops vpfe_ioctl_ops = {
+ .vidioc_querycap = vpfe_querycap,
+ .vidioc_enum_fmt_vid_cap = vpfe_enum_fmt,
+ .vidioc_g_fmt_vid_cap = vpfe_g_fmt,
+ .vidioc_s_fmt_vid_cap = vpfe_s_fmt,
+ .vidioc_try_fmt_vid_cap = vpfe_try_fmt,
+
+ .vidioc_enum_framesizes = vpfe_enum_size,
+
+ .vidioc_enum_input = vpfe_enum_input,
+ .vidioc_g_input = vpfe_g_input,
+ .vidioc_s_input = vpfe_s_input,
+
+ .vidioc_querystd = vpfe_querystd,
+ .vidioc_s_std = vpfe_s_std,
+ .vidioc_g_std = vpfe_g_std,
+
+ .vidioc_reqbufs = vb2_ioctl_reqbufs,
+ .vidioc_create_bufs = vb2_ioctl_create_bufs,
+ .vidioc_prepare_buf = vb2_ioctl_prepare_buf,
+ .vidioc_querybuf = vb2_ioctl_querybuf,
+ .vidioc_qbuf = vb2_ioctl_qbuf,
+ .vidioc_dqbuf = vb2_ioctl_dqbuf,
+ .vidioc_expbuf = vb2_ioctl_expbuf,
+ .vidioc_streamon = vb2_ioctl_streamon,
+ .vidioc_streamoff = vb2_ioctl_streamoff,
+
+ .vidioc_log_status = v4l2_ctrl_log_status,
+ .vidioc_subscribe_event = v4l2_ctrl_subscribe_event,
+ .vidioc_unsubscribe_event = v4l2_event_unsubscribe,
+
+ .vidioc_cropcap = vpfe_cropcap,
+ .vidioc_g_selection = vpfe_g_selection,
+ .vidioc_s_selection = vpfe_s_selection,
+
+ .vidioc_default = vpfe_ioctl_default,
+};
+
+static int
+vpfe_async_bound(struct v4l2_async_notifier *notifier,
+ struct v4l2_subdev *subdev,
+ struct v4l2_async_subdev *asd)
+{
+ struct vpfe_device *vpfe = container_of(notifier->v4l2_dev,
+ struct vpfe_device, v4l2_dev);
+ struct v4l2_subdev_mbus_code_enum mbus_code;
+ struct vpfe_subdev_info *sdinfo;
+ bool found = false;
+ int i, j;
+
+ vpfe_dbg(1, vpfe, "vpfe_async_bound\n");
+
+ for (i = 0; i < ARRAY_SIZE(vpfe->cfg->asd); i++) {
+ sdinfo = &vpfe->cfg->sub_devs[i];
+
+ if (!strcmp(sdinfo->name, subdev->name)) {
+ vpfe->sd[i] = subdev;
+ vpfe_info(vpfe,
+ "v4l2 sub device %s registered\n",
+ subdev->name);
+ vpfe->sd[i]->grp_id =
+ sdinfo->grp_id;
+ /* update tvnorms from the sub devices */
+ for (j = 0; j < 1; j++)
+ vpfe->video_dev->tvnorms |=
+ sdinfo->inputs[j].std;
+
+ found = true;
+ break;
+ }
+ }
+
+ if (!found) {
+ vpfe_info(vpfe, "sub device (%s) not matched\n", subdev->name);
+ return -EINVAL;
+ }
+
+ /* setup the supported formats & indexes */
+ for (j = 0, i = 0; ; ++j) {
+ struct vpfe_fmt *fmt;
+ int ret;
+
+ memset(&mbus_code, 0, sizeof(mbus_code));
+ mbus_code.index = j;
+ ret = v4l2_subdev_call(subdev, pad, enum_mbus_code,
+ NULL, &mbus_code);
+ if (ret)
+ break;
+
+ fmt = find_format_by_code(mbus_code.code);
+ if (!fmt)
+ continue;
+
+ fmt->supported = true;
+ fmt->index = i++;
+ }
+
+ return 0;
+}
+
+static int vpfe_probe_complete(struct vpfe_device *vpfe)
+{
+ struct video_device *vdev;
+ struct vb2_queue *q;
+ int err;
+
+ spin_lock_init(&vpfe->dma_queue_lock);
+ mutex_init(&vpfe->lock);
+
+ vpfe->fmt.type = V4L2_BUF_TYPE_VIDEO_CAPTURE;
+
+ /* set first sub device as current one */
+ vpfe->current_subdev = &vpfe->cfg->sub_devs[0];
+ vpfe->v4l2_dev.ctrl_handler = vpfe->sd[0]->ctrl_handler;
+
+ err = vpfe_set_input(vpfe, 0);
+ if (err)
+ goto probe_out;
+
+ /* Initialize videobuf2 queue as per the buffer type */
+ vpfe->alloc_ctx = vb2_dma_contig_init_ctx(vpfe->pdev);
+ if (IS_ERR(vpfe->alloc_ctx)) {
+ vpfe_err(vpfe, "Failed to get the context\n");
+ err = PTR_ERR(vpfe->alloc_ctx);
+ goto probe_out;
+ }
+
+ q = &vpfe->buffer_queue;
+ q->type = V4L2_BUF_TYPE_VIDEO_CAPTURE;
+ q->io_modes = VB2_MMAP | VB2_DMABUF | VB2_READ;
+ q->drv_priv = vpfe;
+ q->ops = &vpfe_video_qops;
+ q->mem_ops = &vb2_dma_contig_memops;
+ q->buf_struct_size = sizeof(struct vpfe_cap_buffer);
+ q->timestamp_flags = V4L2_BUF_FLAG_TIMESTAMP_MONOTONIC;
+ q->lock = &vpfe->lock;
+ q->min_buffers_needed = 1;
+
+ err = vb2_queue_init(q);
+ if (err) {
+ vpfe_err(vpfe, "vb2_queue_init() failed\n");
+ vb2_dma_contig_cleanup_ctx(vpfe->alloc_ctx);
+ goto probe_out;
+ }
+
+ INIT_LIST_HEAD(&vpfe->dma_queue);
+
+ vdev = vpfe->video_dev;
+ strlcpy(vdev->name, VPFE_MODULE_NAME, sizeof(vdev->name));
+ vdev->release = video_device_release;
+ vdev->fops = &vpfe_fops;
+ vdev->ioctl_ops = &vpfe_ioctl_ops;
+ vdev->v4l2_dev = &vpfe->v4l2_dev;
+ vdev->vfl_dir = VFL_DIR_RX;
+ vdev->queue = q;
+ vdev->lock = &vpfe->lock;
+ video_set_drvdata(vdev, vpfe);
+ err = video_register_device(vpfe->video_dev, VFL_TYPE_GRABBER, -1);
+ if (err) {
+ vpfe_err(vpfe,
+ "Unable to register video device.\n");
+ goto probe_out;
+ }
+
+ return 0;
+
+probe_out:
+ v4l2_device_unregister(&vpfe->v4l2_dev);
+ return err;
+}
+
+static int vpfe_async_complete(struct v4l2_async_notifier *notifier)
+{
+ struct vpfe_device *vpfe = container_of(notifier->v4l2_dev,
+ struct vpfe_device, v4l2_dev);
+
+ return vpfe_probe_complete(vpfe);
+}
+
+static struct vpfe_config *
+vpfe_get_pdata(struct platform_device *pdev)
+{
+ struct device_node *endpoint = NULL, *rem = NULL;
+ struct v4l2_of_endpoint bus_cfg;
+ struct vpfe_subdev_info *sdinfo;
+ struct vpfe_config *pdata;
+ unsigned int flags;
+ unsigned int i;
+ int err;
+
+ dev_dbg(&pdev->dev, "vpfe_get_pdata\n");
+
+ if (!IS_ENABLED(CONFIG_OF) || !pdev->dev.of_node)
+ return pdev->dev.platform_data;
+
+ pdata = devm_kzalloc(&pdev->dev, sizeof(*pdata), GFP_KERNEL);
+ if (!pdata)
+ return NULL;
+
+ for (i = 0; ; i++) {
+ endpoint = of_graph_get_next_endpoint(pdev->dev.of_node,
+ endpoint);
+ if (!endpoint)
+ break;
+
+ sdinfo = &pdata->sub_devs[i];
+ sdinfo->grp_id = 0;
+
+ /* we only support camera */
+ sdinfo->inputs[0].index = i;
+ strcpy(sdinfo->inputs[0].name, "Camera");
+ sdinfo->inputs[0].type = V4L2_INPUT_TYPE_CAMERA;
+ sdinfo->inputs[0].std = V4L2_STD_ALL;
+ sdinfo->inputs[0].capabilities = V4L2_IN_CAP_STD;
+
+ sdinfo->can_route = 0;
+ sdinfo->routes = NULL;
+
+ of_property_read_u32(endpoint, "ti,am437x-vpfe-interface",
+ &sdinfo->vpfe_param.if_type);
+ if (sdinfo->vpfe_param.if_type < 0 ||
+ sdinfo->vpfe_param.if_type > 4) {
+ sdinfo->vpfe_param.if_type = VPFE_RAW_BAYER;
+ }
+
+ err = v4l2_of_parse_endpoint(endpoint, &bus_cfg);
+ if (err) {
+ dev_err(&pdev->dev, "Could not parse the endpoint\n");
+ goto done;
+ }
+
+ sdinfo->vpfe_param.bus_width = bus_cfg.bus.parallel.bus_width;
+
+ if (sdinfo->vpfe_param.bus_width < 8 ||
+ sdinfo->vpfe_param.bus_width > 16) {
+ dev_err(&pdev->dev, "Invalid bus width.\n");
+ goto done;
+ }
+
+ flags = bus_cfg.bus.parallel.flags;
+
+ if (flags & V4L2_MBUS_HSYNC_ACTIVE_HIGH)
+ sdinfo->vpfe_param.hdpol = 1;
+
+ if (flags & V4L2_MBUS_VSYNC_ACTIVE_HIGH)
+ sdinfo->vpfe_param.vdpol = 1;
+
+ rem = of_graph_get_remote_port_parent(endpoint);
+ if (!rem) {
+ dev_err(&pdev->dev, "Remote device at %s not found\n",
+ endpoint->full_name);
+ goto done;
+ }
+
+ strncpy(sdinfo->name, rem->name, sizeof(sdinfo->name));
+
+ pdata->asd[i] = devm_kzalloc(&pdev->dev,
+ sizeof(struct v4l2_async_subdev),
+ GFP_KERNEL);
+ pdata->asd[i]->match_type = V4L2_ASYNC_MATCH_OF;
+ pdata->asd[i]->match.of.node = rem;
+ of_node_put(endpoint);
+ of_node_put(rem);
+ }
+
+ of_node_put(endpoint);
+ return pdata;
+
+done:
+ of_node_put(endpoint);
+ of_node_put(rem);
+ return NULL;
+}
+
+/*
+ * vpfe_probe : This function creates device entries by register
+ * itself to the V4L2 driver and initializes fields of each
+ * device objects
+ */
+static int vpfe_probe(struct platform_device *pdev)
+{
+ struct vpfe_config *vpfe_cfg = vpfe_get_pdata(pdev);
+ struct vpfe_device *vpfe;
+ struct vpfe_ccdc *ccdc;
+ struct resource *res;
+ int ret;
+
+ if (!vpfe_cfg) {
+ dev_err(&pdev->dev, "No platform data\n");
+ return -EINVAL;
+ }
+
+ vpfe = devm_kzalloc(&pdev->dev, sizeof(*vpfe), GFP_KERNEL);
+ if (!vpfe)
+ return -ENOMEM;
+
+ vpfe->pdev = &pdev->dev;
+ vpfe->cfg = vpfe_cfg;
+ ccdc = &vpfe->ccdc;
+
