From: Jeremy Sowden <jeremy@azazel.net>
To: Jan Engelhardt <jengelh@inai.de>
Cc: "Netfilter Devel" <netfilter-devel@vger.kernel.org>,
"Franta Hanzlík" <franta@hanzlici.cz>
Subject: Re: [PATCH xtables-addons v2 1/2] xt_pknock, xt_SYSRQ: don't set shash_desc::flags.
Date: Mon, 12 Aug 2019 17:57:31 +0100 [thread overview]
Message-ID: <20190812165731.GC5190@azazel.net> (raw)
In-Reply-To: <nycvar.YFH.7.76.1908122317330.19510@n3.vanv.qr>
[-- Attachment #1.1: Type: text/plain, Size: 1632 bytes --]
On 2019-08-12, at 23:17:52 +0800, Jan Engelhardt wrote:
> On Monday 2019-08-12 19:57, Jeremy Sowden wrote:
> >shash_desc::flags was removed from the kernel in 5.1.
> >
> >Signed-off-by: Jeremy Sowden <jeremy@azazel.net>
> >---
> > extensions/pknock/xt_pknock.c | 1 -
> > extensions/xt_SYSRQ.c | 1 -
> > 2 files changed, 2 deletions(-)
> >
> >diff --git a/extensions/pknock/xt_pknock.c b/extensions/pknock/xt_pknock.c
> >index c76901ac4c1a..8021ea07e1b9 100644
> >--- a/extensions/pknock/xt_pknock.c
> >+++ b/extensions/pknock/xt_pknock.c
> >@@ -1125,7 +1125,6 @@ static int __init xt_pknock_mt_init(void)
> >
> > crypto.size = crypto_shash_digestsize(crypto.tfm);
> > crypto.desc.tfm = crypto.tfm;
> >- crypto.desc.flags = 0;
>
> But this will still be needed for 5.0 I guess, so it cannot just be
> unconditionally removed.
That assignment was actually superfluous anyway, because crypto.desc is
zero-initialized when crypto is initialized (xt_pknock.c, ll. 110ff.):
static struct {
const char *algo;
struct crypto_shash *tfm;
unsigned int size;
struct shash_desc desc;
} crypto = {
.algo = "hmac(sha256)",
.tfm = NULL,
.size = 0
};
In fact the explicit zero-initialization of .tfm and .size is also
superfluous and can be removed:
static struct {
const char *algo;
struct crypto_shash *tfm;
unsigned int size;
struct shash_desc desc;
} crypto = {
.algo = "hmac(sha256)",
};
Adding an initializer to the variable declaration in xt_SYSRQ.c will do
the same thing. Patch attached.
J.
[-- Attachment #1.2: 0001-xt_pknock-xt_SYSRQ-don-t-set-shash_desc-flags.patch --]
[-- Type: text/x-diff, Size: 2001 bytes --]
From ea440005076686ba946da433049d4e68c4672984 Mon Sep 17 00:00:00 2001
From: Jeremy Sowden <jeremy@azazel.net>
Date: Sun, 11 Aug 2019 14:08:42 +0100
Subject: [PATCH] xt_pknock, xt_SYSRQ: don't set shash_desc::flags.
shash_desc::flags was removed from the kernel in 5.1, so removed the
explicit assignment of zero to it.
In the case of xt_pknock.c, the change is backwards-compatible because
the shash_desc was already zero-initialized when the enclosing crypto
struct was initialized. In the case of xt_SYSRQ.c, we add an
initializer for the shash_desc which will ensure that all members which
are not explicitly initialized will be initialized to zero, including
.flags in the case of older kernels.
Signed-off-by: Jeremy Sowden <jeremy@azazel.net>
---
extensions/pknock/xt_pknock.c | 1 -
extensions/xt_SYSRQ.c | 4 +---
2 files changed, 1 insertion(+), 4 deletions(-)
diff --git a/extensions/pknock/xt_pknock.c b/extensions/pknock/xt_pknock.c
index c76901ac4c1a..8021ea07e1b9 100644
--- a/extensions/pknock/xt_pknock.c
+++ b/extensions/pknock/xt_pknock.c
@@ -1125,7 +1125,6 @@ static int __init xt_pknock_mt_init(void)
crypto.size = crypto_shash_digestsize(crypto.tfm);
crypto.desc.tfm = crypto.tfm;
- crypto.desc.flags = 0;
pde = proc_mkdir("xt_pknock", init_net.proc_net);
if (pde == NULL) {
diff --git a/extensions/xt_SYSRQ.c b/extensions/xt_SYSRQ.c
index c386c7e2db5d..f04bd2cdc0f2 100644
--- a/extensions/xt_SYSRQ.c
+++ b/extensions/xt_SYSRQ.c
@@ -74,7 +74,7 @@ static unsigned int sysrq_tg(const void *pdata, uint16_t len)
{
const char *data = pdata;
int i, n;
- struct shash_desc desc;
+ struct shash_desc desc = { .tfm = sysrq_tfm };
int ret;
long new_seqno = 0;
@@ -113,8 +113,6 @@ static unsigned int sysrq_tg(const void *pdata, uint16_t len)
return NF_DROP;
}
- desc.tfm = sysrq_tfm;
- desc.flags = 0;
ret = crypto_shash_init(&desc);
if (ret != 0)
goto hash_fail;
--
2.20.1
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]
next prev parent reply other threads:[~2019-08-12 16:57 UTC|newest]
Thread overview: 15+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-08-11 9:40 xtables addons build on 5.2.6 ends with error: 'struct shash_desc' has no member named 'flags' Franta Hanzlík
2019-08-11 12:42 ` Jeremy Sowden
2019-08-11 13:16 ` [PATCH xtables-addons 1/2] xt_pknock, xt_SYSRQ: don't set shash_desc::flags Jeremy Sowden
2019-08-11 13:16 ` [PATCH xtables-addons 2/2] xt_DHCPMAC: replaced skb_make_writable with skb_ensure_writable Jeremy Sowden
2019-08-11 18:42 ` Florian Westphal
2019-08-12 11:06 ` Jeremy Sowden
2019-08-12 11:57 ` [PATCH xtables-addons v2 0/2] Kernel API updates Jeremy Sowden
2019-08-12 11:57 ` [PATCH xtables-addons v2 1/2] xt_pknock, xt_SYSRQ: don't set shash_desc::flags Jeremy Sowden
2019-08-12 15:17 ` Jan Engelhardt
2019-08-12 16:57 ` Jeremy Sowden [this message]
2019-08-19 19:34 ` Franta Hanzlík
2019-09-01 17:04 ` Jeremy Sowden
2019-09-06 8:38 ` Jan Engelhardt
2019-08-12 11:57 ` [PATCH xtables-addons v2 2/2] xt_DHCPMAC: replaced skb_make_writable with skb_ensure_writable Jeremy Sowden
2019-08-12 14:54 ` [PATCH xtables-addons v2 0/2] Kernel API updates Jeremy Sowden
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20190812165731.GC5190@azazel.net \
--to=jeremy@azazel.net \
--cc=franta@hanzlici.cz \
--cc=jengelh@inai.de \
--cc=netfilter-devel@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.