+ res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+ ccdc->ccdc_cfg.base_addr = devm_ioremap_resource(&pdev->dev, res);
+ if (IS_ERR(ccdc->ccdc_cfg.base_addr))
+ return PTR_ERR(ccdc->ccdc_cfg.base_addr);
+
+ vpfe->irq = platform_get_irq(pdev, 0);
+ if (vpfe->irq <= 0) {
+ dev_err(&pdev->dev, "No IRQ resource\n");
+ return -ENODEV;
+ }
+
+ ret = devm_request_irq(vpfe->pdev, vpfe->irq, vpfe_isr, 0,
+ "vpfe_capture0", vpfe);
+ if (ret) {
+ dev_err(&pdev->dev, "Unable to request interrupt\n");
+ return -EINVAL;
+ }
+
+ vpfe->video_dev = video_device_alloc();
+ if (!vpfe->video_dev) {
+ dev_err(&pdev->dev, "Unable to allocate video device\n");
+ return -ENOMEM;
+ }
+
+ ret = v4l2_device_register(&pdev->dev, &vpfe->v4l2_dev);
+ if (ret) {
+ vpfe_err(vpfe,
+ "Unable to register v4l2 device.\n");
+ goto probe_out_video_release;
+ }
+
+ /* set the driver data in platform device */
+ platform_set_drvdata(pdev, vpfe);
+ /* Enabling module functional clock */
+ pm_runtime_enable(&pdev->dev);
+
+ /* for now just enable it here instead of waiting for the open */
+ pm_runtime_get_sync(&pdev->dev);
+
+ vpfe_ccdc_config_defaults(ccdc);
+
+ pm_runtime_put_sync(&pdev->dev);
+
+ vpfe->sd = devm_kzalloc(&pdev->dev, sizeof(struct v4l2_subdev *) *
+ ARRAY_SIZE(vpfe->cfg->asd), GFP_KERNEL);
+ if (!vpfe->sd) {
+ ret = -ENOMEM;
+ goto probe_out_v4l2_unregister;
+ }
+
+ vpfe->notifier.subdevs = vpfe->cfg->asd;
+ vpfe->notifier.num_subdevs = ARRAY_SIZE(vpfe->cfg->asd);
+ vpfe->notifier.bound = vpfe_async_bound;
+ vpfe->notifier.complete = vpfe_async_complete;
+ ret = v4l2_async_notifier_register(&vpfe->v4l2_dev,
+ &vpfe->notifier);
+ if (ret) {
+ vpfe_err(vpfe, "Error registering async notifier\n");
+ ret = -EINVAL;
+ goto probe_out_v4l2_unregister;
+ }
+
+ return 0;
+
+probe_out_v4l2_unregister:
+ v4l2_device_unregister(&vpfe->v4l2_dev);
+probe_out_video_release:
+ if (!video_is_registered(vpfe->video_dev))
+ video_device_release(vpfe->video_dev);
+ return ret;
+}
+
+/*
+ * vpfe_remove : It un-register device from V4L2 driver
+ */
+static int vpfe_remove(struct platform_device *pdev)
+{
+ struct vpfe_device *vpfe = platform_get_drvdata(pdev);
+
+ vpfe_dbg(2, vpfe, "vpfe_remove\n");
+
+ pm_runtime_disable(&pdev->dev);
+
+ v4l2_async_notifier_unregister(&vpfe->notifier);
+ v4l2_device_unregister(&vpfe->v4l2_dev);
+ video_unregister_device(vpfe->video_dev);
+
+ return 0;
+}
+
+#ifdef CONFIG_PM_SLEEP
+
+static void vpfe_save_context(struct vpfe_ccdc *ccdc)
+{
+ ccdc->ccdc_ctx[VPFE_PCR >> 2] = vpfe_reg_read(ccdc, VPFE_PCR);
+ ccdc->ccdc_ctx[VPFE_SYNMODE >> 2] = vpfe_reg_read(ccdc, VPFE_SYNMODE);
+ ccdc->ccdc_ctx[VPFE_SDOFST >> 2] = vpfe_reg_read(ccdc, VPFE_SDOFST);
+ ccdc->ccdc_ctx[VPFE_SDR_ADDR >> 2] = vpfe_reg_read(ccdc, VPFE_SDR_ADDR);
+ ccdc->ccdc_ctx[VPFE_CLAMP >> 2] = vpfe_reg_read(ccdc, VPFE_CLAMP);
+ ccdc->ccdc_ctx[VPFE_DCSUB >> 2] = vpfe_reg_read(ccdc, VPFE_DCSUB);
+ ccdc->ccdc_ctx[VPFE_COLPTN >> 2] = vpfe_reg_read(ccdc, VPFE_COLPTN);
+ ccdc->ccdc_ctx[VPFE_BLKCMP >> 2] = vpfe_reg_read(ccdc, VPFE_BLKCMP);
+ ccdc->ccdc_ctx[VPFE_VDINT >> 2] = vpfe_reg_read(ccdc, VPFE_VDINT);
+ ccdc->ccdc_ctx[VPFE_ALAW >> 2] = vpfe_reg_read(ccdc, VPFE_ALAW);
+ ccdc->ccdc_ctx[VPFE_REC656IF >> 2] = vpfe_reg_read(ccdc, VPFE_REC656IF);
+ ccdc->ccdc_ctx[VPFE_CCDCFG >> 2] = vpfe_reg_read(ccdc, VPFE_CCDCFG);
+ ccdc->ccdc_ctx[VPFE_CULLING >> 2] = vpfe_reg_read(ccdc, VPFE_CULLING);
+ ccdc->ccdc_ctx[VPFE_HD_VD_WID >> 2] = vpfe_reg_read(ccdc,
+ VPFE_HD_VD_WID);
+ ccdc->ccdc_ctx[VPFE_PIX_LINES >> 2] = vpfe_reg_read(ccdc,
+ VPFE_PIX_LINES);
+ ccdc->ccdc_ctx[VPFE_HORZ_INFO >> 2] = vpfe_reg_read(ccdc,
+ VPFE_HORZ_INFO);
+ ccdc->ccdc_ctx[VPFE_VERT_START >> 2] = vpfe_reg_read(ccdc,
+ VPFE_VERT_START);
+ ccdc->ccdc_ctx[VPFE_VERT_LINES >> 2] = vpfe_reg_read(ccdc,
+ VPFE_VERT_LINES);
+ ccdc->ccdc_ctx[VPFE_HSIZE_OFF >> 2] = vpfe_reg_read(ccdc,
+ VPFE_HSIZE_OFF);
+}
+
+static int vpfe_suspend(struct device *dev)
+{
+ struct platform_device *pdev = to_platform_device(dev);
+ struct vpfe_device *vpfe = platform_get_drvdata(pdev);
+ struct vpfe_ccdc *ccdc = &vpfe->ccdc;
+
+ /* if streaming has not started we don't care */
+ if (!vb2_start_streaming_called(&vpfe->buffer_queue))
+ return 0;
+
+ pm_runtime_get_sync(dev);
+ vpfe_config_enable(ccdc, 1);
+
+ /* Save VPFE context */
+ vpfe_save_context(ccdc);
+
+ /* Disable CCDC */
+ vpfe_pcr_enable(ccdc, 0);
+ vpfe_config_enable(ccdc, 0);
+
+ /* Disable both master and slave clock */
+ pm_runtime_put_sync(dev);
+
+ /* Select sleep pin state */
+ pinctrl_pm_select_sleep_state(dev);
+
+ return 0;
+}
+
+static void vpfe_restore_context(struct vpfe_ccdc *ccdc)
+{
+ vpfe_reg_write(ccdc, ccdc->ccdc_ctx[VPFE_SYNMODE >> 2], VPFE_SYNMODE);
+ vpfe_reg_write(ccdc, ccdc->ccdc_ctx[VPFE_CULLING >> 2], VPFE_CULLING);
+ vpfe_reg_write(ccdc, ccdc->ccdc_ctx[VPFE_SDOFST >> 2], VPFE_SDOFST);
+ vpfe_reg_write(ccdc, ccdc->ccdc_ctx[VPFE_SDR_ADDR >> 2], VPFE_SDR_ADDR);
+ vpfe_reg_write(ccdc, ccdc->ccdc_ctx[VPFE_CLAMP >> 2], VPFE_CLAMP);
+ vpfe_reg_write(ccdc, ccdc->ccdc_ctx[VPFE_DCSUB >> 2], VPFE_DCSUB);
+ vpfe_reg_write(ccdc, ccdc->ccdc_ctx[VPFE_COLPTN >> 2], VPFE_COLPTN);
+ vpfe_reg_write(ccdc, ccdc->ccdc_ctx[VPFE_BLKCMP >> 2], VPFE_BLKCMP);
+ vpfe_reg_write(ccdc, ccdc->ccdc_ctx[VPFE_VDINT >> 2], VPFE_VDINT);
+ vpfe_reg_write(ccdc, ccdc->ccdc_ctx[VPFE_ALAW >> 2], VPFE_ALAW);
+ vpfe_reg_write(ccdc, ccdc->ccdc_ctx[VPFE_REC656IF >> 2], VPFE_REC656IF);
+ vpfe_reg_write(ccdc, ccdc->ccdc_ctx[VPFE_CCDCFG >> 2], VPFE_CCDCFG);
+ vpfe_reg_write(ccdc, ccdc->ccdc_ctx[VPFE_PCR >> 2], VPFE_PCR);
+ vpfe_reg_write(ccdc, ccdc->ccdc_ctx[VPFE_HD_VD_WID >> 2],
+ VPFE_HD_VD_WID);
+ vpfe_reg_write(ccdc, ccdc->ccdc_ctx[VPFE_PIX_LINES >> 2],
+ VPFE_PIX_LINES);
+ vpfe_reg_write(ccdc, ccdc->ccdc_ctx[VPFE_HORZ_INFO >> 2],
+ VPFE_HORZ_INFO);
+ vpfe_reg_write(ccdc, ccdc->ccdc_ctx[VPFE_VERT_START >> 2],
+ VPFE_VERT_START);
+ vpfe_reg_write(ccdc, ccdc->ccdc_ctx[VPFE_VERT_LINES >> 2],
+ VPFE_VERT_LINES);
+ vpfe_reg_write(ccdc, ccdc->ccdc_ctx[VPFE_HSIZE_OFF >> 2],
+ VPFE_HSIZE_OFF);
+}
+
+static int vpfe_resume(struct device *dev)
+{
+ struct platform_device *pdev = to_platform_device(dev);
+ struct vpfe_device *vpfe = platform_get_drvdata(pdev);
+ struct vpfe_ccdc *ccdc = &vpfe->ccdc;
+
+ /* if streaming has not started we don't care */
+ if (!vb2_start_streaming_called(&vpfe->buffer_queue))
+ return 0;
+
+ /* Enable both master and slave clock */
+ pm_runtime_get_sync(dev);
+ vpfe_config_enable(ccdc, 1);
+
+ /* Restore VPFE context */
+ vpfe_restore_context(ccdc);
+
+ vpfe_config_enable(ccdc, 0);
+ pm_runtime_put_sync(dev);
+
+ /* Select default pin state */
+ pinctrl_pm_select_default_state(dev);
+
+ return 0;
+}
+
+#endif
+
+static SIMPLE_DEV_PM_OPS(vpfe_pm_ops, vpfe_suspend, vpfe_resume);
+
+static const struct of_device_id vpfe_of_match[] = {
+ { .compatible = "ti,am437x-vpfe", },
+ { /* sentinel */ },
+};
+MODULE_DEVICE_TABLE(of, vpfe_of_match);
+
+static struct platform_driver vpfe_driver = {
+ .probe = vpfe_probe,
+ .remove = vpfe_remove,
+ .driver = {
+ .name = VPFE_MODULE_NAME,
+ .owner = THIS_MODULE,
+ .pm = &vpfe_pm_ops,
+ .of_match_table = of_match_ptr(vpfe_of_match),
+ },
+};
+
+module_platform_driver(vpfe_driver);
+
+MODULE_AUTHOR("Texas Instruments");
+MODULE_DESCRIPTION("TI AM437x VPFE driver");
+MODULE_LICENSE("GPL");
+MODULE_VERSION(VPFE_VERSION);
diff --git a/drivers/media/platform/am437x/am437x-vpfe.h b/drivers/media/platform/am437x/am437x-vpfe.h
new file mode 100644
index 0000000..0f55735
--- /dev/null
+++ b/drivers/media/platform/am437x/am437x-vpfe.h
@@ -0,0 +1,283 @@
+/*
+ * Copyright (C) 2013 - 2014 Texas Instruments, Inc.
+ *
+ * Benoit Parrot <bparrot@ti.com>
+ * Lad, Prabhakar <prabhakar.csengg@gmail.com>
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#ifndef AM437X_VPFE_H
+#define AM437X_VPFE_H
+
+#include <linux/am437x-vpfe.h>
+#include <linux/clk.h>
+#include <linux/device.h>
+#include <linux/io.h>
+#include <linux/i2c.h>
+#include <linux/videodev2.h>
+
+#include <media/v4l2-dev.h>
+#include <media/v4l2-device.h>
+#include <media/v4l2-ioctl.h>
+#include <media/videobuf2-dma-contig.h>
+
+#include "am437x-vpfe_regs.h"
+
+enum vpfe_pin_pol {
+ VPFE_PINPOL_POSITIVE = 0,
+ VPFE_PINPOL_NEGATIVE,
+};
+
+enum vpfe_hw_if_type {
+ /* Raw Bayer */
+ VPFE_RAW_BAYER = 0,
+ /* BT656 - 8 bit */
+ VPFE_BT656,
+ /* BT656 - 10 bit */
+ VPFE_BT656_10BIT,
+ /* YCbCr - 8 bit with external sync */
+ VPFE_YCBCR_SYNC_8,
+ /* YCbCr - 16 bit with external sync */
+ VPFE_YCBCR_SYNC_16,
+};
+
+/* interface description */
+struct vpfe_hw_if_param {
+ enum vpfe_hw_if_type if_type;
+ enum vpfe_pin_pol hdpol;
+ enum vpfe_pin_pol vdpol;
+ unsigned int bus_width;
+};
+
+#define VPFE_MAX_SUBDEV 1
+#define VPFE_MAX_INPUTS 1
+
+struct vpfe_pixel_format {
+ struct v4l2_fmtdesc fmtdesc;
+ /* bytes per pixel */
+ int bpp;
+};
+
+struct vpfe_std_info {
+ int active_pixels;
+ int active_lines;
+ /* current frame format */
+ int frame_format;
+};
+
+struct vpfe_route {
+ u32 input;
+ u32 output;
+};
+
+struct vpfe_subdev_info {
+ char name[32];
+ /* Sub device group id */
+ int grp_id;
+ /* inputs available at the sub device */
+ struct v4l2_input inputs[VPFE_MAX_INPUTS];
+ /* Sub dev routing information for each input */
+ struct vpfe_route *routes;
+ /* check if sub dev supports routing */
+ int can_route;
+ /* ccdc bus/interface configuration */
+ struct vpfe_hw_if_param vpfe_param;
+ struct v4l2_subdev *sd;
+};
+
+struct vpfe_config {
+ /* information about each subdev */
+ struct vpfe_subdev_info sub_devs[VPFE_MAX_SUBDEV];
+ /* Flat array, arranged in groups */
+ struct v4l2_async_subdev *asd[VPFE_MAX_SUBDEV];
+};
+
+struct vpfe_cap_buffer {
+ struct vb2_buffer vb;
+ struct list_head list;
+};
+
+enum ccdc_pixfmt {
+ CCDC_PIXFMT_RAW = 0,
+ CCDC_PIXFMT_YCBCR_16BIT,
+ CCDC_PIXFMT_YCBCR_8BIT,
+};
+
+enum ccdc_frmfmt {
+ CCDC_FRMFMT_PROGRESSIVE = 0,
+ CCDC_FRMFMT_INTERLACED,
+};
+
+/* PIXEL ORDER IN MEMORY from LSB to MSB */
+/* only applicable for 8-bit input mode */
+enum ccdc_pixorder {
+ CCDC_PIXORDER_YCBYCR,
+ CCDC_PIXORDER_CBYCRY,
+};
+
+enum ccdc_buftype {
+ CCDC_BUFTYPE_FLD_INTERLEAVED,
+ CCDC_BUFTYPE_FLD_SEPARATED
+};
+
+
+/* returns the highest bit used for the gamma */
+static inline u8 ccdc_gamma_width_max_bit(enum vpfe_ccdc_gamma_width width)
+{
+ return 15 - width;
+}
+
+/* returns the highest bit used for this data size */
+static inline u8 ccdc_data_size_max_bit(enum vpfe_ccdc_data_size sz)
+{
+ return sz == VPFE_CCDC_DATA_8BITS ? 7 : 15 - sz;
+}
+
+/* Structure for CCDC configuration parameters for raw capture mode */
+struct ccdc_params_raw {
+ /* pixel format */
+ enum ccdc_pixfmt pix_fmt;
+ /* progressive or interlaced frame */
+ enum ccdc_frmfmt frm_fmt;
+ struct v4l2_rect win;
+ /* Current Format Bytes Per Pixels */
+ unsigned int bytesperpixel;
+ /* Current Format Bytes per Lines
+ * (Aligned to 32 bytes) used for HORZ_INFO
+ */
+ unsigned int bytesperline;
+ /* field id polarity */
+ enum vpfe_pin_pol fid_pol;
+ /* vertical sync polarity */
+ enum vpfe_pin_pol vd_pol;
+ /* horizontal sync polarity */
+ enum vpfe_pin_pol hd_pol;
+ /* interleaved or separated fields */
+ enum ccdc_buftype buf_type;
+ /*
+ * enable to store the image in inverse
+ * order in memory(bottom to top)
+ */
+ unsigned char image_invert_enable;
+ /* configurable parameters */
+ struct vpfe_ccdc_config_params_raw config_params;
+};
+
+struct ccdc_params_ycbcr {
+ /* pixel format */
+ enum ccdc_pixfmt pix_fmt;
+ /* progressive or interlaced frame */
+ enum ccdc_frmfmt frm_fmt;
+ struct v4l2_rect win;
+ /* Current Format Bytes Per Pixels */
+ unsigned int bytesperpixel;
+ /* Current Format Bytes per Lines
+ * (Aligned to 32 bytes) used for HORZ_INFO
+ */
+ unsigned int bytesperline;
+ /* field id polarity */
+ enum vpfe_pin_pol fid_pol;
+ /* vertical sync polarity */
+ enum vpfe_pin_pol vd_pol;
+ /* horizontal sync polarity */
+ enum vpfe_pin_pol hd_pol;
+ /* enable BT.656 embedded sync mode */
+ int bt656_enable;
+ /* cb:y:cr:y or y:cb:y:cr in memory */
+ enum ccdc_pixorder pix_order;
+ /* interleaved or separated fields */
+ enum ccdc_buftype buf_type;
+};
+
+/*
+ * CCDC operational configuration
+ */
+struct ccdc_config {
+ /* CCDC interface type */
+ enum vpfe_hw_if_type if_type;
+ /* Raw Bayer configuration */
+ struct ccdc_params_raw bayer;
+ /* YCbCr configuration */
+ struct ccdc_params_ycbcr ycbcr;
+ /* ccdc base address */
+ void __iomem *base_addr;
+};
+
+struct vpfe_ccdc {
+ struct ccdc_config ccdc_cfg;
+ u32 ccdc_ctx[VPFE_REG_END / sizeof(u32)];
+};
+
+struct vpfe_device {
+ /* V4l2 specific parameters */
+ /* Identifies video device for this channel */
+ struct video_device *video_dev;
+ /* sub devices */
+ struct v4l2_subdev **sd;
+ /* vpfe cfg */
+ struct vpfe_config *cfg;
+ /* V4l2 device */
+ struct v4l2_device v4l2_dev;
+ /* parent device */
+ struct device *pdev;
+ /* subdevice async Notifier */
+ struct v4l2_async_notifier notifier;
+ /* Indicates id of the field which is being displayed */
+ unsigned field;
+ unsigned sequence;
+ /* current interface type */
+ struct vpfe_hw_if_param vpfe_if_params;
+ /* ptr to currently selected sub device */
+ struct vpfe_subdev_info *current_subdev;
+ /* current input at the sub device */
+ int current_input;
+ /* Keeps track of the information about the standard */
+ struct vpfe_std_info std_info;
+ /* std index into std table */
+ int std_index;
+ /* IRQs used when CCDC output to SDRAM */
+ unsigned int irq;
+ /* Pointer pointing to current v4l2_buffer */
+ struct vpfe_cap_buffer *cur_frm;
+ /* Pointer pointing to next v4l2_buffer */
+ struct vpfe_cap_buffer *next_frm;
+ /* Used to store pixel format */
+ struct v4l2_format fmt;
+ /* Used to store current bytes per pixel based on current format */
+ unsigned int bpp;
+ /*
+ * used when IMP is chained to store the crop window which
+ * is different from the image window
+ */
+ struct v4l2_rect crop;
+ /* Buffer queue used in video-buf */
+ struct vb2_queue buffer_queue;
+ /* Allocator-specific contexts for each plane */
+ struct vb2_alloc_ctx *alloc_ctx;
+ /* Queue of filled frames */
+ struct list_head dma_queue;
+ /* IRQ lock for DMA queue */
+ spinlock_t dma_queue_lock;
+ /* lock used to access this structure */
+ struct mutex lock;
+ /*
+ * offset where second field starts from the starting of the
+ * buffer for field separated YCbCr formats
+ */
+ u32 field_off;
+ struct vpfe_ccdc ccdc;
+};
+
+#endif /* AM437X_VPFE_H */
diff --git a/drivers/media/platform/am437x/am437x-vpfe_regs.h b/drivers/media/platform/am437x/am437x-vpfe_regs.h
new file mode 100644
index 0000000..4a0ed29
--- /dev/null
+++ b/drivers/media/platform/am437x/am437x-vpfe_regs.h
@@ -0,0 +1,140 @@
+/*
+ * TI AM437x Image Sensor Interface Registers
+ *
+ * Copyright (C) 2013 - 2014 Texas Instruments, Inc.
+ *
+ * Benoit Parrot <bparrot@ti.com>
+ * Lad, Prabhakar <prabhakar.csengg@gmail.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ */
+
+#ifndef AM437X_VPFE_REGS_H
+#define AM437X_VPFE_REGS_H
+
+/* VPFE module register offset */
+#define VPFE_REVISION 0x0
+#define VPFE_PCR 0x4
+#define VPFE_SYNMODE 0x8
+#define VPFE_HD_VD_WID 0xc
+#define VPFE_PIX_LINES 0x10
+#define VPFE_HORZ_INFO 0x14
+#define VPFE_VERT_START 0x18
+#define VPFE_VERT_LINES 0x1c
+#define VPFE_CULLING 0x20
+#define VPFE_HSIZE_OFF 0x24
+#define VPFE_SDOFST 0x28
+#define VPFE_SDR_ADDR 0x2c
+#define VPFE_CLAMP 0x30
+#define VPFE_DCSUB 0x34
+#define VPFE_COLPTN 0x38
+#define VPFE_BLKCMP 0x3c
+#define VPFE_VDINT 0x48
+#define VPFE_ALAW 0x4c
+#define VPFE_REC656IF 0x50
+#define VPFE_CCDCFG 0x54
+#define VPFE_DMA_CNTL 0x98
+#define VPFE_SYSCONFIG 0x104
+#define VPFE_CONFIG 0x108
+#define VPFE_IRQ_EOI 0x110
+#define VPFE_IRQ_STS_RAW 0x114
+#define VPFE_IRQ_STS 0x118
+#define VPFE_IRQ_EN_SET 0x11c
+#define VPFE_IRQ_EN_CLR 0x120
+#define VPFE_REG_END 0x124
+
+/* Define bit fields within selected registers */
+#define VPFE_FID_POL_MASK 1
+#define VPFE_FID_POL_SHIFT 4
+#define VPFE_HD_POL_MASK 1
+#define VPFE_HD_POL_SHIFT 3
+#define VPFE_VD_POL_MASK 1
+#define VPFE_VD_POL_SHIFT 2
+#define VPFE_HSIZE_OFF_MASK 0xffffffe0
+#define VPFE_32BYTE_ALIGN_VAL 31
+#define VPFE_FRM_FMT_MASK 0x1
+#define VPFE_FRM_FMT_SHIFT 7
+#define VPFE_DATA_SZ_MASK 7
+#define VPFE_DATA_SZ_SHIFT 8
+#define VPFE_PIX_FMT_MASK 3
+#define VPFE_PIX_FMT_SHIFT 12
+#define VPFE_VP2SDR_DISABLE 0xfffbffff
+#define VPFE_WEN_ENABLE (1 << 17)
+#define VPFE_SDR2RSZ_DISABLE 0xfff7ffff
+#define VPFE_VDHDEN_ENABLE (1 << 16)
+#define VPFE_LPF_ENABLE (1 << 14)
+#define VPFE_ALAW_ENABLE (1 << 3)
+#define VPFE_ALAW_GAMMA_WD_MASK 7
+#define VPFE_BLK_CLAMP_ENABLE (1 << 31)
+#define VPFE_BLK_SGAIN_MASK 0x1f
+#define VPFE_BLK_ST_PXL_MASK 0x7fff
+#define VPFE_BLK_ST_PXL_SHIFT 10
+#define VPFE_BLK_SAMPLE_LN_MASK 7
+#define VPFE_BLK_SAMPLE_LN_SHIFT 28
+#define VPFE_BLK_SAMPLE_LINE_MASK 7
+#define VPFE_BLK_SAMPLE_LINE_SHIFT 25
+#define VPFE_BLK_DC_SUB_MASK 0x03fff
+#define VPFE_BLK_COMP_MASK 0xff
+#define VPFE_BLK_COMP_GB_COMP_SHIFT 8
+#define VPFE_BLK_COMP_GR_COMP_SHIFT 16
+#define VPFE_BLK_COMP_R_COMP_SHIFT 24
+#define VPFE_LATCH_ON_VSYNC_DISABLE (1 << 15)
+#define VPFE_DATA_PACK_ENABLE (1 << 11)
+#define VPFE_HORZ_INFO_SPH_SHIFT 16
+#define VPFE_VERT_START_SLV0_SHIFT 16
+#define VPFE_VDINT_VDINT0_SHIFT 16
+#define VPFE_VDINT_VDINT1_MASK 0xffff
+#define VPFE_PPC_RAW 1
+#define VPFE_DCSUB_DEFAULT_VAL 0
+#define VPFE_CLAMP_DEFAULT_VAL 0
+#define VPFE_COLPTN_VAL 0xbb11bb11
+#define VPFE_TWO_BYTES_PER_PIXEL 2
+#define VPFE_INTERLACED_IMAGE_INVERT 0x4b6d
+#define VPFE_INTERLACED_NO_IMAGE_INVERT 0x0249
+#define VPFE_PROGRESSIVE_IMAGE_INVERT 0x4000
+#define VPFE_PROGRESSIVE_NO_IMAGE_INVERT 0
+#define VPFE_INTERLACED_HEIGHT_SHIFT 1
+#define VPFE_SYN_MODE_INPMOD_SHIFT 12
+#define VPFE_SYN_MODE_INPMOD_MASK 3
+#define VPFE_SYN_MODE_8BITS (7 << 8)
+#define VPFE_SYN_MODE_10BITS (6 << 8)
+#define VPFE_SYN_MODE_11BITS (5 << 8)
+#define VPFE_SYN_MODE_12BITS (4 << 8)
+#define VPFE_SYN_MODE_13BITS (3 << 8)
+#define VPFE_SYN_MODE_14BITS (2 << 8)
+#define VPFE_SYN_MODE_15BITS (1 << 8)
+#define VPFE_SYN_MODE_16BITS (0 << 8)
+#define VPFE_SYN_FLDMODE_MASK 1
+#define VPFE_SYN_FLDMODE_SHIFT 7
+#define VPFE_REC656IF_BT656_EN 3
+#define VPFE_SYN_MODE_VD_POL_NEGATIVE (1 << 2)
+#define VPFE_CCDCFG_Y8POS_SHIFT 11
+#define VPFE_CCDCFG_BW656_10BIT (1 << 5)
+#define VPFE_SDOFST_FIELD_INTERLEAVED 0x249
+#define VPFE_NO_CULLING 0xffff00ff
+#define VPFE_VDINT0 (1 << 0)
+#define VPFE_VDINT1 (1 << 1)
+#define VPFE_VDINT2 (1 << 2)
+#define VPFE_DMA_CNTL_OVERFLOW (1 << 31)
+
+#define VPFE_CONFIG_PCLK_INV_SHIFT 0
+#define VPFE_CONFIG_PCLK_INV_MASK 1
+#define VPFE_CONFIG_PCLK_INV_NOT_INV 0
+#define VPFE_CONFIG_PCLK_INV_INV 1
+#define VPFE_CONFIG_EN_SHIFT 1
+#define VPFE_CONFIG_EN_MASK 2
+#define VPFE_CONFIG_EN_DISABLE 0
+#define VPFE_CONFIG_EN_ENABLE 1
+#define VPFE_CONFIG_ST_SHIFT 2
+#define VPFE_CONFIG_ST_MASK 4
+#define VPFE_CONFIG_ST_OCP_ACTIVE 0
+#define VPFE_CONFIG_ST_OCP_STANDBY 1
+
+#endif /* AM437X_VPFE_REGS_H */
diff --git a/include/uapi/linux/Kbuild b/include/uapi/linux/Kbuild
index ed39ac8..65495f0 100644
--- a/include/uapi/linux/Kbuild
+++ b/include/uapi/linux/Kbuild
@@ -34,6 +34,7 @@ header-y += adfs_fs.h
header-y += affs_hardblocks.h
header-y += agpgart.h
header-y += aio_abi.h
+header-y += am437x-vpfe.h
header-y += apm_bios.h
header-y += arcfb.h
header-y += atalk.h
diff --git a/include/uapi/linux/am437x-vpfe.h b/include/uapi/linux/am437x-vpfe.h
new file mode 100644
index 0000000..9b03033f
--- /dev/null
+++ b/include/uapi/linux/am437x-vpfe.h
@@ -0,0 +1,122 @@
+/*
+ * Copyright (C) 2013 - 2014 Texas Instruments, Inc.
+ *
+ * Benoit Parrot <bparrot@ti.com>
+ * Lad, Prabhakar <prabhakar.csengg@gmail.com>
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#ifndef AM437X_VPFE_USER_H
+#define AM437X_VPFE_USER_H
+
+enum vpfe_ccdc_data_size {
+ VPFE_CCDC_DATA_16BITS = 0,
+ VPFE_CCDC_DATA_15BITS,
+ VPFE_CCDC_DATA_14BITS,
+ VPFE_CCDC_DATA_13BITS,
+ VPFE_CCDC_DATA_12BITS,
+ VPFE_CCDC_DATA_11BITS,
+ VPFE_CCDC_DATA_10BITS,
+ VPFE_CCDC_DATA_8BITS,
+};
+
+/* enum for No of pixel per line to be avg. in Black Clamping*/
+enum vpfe_ccdc_sample_length {
+ VPFE_CCDC_SAMPLE_1PIXELS = 0,
+ VPFE_CCDC_SAMPLE_2PIXELS,
+ VPFE_CCDC_SAMPLE_4PIXELS,
+ VPFE_CCDC_SAMPLE_8PIXELS,
+ VPFE_CCDC_SAMPLE_16PIXELS,
+};
+
+/* enum for No of lines in Black Clamping */
+enum vpfe_ccdc_sample_line {
+ VPFE_CCDC_SAMPLE_1LINES = 0,
+ VPFE_CCDC_SAMPLE_2LINES,
+ VPFE_CCDC_SAMPLE_4LINES,
+ VPFE_CCDC_SAMPLE_8LINES,
+ VPFE_CCDC_SAMPLE_16LINES,
+};
+
+/* enum for Alaw gamma width */
+enum vpfe_ccdc_gamma_width {
+ VPFE_CCDC_GAMMA_BITS_15_6 = 0, /* use bits 15-6 for gamma */
+ VPFE_CCDC_GAMMA_BITS_14_5,
+ VPFE_CCDC_GAMMA_BITS_13_4,
+ VPFE_CCDC_GAMMA_BITS_12_3,
+ VPFE_CCDC_GAMMA_BITS_11_2,
+ VPFE_CCDC_GAMMA_BITS_10_1,
+ VPFE_CCDC_GAMMA_BITS_09_0, /* use bits 9-0 for gamma */
+};
+
+/* structure for ALaw */
+struct vpfe_ccdc_a_law {
+ /* Enable/disable A-Law */
+ unsigned char enable;
+ /* Gamma Width Input */
+ enum vpfe_ccdc_gamma_width gamma_wd;
+};
+
+/* structure for Black Clamping */
+struct vpfe_ccdc_black_clamp {
+ unsigned char enable;
+ /* only if bClampEnable is TRUE */
+ enum vpfe_ccdc_sample_length sample_pixel;
+ /* only if bClampEnable is TRUE */
+ enum vpfe_ccdc_sample_line sample_ln;
+ /* only if bClampEnable is TRUE */
+ unsigned short start_pixel;
+ /* only if bClampEnable is TRUE */
+ unsigned short sgain;
+ /* only if bClampEnable is FALSE */
+ unsigned short dc_sub;
+};
+
+/* structure for Black Level Compensation */
+struct vpfe_ccdc_black_compensation {
+ /* Constant value to subtract from Red component */
+ char r;
+ /* Constant value to subtract from Gr component */
+ char gr;
+ /* Constant value to subtract from Blue component */
+ char b;
+ /* Constant value to subtract from Gb component */
+ char gb;
+};
+
+/* Structure for CCDC configuration parameters for raw capture mode passed
+ * by application
+ */
+struct vpfe_ccdc_config_params_raw {
+ /* data size value from 8 to 16 bits */
+ enum vpfe_ccdc_data_size data_sz;
+ /* Structure for Optional A-Law */
+ struct vpfe_ccdc_a_law alaw;
+ /* Structure for Optical Black Clamp */
+ struct vpfe_ccdc_black_clamp blk_clamp;
+ /* Structure for Black Compensation */
+ struct vpfe_ccdc_black_compensation blk_comp;
+};
+
+/*
+ * Private IOCTL
+ * VIDIOC_AM437X_CCDC_CFG - Set CCDC configuration for raw capture
+ * This is an experimental ioctl that will change in future kernels. So use
+ * this ioctl with care !
+ **/
+#define VIDIOC_AM437X_CCDC_CFG \
+ _IOW('V', BASE_VIDIOC_PRIVATE + 1, void *)
+
+#endif /* AM437X_VPFE_USER_H */
--
1.9.1
^ permalink raw reply related
* Re: [PATCH v4 2/5] crypto: AF_ALG: add AEAD support
From: Stephan Mueller @ 2014-12-06 20:08 UTC (permalink / raw)
To: Herbert Xu
Cc: Daniel Borkmann, 'Quentin Gouchet', 'LKML',
linux-crypto, linux-api
In-Reply-To: <45756810.oFuGYtKPb9@tachyon.chronox.de>
Am Freitag, 5. Dezember 2014, 22:51:51 schrieb Stephan Mueller:
Hi Herbert,
> > > +static struct proto_ops algif_aead_ops = {
> > > + .family = PF_ALG,
> > > +
> > > + .connect = sock_no_connect,
> > > + .socketpair = sock_no_socketpair,
> > > + .getname = sock_no_getname,
> > > + .ioctl = sock_no_ioctl,
> > > + .listen = sock_no_listen,
> > > + .shutdown = sock_no_shutdown,
> > > + .getsockopt = sock_no_getsockopt,
> > > + .mmap = sock_no_mmap,
> > > + .bind = sock_no_bind,
> > > + .accept = sock_no_accept,
> > > +
> > > + .release = af_alg_release,
> > > + .sendmsg = aead_sendmsg,
> > > + .sendpage = aead_sendpage,
> > > + .recvmsg = aead_recvmsg,
> > > + .poll = aead_poll,
> > > + .setsockopt = aead_setsockopt,
> >
> > No it should go into the parent setsockopt. Perhaps add a setsockopt
> > to af_alg_type in order to keep this out of the generic code.
>
> I was thinking about that for quite a while. My thought for the current
> approach was that the actual cipher operation happens in the child FD (i.e.
> after accept). AAD is delivered to that FD. Therefore, I thought that the
> size of the AAD can be specific to that operational FD.
>
> If we move it to the parent setsockopt, all child FDs have the same AAD
> size. If you think that this is the right course of action, I can surely
> implement that.
>
> Would you please be so kind and help me understand when some operations are
> intended for the parent FD and when for the child FD?
While implementing that request, I thought about setting the auth size as part
of the msg control in sendmsg instead of setsockopt. This would save us a
system call and thus CPU cycles.
Thus may I propose that instead of using setsockopt, ALG_SET_AEAD_AUTHSIZE
should be added to af_alg_cmsg_send the same way as ALG_SET_AEAD_ASSOCLEN.
Along that line, wouldn't it make sense to get rid of the setsockopt in
general and move even setting the key into the msg control and thus into
af_alg_cmsg_send? I understand that this is a change to the user space API.
Thanks
--
Ciao
Stephan
^ permalink raw reply
* Re: [PATCH v4 2/5] crypto: AF_ALG: add AEAD support
From: Herbert Xu @ 2014-12-06 23:27 UTC (permalink / raw)
To: Stephan Mueller
Cc: Daniel Borkmann, 'Quentin Gouchet', 'LKML',
linux-crypto-u79uwXL29TY76Z2rM5mHXA,
linux-api-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <1677221.HN3xC7fa4q-PJstQz4BMNNP20K/wil9xYQuADTiUCJX@public.gmane.org>
On Sat, Dec 06, 2014 at 09:08:54PM +0100, Stephan Mueller wrote:
>
> While implementing that request, I thought about setting the auth size as part
> of the msg control in sendmsg instead of setsockopt. This would save us a
> system call and thus CPU cycles.
No, this is a tfm property and therefore you can't set them on the
child sockets sicne two child sockets would share the same tfm.
Cheers,
--
Email: Herbert Xu <herbert-lOAM2aK0SrRLBo1qDEOMRrpzq4S04n8Q@public.gmane.org>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
^ permalink raw reply
* Re: [PATCH v4 2/5] crypto: AF_ALG: add AEAD support
From: Herbert Xu @ 2014-12-06 23:28 UTC (permalink / raw)
To: Stephan Mueller
Cc: Daniel Borkmann, 'Quentin Gouchet', 'LKML',
linux-crypto-u79uwXL29TY76Z2rM5mHXA,
linux-api-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <45756810.oFuGYtKPb9-PJstQz4BMNNP20K/wil9xYQuADTiUCJX@public.gmane.org>
On Fri, Dec 05, 2014 at 10:51:51PM +0100, Stephan Mueller wrote:
>
> Would you please be so kind and help me understand when some operations are
> intended for the parent FD and when for the child FD?
If it changes the tfm it goes in the parent FD otherwise it's the
child FD.
Cheers,
--
Email: Herbert Xu <herbert-lOAM2aK0SrRLBo1qDEOMRrpzq4S04n8Q@public.gmane.org>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
^ permalink raw reply
* Re: [PATCH v4 2/5] crypto: AF_ALG: add AEAD support
From: Stephan Mueller @ 2014-12-07 12:56 UTC (permalink / raw)
To: Herbert Xu
Cc: Daniel Borkmann, 'Quentin Gouchet', 'LKML',
linux-crypto-u79uwXL29TY76Z2rM5mHXA,
linux-api-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <20141205154606.GA30180-lOAM2aK0SrRLBo1qDEOMRrpzq4S04n8Q@public.gmane.org>
Am Freitag, 5. Dezember 2014, 23:46:06 schrieb Herbert Xu:
Hi Herbert,
> > +static struct proto_ops algif_aead_ops = {
> > + .family = PF_ALG,
> > +
> > + .connect = sock_no_connect,
> > + .socketpair = sock_no_socketpair,
> > + .getname = sock_no_getname,
> > + .ioctl = sock_no_ioctl,
> > + .listen = sock_no_listen,
> > + .shutdown = sock_no_shutdown,
> > + .getsockopt = sock_no_getsockopt,
> > + .mmap = sock_no_mmap,
> > + .bind = sock_no_bind,
> > + .accept = sock_no_accept,
> > +
> > + .release = af_alg_release,
> > + .sendmsg = aead_sendmsg,
> > + .sendpage = aead_sendpage,
> > + .recvmsg = aead_recvmsg,
> > + .poll = aead_poll,
> > + .setsockopt = aead_setsockopt,
>
> No it should go into the parent setsockopt. Perhaps add a setsockopt
> to af_alg_type in order to keep this out of the generic code.
What about adding a setauthsize to af_alg_type that is called from the parent
setsockopt? I would like to keep all user space interfaces as close together
as possible. With a new setsockopt callback, I fear that user space interface
handling may be scattered around.
Currently I am simply thinking about the following:
partent setsockopt:
case ALG_SET_AEAD_AUTHSIZE:
if (sock->state == SS_CONNECTED)
goto unlock;
if (!type->setauthsize)
goto unlock;
err = type->setauthsize(ask->private, optlen);
af_alg_type:
int (*setauthsize)(void *private, unsigned int authsize);
--
Ciao
Stephan
^ permalink raw reply
* Re: [PATCH v6 4/7] fs/fuse: support compiling out splice
From: Pieter Smith @ 2014-12-07 13:00 UTC (permalink / raw)
To: Miklos Szeredi
Cc: Kernel Mailing List, open list:FUSE: FILESYSTEM...,
open list:ABI/API, Linux-Fsdevel
In-Reply-To: <CAJfpegu+NQQ_+af=6UXi2=fyTJK1CsQBZ2++Nysu4mume4UEOQ@mail.gmail.com>
On Fri, Dec 05, 2014 at 11:44:36AM +0100, Miklos Szeredi wrote:
> On Thu, Dec 4, 2014 at 6:50 PM, Pieter Smith <pieter@boesman.nl> wrote:
> > To implement splice support, fs/fuse makes use of nosteal_pipe_buf_ops. This
> > struct is exported by fs/splice. The goal of the larger patch set is to
> > completely compile out fs/splice, so uses of the exported struct need to be
> > compiled out along with fs/splice.
> >
> > This patch therefore compiles out splice support in fs/fuse when
> > CONFIG_SYSCALL_SPLICE is undefined.
> >
> > Signed-off-by: Pieter Smith <pieter@boesman.nl>
>
>
> In the future could you PLEASE PLEASE cut the fuse-devel Cc from the
> non-fuse specific patches (and I guess that goes for any other
> subsystem specific lists and persons as well)?
>
> Otherwise:
>
> Acked-by: Miklos Szeredi <mszeredi@suse.cz>
>
> Thanks,
> Miklos
I added your Acked-by and removed you and fuse-dev from postings on future
versions of this patch-set.
Thanks,
Pieter
^ permalink raw reply
* [PATCH] tools: testing: selftests: powerpc: pmu: ebb: ebb.c: Remove unused function
From: Rickard Strandqvist @ 2014-12-07 20:50 UTC (permalink / raw)
To: Shuah Khan, Benjamin Herrenschmidt
Cc: Rickard Strandqvist, Michael Ellerman,
linux-api-u79uwXL29TY76Z2rM5mHXA,
linux-kernel-u79uwXL29TY76Z2rM5mHXA
Remove the function event_bhrb_init() that is not used anywhere.
This was partially found by using a static code analysis program called cppcheck.
Signed-off-by: Rickard Strandqvist <rickard_strandqvist-IW2WV5XWFqGZkjO+N0TKoMugMpMbD5Xr@public.gmane.org>
---
tools/testing/selftests/powerpc/pmu/ebb/ebb.c | 5 -----
tools/testing/selftests/powerpc/pmu/ebb/ebb.h | 1 -
2 files changed, 6 deletions(-)
diff --git a/tools/testing/selftests/powerpc/pmu/ebb/ebb.c b/tools/testing/selftests/powerpc/pmu/ebb/ebb.c
index d7a72ce..75af212 100644
--- a/tools/testing/selftests/powerpc/pmu/ebb/ebb.c
+++ b/tools/testing/selftests/powerpc/pmu/ebb/ebb.c
@@ -324,11 +324,6 @@ void event_ebb_init(struct event *e)
e->attr.config |= (1ull << 63);
}
-void event_bhrb_init(struct event *e, unsigned ifm)
-{
- e->attr.config |= (1ull << 62) | ((u64)ifm << 60);
-}
-
void event_leader_ebb_init(struct event *e)
{
event_ebb_init(e);
diff --git a/tools/testing/selftests/powerpc/pmu/ebb/ebb.h b/tools/testing/selftests/powerpc/pmu/ebb/ebb.h
index e44eee5..a71596a 100644
--- a/tools/testing/selftests/powerpc/pmu/ebb/ebb.h
+++ b/tools/testing/selftests/powerpc/pmu/ebb/ebb.h
@@ -46,7 +46,6 @@ static inline void ebb_enable_pmc_counting(int pmc)
bool ebb_check_count(int pmc, u64 sample_period, int fudge);
void event_leader_ebb_init(struct event *e);
void event_ebb_init(struct event *e);
-void event_bhrb_init(struct event *e, unsigned ifm);
void setup_ebb_handler(void (*callee)(void));
void standard_ebb_callee(void);
int ebb_event_enable(struct event *e);
--
1.7.10.4
^ permalink raw reply related
* [PATCH v5 0/8] crypto: AF_ALG: add AEAD and RNG support
From: Stephan Mueller @ 2014-12-07 22:20 UTC (permalink / raw)
To: Herbert Xu
Cc: Daniel Borkmann, 'Quentin Gouchet', 'LKML',
linux-crypto-u79uwXL29TY76Z2rM5mHXA,
linux-api-u79uwXL29TY76Z2rM5mHXA
Hi,
This patch set adds AEAD and RNG support to the AF_ALG interface
exported by the kernel crypto API. By extending AF_ALG with AEAD and RNG
support, all cipher types the kernel crypto API allows access to are
now accessible from userspace.
Both, AEAD and RNG implementations are stand-alone and do not depend
other AF_ALG interfaces (like hash or skcipher).
The AEAD implementation uses the same approach as provided with
skcipher by offering the following interfaces:
* sendmsg and recvmsg interfaces allowing multiple
invocations supporting a threaded user space. To support
multi-threaded user space, kernel-side buffering
is implemented similarly to skcipher.
* splice / vmsplice interfaces allowing a zero-copy
invocation
The RNG interface only implements the recvmsg interface as
zero-copy is not applicable.
The new AEAD and RNG interfaces are fully tested with the test application
provided at [1]. That test application exercises all newly added user space
interfaces. The testing covers:
* use of the sendmsg/recvmsg interface
* use of the splice / vmsplice interface
* invocation of all AF_ALG types (aead, rng, skcipher, hash)
* using all types of operation (encryption, decryption, keyed MD,
MD, random numbers, AEAD decryption with positive and negative
authentication verification)
* stress testing by running all tests for 30 minutes in an
endless loop
* test execution on 64 bit and 32 bit
[1] http://www.chronox.de/libkcapi.html
Changes v2:
* rebase to current cryptodev-2.6 tree
* use memzero_explicit to zeroize AEAD associated data
* use sizeof for determining length of AEAD associated data
* update algif_rng.c covering all suggestions from Daniel Borkmann
<dborkman-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
* addition of patch 9: add digestsize interface for hashes
* addition of patch to update documentation covering the userspace interface
* change numbers of getsockopt options: separate them from sendmsg interface
definitions
Changes v3:
* remove getsockopt interface
* AEAD: associated data is set prepended to the plain/ciphertext
* AEAD: allowing arbitrary associated data lengths
* remove setkey patch as protection was already in the existing code
Changes v4:
* stand-alone implementation of AEAD
* testing of all interfaces offered by AEAD
* stress testing of AEAD and RNG
Changes v5:
* AEAD: add outer while(size) loop in aead_sendmsg to ensure all data is
copied into the kernel (reporter Herbert Xu)
* AEAD: aead_sendmsg bug fix: change size -= len; to size -= plen;
* AF_ALG / AEAD: add aead_setauthsize and associated extension to
struct af_alg_type as well as alg_setsockopt (reporter Herbert Xu)
* RNG: rng_recvmsg: use 128 byte stack variable for output of RNG instead
of ctx->result (reporter Herbert Xu)
* RNG / AF_ALG: allow user space to seed RNG via setsockopt
* RNG: rng_recvmsg bug fix: use genlen as result variable for
crypto_rng_get_bytes as previously no negative errors were obtained
* AF_ALG: alg_setop: zeroize buffer before free
Stephan Mueller (8):
crypto: AF_ALG: add user space interface for AEAD
crypto: AF_ALG: add setsockopt for auth tag size
crypto: AF_ALG: add AEAD support
crypto: AF_ALG: enable AEAD interface compilation
crypto: AF_ALG: add user space interface for RNG
crypto: AF_ALG: zeroize key / seed data
crypto: AF_ALG: add random number generator support
crypto: AF_ALG: enable RNG interface compilation
crypto/Kconfig | 18 ++
crypto/Makefile | 2 +
crypto/af_alg.c | 42 ++-
crypto/algif_aead.c | 650 ++++++++++++++++++++++++++++++++++++++++++++
crypto/algif_rng.c | 192 +++++++++++++
include/crypto/if_alg.h | 3 +
include/uapi/linux/if_alg.h | 3 +
7 files changed, 900 insertions(+), 10 deletions(-)
create mode 100644 crypto/algif_aead.c
create mode 100644 crypto/algif_rng.c
--
2.1.0
^ permalink raw reply
* [PATCH v5 1/8] crypto: AF_ALG: add user space interface for AEAD
From: Stephan Mueller @ 2014-12-07 22:21 UTC (permalink / raw)
To: Herbert Xu
Cc: Daniel Borkmann, 'Quentin Gouchet', 'LKML',
linux-crypto, linux-api
In-Reply-To: <56740432.V2v4gLHrzS@tachyon.chronox.de>
AEAD requires the caller to specify the following information separate
from the data stream. This information allows the AEAD interface handler
to identify the AAD, ciphertext/plaintext and the authentication tag:
* Associated authentication data of arbitrary length and
length
* Length of authentication tag for encryption
Signed-off-by: Stephan Mueller <smueller@chronox.de>
---
crypto/af_alg.c | 6 ++++++
include/crypto/if_alg.h | 1 +
include/uapi/linux/if_alg.h | 2 ++
3 files changed, 9 insertions(+)
diff --git a/crypto/af_alg.c b/crypto/af_alg.c
index 6a3ad80..68ff113 100644
--- a/crypto/af_alg.c
+++ b/crypto/af_alg.c
@@ -421,6 +421,12 @@ int af_alg_cmsg_send(struct msghdr *msg, struct af_alg_control *con)
con->op = *(u32 *)CMSG_DATA(cmsg);
break;
+ case ALG_SET_AEAD_ASSOCLEN:
+ if (cmsg->cmsg_len < CMSG_LEN(sizeof(u32)))
+ return -EINVAL;
+ con->aead_assoclen = *(u32 *)CMSG_DATA(cmsg);
+ break;
+
default:
return -EINVAL;
}
diff --git a/include/crypto/if_alg.h b/include/crypto/if_alg.h
index d61c111..cd62bf4 100644
--- a/include/crypto/if_alg.h
+++ b/include/crypto/if_alg.h
@@ -42,6 +42,7 @@ struct af_alg_completion {
struct af_alg_control {
struct af_alg_iv *iv;
int op;
+ unsigned int aead_assoclen;
};
struct af_alg_type {
diff --git a/include/uapi/linux/if_alg.h b/include/uapi/linux/if_alg.h
index 0f9acce..f2acd2f 100644
--- a/include/uapi/linux/if_alg.h
+++ b/include/uapi/linux/if_alg.h
@@ -32,6 +32,8 @@ struct af_alg_iv {
#define ALG_SET_KEY 1
#define ALG_SET_IV 2
#define ALG_SET_OP 3
+#define ALG_SET_AEAD_ASSOCLEN 4
+#define ALG_SET_AEAD_AUTHSIZE 5
/* Operations */
#define ALG_OP_DECRYPT 0
--
2.1.0
^ permalink raw reply related
* [PATCH v5 2/8] crypto: AF_ALG: add setsockopt for auth tag size
From: Stephan Mueller @ 2014-12-07 22:21 UTC (permalink / raw)
To: Herbert Xu
Cc: Daniel Borkmann, 'Quentin Gouchet', 'LKML',
linux-crypto, linux-api
In-Reply-To: <56740432.V2v4gLHrzS@tachyon.chronox.de>
Use setsockopt on the tfm FD to provide the authentication tag size for
an AEAD cipher. This is achieved by adding a callback function which is
intended to be used by the AEAD AF_ALG implementation.
The optlen argument of the setsockopt specifies the authentication tag
size to be used with the AEAD tfm.
Signed-off-by: Stephan Mueller <smueller@chronox.de>
---
crypto/af_alg.c | 7 +++++++
include/crypto/if_alg.h | 1 +
2 files changed, 8 insertions(+)
diff --git a/crypto/af_alg.c b/crypto/af_alg.c
index 68ff113..547c369 100644
--- a/crypto/af_alg.c
+++ b/crypto/af_alg.c
@@ -215,6 +215,13 @@ static int alg_setsockopt(struct socket *sock, int level, int optname,
goto unlock;
err = alg_setkey(sk, optval, optlen);
+ break;
+ case ALG_SET_AEAD_AUTHSIZE:
+ if (sock->state == SS_CONNECTED)
+ goto unlock;
+ if (!type->setauthsize)
+ goto unlock;
+ err = type->setauthsize(ask->private, optlen);
}
unlock:
diff --git a/include/crypto/if_alg.h b/include/crypto/if_alg.h
index cd62bf4..5c7b6c5 100644
--- a/include/crypto/if_alg.h
+++ b/include/crypto/if_alg.h
@@ -50,6 +50,7 @@ struct af_alg_type {
void (*release)(void *private);
int (*setkey)(void *private, const u8 *key, unsigned int keylen);
int (*accept)(void *private, struct sock *sk);
+ int (*setauthsize)(void *private, unsigned int authsize);
struct proto_ops *ops;
struct module *owner;
--
2.1.0
^ permalink raw reply related
* [PATCH v5 3/8] crypto: AF_ALG: add AEAD support
From: Stephan Mueller @ 2014-12-07 22:22 UTC (permalink / raw)
To: Herbert Xu
Cc: Daniel Borkmann, 'Quentin Gouchet', 'LKML',
linux-crypto, linux-api
In-Reply-To: <56740432.V2v4gLHrzS@tachyon.chronox.de>
This patch adds the AEAD support for AF_ALG.
The implementation is based on algif_skcipher, but contains heavy
modifications to streamline the interface for AEAD uses.
To use AEAD, the user space consumer has to use the salg_type named
"aead".
The AEAD implementation includes some overhead to calculate the size of
the ciphertext, because the AEAD implementation of the kernel crypto API
makes implied assumption on the location of the authentication tag. When
performing an encryption, the tag will be added to the created
ciphertext (note, the tag is placed adjacent to the ciphertext). For
decryption, the caller must hand in the ciphertext with the tag appended
to the ciphertext. Therefore, the selection of the used memory
needs to add/subtract the tag size from the source/destination buffers
depending on the encryption type. The code is provided with comments
explaining when and how that operation is performed.
A fully working example using all aspects of AEAD is provided at
http://www.chronox.de/libkcapi.html
Signed-off-by: Stephan Mueller <smueller@chronox.de>
---
crypto/algif_aead.c | 650 ++++++++++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 650 insertions(+)
create mode 100644 crypto/algif_aead.c
diff --git a/crypto/algif_aead.c b/crypto/algif_aead.c
new file mode 100644
index 0000000..97d6a0e
--- /dev/null
+++ b/crypto/algif_aead.c
@@ -0,0 +1,650 @@
+/*
+ * algif_aeadr: User-space interface for AEAD algorithms
+ *
+ * Copyright (C) 2014, Stephan Mueller <smueller@chronox.de>
+ *
+ * This file provides the user-space API for AEAD ciphers.
+ *
+ * This file is derived from algif_skcipher.c.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation; either version 2 of the License, or (at your option)
+ * any later version.
+ *
+ */
+
+#include <crypto/scatterwalk.h>
+#include <crypto/if_alg.h>
+#include <linux/init.h>
+#include <linux/list.h>
+#include <linux/kernel.h>
+#include <linux/mm.h>
+#include <linux/module.h>
+#include <linux/net.h>
+#include <net/sock.h>
+
+struct aead_sg_list {
+ unsigned int cur;
+ struct scatterlist sg[ALG_MAX_PAGES];
+};
+
+struct aead_ctx {
+ struct aead_sg_list tsgl;
+ struct af_alg_sgl rsgl;
+
+ void *iv;
+
+ struct af_alg_completion completion;
+
+ unsigned long used;
+
+ unsigned int len;
+ bool more;
+ bool merge;
+ bool enc;
+
+ size_t aead_assoclen;
+ struct aead_request aead_req;
+};
+
+static inline int aead_sndbuf(struct sock *sk)
+{
+ struct alg_sock *ask = alg_sk(sk);
+ struct aead_ctx *ctx = ask->private;
+
+ return max_t(int, max_t(int, sk->sk_sndbuf & PAGE_MASK, PAGE_SIZE) -
+ ctx->used, 0);
+}
+
+static inline bool aead_writable(struct sock *sk)
+{
+ return PAGE_SIZE <= aead_sndbuf(sk);
+}
+
+static inline bool aead_sufficient_data(struct aead_ctx *ctx)
+{
+ unsigned as = crypto_aead_authsize(crypto_aead_reqtfm(&ctx->aead_req));
+
+ return (ctx->used >= (ctx->aead_assoclen + ctx->enc ? : as ));
+}
+static inline bool aead_readable(struct aead_ctx *ctx)
+{
+ /*
+ * Ensure that assoc data is present, the plaintext / ciphertext
+ * is non-zero and that the authentication tag is also present
+ * in case of a decryption operation.
+ *
+ * Also, wait until all data is received before processing.
+ */
+ return (aead_sufficient_data(ctx) && !ctx->more);
+}
+
+static void aead_put_sgl(struct sock *sk)
+{
+ struct alg_sock *ask = alg_sk(sk);
+ struct aead_ctx *ctx = ask->private;
+ struct aead_sg_list *sgl = &ctx->tsgl;
+ struct scatterlist *sg = sgl->sg;
+ unsigned int i;
+
+ for (i = 0; i < sgl->cur; i++) {
+ if (!sg_page(sg + i))
+ continue;
+
+ put_page(sg_page(sg + i));
+ sg_assign_page(sg + i, NULL);
+ }
+ sgl->cur = 0;
+ ctx->used = 0;
+ ctx->more = 0;
+ ctx->merge = 0;
+}
+
+static int aead_wait_for_wmem(struct sock *sk, unsigned flags)
+{
+ long timeout;
+ DEFINE_WAIT(wait);
+ int err = -ERESTARTSYS;
+
+ if (flags & MSG_DONTWAIT)
+ return -EAGAIN;
+
+ set_bit(SOCK_ASYNC_NOSPACE, &sk->sk_socket->flags);
+
+ for (;;) {
+ if (signal_pending(current))
+ break;
+ prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE);
+ timeout = MAX_SCHEDULE_TIMEOUT;
+ if (sk_wait_event(sk, &timeout, aead_writable(sk))) {
+ err = 0;
+ break;
+ }
+ }
+ finish_wait(sk_sleep(sk), &wait);
+
+ return err;
+}
+
+static void aead_wmem_wakeup(struct sock *sk)
+{
+ struct socket_wq *wq;
+
+ if (!aead_writable(sk))
+ return;
+
+ rcu_read_lock();
+ wq = rcu_dereference(sk->sk_wq);
+ if (wq_has_sleeper(wq))
+ wake_up_interruptible_sync_poll(&wq->wait, POLLIN |
+ POLLRDNORM |
+ POLLRDBAND);
+ sk_wake_async(sk, SOCK_WAKE_WAITD, POLL_IN);
+ rcu_read_unlock();
+}
+
+static int aead_wait_for_data(struct sock *sk, unsigned flags)
+{
+ struct alg_sock *ask = alg_sk(sk);
+ struct aead_ctx *ctx = ask->private;
+ long timeout;
+ DEFINE_WAIT(wait);
+ int err = -ERESTARTSYS;
+
+ if (flags & MSG_DONTWAIT) {
+ return -EAGAIN;
+ }
+
+ set_bit(SOCK_ASYNC_WAITDATA, &sk->sk_socket->flags);
+
+ for (;;) {
+ if (signal_pending(current))
+ break;
+ prepare_to_wait(sk_sleep(sk), &wait, TASK_INTERRUPTIBLE);
+ timeout = MAX_SCHEDULE_TIMEOUT;
+ if (sk_wait_event(sk, &timeout, aead_readable(ctx))) {
+ err = 0;
+ break;
+ }
+ }
+ finish_wait(sk_sleep(sk), &wait);
+
+ clear_bit(SOCK_ASYNC_WAITDATA, &sk->sk_socket->flags);
+
+ return err;
+}
+
+static void aead_data_wakeup(struct sock *sk)
+{
+ struct alg_sock *ask = alg_sk(sk);
+ struct aead_ctx *ctx = ask->private;
+ struct socket_wq *wq;
+
+ if (!aead_readable(ctx))
+ return;
+
+ rcu_read_lock();
+ wq = rcu_dereference(sk->sk_wq);
+ if (wq_has_sleeper(wq))
+ wake_up_interruptible_sync_poll(&wq->wait, POLLOUT |
+ POLLRDNORM |
+ POLLRDBAND);
+ sk_wake_async(sk, SOCK_WAKE_SPACE, POLL_OUT);
+ rcu_read_unlock();
+}
+
+static int aead_sendmsg(struct kiocb *unused, struct socket *sock,
+ struct msghdr *msg, size_t size)
+{
+ struct sock *sk = sock->sk;
+ struct alg_sock *ask = alg_sk(sk);
+ struct aead_ctx *ctx = ask->private;
+ unsigned ivsize =
+ crypto_aead_ivsize(crypto_aead_reqtfm(&ctx->aead_req));
+ struct aead_sg_list *sgl = &ctx->tsgl;
+ struct af_alg_control con = {};
+ long copied = 0;
+ bool enc = 0;
+ bool init = 0;
+ int err = -EINVAL;
+
+ if (msg->msg_controllen) {
+ err = af_alg_cmsg_send(msg, &con);
+ if (err)
+ return err;
+
+ init = 1;
+ switch (con.op) {
+ case ALG_OP_ENCRYPT:
+ enc = 1;
+ break;
+ case ALG_OP_DECRYPT:
+ enc = 0;
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ if (con.iv && con.iv->ivlen != ivsize)
+ return -EINVAL;
+
+ if (!con.aead_assoclen)
+ return -EINVAL;
+
+ /* aead_recvmsg limits the maximum AD size to one page */
+ if (con.aead_assoclen > PAGE_SIZE)
+ return -E2BIG;
+ }
+
+ lock_sock(sk);
+ if (!ctx->more && ctx->used)
+ goto unlock;
+
+ if (init) {
+ ctx->enc = enc;
+ if (con.iv)
+ memcpy(ctx->iv, con.iv->iv, ivsize);
+
+ ctx->aead_assoclen = con.aead_assoclen;
+ }
+
+ while (size) {
+ unsigned long len = size;
+ struct scatterlist *sg = NULL;
+
+ if (ctx->merge) {
+ sg = sgl->sg + sgl->cur - 1;
+ len = min_t(unsigned long, len,
+ PAGE_SIZE - sg->offset - sg->length);
+ err = memcpy_fromiovec(page_address(sg_page(sg)) +
+ sg->offset + sg->length,
+ msg->msg_iov, len);
+ if (err)
+ goto unlock;
+
+ sg->length += len;
+ ctx->merge = (sg->offset + sg->length) &
+ (PAGE_SIZE - 1);
+
+ ctx->used += len;
+ copied += len;
+ size -= len;
+ }
+
+ if (!aead_writable(sk)) {
+ /*
+ * If there is more data to be expected, but we cannot
+ * write more data, forcefully define that we do not
+ * expect more data to invoke the AEAD operation. This
+ * prevents a deadlock in user space.
+ */
+ ctx->more = 0;
+ err = aead_wait_for_wmem(sk, msg->msg_flags);
+ if (err)
+ goto unlock;
+ }
+
+ len = min_t(unsigned long, size, aead_sndbuf(sk));
+ while (len && sgl->cur < ALG_MAX_PAGES) {
+ int plen = 0;
+
+ sg = sgl->sg + sgl->cur;
+ plen = min_t(int, len, PAGE_SIZE);
+
+ if (sgl->cur >= ALG_MAX_PAGES) {
+ err = -E2BIG;
+ goto unlock;
+ }
+
+ sg_assign_page(sg, alloc_page(GFP_KERNEL));
+ err = -ENOMEM;
+ if (!sg_page(sg))
+ goto unlock;
+
+ err = memcpy_fromiovec(page_address(sg_page(sg)),
+ msg->msg_iov, plen);
+ if (err) {
+ __free_page(sg_page(sg));
+ sg_assign_page(sg, NULL);
+ goto unlock;
+ }
+
+ sg->length = plen;
+ len -= plen;
+ ctx->used += plen;
+ copied += plen;
+ sgl->cur++;
+ size -= plen;
+ ctx->merge = plen & (PAGE_SIZE - 1);
+ }
+ }
+
+ err = 0;
+
+ ctx->more = msg->msg_flags & MSG_MORE;
+
+unlock:
+ aead_data_wakeup(sk);
+ release_sock(sk);
+
+ return copied ?: err;
+}
+
+static ssize_t aead_sendpage(struct socket *sock, struct page *page,
+ int offset, size_t size, int flags)
+{
+ struct sock *sk = sock->sk;
+ struct alg_sock *ask = alg_sk(sk);
+ struct aead_ctx *ctx = ask->private;
+ struct aead_sg_list *sgl = &ctx->tsgl;
+ int err = -EINVAL;
+
+ if (flags & MSG_SENDPAGE_NOTLAST)
+ flags |= MSG_MORE;
+
+ if (sgl->cur >= ALG_MAX_PAGES)
+ return -E2BIG;
+
+ lock_sock(sk);
+ if (!ctx->more && ctx->used)
+ goto unlock;
+
+ if (!size)
+ goto done;
+
+ if (!aead_writable(sk)) {
+ /* see aead_sendmsg why more is set to 0 */
+ ctx->more = 0;
+ err = aead_wait_for_wmem(sk, flags);
+ if (err)
+ goto unlock;
+ }
+
+ ctx->merge = 0;
+
+ get_page(page);
+ sg_set_page(sgl->sg + sgl->cur, page, size, offset);
+ sgl->cur++;
+ ctx->used += size;
+
+ err = 0;
+
+done:
+ ctx->more = flags & MSG_MORE;
+
+unlock:
+ aead_data_wakeup(sk);
+ release_sock(sk);
+
+ return err ?: size;
+}
+
+static int aead_recvmsg(struct kiocb *unused, struct socket *sock,
+ struct msghdr *msg, size_t ignored, int flags)
+{
+ struct sock *sk = sock->sk;
+ struct alg_sock *ask = alg_sk(sk);
+ struct aead_ctx *ctx = ask->private;
+ unsigned bs = crypto_aead_blocksize(crypto_aead_reqtfm(&ctx->aead_req));
+ unsigned as = crypto_aead_authsize(crypto_aead_reqtfm(&ctx->aead_req));
+ struct aead_sg_list *sgl = &ctx->tsgl;
+ struct scatterlist *sg = sgl->sg;
+ struct scatterlist assoc;
+ size_t assoclen = 0;
+ unsigned int i = 0;
+ int err = -EAGAIN;
+ unsigned long used = 0;
+ unsigned long outlen = 0;
+
+ /*
+ * Require exactly one IOV block as the AEAD operation is a one shot
+ * due to the authentication tag.
+ */
+ if (msg->msg_iovlen != 1)
+ return -ENOMSG;
+
+ lock_sock(sk);
+ /*
+ * AEAD memory structure: For encryption, the tag is appended to the
+ * ciphertext which implies that the memory allocated for the ciphertext
+ * must be increased by the tag length. For decryption, the tag
+ * is expected to be concatenated to the ciphertext. The plaintext
+ * therefore has a memory size of the ciphertext minus the tag length.
+ *
+ * The memory structure for cipher operation has the following
+ * structure:
+ * AEAD encryption input: assoc data || plaintext
+ * AEAD encryption output: cipherntext || auth tag
+ * AEAD decryption input: assoc data || ciphertext || auth tag
+ * AEAD decryption output: plaintext
+ */
+
+ if (!aead_readable(ctx)) {
+ err = aead_wait_for_data(sk, flags);
+ if (err)
+ goto unlock;
+ }
+
+ used = ctx->used;
+
+ err = -ENOMEM;
+ if (!aead_sufficient_data(ctx))
+ goto unlock;
+ /*
+ * The cipher operation input data is reduced by the associated data
+ * length as this data is processed separately later on.
+ */
+ used -= ctx->aead_assoclen;
+
+ if (ctx->enc) {
+ /* round up output buffer to multiple of block size */
+ outlen = ((used + bs - 1) / bs * bs);
+ /* add the size needed for the auth tag to be created */
+ outlen += as;
+ } else {
+ /* output data size is input without the authentication tag */
+ outlen = used - as;
+ /* round up output buffer to multiple of block size */
+ outlen = ((outlen + bs - 1) / bs * bs);
+ }
+
+ /* ensure output buffer is sufficiently large */
+ if (msg->msg_iov->iov_len < outlen)
+ goto unlock;
+
+ outlen = af_alg_make_sg(&ctx->rsgl, msg->msg_iov->iov_base, outlen, 1);
+ err = outlen;
+ if (err < 0)
+ goto unlock;
+
+ err = -EINVAL;
+ /*
+ * first chunk of input is AD -- one scatterlist entry is one page,
+ * and we process only one scatterlist, the maximum size of AD is
+ * one page
+ */
+ sg_init_table(&assoc, 1);
+ sg_set_page(&assoc, sg_page(sg), ctx->aead_assoclen, sg->offset);
+ aead_request_set_assoc(&ctx->aead_req, &assoc, ctx->aead_assoclen);
+
+ /* point sg to cipher/plaintext start */
+ assoclen = ctx->aead_assoclen;
+ for(i = 0; i < ctx->tsgl.cur; i++) {
+ sg = sgl->sg + i;
+ if (sg->length <= assoclen) {
+ assoclen -= sg->length;
+ if (i >= ctx->tsgl.cur)
+ goto unlock;
+ } else {
+ sg->length -= assoclen;
+ sg->offset += assoclen;
+ break;
+ }
+ }
+
+ aead_request_set_crypt(&ctx->aead_req, sg, ctx->rsgl.sg, used, ctx->iv);
+
+ err = af_alg_wait_for_completion(ctx->enc ?
+ crypto_aead_encrypt(&ctx->aead_req) :
+ crypto_aead_decrypt(&ctx->aead_req),
+ &ctx->completion);
+
+ af_alg_free_sg(&ctx->rsgl);
+
+ if (err)
+ goto unlock;
+
+ aead_put_sgl(sk);
+
+ err = 0;
+
+unlock:
+ aead_wmem_wakeup(sk);
+ release_sock(sk);
+
+ return err ? err : outlen;
+}
+
+static unsigned int aead_poll(struct file *file, struct socket *sock,
+ poll_table *wait)
+{
+ struct sock *sk = sock->sk;
+ struct alg_sock *ask = alg_sk(sk);
+ struct aead_ctx *ctx = ask->private;
+ unsigned int mask;
+
+ sock_poll_wait(file, sk_sleep(sk), wait);
+ mask = 0;
+
+ if (aead_readable(ctx))
+ mask |= POLLIN | POLLRDNORM;
+
+ if (aead_writable(sk))
+ mask |= POLLOUT | POLLWRNORM | POLLWRBAND;
+
+ return mask;
+}
+
+static struct proto_ops algif_aead_ops = {
+ .family = PF_ALG,
+
+ .connect = sock_no_connect,
+ .socketpair = sock_no_socketpair,
+ .getname = sock_no_getname,
+ .ioctl = sock_no_ioctl,
+ .listen = sock_no_listen,
+ .shutdown = sock_no_shutdown,
+ .getsockopt = sock_no_getsockopt,
+ .mmap = sock_no_mmap,
+ .bind = sock_no_bind,
+ .accept = sock_no_accept,
+ .setsockopt = sock_no_setsockopt,
+
+ .release = af_alg_release,
+ .sendmsg = aead_sendmsg,
+ .sendpage = aead_sendpage,
+ .recvmsg = aead_recvmsg,
+ .poll = aead_poll,
+};
+
+static void *aead_bind(const char *name, u32 type, u32 mask)
+{
+ return crypto_alloc_aead(name, type, mask);
+}
+
+static void aead_release(void *private)
+{
+ crypto_free_aead(private);
+}
+
+static int aead_setauthsize(void *private, unsigned int authsize)
+{
+ return crypto_aead_setauthsize(private, authsize);
+}
+
+static int aead_setkey(void *private, const u8 *key, unsigned int keylen)
+{
+ return crypto_aead_setkey(private, key, keylen);
+}
+
+static void aead_sock_destruct(struct sock *sk)
+{
+ struct alg_sock *ask = alg_sk(sk);
+ struct aead_ctx *ctx = ask->private;
+ unsigned int ivlen = crypto_aead_ivsize(
+ crypto_aead_reqtfm(&ctx->aead_req));
+
+ aead_put_sgl(sk);
+ sock_kzfree_s(sk, ctx->iv, ivlen);
+ sock_kfree_s(sk, ctx, ctx->len);
+ af_alg_release_parent(sk);
+}
+
+static int aead_accept_parent(void *private, struct sock *sk)
+{
+ struct aead_ctx *ctx;
+ struct alg_sock *ask = alg_sk(sk);
+ unsigned int len = sizeof(*ctx) + crypto_aead_reqsize(private);
+ unsigned int ivlen = crypto_aead_ivsize(private);
+
+ ctx = sock_kmalloc(sk, len, GFP_KERNEL);
+ if (!ctx)
+ return -ENOMEM;
+ memset(ctx, 0, len);
+
+ ctx->iv = sock_kmalloc(sk, ivlen, GFP_KERNEL);
+ if (!ctx->iv) {
+ sock_kfree_s(sk, ctx, len);
+ return -ENOMEM;
+ }
+ memset(ctx->iv, 0, ivlen);
+
+ ctx->len = len;
+ ctx->used = 0;
+ ctx->more = 0;
+ ctx->merge = 0;
+ ctx->enc = 0;
+ ctx->tsgl.cur = 0;
+ ctx->aead_assoclen = 0;
+ af_alg_init_completion(&ctx->completion);
+ sg_init_table(ctx->tsgl.sg, ALG_MAX_PAGES);
+
+ ask->private = ctx;
+
+ aead_request_set_tfm(&ctx->aead_req, private);
+ aead_request_set_callback(&ctx->aead_req, CRYPTO_TFM_REQ_MAY_BACKLOG,
+ af_alg_complete, &ctx->completion);
+
+ sk->sk_destruct = aead_sock_destruct;
+
+ return 0;
+}
+
+static const struct af_alg_type algif_type_aead = {
+ .bind = aead_bind,
+ .release = aead_release,
+ .setkey = aead_setkey,
+ .setauthsize = aead_setauthsize,
+ .accept = aead_accept_parent,
+ .ops = &algif_aead_ops,
+ .name = "aead",
+ .owner = THIS_MODULE
+};
+
+static int __init algif_aead_init(void)
+{
+ return af_alg_register_type(&algif_type_aead);
+}
+
+static void __exit algif_aead_exit(void)
+{
+ int err = af_alg_unregister_type(&algif_type_aead);
+ BUG_ON(err);
+}
+
+module_init(algif_aead_init);
+module_exit(algif_aead_exit);
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Stephan Mueller <smueller@chronox.de>");
+MODULE_DESCRIPTION("AEAD kernel crypto API user space interface");
--
2.1.0
^ permalink raw reply related
* [PATCH v5 4/8] crypto: AF_ALG: enable AEAD interface compilation
From: Stephan Mueller @ 2014-12-07 22:23 UTC (permalink / raw)
To: Herbert Xu
Cc: Daniel Borkmann, 'Quentin Gouchet', 'LKML',
linux-crypto, linux-api
In-Reply-To: <56740432.V2v4gLHrzS@tachyon.chronox.de>
Enable compilation of the AEAD AF_ALG support and provide a Kconfig
option to compile the AEAD AF_ALG support.
Signed-off-by: Stephan Mueller <smueller@chronox.de>
---
crypto/Kconfig | 9 +++++++++
crypto/Makefile | 1 +
2 files changed, 10 insertions(+)
diff --git a/crypto/Kconfig b/crypto/Kconfig
index 87bbc9c..0b2affc 100644
--- a/crypto/Kconfig
+++ b/crypto/Kconfig
@@ -1505,6 +1505,15 @@ config CRYPTO_USER_API_SKCIPHER
This option enables the user-spaces interface for symmetric
key cipher algorithms.
+config CRYPTO_USER_API_AEAD
+ tristate "User-space interface for AEAD cipher algorithms"
+ depends on NET
+ select CRYPTO_AEAD
+ select CRYPTO_USER_API
+ help
+ This option enables the user-spaces interface for AEAD
+ cipher algorithms.
+
config CRYPTO_HASH_INFO
bool
diff --git a/crypto/Makefile b/crypto/Makefile
index 1445b91..593fd3c 100644
--- a/crypto/Makefile
+++ b/crypto/Makefile
@@ -99,6 +99,7 @@ obj-$(CONFIG_CRYPTO_GHASH) += ghash-generic.o
obj-$(CONFIG_CRYPTO_USER_API) += af_alg.o
obj-$(CONFIG_CRYPTO_USER_API_HASH) += algif_hash.o
obj-$(CONFIG_CRYPTO_USER_API_SKCIPHER) += algif_skcipher.o
+obj-$(CONFIG_CRYPTO_USER_API_AEAD) += algif_aead.o
#
# generic algorithms and the async_tx api
--
2.1.0
^ permalink raw reply related
* [PATCH v5 5/8] crypto: AF_ALG: add user space interface for RNG
From: Stephan Mueller @ 2014-12-07 22:23 UTC (permalink / raw)
To: Herbert Xu
Cc: Daniel Borkmann, 'Quentin Gouchet', 'LKML',
linux-crypto, linux-api
In-Reply-To: <56740432.V2v4gLHrzS@tachyon.chronox.de>
Allow user space to seed / reset the RNG via a setsockopt.
This patch reuses alg_setkey to copy data into the kernel. The
alg_setkey is now used for two mechanisms: setkey and seeding.
The function is extended by the providing the function pointer
to the function handling the copied data.
As the alg_setkey is now usable for more than just setkey, it is renamed
to alg_setop.
Signed-off-by: Stephan Mueller <smueller@chronox.de>
---
crypto/af_alg.c | 29 +++++++++++++++++++----------
include/crypto/if_alg.h | 1 +
include/uapi/linux/if_alg.h | 1 +
3 files changed, 21 insertions(+), 10 deletions(-)
diff --git a/crypto/af_alg.c b/crypto/af_alg.c
index 547c369..d1eb1e9 100644
--- a/crypto/af_alg.c
+++ b/crypto/af_alg.c
@@ -169,26 +169,27 @@ static int alg_bind(struct socket *sock, struct sockaddr *uaddr, int addr_len)
return 0;
}
-static int alg_setkey(struct sock *sk, char __user *ukey,
- unsigned int keylen)
+static int alg_setop(struct sock *sk, char __user *udata,
+ unsigned int datalen,
+ int (*handler)(void *private, const u8 *data,
+ unsigned int datalen))
{
struct alg_sock *ask = alg_sk(sk);
- const struct af_alg_type *type = ask->type;
- u8 *key;
+ u8 *data;
int err;
- key = sock_kmalloc(sk, keylen, GFP_KERNEL);
- if (!key)
+ data = sock_kmalloc(sk, datalen, GFP_KERNEL);
+ if (!data)
return -ENOMEM;
err = -EFAULT;
- if (copy_from_user(key, ukey, keylen))
+ if (copy_from_user(data, udata, datalen))
goto out;
- err = type->setkey(ask->private, key, keylen);
+ err = handler(ask->private, data, datalen);
out:
- sock_kfree_s(sk, key, keylen);
+ sock_kfree_s(sk, data, datalen);
return err;
}
@@ -214,7 +215,7 @@ static int alg_setsockopt(struct socket *sock, int level, int optname,
if (!type->setkey)
goto unlock;
- err = alg_setkey(sk, optval, optlen);
+ err = alg_setop(sk, optval, optlen, type->setkey);
break;
case ALG_SET_AEAD_AUTHSIZE:
if (sock->state == SS_CONNECTED)
@@ -222,6 +223,14 @@ static int alg_setsockopt(struct socket *sock, int level, int optname,
if (!type->setauthsize)
goto unlock;
err = type->setauthsize(ask->private, optlen);
+ break;
+ case ALG_SET_RNG_SEED:
+ if (sock->state == SS_CONNECTED)
+ goto unlock;
+ if (!type->setseed)
+ goto unlock;
+
+ err = alg_setop(sk, optval, optlen, type->setseed);
}
unlock:
diff --git a/include/crypto/if_alg.h b/include/crypto/if_alg.h
index 5c7b6c5..0808ed1 100644
--- a/include/crypto/if_alg.h
+++ b/include/crypto/if_alg.h
@@ -51,6 +51,7 @@ struct af_alg_type {
int (*setkey)(void *private, const u8 *key, unsigned int keylen);
int (*accept)(void *private, struct sock *sk);
int (*setauthsize)(void *private, unsigned int authsize);
+ int (*setseed)(void *private, const u8 *seed, unsigned int seedlen);
struct proto_ops *ops;
struct module *owner;
diff --git a/include/uapi/linux/if_alg.h b/include/uapi/linux/if_alg.h
index f2acd2f..973dcca 100644
--- a/include/uapi/linux/if_alg.h
+++ b/include/uapi/linux/if_alg.h
@@ -34,6 +34,7 @@ struct af_alg_iv {
#define ALG_SET_OP 3
#define ALG_SET_AEAD_ASSOCLEN 4
#define ALG_SET_AEAD_AUTHSIZE 5
+#define ALG_SET_RNG_SEED 6
/* Operations */
#define ALG_OP_DECRYPT 0
--
2.1.0
^ permalink raw reply related
* [PATCH v5 6/8] crypto: AF_ALG: zeroize key / seed data
From: Stephan Mueller @ 2014-12-07 22:24 UTC (permalink / raw)
To: Herbert Xu
Cc: Daniel Borkmann, 'Quentin Gouchet', 'LKML',
linux-crypto-u79uwXL29TY76Z2rM5mHXA,
linux-api-u79uwXL29TY76Z2rM5mHXA
In-Reply-To: <56740432.V2v4gLHrzS-PJstQz4BMNNP20K/wil9xYQuADTiUCJX@public.gmane.org>
alg_setop shall zeroize the sensitive data after use.
Signed-off-by: Stephan Mueller <smueller-T9tCv8IpfcWELgA04lAiVw@public.gmane.org>
---
crypto/af_alg.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/crypto/af_alg.c b/crypto/af_alg.c
index d1eb1e9..836fa4a 100644
--- a/crypto/af_alg.c
+++ b/crypto/af_alg.c
@@ -189,7 +189,7 @@ static int alg_setop(struct sock *sk, char __user *udata,
err = handler(ask->private, data, datalen);
out:
- sock_kfree_s(sk, data, datalen);
+ sock_kzfree_s(sk, data, datalen);
return err;
}
--
2.1.0
^ permalink raw reply related
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox