Hi Linus:
Here is the crypto update for 2.6.28:
The highlight of this batch is the new run-time self-test framework
for algorithms. This reorganises the same tests that we've always
had such that they're run at algorithm registration time so that
only algorithms passing the tests of their type are made available
to users. Hopefully that means people won't be sending me any
more algorithm updates that fail these tests :)
Apart from that a new algorithm type for random number generators.
The only two algorithms in that type are the existing kernel RNG
plus a PRNG based on AES. This is used by IPsec for IV generation.
We also have a new implementation of CRC32c that uses the Intel
instruction for CRC32c.
Adrian Bunk (1):
crypto: Kconfig - Replace leading spaces with tabs
Austin Zhang (1):
crypto: crc32c - Use Intel CRC32 instruction
Herbert Xu (10):
crypto: api - Export crypto_alg_lookup instead of __crypto_alg_lookup
crypto: api - Display larval objects properly
crypto: tcrypt - Avoid using contiguous pages
crypto: tcrypt - Abort and only log if there is an error
crypto: tcrypt - Add alg_test interface
crypto: cryptomgr - Add test infrastructure
crypto: api - Use test infrastructure
crypto: cryptomgr - Test ciphers using ECB
crypto: skcipher - Move IV generators into their own modules
crypto: skcipher - Use RNG interface instead of get_random_bytes
Neil Horman (2):
crypto: api - Add fips_enable flag
crypto: rng - RNG interface and implementation
arch/x86/crypto/Makefile | 2
arch/x86/crypto/crc32c-intel.c | 197
crypto/Kconfig | 127
crypto/Makefile | 12
crypto/algapi.c | 147
crypto/algboss.c | 300 +
crypto/ansi_cprng.c | 417 +
crypto/api.c | 81
crypto/blkcipher.c | 29
crypto/chainiv.c | 44
crypto/cryptomgr.c | 222
crypto/eseqiv.c | 35
crypto/fips.c | 27
crypto/internal.h | 18
crypto/krng.c | 66
crypto/proc.c | 57
crypto/rng.c | 126
crypto/seqiv.c | 27
crypto/tcrypt.c | 1347 -----
crypto/tcrypt.h | 8709 ------------------------------------
crypto/testmgr.c | 1868 +++++++
crypto/testmgr.h | 8738 +++++++++++++++++++++++++++++++++++++
include/crypto/internal/rng.h | 26
include/crypto/internal/skcipher.h | 6
include/crypto/rng.h | 75
include/linux/crypto.h | 35
26 files changed, 12464 insertions(+), 10274 deletions(-)
Thanks,
--
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
On Fri, 10 Oct 2008, Herbert Xu wrote:
>
> Here is the crypto update for 2.6.28:
Umm. "Here" was empty.
Mind adding where it actually is, again?
Yeah, yeah, I can look at my old merges (you didn't think I _remember_
things, did you?), and I can see that it's going to be
git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6
but really, please do always say where things are supposed to come from!
Also, please us
git diff -M --stat --summary
for the diffstat. That way I would have seen the
crypto/{cryptomgr.c => algboss.c} | 92 +-
as a rename in the email, instead of having to look why my numbers don't
match the numbers you claimed..
Linus
On Fri, 10 Oct 2008 11:22:44 -0700 (PDT) Linus Torvalds wrote:
>
>
> On Fri, 10 Oct 2008, Herbert Xu wrote:
> >
> > Here is the crypto update for 2.6.28:
>
> Umm. "Here" was empty.
>
> Mind adding where it actually is, again?
>
> Yeah, yeah, I can look at my old merges (you didn't think I _remember_
> things, did you?), and I can see that it's going to be
>
> git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6
>
> but really, please do always say where things are supposed to come from!
>
> Also, please us
>
> git diff -M --stat --summary
>
> for the diffstat. That way I would have seen the
>
> crypto/{cryptomgr.c => algboss.c} | 92 +-
>
> as a rename in the email, instead of having to look why my numbers don't
> match the numbers you claimed..
Herbert, see section 16 of Documentation/SubmittingPatches.
---
~Randy
On Fri, Oct 10, 2008 at 11:22:44AM -0700, Linus Torvalds wrote:
>
>
> On Fri, 10 Oct 2008, Herbert Xu wrote:
> >
> > Here is the crypto update for 2.6.28:
>
> Umm. "Here" was empty.
Sorry, I somehow deleted that bit while adding the changelog :)
> Also, please us
>
> git diff -M --stat --summary
>
> for the diffstat. That way I would have seen the
Thanks, I'll run that from now on.
Here it is again:
The highlight of this batch is the new run-time self-test framework
for algorithms. This reorganises the same tests that we've always
had such that they're run at algorithm registration time so that
only algorithms passing the tests of their type are made available
to users. Hopefully that means people won't be sending me any
more algorithm updates that fail these tests :)
Apart from that a new algorithm type for random number generators.
The only two algorithms in that type are the existing kernel RNG
plus a PRNG based on AES. This is used by IPsec for IV generation.
We also have a new implementation of CRC32c that uses the Intel
instruction for CRC32c.
Please pull from
git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6.git
or
master.kernel.org:/pub/scm/linux/kernel/git/herbert/crypto-2.6.git
Adrian Bunk (1):
crypto: Kconfig - Replace leading spaces with tabs
Austin Zhang (1):
crypto: crc32c - Use Intel CRC32 instruction
Herbert Xu (10):
crypto: api - Export crypto_alg_lookup instead of __crypto_alg_lookup
crypto: api - Display larval objects properly
crypto: tcrypt - Avoid using contiguous pages
crypto: tcrypt - Abort and only log if there is an error
crypto: tcrypt - Add alg_test interface
crypto: cryptomgr - Add test infrastructure
crypto: api - Use test infrastructure
crypto: cryptomgr - Test ciphers using ECB
crypto: skcipher - Move IV generators into their own modules
crypto: skcipher - Use RNG interface instead of get_random_bytes
Neil Horman (2):
crypto: api - Add fips_enable flag
crypto: rng - RNG interface and implementation
arch/x86/crypto/Makefile | 2 +
arch/x86/crypto/crc32c-intel.c | 197 +
crypto/Kconfig | 127 +-
crypto/Makefile | 12 +-
crypto/algapi.c | 147 +-
crypto/{cryptomgr.c => algboss.c} | 92 +-
crypto/ansi_cprng.c | 417 ++
crypto/api.c | 81 +-
crypto/blkcipher.c | 29 -
crypto/chainiv.c | 44 +-
crypto/eseqiv.c | 35 +-
crypto/fips.c | 27 +
crypto/internal.h | 18 +-
crypto/krng.c | 66 +
crypto/proc.c | 57 +-
crypto/rng.c | 126 +
crypto/seqiv.c | 27 +-
crypto/tcrypt.c | 1347 +-----
crypto/tcrypt.h | 8709 +-----------------------------------
crypto/testmgr.c | 1868 ++++++++
crypto/{tcrypt.h => testmgr.h} | 85 +-
include/crypto/internal/rng.h | 26 +
include/crypto/internal/skcipher.h | 6 -
include/crypto/rng.h | 75 +
include/linux/crypto.h | 35 +
25 files changed, 3531 insertions(+), 10124 deletions(-)
create mode 100644 arch/x86/crypto/crc32c-intel.c
rename crypto/{cryptomgr.c => algboss.c} (69%)
create mode 100644 crypto/ansi_cprng.c
create mode 100644 crypto/fips.c
create mode 100644 crypto/krng.c
create mode 100644 crypto/rng.c
create mode 100644 crypto/testmgr.c
copy crypto/{tcrypt.h => testmgr.h} (99%)
create mode 100644 include/crypto/internal/rng.h
create mode 100644 include/crypto/rng.h
Thanks,
--
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
Noticed the bug below while running our -Werror x86 randconfig build
tests against tip/master. Please double check my control flow
analysis.
Ingo
---------->
From e5c3b90a18b73d593bba2a3f927a79469b6218d7 Mon Sep 17 00:00:00 2001
From: Ingo Molnar <[email protected]>
Date: Tue, 25 Nov 2008 09:47:28 +0100
Subject: [PATCH] crypto: fix error flow of crypto/testmgr.c:test_comp()
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 8bit
this warning:
crypto/testmgr.c: In function ‘test_comp’:
crypto/testmgr.c:829: warning: ‘ret’ may be used uninitialized in this function
triggers because GCC correctly notices that in the ctcount == 0 &&
dtcount != 0 input condition case this function can return an undefined
value, if the second loop fails.
Remove the shadowed 'ret' variable from the second loop that was probably
unintended.
Signed-off-by: Ingo Molnar <[email protected]>
---
crypto/testmgr.c | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/crypto/testmgr.c b/crypto/testmgr.c
index b828c6c..0cf0ae3 100644
--- a/crypto/testmgr.c
+++ b/crypto/testmgr.c
@@ -853,7 +853,7 @@ static int test_comp(struct crypto_comp *tfm, struct comp_testvec *ctemplate,
}
for (i = 0; i < dtcount; i++) {
- int ilen, ret, dlen = COMP_BUF_SIZE;
+ int ilen, dlen = COMP_BUF_SIZE;
memset(result, 0, sizeof (result));
On Tue, Nov 25, 2008 at 09:58:21AM +0100, Ingo Molnar wrote:
>
> Noticed the bug below while running our -Werror x86 randconfig build
> tests against tip/master. Please double check my control flow
> analysis.
Good catch, I'll apply this patch. Thanks Ingo!
--
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
* Herbert Xu <[email protected]> wrote:
> On Tue, Nov 25, 2008 at 09:58:21AM +0100, Ingo Molnar wrote:
> >
> > Noticed the bug below while running our -Werror x86 randconfig build
> > tests against tip/master. Please double check my control flow
> > analysis.
>
> Good catch, I'll apply this patch. Thanks Ingo!
Thanks.
btw, on a related note, there's a couple of networking related warning
fixes/annotations i've got queued up.
I've got 37 commits of that nature in the tip/warnings/* topic
branches:
d94b43b: fix warning in drivers/net/s2io.c
78f065d: fix warnings in drivers/net/niu.c
9cc0c65: fix warning in drivers/net/niu.c
32fd3b5: fix warning in drivers/net/qla3xxx.c
ba58621: fix warning in net/sunrpc/xprtrdma/svc_rdma_transport.c
b91eb99: fix warning in net/sched/sch_hfsc.c
c390a37: fix warning in net/netfilter/nf_conntrack_ftp.c
8fecea6: fix warning in net/ipv6/ip6_flowlabel.c
e5e5492: fix warning in drivers/net/wan/z85230.c
67230f2: fix warning in net/dsa/mv88e6xxx.c
948158e: fix warning in net/dsa/mv88e6060.c
8574b3b: fix warning in fs/dlm/netlink.c
9b9c8fd: fix warning in drivers/net/sis900.c
924a289: fix warning in drivers/net/wireless/iwlwifi/iwl-3945.c
2267239: fix warning in drivers/net/wireless/b43/main.c
de96506: fix warning in drivers/net/wireless/ath9k/rc.c
742cddc: fix error path in drivers/net/mlx4/en_rx.c
181787d: fix warning in drivers/net/cassini.c
96c7925: fix warning in net/dccp/options.c
193e56a: fix warning in drivers/net/mlx4/mcg.c
bcc35f7: fix warning in net/ax25/sysctl_net_ax25.c
59681c6: fix warning in net/sunrpc/xprtrdma/verbs.c
ad8d03b: fix warning in net/bluetooth/rfcomm/sock.c
320451a: fix warning in drivers/net/wireless/ray_cs.c
decc165: fix warning in net/netfilter/nf_conntrack_proto_sctp.c
b547d51: fix warning in net/netfilter/ipvs/ip_vs_ctl.c
2281d27: fix warning in drivers/net/atlx/atl2.c
b5a7c2b: fix warning in net/netfilter/nf_conntrack_proto_tcp.c
822c18d: fix warning in drivers/net/wireless/iwlwifi/iwl-scan.c
69f38f6: fix warning in drivers/net/depca.c
557073e: fix warning in drivers/net/3c523.c
d138e44: drivers/net/wireless/b43/phy_g.c: type check debug printouts
7e4f788: fix warning in drivers/net/sky2.c
deb797f: fix warning in net/sunrpc/svcauth_unix.c
923a5a9: fix warning in security/selinux/netnode.c
c62cda5: fix warning in net/packet/af_packet.c
d581867: fix warning in net/rfkill/rfkill.c
Note that the reason there's so many annotations is that this a full
range of fixes for allyes/allno/allmod and randconfig kernels on x86.
That's a lot of build configurations and a lot of warnings to
annotate.
Another one of them is a real bugfix:
742cddc: fix error path in drivers/net/mlx4/en_rx.c
these are side products of -tip's -Werror testing, i.e. not merged
anywhere in the direction of linux-next or -git (obviously).
They are done mainly so that we can notice the warnings we introduce
via any of the -tip trees - but we analyze and look at all other types
of upstream kernel warnings as well.
I'll thus keep these commits indefinitely - but if someone from the
networking side would like to work with us to pick them up and
eventually nurse them upstream, that would be great!
One thing is not really feasible: there's just too many of them for me
to spend 10 minutes on each to submit them with a proper Cc:, etc.
(such overhead mounts up quickly if you do the numbers)
So i've attached them below as a dump of patches. If someone wants to
cherry-pick bits of them, they are also available in
tip/auto-warnings-next:
git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git auto-warnings-next
Warning: some of them might be bad, so please double check the
analysis.
Ingo
-------------->
commit d94b43b196712688c7b018edd69a4ca1650f61e0
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 12:55:10 2008 +0100
fix warning in drivers/net/s2io.c
this warning:
drivers/net/s2io.c: In function ‘rx_intr_handler’:
drivers/net/s2io.c:7369: warning: ‘lro’ may be used uninitialized in this function
triggers because GCC does not recognize the (correct) error flow
between:
- s2io_club_tcp_session()and 'lro'
Annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/s2io.c b/drivers/net/s2io.c
index 6a1375f..a150575 100644
--- a/drivers/net/s2io.c
+++ b/drivers/net/s2io.c
@@ -7366,7 +7366,7 @@ static int rx_osm_handler(struct ring_info *ring_data, struct RxD_t * rxdp)
int ring_no = ring_data->ring_no;
u16 l3_csum, l4_csum;
unsigned long long err = rxdp->Control_1 & RXD_T_CODE;
- struct lro *lro;
+ struct lro *uninitialized_var(lro);
u8 err_mask;
skb->dev = dev;
commit 78f065dd3b7f8116dafc7eb6bf966e8276580c15
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 12:58:10 2008 +0100
fix warnings in drivers/net/niu.c
these warnings:
drivers/net/niu.c: In function ‘serdes_init_niu_1g_serdes’:
drivers/net/niu.c:451: warning: ‘sig’ may be used uninitialized in this function
drivers/net/niu.c: In function ‘serdes_init_niu_10g_serdes’:
drivers/net/niu.c:550: warning: ‘sig’ may be used uninitialized in this function
triggers because GCC does not recognize that the max_retry loop
always initializes 'sig', due to max_retry != 0.
Annotate them.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/niu.c b/drivers/net/niu.c
index afa399c..7e25142 100644
--- a/drivers/net/niu.c
+++ b/drivers/net/niu.c
@@ -448,7 +448,7 @@ static int serdes_init_niu_1g_serdes(struct niu *np)
struct niu_link_config *lp = &np->link_config;
u16 pll_cfg, pll_sts;
int max_retry = 100;
- u64 sig, mask, val;
+ u64 uninitialized_var(sig), mask, val;
u32 tx_cfg, rx_cfg;
unsigned long i;
int err;
@@ -547,7 +547,7 @@ static int serdes_init_niu_10g_serdes(struct niu *np)
struct niu_link_config *lp = &np->link_config;
u32 tx_cfg, rx_cfg, pll_cfg, pll_sts;
int max_retry = 100;
- u64 sig, mask, val;
+ u64 uninitialized_var(sig), mask, val;
unsigned long i;
int err;
commit 9cc0c65af74e02da2c3d621cb061bd61b64cc24e
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 12:54:50 2008 +0100
fix warning in drivers/net/niu.c
this warning:
drivers/net/niu.c: In function ‘esr_reset’:
drivers/net/niu.c:741: warning: ‘reset’ may be used uninitialized in this function
triggers because GCC does not recognize the (correct) error flow
between:
- esr_read_reset() and 'reset'
Annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/niu.c b/drivers/net/niu.c
index 1b6f548..afa399c 100644
--- a/drivers/net/niu.c
+++ b/drivers/net/niu.c
@@ -738,7 +738,7 @@ static int esr_write_glue0(struct niu *np, unsigned long chan, u32 val)
static int esr_reset(struct niu *np)
{
- u32 reset;
+ u32 uninitialized_var(reset);
int err;
err = mdio_write(np, np->port, NIU_ESR_DEV_ADDR,
commit 32fd3b535aa43e290c33873124f91fec6a4329f7
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 12:37:40 2008 +0100
fix warning in drivers/net/qla3xxx.c
this warning:
drivers/net/qla3xxx.c: In function ‘ql3xxx_probe’:
drivers/net/qla3xxx.c:3912: warning: ‘pci_using_dac’ may be used uninitialized in this function
triggers because GCC does not recognize the (correct) error flow
between 'pci_using_dac' and 'err'.
Annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/qla3xxx.c b/drivers/net/qla3xxx.c
index 508452c..3ae623e 100644
--- a/drivers/net/qla3xxx.c
+++ b/drivers/net/qla3xxx.c
@@ -3909,7 +3909,7 @@ static int __devinit ql3xxx_probe(struct pci_dev *pdev,
struct net_device *ndev = NULL;
struct ql3_adapter *qdev = NULL;
static int cards_found = 0;
- int pci_using_dac, err;
+ int uninitialized_var(pci_using_dac), err;
err = pci_enable_device(pdev);
if (err) {
commit ba58621694ef08b59b753bc5042b3db9472c5aff
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 12:03:21 2008 +0100
fix warning in net/sunrpc/xprtrdma/svc_rdma_transport.c
this warning:
net/sunrpc/xprtrdma/svc_rdma_transport.c: In function ‘svc_rdma_accept’:
net/sunrpc/xprtrdma/svc_rdma_transport.c:830: warning: ‘dma_mr_acc’ may be used uninitialized in this function
triggers because GCC does not recognize the (correct) flow connection
between need_dma_mr and dma_mr_acc.
Annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/net/sunrpc/xprtrdma/svc_rdma_transport.c b/net/sunrpc/xprtrdma/svc_rdma_transport.c
index 6fb493c..636095c 100644
--- a/net/sunrpc/xprtrdma/svc_rdma_transport.c
+++ b/net/sunrpc/xprtrdma/svc_rdma_transport.c
@@ -827,7 +827,7 @@ static struct svc_xprt *svc_rdma_accept(struct svc_xprt *xprt)
struct rdma_conn_param conn_param;
struct ib_qp_init_attr qp_attr;
struct ib_device_attr devattr;
- int dma_mr_acc;
+ int uninitialized_var(dma_mr_acc);
int need_dma_mr;
int ret;
int i;
commit b91eb990a0d839e810bb15048af08a69ab1b6ffe
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 12:03:12 2008 +0100
fix warning in net/sched/sch_hfsc.c
this warning:
net/sched/sch_hfsc.c: In function ‘hfsc_enqueue’:
net/sched/sch_hfsc.c:1577: warning: ‘err’ may be used uninitialized in this function
triggers because GCC does not recognize the (correct) error flow
between hfsc_classify(), 'cl' and 'err'.
Annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c
index c1e77da..b39e3b5 100644
--- a/net/sched/sch_hfsc.c
+++ b/net/sched/sch_hfsc.c
@@ -1574,7 +1574,7 @@ static int
hfsc_enqueue(struct sk_buff *skb, struct Qdisc *sch)
{
struct hfsc_class *cl;
- int err;
+ int uninitialized_var(err);
cl = hfsc_classify(skb, sch, &err);
if (cl == NULL) {
commit c390a3784564c1fb2183e4a31ea39a9beb47d034
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 11:31:28 2008 +0100
fix warning in net/netfilter/nf_conntrack_ftp.c
this warning:
net/netfilter/nf_conntrack_ftp.c: In function ‘help’:
net/netfilter/nf_conntrack_ftp.c:360: warning: ‘matchoff’ may be used uninitialized in this function
net/netfilter/nf_conntrack_ftp.c:360: warning: ‘matchlen’ may be used uninitialized in this function
triggers because GCC does not recognize the (correct) error flow
between find_pattern(), 'found', 'matchoff' and 'matchlen'.
Annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/net/netfilter/nf_conntrack_ftp.c b/net/netfilter/nf_conntrack_ftp.c
index 4f71071..af7ad91 100644
--- a/net/netfilter/nf_conntrack_ftp.c
+++ b/net/netfilter/nf_conntrack_ftp.c
@@ -357,7 +357,7 @@ static int help(struct sk_buff *skb,
int ret;
u32 seq;
int dir = CTINFO2DIR(ctinfo);
- unsigned int matchlen, matchoff;
+ unsigned int uninitialized_var(matchlen), uninitialized_var(matchoff);
struct nf_ct_ftp_master *ct_ftp_info = &nfct_help(ct)->help.ct_ftp_info;
struct nf_conntrack_expect *exp;
union nf_inet_addr *daddr;
@@ -592,3 +592,4 @@ static int __init nf_conntrack_ftp_init(void)
module_init(nf_conntrack_ftp_init);
module_exit(nf_conntrack_ftp_fini);
+
commit 8fecea67aa434dd1653c4a275532e29903caeb88
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 11:31:20 2008 +0100
fix warning in net/ipv6/ip6_flowlabel.c
this warning:
net/ipv6/ip6_flowlabel.c: In function ‘ipv6_flowlabel_opt’:
net/ipv6/ip6_flowlabel.c:467: warning: ‘err’ may be used uninitialized in this function
triggers because GCC does not recognize the (correct) error flow
between fl_create() and 'err'.
Annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/net/ipv6/ip6_flowlabel.c b/net/ipv6/ip6_flowlabel.c
index 37a4e77..62809ec 100644
--- a/net/ipv6/ip6_flowlabel.c
+++ b/net/ipv6/ip6_flowlabel.c
@@ -464,7 +464,7 @@ static inline void fl_link(struct ipv6_pinfo *np, struct ipv6_fl_socklist *sfl,
int ipv6_flowlabel_opt(struct sock *sk, char __user *optval, int optlen)
{
- int err;
+ int uninitialized_var(err);
struct net *net = sock_net(sk);
struct ipv6_pinfo *np = inet6_sk(sk);
struct in6_flowlabel_req freq;
commit 67230f214ec140d07fcd32535841ee6eb4437364
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 11:31:12 2008 +0100
fix warning in net/dsa/mv88e6xxx.c
this warning:
net/dsa/mv88e6xxx.c: In function ‘mv88e6xxx_poll_link’:
net/dsa/mv88e6xxx.c:361: warning: ‘port_status’ may be used uninitialized in this function
triggers because GCC does not recognize the (correct) error flow
between 'link' and 'port_status'.
Annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/net/dsa/mv88e6xxx.c b/net/dsa/mv88e6xxx.c
index aa6c609..4e4d8b5 100644
--- a/net/dsa/mv88e6xxx.c
+++ b/net/dsa/mv88e6xxx.c
@@ -358,7 +358,7 @@ void mv88e6xxx_poll_link(struct dsa_switch *ds)
for (i = 0; i < DSA_MAX_PORTS; i++) {
struct net_device *dev;
- int port_status;
+ int uninitialized_var(port_status);
int link;
int speed;
int duplex;
commit 948158e35984f86e565db303e0f026e536edca85
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 11:31:04 2008 +0100
fix warning in net/dsa/mv88e6060.c
this warning:
net/dsa/mv88e6060.c: In function ‘mv88e6060_poll_link’:
net/dsa/mv88e6060.c:225: warning: ‘port_status’ may be used uninitialized in this function
triggers because GCC does not recognize the (correct) error flow
between 'link' and 'port_status'.
Annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/net/dsa/mv88e6060.c b/net/dsa/mv88e6060.c
index 54068ef..bfb4a64 100644
--- a/net/dsa/mv88e6060.c
+++ b/net/dsa/mv88e6060.c
@@ -222,7 +222,7 @@ static void mv88e6060_poll_link(struct dsa_switch *ds)
for (i = 0; i < DSA_MAX_PORTS; i++) {
struct net_device *dev;
- int port_status;
+ int uninitialized_var(port_status);
int link;
int speed;
int duplex;
commit 8574b3b4834c740d37d6eb2d6839037779b56da4
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 11:30:57 2008 +0100
fix warning in fs/dlm/netlink.c
this warning:
fs/dlm/netlink.c: In function ‘dlm_timeout_warn’:
fs/dlm/netlink.c:131: warning: ‘send_skb’ may be used uninitialized in this function
triggers because GCC does not recognize the (correct) error flow
between prepare_data() and send_skb.
Annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/fs/dlm/netlink.c b/fs/dlm/netlink.c
index 18bda83..aa2a577 100644
--- a/fs/dlm/netlink.c
+++ b/fs/dlm/netlink.c
@@ -127,8 +127,8 @@ static void fill_data(struct dlm_lock_data *data, struct dlm_lkb *lkb)
void dlm_timeout_warn(struct dlm_lkb *lkb)
{
+ struct sk_buff *uninitialized_var(send_skb);
struct dlm_lock_data *data;
- struct sk_buff *send_skb;
size_t size;
int rv;
commit 9b9c8fde6c9d175bb7a108e3d231626984364451
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 11:30:15 2008 +0100
fix warning in drivers/net/sis900.c
this warning:
drivers/net/sis900.c: In function ‘sis900_timer’:
drivers/net/sis900.c:1280: warning: ‘speed’ may be used uninitialized in this function
triggers because GCC does not recognize the (correct) error flow
between sis900_read_mode(), 'speed' and 'duplex'.
Annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/sis900.c b/drivers/net/sis900.c
index fa3a460..8be433e 100644
--- a/drivers/net/sis900.c
+++ b/drivers/net/sis900.c
@@ -1277,7 +1277,7 @@ static void sis900_timer(unsigned long data)
u16 status;
if (!sis_priv->autong_complete){
- int speed, duplex = 0;
+ int uninitialized_var(speed), duplex = 0;
sis900_read_mode(net_dev, &speed, &duplex);
if (duplex){
commit 924a289446dadc9c3c9896c87446e0403c746537
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 10:38:24 2008 +0100
fix warning in drivers/net/wireless/iwlwifi/iwl-3945.c
this warning:
drivers/net/wireless/iwlwifi/iwl-3945.c: In function ‘iwl3945_txpower_set_from_eeprom’:
drivers/net/wireless/iwlwifi/iwl-3945.c:2182: warning: ‘power_idx’ may be used uninitialized in this function
triggers because GCC does not recognize the (correct) error flow
between iwl3945_hw_reg_get_matched_power_index() and power_idx.
Annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/wireless/iwlwifi/iwl-3945.c b/drivers/net/wireless/iwlwifi/iwl-3945.c
index 7ca5627..e9e0b1a 100644
--- a/drivers/net/wireless/iwlwifi/iwl-3945.c
+++ b/drivers/net/wireless/iwlwifi/iwl-3945.c
@@ -2179,7 +2179,7 @@ int iwl3945_txpower_set_from_eeprom(struct iwl3945_priv *priv)
/* set tx power value for all OFDM rates */
for (rate_index = 0; rate_index < IWL_OFDM_RATES;
rate_index++) {
- s32 power_idx;
+ s32 uninitialized_var(power_idx);
int rc;
/* use channel group's clip-power table,
commit 22672393d84dc3c74b7cd46eebdb44d43747318b
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 10:38:08 2008 +0100
fix warning in drivers/net/wireless/b43/main.c
this warning:
drivers/net/wireless/b43/main.c: In function ‘b43_op_config’:
drivers/net/wireless/b43/main.c:3258: warning: ‘gmode’ may be used uninitialized in this function
triggers because GCC does not recognize the (correct) error flow
between the list_for_each_entry() loop, up_dev and gmode.
Annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/wireless/b43/main.c b/drivers/net/wireless/b43/main.c
index 14c44df..bd70b27 100644
--- a/drivers/net/wireless/b43/main.c
+++ b/drivers/net/wireless/b43/main.c
@@ -3255,7 +3255,7 @@ static int b43_switch_band(struct b43_wl *wl, struct ieee80211_channel *chan)
struct b43_wldev *down_dev;
struct b43_wldev *d;
int err;
- bool gmode;
+ bool uninitialized_var(gmode);
int prev_status;
/* Find a device and PHY which supports the band. */
commit de965062d615afc885bd800976e488bd3c2d1d6f
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 10:37:58 2008 +0100
fix warning in drivers/net/wireless/ath9k/rc.c
this warning:
drivers/net/wireless/ath9k/rc.c: In function ‘ath_rc_rate_getidx’:
drivers/net/wireless/ath9k/rc.c:1033: warning: ‘nextindex’ may be used uninitialized in this function
triggers because GCC does not recognize the (correct) error flow
between ath_rc_get_nextlowervalid_txrate() and &nextindex.
Annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/wireless/ath9k/rc.c b/drivers/net/wireless/ath9k/rc.c
index cca2fc5..27b35d3 100644
--- a/drivers/net/wireless/ath9k/rc.c
+++ b/drivers/net/wireless/ath9k/rc.c
@@ -1030,7 +1030,7 @@ static u8 ath_rc_rate_getidx(struct ath_softc *sc,
u16 min_rate)
{
u32 j;
- u8 nextindex;
+ u8 uninitialized_var(nextindex);
struct ath_tx_ratectrl *rate_ctrl =
(struct ath_tx_ratectrl *)(ath_rc_priv);
commit e5e5492e7094ed4bea8daf0302f60fbb21daa4c7
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 14:52:20 2008 +0100
fix warning in drivers/net/wan/z85230.c
this warning:
drivers/net/wan/z85230.c: In function ‘z8530_interrupt’:
drivers/net/wan/z85230.c:713: warning: ‘intr’ may be used uninitialized in this function
is clearly bogus - annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/wan/z85230.c b/drivers/net/wan/z85230.c
index 5bf7e01..d7bf53a 100644
--- a/drivers/net/wan/z85230.c
+++ b/drivers/net/wan/z85230.c
@@ -710,7 +710,7 @@ EXPORT_SYMBOL(z8530_nop);
irqreturn_t z8530_interrupt(int irq, void *dev_id)
{
struct z8530_dev *dev=dev_id;
- u8 intr;
+ u8 uninitialized_var(intr);
static volatile int locker=0;
int work=0;
struct z8530_irqhandler *irqs;
commit 742cddc749294be3c0fa909e6255c5a338a1ac40
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 10:37:47 2008 +0100
fix error path in drivers/net/mlx4/en_rx.c
this warning:
drivers/net/mlx4/en_rx.c: In function ‘mlx4_en_activate_rx_rings’:
drivers/net/mlx4/en_rx.c:412: warning: ‘err’ may be used uninitialized in this function
Triggers because 'err' is uninitialized in the following input
conditions: priv->rx_ring_num is zero and mlx4_en_fill_rx_buffers()
fails.
But even if ->rx_ring_num is nonzero, 'err' will be zero if
mlx4_en_fill_rx_buffers() fails and mlx4_en_activate_rx_rings() returns
success - incorrectly.
So it's best to keep the error code uptodate on mlx4_en_fill_rx_buffers()
calls as well.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/mlx4/en_rx.c b/drivers/net/mlx4/en_rx.c
index 6232227..aed7739 100644
--- a/drivers/net/mlx4/en_rx.c
+++ b/drivers/net/mlx4/en_rx.c
@@ -443,7 +443,8 @@ int mlx4_en_activate_rx_rings(struct mlx4_en_priv *priv)
/* Fill Rx buffers */
ring->full = 0;
}
- if (mlx4_en_fill_rx_buffers(priv))
+ err = mlx4_en_fill_rx_buffers(priv);
+ if (err)
goto err_buffers;
for (ring_ind = 0; ring_ind < priv->rx_ring_num; ring_ind++) {
commit 181787d1c30af364ffd15428fe6fab7798c3f3c1
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 10:37:35 2008 +0100
fix warning in drivers/net/cassini.c
this warning:
drivers/net/cassini.c: In function ‘cas_rx_ringN’:
drivers/net/cassini.c:2350: warning: ‘skb’ may be used uninitialized in this function
triggers because GCC does not recognize the (correct) error flow
between cas_rx_process_pkt() and 'skb'.
Annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/cassini.c b/drivers/net/cassini.c
index 86909cf..e083e1c 100644
--- a/drivers/net/cassini.c
+++ b/drivers/net/cassini.c
@@ -2347,7 +2347,7 @@ static int cas_rx_ringN(struct cas *cp, int ring, int budget)
drops = 0;
while (1) {
struct cas_rx_comp *rxc = rxcs + entry;
- struct sk_buff *skb;
+ struct sk_buff *uninitialized_var(skb);
int type, len;
u64 words[4];
int i, dring;
commit 96c79254264033f1505198d5d362b68d88d1bbaf
Author: Ingo Molnar <[email protected]>
Date: Fri Oct 17 12:41:30 2008 +0200
fix warning in net/dccp/options.c
this warning:
net/dccp/options.c: In function ‘dccp_parse_options’:
net/dccp/options.c:67: warning: ‘value’ may be used uninitialized in this function
is a bogus GCC warning. The compiler does not recognize the relation
between "value" and "mandatory" variables: the code flow can ever reach
the "out_invalid_option:" label if 'mandatory' is set to 1, and when
'mandatory' is non-zero, we'll always have 'value' initialized.
Help out the compiler by annotating the variable.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/net/dccp/options.c b/net/dccp/options.c
index 0809b63..18dcfb9 100644
--- a/net/dccp/options.c
+++ b/net/dccp/options.c
@@ -64,7 +64,7 @@ int dccp_parse_options(struct sock *sk, struct dccp_request_sock *dreq,
(dh->dccph_doff * 4);
struct dccp_options_received *opt_recv = &dp->dccps_options_received;
unsigned char opt, len;
- unsigned char *value;
+ unsigned char *uninitialized_var(value);
u32 elapsed_time;
__be32 opt_val;
int rc;
commit 193e56a10e56505dc7eac9e5632834c72b55613d
Author: Ingo Molnar <[email protected]>
Date: Mon Aug 18 16:18:34 2008 +0200
fix warning in drivers/net/mlx4/mcg.c
fix warning:
drivers/net/mlx4/mcg.c: In function ‘mlx4_multicast_attach’:
drivers/net/mlx4/mcg.c:217: warning: integer overflow in expression
there was no real danger of overflow here though.
md5:
db8eb55620f886c03854a2abb2ce6c3f mcg.o.before.asm
db8eb55620f886c03854a2abb2ce6c3f mcg.o.after.asm
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/mlx4/mcg.c b/drivers/net/mlx4/mcg.c
index 592c01a..224ef0f 100644
--- a/drivers/net/mlx4/mcg.c
+++ b/drivers/net/mlx4/mcg.c
@@ -215,7 +215,7 @@ int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
if (block_mcast_loopback)
mgm->qp[members_count++] = cpu_to_be32((qp->qpn & MGM_QPN_MASK) |
- (1 << MGM_BLCK_LB_BIT));
+ (1U << MGM_BLCK_LB_BIT));
else
mgm->qp[members_count++] = cpu_to_be32(qp->qpn & MGM_QPN_MASK);
commit bcc35f75a3b9b96c53a74176b135304fc7ffe052
Author: Ingo Molnar <[email protected]>
Date: Fri Oct 17 15:49:15 2008 +0200
fix warning in net/ax25/sysctl_net_ax25.c
fix this warning:
net/ax25/sysctl_net_ax25.c:27: warning: ‘min_ds_timeout’ defined but not used
net/ax25/sysctl_net_ax25.c:27: warning: ‘max_ds_timeout’ defined but not used
These are only used in the CONFIG_AX25_DAMA_SLAVE case.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/net/ax25/sysctl_net_ax25.c b/net/ax25/sysctl_net_ax25.c
index f288fc4..735ceef 100644
--- a/net/ax25/sysctl_net_ax25.c
+++ b/net/ax25/sysctl_net_ax25.c
@@ -24,7 +24,9 @@ static int min_idle[1], max_idle[] = {65535000};
static int min_n2[] = {1}, max_n2[] = {31};
static int min_paclen[] = {1}, max_paclen[] = {512};
static int min_proto[1], max_proto[] = { AX25_PROTO_MAX };
+#ifdef CONFIG_AX25_DAMA_SLAVE
static int min_ds_timeout[1], max_ds_timeout[] = {65535000};
+#endif
static struct ctl_table_header *ax25_table_header;
commit 59681c6c30e3f700af8d716f66e41ad18f726b80
Author: Ingo Molnar <[email protected]>
Date: Mon Oct 20 18:50:11 2008 +0200
fix warning in net/sunrpc/xprtrdma/verbs.c
fix this warning:
net/sunrpc/xprtrdma/verbs.c: In function ‘rpcrdma_conn_upcall’:
net/sunrpc/xprtrdma/verbs.c:279: warning: unused variable ‘addr’
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c
index a5fef5e..900e595 100644
--- a/net/sunrpc/xprtrdma/verbs.c
+++ b/net/sunrpc/xprtrdma/verbs.c
@@ -276,7 +276,9 @@ rpcrdma_conn_upcall(struct rdma_cm_id *id, struct rdma_cm_event *event)
struct rpcrdma_xprt *xprt = id->context;
struct rpcrdma_ia *ia = &xprt->rx_ia;
struct rpcrdma_ep *ep = &xprt->rx_ep;
+#ifdef RPC_DEBUG
struct sockaddr_in *addr = (struct sockaddr_in *) &ep->rep_remote_addr;
+#endif
struct ib_qp_attr attr;
struct ib_qp_init_attr iattr;
int connstate = 0;
commit ad8d03b315507a954160fcf04dc94cef9f118e77
Author: Ingo Molnar <[email protected]>
Date: Mon Oct 20 17:34:02 2008 +0200
fix warning in net/bluetooth/rfcomm/sock.c
fix this warning:
net/bluetooth/rfcomm/sock.c: In function ‘rfcomm_sock_ioctl’:
net/bluetooth/rfcomm/sock.c:795: warning: unused variable ‘sk’
perhaps BT_DEBUG() should be improved to do printf format checking
instead of the #ifdef, but that looks quite intrusive: each bluetooth
.c file undefines the macro.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/net/bluetooth/rfcomm/sock.c b/net/bluetooth/rfcomm/sock.c
index 8a972b6..bc0d4a7 100644
--- a/net/bluetooth/rfcomm/sock.c
+++ b/net/bluetooth/rfcomm/sock.c
@@ -792,7 +792,9 @@ static int rfcomm_sock_getsockopt(struct socket *sock, int level, int optname, c
static int rfcomm_sock_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
{
+#if defined(CONFIG_BT_RFCOMM_TTY) || defined(CONFIG_BT_RFCOMM_DEBUG)
struct sock *sk = sock->sk;
+#endif
int err;
BT_DBG("sk %p cmd %x arg %lx", sk, cmd, arg);
commit 320451a3f79098ca800f2e58674074c26efee020
Author: Ingo Molnar <[email protected]>
Date: Sat Oct 18 21:33:06 2008 +0200
fix warning in drivers/net/wireless/ray_cs.c
fix this warning:
drivers/net/wireless/ray_cs.c:297: warning: ‘rcsid’ defined but not used
'rcsid' is only used in the CONFIG_PROC_FS=y case.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/wireless/ray_cs.c b/drivers/net/wireless/ray_cs.c
index 1404a57..c23a811 100644
--- a/drivers/net/wireless/ray_cs.c
+++ b/drivers/net/wireless/ray_cs.c
@@ -294,7 +294,9 @@ static char hop_pattern_length[] = { 1,
JAPAN_TEST_HOP_MOD
};
+#ifdef CONFIG_PROC_FS
static char rcsid[] = "Raylink/WebGear wireless LAN - Corey <Thomas [email protected]>";
+#endif
/*=============================================================================
ray_attach() creates an "instance" of the driver, allocating
commit decc1656e870c9e5bd94f2357b8496b8da6c2e73
Author: Ingo Molnar <[email protected]>
Date: Sat Oct 18 19:14:40 2008 +0200
fix warning in net/netfilter/nf_conntrack_proto_sctp.c
fix this warning:
net/netfilter/nf_conntrack_proto_sctp.c: In function ‘sctp_packet’:
net/netfilter/nf_conntrack_proto_sctp.c:376: warning: array subscript is above array bounds
GCC noticed that new_state could in theory be SCTP_CONNTRACK_MAX,
which is beyond the limit of the sctp_timeouts[] array.
This situation could happen if for_each_sctp_chunk() exits immediately.
That should not normally happen - but it's not obvious to me and it's
better to insert an extra check to the end of the loop for this case.
This gets rid of warning as well.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/net/netfilter/nf_conntrack_proto_sctp.c b/net/netfilter/nf_conntrack_proto_sctp.c
index ae8c260..d31ced4 100644
--- a/net/netfilter/nf_conntrack_proto_sctp.c
+++ b/net/netfilter/nf_conntrack_proto_sctp.c
@@ -373,6 +373,9 @@ static int sctp_packet(struct nf_conn *ct,
}
write_unlock_bh(&sctp_lock);
+ if (new_state == SCTP_CONNTRACK_MAX)
+ goto out;
+
nf_ct_refresh_acct(ct, ctinfo, skb, sctp_timeouts[new_state]);
if (old_state == SCTP_CONNTRACK_COOKIE_ECHOED &&
commit b547d51a26c5ed23571df50c7254439e2f5dbebe
Author: Ingo Molnar <[email protected]>
Date: Sat Oct 18 19:02:37 2008 +0200
fix warning in net/netfilter/ipvs/ip_vs_ctl.c
Impact: fix potential kernel stack information exposure to user-space.
fix this warning:
net/netfilter/ipvs/ip_vs_ctl.c: In function ‘ip_vs_genl_set_cmd’:
net/netfilter/ipvs/ip_vs_ctl.c:3034: warning: ‘t.udp_timeout’ may be used uninitialized in this function
net/netfilter/ipvs/ip_vs_ctl.c:3034: note: ‘t.udp_timeout’ was declared here
GCC seems to be correct here: in the !CONFIG_IP_VS_PROTO_UDP case
__ip_vs_get_timeouts keeps the timeout structure uninitialized. But
in a couple of places we use and copy this structure to user-space,
potentially exposing kernel-stack data to user-space.
While it's a minor issue that only triggers with certain config variations,
it's best to always initialize this structure in __ip_vs_get_timeouts().
(GCC will eliminate this memset)
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c
index 0302cf3..7e3a0c7 100644
--- a/net/netfilter/ipvs/ip_vs_ctl.c
+++ b/net/netfilter/ipvs/ip_vs_ctl.c
@@ -2327,6 +2327,7 @@ __ip_vs_get_dest_entries(const struct ip_vs_get_dests *get,
static inline void
__ip_vs_get_timeouts(struct ip_vs_timeout_user *u)
{
+ memset(u, 0, sizeof(*u));
#ifdef CONFIG_IP_VS_PROTO_TCP
u->tcp_timeout =
ip_vs_protocol_tcp.timeout_table[IP_VS_TCP_S_ESTABLISHED] / HZ;
commit 2281d27663a5f17689ec49e029b4962cba19d614
Author: Ingo Molnar <[email protected]>
Date: Sat Oct 18 16:27:37 2008 +0200
fix warning in drivers/net/atlx/atl2.c
fix this warning:
drivers/net/atlx/atl2.c: In function ‘atl2_request_irq’:
drivers/net/atlx/atl2.c:644: warning: unused variable ‘err’
'err' is unused in the !CONFIG_PCI_MSI case.
Instead of further increasing the #ifdeffery in this function,
restructure the code a bit and get rid of the #ifdef. This
relies on the fact that pci_enable_msi() will always fail in
the !CONFIG_PCI_MSI case.
There should be no change in driver behavior.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/atlx/atl2.c b/drivers/net/atlx/atl2.c
index 8571e8c..e02bd7e 100644
--- a/drivers/net/atlx/atl2.c
+++ b/drivers/net/atlx/atl2.c
@@ -644,7 +644,6 @@ static int atl2_request_irq(struct atl2_adapter *adapter)
int flags, err = 0;
flags = IRQF_SHARED;
-#ifdef CONFIG_PCI_MSI
adapter->have_msi = true;
err = pci_enable_msi(adapter->pdev);
if (err)
@@ -652,7 +651,6 @@ static int atl2_request_irq(struct atl2_adapter *adapter)
if (adapter->have_msi)
flags &= ~IRQF_SHARED;
-#endif
return request_irq(adapter->pdev->irq, &atl2_intr, flags, netdev->name,
netdev);
commit b5a7c2b9e1a1b4d80da5d66b330545fdeafe1dad
Author: Ingo Molnar <[email protected]>
Date: Fri Oct 17 19:48:51 2008 +0200
fix warning in net/netfilter/nf_conntrack_proto_tcp.c
fix this warning:
net/netfilter/nf_conntrack_proto_tcp.c: In function ‘tcp_in_window’:
net/netfilter/nf_conntrack_proto_tcp.c:491: warning: unused variable ‘net’
net/netfilter/nf_conntrack_proto_tcp.c: In function ‘tcp_packet’:
net/netfilter/nf_conntrack_proto_tcp.c:812: warning: unused variable ‘net’
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/include/net/netfilter/nf_conntrack_l4proto.h b/include/net/netfilter/nf_conntrack_l4proto.h
index 7f2f43c..debdaf7 100644
--- a/include/net/netfilter/nf_conntrack_l4proto.h
+++ b/include/net/netfilter/nf_conntrack_l4proto.h
@@ -129,7 +129,7 @@ extern const struct nla_policy nf_ct_port_nla_policy[];
&& net_ratelimit())
#endif
#else
-#define LOG_INVALID(net, proto) 0
+static inline int LOG_INVALID(struct net *net, int proto) { return 0; }
#endif /* CONFIG_SYSCTL */
#endif /*_NF_CONNTRACK_PROTOCOL_H*/
commit 822c18db871bc88b2a60d7a9618ec79a8b5e1ef5
Author: Ingo Molnar <[email protected]>
Date: Fri Oct 17 18:31:33 2008 +0200
fix warning in drivers/net/wireless/iwlwifi/iwl-scan.c
fix this warning:
drivers/net/wireless/iwlwifi/iwl-scan.c:92: warning: ‘iwl_escape_essid’ defined but not used
Convert the IWL debug macros to proper printf format checking.
Besides better type checking, this gets rid of the warning as well.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/wireless/iwlwifi/iwl-debug.h b/drivers/net/wireless/iwlwifi/iwl-debug.h
index e548d67..ddc1fbf 100644
--- a/drivers/net/wireless/iwlwifi/iwl-debug.h
+++ b/drivers/net/wireless/iwlwifi/iwl-debug.h
@@ -68,8 +68,10 @@ void iwl_dbgfs_unregister(struct iwl_priv *priv);
#endif
#else
-#define IWL_DEBUG(level, fmt, args...)
-#define IWL_DEBUG_LIMIT(level, fmt, args...)
+static inline void __attribute__ ((format(printf, 2, 3)))
+IWL_DEBUG(int level, const char *fmt, ...) { }
+static inline void __attribute__ ((format(printf, 2, 3)))
+IWL_DEBUG_LIMIT(int level, const char *fmt, ...) { }
#endif /* CONFIG_IWLWIFI_DEBUG */
diff --git a/drivers/net/wireless/iwlwifi/iwl-led.c b/drivers/net/wireless/iwlwifi/iwl-led.c
index 4eee1b1..f4f0664 100644
--- a/drivers/net/wireless/iwlwifi/iwl-led.c
+++ b/drivers/net/wireless/iwlwifi/iwl-led.c
@@ -43,7 +43,6 @@
#include "iwl-io.h"
#include "iwl-helpers.h"
-#ifdef CONFIG_IWLWIFI_DEBUG
static const char *led_type_str[] = {
__stringify(IWL_LED_TRG_TX),
__stringify(IWL_LED_TRG_RX),
@@ -51,7 +50,6 @@ static const char *led_type_str[] = {
__stringify(IWL_LED_TRG_RADIO),
NULL
};
-#endif /* CONFIG_IWLWIFI_DEBUG */
static const struct {
commit 69f38f617c3f832e088be9d35242d929aa623fd4
Author: Ingo Molnar <[email protected]>
Date: Wed Aug 27 19:06:48 2008 +0200
fix warning in drivers/net/depca.c
fix warning:
drivers/net/depca.c: In function ‘depca_eisa_probe’:
drivers/net/depca.c:1564: warning: ‘mem_start’ may be used uninitialized in this function
this seems to be a real bug - depca_eisa_probe() does not check
for failure. Add it, symmetric to depca_isa_probe().
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/depca.c b/drivers/net/depca.c
index ace39ec..f00bd64 100644
--- a/drivers/net/depca.c
+++ b/drivers/net/depca.c
@@ -1558,6 +1558,7 @@ static int __init depca_isa_probe (struct platform_device *device)
#ifdef CONFIG_EISA
static int __init depca_eisa_probe (struct device *device)
{
+ enum depca_type adapter = unknown;
struct eisa_device *edev;
struct net_device *dev;
struct depca_private *lp;
@@ -1576,7 +1577,11 @@ static int __init depca_eisa_probe (struct device *device)
* the EISA configuration structures (yet... :-), just rely on
* the ISA probing to sort it out... */
- depca_shmem_probe (&mem_start);
+ adapter = depca_shmem_probe (&mem_start);
+ if (adapter == unknown) {
+ status = -ENODEV;
+ goto out_free;
+ }
dev->base_addr = ioaddr;
dev->irq = irq;
commit 557073e3dac7b8479df69c140a5fd2b8f4a9cfad
Author: Ingo Molnar <[email protected]>
Date: Wed Aug 27 19:05:19 2008 +0200
fix warning in drivers/net/3c523.c
fix warning:
drivers/net/3c523.c:582: warning: ‘cleanup_card’ defined but not used
No code changed:
md5:
ebe4a1b27d3f21b0b12a78c58463b0d7 3c523.o.before.asm
ebe4a1b27d3f21b0b12a78c58463b0d7 3c523.o.after.asm
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/3c523.c b/drivers/net/3c523.c
index e2ce41d..7b9d427 100644
--- a/drivers/net/3c523.c
+++ b/drivers/net/3c523.c
@@ -578,13 +578,13 @@ err_out:
return retval;
}
+#ifdef MODULE
static void cleanup_card(struct net_device *dev)
{
mca_set_adapter_procfn(((struct priv *) (dev->priv))->slot, NULL, NULL);
release_region(dev->base_addr, ELMC_IO_EXTENT);
}
-
-#ifndef MODULE
+#else
struct net_device * __init elmc_probe(int unit)
{
struct net_device *dev = alloc_etherdev(sizeof(struct priv));
commit d138e44e2f8d0f26e04b0386d328d9cc7e33d82e
Author: Ingo Molnar <[email protected]>
Date: Fri Oct 17 14:30:37 2008 +0200
drivers/net/wireless/b43/phy_g.c: type check debug printouts
this warning:
drivers/net/wireless/b43/phy_g.c: In function ‘b43_gphy_op_recalc_txpower’:
drivers/net/wireless/b43/phy_g.c:3191: warning: unused variable ‘dbm’
is caused because b43dbg() is a macro, hence in the !B43_DEBUG
case the compiler does not know that the variables are used.
Convert it to a proper inline instead. This also improves type checking
in the !B43_DEBUG case.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/wireless/b43/b43.h b/drivers/net/wireless/b43/b43.h
index 427b820..ac55b62 100644
--- a/drivers/net/wireless/b43/b43.h
+++ b/drivers/net/wireless/b43/b43.h
@@ -853,7 +853,8 @@ void b43warn(struct b43_wl *wl, const char *fmt, ...)
void b43dbg(struct b43_wl *wl, const char *fmt, ...)
__attribute__ ((format(printf, 2, 3)));
#else /* DEBUG */
-# define b43dbg(wl, fmt...) do { /* nothing */ } while (0)
+static inline void __attribute__ ((format(printf, 2, 3)))
+b43dbg(struct b43_wl *wl, const char *fmt, ...) { }
#endif /* DEBUG */
/* A WARN_ON variant that vanishes when b43 debugging is disabled.
commit 7e4f7889aca92f2cf95ba31e005fc1368c1cbc5a
Author: Ingo Molnar <[email protected]>
Date: Fri Oct 17 17:47:04 2008 +0200
fix warning in drivers/net/sky2.c
fix this warning:
drivers/net/sky2.c: In function ‘sky2_mhz’:
drivers/net/sky2.c:2753: warning: control reaches end of non-void function
BUG() can be NOP on !CONFIG_BUG.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c
index 3813d15..85fdacc 100644
--- a/drivers/net/sky2.c
+++ b/drivers/net/sky2.c
@@ -2748,7 +2748,7 @@ static u32 sky2_mhz(const struct sky2_hw *hw)
return 156;
default:
- BUG();
+ panic("sky2_mhz: unknown chip id!");
}
}
commit deb797f96d920083ba8a0bcfa6f57b19fc593571
Author: Ingo Molnar <[email protected]>
Date: Fri Oct 17 17:34:27 2008 +0200
fix warning in net/sunrpc/svcauth_unix.c
fix this warning:
net/sunrpc/svcauth_unix.c: In function ‘svcauth_unix_set_client’:
net/sunrpc/svcauth_unix.c:681: warning: ‘sin6’ may be used uninitialized in this function
caused by BUG() being a NOP on !CONFIG_BUG.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/net/sunrpc/svcauth_unix.c b/net/sunrpc/svcauth_unix.c
index f24800f..200ed93 100644
--- a/net/sunrpc/svcauth_unix.c
+++ b/net/sunrpc/svcauth_unix.c
@@ -692,7 +692,7 @@ svcauth_unix_set_client(struct svc_rqst *rqstp)
sin6 = svc_addr_in6(rqstp);
break;
default:
- BUG();
+ panic("svcauth_unix_set_client: bad address family!");
}
rqstp->rq_client = NULL;
@@ -873,3 +873,4 @@ struct auth_ops svcauth_unix = {
.set_client = svcauth_unix_set_client,
};
+
commit 923a5a960577dcc5189b6113300a122c0212b292
Author: Ingo Molnar <[email protected]>
Date: Fri Oct 17 16:55:34 2008 +0200
fix warning in security/selinux/netnode.c
fix this warning:
security/selinux/netnode.c: In function ‘sel_netnode_find’:
security/selinux/netnode.c:131: warning: ‘idx’ may be used uninitialized in this function
BUG() can be a NOP in the !CONFIG_BUG case. Return NULL in that case,
instead of running in an undefined way.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/security/selinux/netnode.c b/security/selinux/netnode.c
index 7100072..a307122 100644
--- a/security/selinux/netnode.c
+++ b/security/selinux/netnode.c
@@ -140,6 +140,7 @@ static struct sel_netnode *sel_netnode_find(const void *addr, u16 family)
break;
default:
BUG();
+ return NULL;
}
list_for_each_entry_rcu(node, &sel_netnode_hash[idx].list, list)
commit c62cda5dfaf7630edefe097a3e23af7732060b08
Author: Ingo Molnar <[email protected]>
Date: Fri Oct 17 16:26:02 2008 +0200
fix warning in net/packet/af_packet.c
fix this warning:
net/packet/af_packet.c: In function ‘tpacket_rcv’:
net/packet/af_packet.c:604: warning: ‘hdrlen’ may be used uninitialized in this function
This is because in the !CONFIG_BUG case BUG() is a NOP.
There is no other sane way to continue but to panic the system
unconditionally, regardless of the CONFIG_BUG setting.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index c718e7e..1478fb8 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -709,7 +709,7 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, struct packe
hdrlen = sizeof(*h.h2);
break;
default:
- BUG();
+ panic("AF_PACKET: bad tp->version");
}
sll = h.raw + TPACKET_ALIGN(hdrlen);
commit d581867031ebe3c6f4e018ea32ed8b9ee5141ead
Author: Ingo Molnar <[email protected]>
Date: Fri Oct 17 16:15:59 2008 +0200
fix warning in net/rfkill/rfkill.c
fix this warning:
net/rfkill/rfkill.c: In function ‘rfkill_get_type_str’:
net/rfkill/rfkill.c:393: warning: control reaches end of non-void function
in the embedded !CONFIG_BUG case BUG() is a NOP.
Return NULL to solve this (unlikely) scenario. All callers feed this
pointer into sprintf %s so NULL should be fine.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/net/rfkill/rfkill.c b/net/rfkill/rfkill.c
index f949a48..eecb9a8 100644
--- a/net/rfkill/rfkill.c
+++ b/net/rfkill/rfkill.c
@@ -389,6 +389,7 @@ static const char *rfkill_get_type_str(enum rfkill_type type)
return "wwan";
default:
BUG();
+ return NULL;
}
}
Ingo Molnar wrote:
> btw, on a related note, there's a couple of networking related warning
> fixes/annotations i've got queued up.
>
> I've got 37 commits of that nature in the tip/warnings/* topic
> branches:
> [...]
>
> They are done mainly so that we can notice the warnings we introduce
> via any of the -tip trees - but we analyze and look at all other types
> of upstream kernel warnings as well.
>
> I'll thus keep these commits indefinitely - but if someone from the
> networking side would like to work with us to pick them up and
> eventually nurse them upstream, that would be great!
>
> One thing is not really feasible: there's just too many of them for me
> to spend 10 minutes on each to submit them with a proper Cc:, etc.
> (such overhead mounts up quickly if you do the numbers)
>
> So i've attached them below as a dump of patches. If someone wants to
> cherry-pick bits of them, they are also available in
> tip/auto-warnings-next:
>
> git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git auto-warnings-next
>
> Warning: some of them might be bad, so please double check the
> analysis.
> commit c390a3784564c1fb2183e4a31ea39a9beb47d034
> Author: Ingo Molnar <[email protected]>
> Date: Tue Nov 25 11:31:28 2008 +0100
>
> fix warning in net/netfilter/nf_conntrack_ftp.c
I've picked up this one in my nf-next-2.6.git tree.
> commit decc1656e870c9e5bd94f2357b8496b8da6c2e73
> Author: Ingo Molnar <[email protected]>
> Date: Sat Oct 18 19:14:40 2008 +0200
>
> fix warning in net/netfilter/nf_conntrack_proto_sctp.c
This one is already fixed differently in my tree.
> commit b5a7c2b9e1a1b4d80da5d66b330545fdeafe1dad
> Author: Ingo Molnar <[email protected]>
> Date: Fri Oct 17 19:48:51 2008 +0200
>
> fix warning in net/netfilter/nf_conntrack_proto_tcp.c
I've also picked up this one.
From: Ingo Molnar <[email protected]>
Date: Tue, 25 Nov 2008 18:14:38 +0100
> btw, on a related note, there's a couple of networking related warning
> fixes/annotations i've got queued up.
I see these now, but they would get properly logged and
tracked at:
http://patchwork.ozlabs.org/project/netdev/
if you would simply CC: netdev instead of sending them into
the ether by informing lkml and the crypto list about them.
From: Ingo Molnar <[email protected]>
Date: Tue, 25 Nov 2008 18:14:38 +0100
> fix warning in drivers/net/s2io.c
> fix warnings in drivers/net/niu.c
> fix warning in drivers/net/niu.c
> fix warning in drivers/net/qla3xxx.c
> fix warning in net/sunrpc/xprtrdma/svc_rdma_transport.c
> fix warning in net/sched/sch_hfsc.c
Applied.
> fix warning in net/netfilter/nf_conntrack_ftp.c
Patrick got this one.
> fix warning in net/ipv6/ip6_flowlabel.c
> fix warning in net/dsa/mv88e6xxx.c
> fix warning in net/dsa/mv88e6060.c
> fix warning in fs/dlm/netlink.c
> fix warning in drivers/net/sis900.c
Applied
> fix warning in drivers/net/wireless/iwlwifi/iwl-3945.c
> fix warning in drivers/net/wireless/b43/main.c
> fix warning in drivers/net/wireless/ath9k/rc.c
Needs to be submitted via wireless folks at linux-wireless
> fix warning in drivers/net/wan/z85230.c
> fix error path in drivers/net/mlx4/en_rx.c
> fix warning in drivers/net/cassini.c
> fix warning in net/dccp/options.c
> fix warning in drivers/net/mlx4/mcg.c
> fix warning in net/ax25/sysctl_net_ax25.c
> fix warning in net/sunrpc/xprtrdma/verbs.c
> fix warning in net/bluetooth/rfcomm/sock.c
Applied.
> fix warning in drivers/net/wireless/ray_cs.c
Submit to wireless maintainer please.
> fix warning in net/netfilter/nf_conntrack_proto_sctp.c
> fix warning in net/netfilter/ipvs/ip_vs_ctl.c
Patrick got these.
> fix warning in drivers/net/atlx/atl2.c
Applied.
> fix warning in net/netfilter/nf_conntrack_proto_tcp.c
Patrick got it.
> fix warning in drivers/net/wireless/iwlwifi/iwl-scan.c
Wireless submission please.
> fix warning in drivers/net/depca.c
> fix warning in drivers/net/3c523.c
Applied.
> drivers/net/wireless/b43/phy_g.c: type check debug printouts
Submit to wireless folks please.
> fix warning in drivers/net/sky2.c
> fix warning in net/sunrpc/svcauth_unix.c
> fix warning in security/selinux/netnode.c
> fix warning in net/packet/af_packet.c
> fix warning in net/rfkill/rfkill.c
Not applied. Changing this from a BUG() to a panic() is wrong.
BUG() kills the current execution context, and allows the system
to potentially continue. Panic() takes down the entire machine.
The !CONFIG_BUG BUG() definition should be marked in such a way
that gcc thinks it is noreturn. Otherwise BUG() isn't really
BUG() anymore. The other option is to never allow BUG to be
disabled or have it always evaluate to a simple bug trap.
* David Miller <[email protected]> wrote:
> From: Ingo Molnar <[email protected]>
> Date: Tue, 25 Nov 2008 18:14:38 +0100
>
> > btw, on a related note, there's a couple of networking related warning
> > fixes/annotations i've got queued up.
>
> I see these now, but they would get properly logged and
> tracked at:
>
> http://patchwork.ozlabs.org/project/netdev/
>
> if you would simply CC: netdev instead of sending them into
> the ether by informing lkml and the crypto list about them.
sure - netdev Cc:-ed, subject line changed - mail repeated below.
This one seems to be a real bug too (subject line nonwithstanding):
69f38f6: fix warning in drivers/net/depca.c
besides the bugfixes, the annotations would be nice to have too - i
got the warnings with latest-gcc (4.3.2) and once we reach a baseline
of no warnings it will be much easier to go forward in a delta manner.
At that points commits that actually cause new warnings will be much
fresher - and the warnings will be a lot more interesting in general
as well.
Right now we've got a baseline noise where it's mostly GCC that is
wrong. Obviously those are the warnings that tend to linger around in
the kernel source: people see the warnings and determine that GCC is
wrong and keep the source unchanged.
The arch/x86/* files already have a zero warnings baseline (clean
-Werror builds), and my experience is that the new warnings that pop
up are generally quite healthy and interesting, with a low false
positive rate.
Hm?
Ingo
---------------->
I've got 37 commits of that nature in the tip/warnings/* topic
branches:
d94b43b: fix warning in drivers/net/s2io.c
78f065d: fix warnings in drivers/net/niu.c
9cc0c65: fix warning in drivers/net/niu.c
32fd3b5: fix warning in drivers/net/qla3xxx.c
ba58621: fix warning in net/sunrpc/xprtrdma/svc_rdma_transport.c
b91eb99: fix warning in net/sched/sch_hfsc.c
c390a37: fix warning in net/netfilter/nf_conntrack_ftp.c
8fecea6: fix warning in net/ipv6/ip6_flowlabel.c
e5e5492: fix warning in drivers/net/wan/z85230.c
67230f2: fix warning in net/dsa/mv88e6xxx.c
948158e: fix warning in net/dsa/mv88e6060.c
8574b3b: fix warning in fs/dlm/netlink.c
9b9c8fd: fix warning in drivers/net/sis900.c
924a289: fix warning in drivers/net/wireless/iwlwifi/iwl-3945.c
2267239: fix warning in drivers/net/wireless/b43/main.c
de96506: fix warning in drivers/net/wireless/ath9k/rc.c
742cddc: fix error path in drivers/net/mlx4/en_rx.c
181787d: fix warning in drivers/net/cassini.c
96c7925: fix warning in net/dccp/options.c
193e56a: fix warning in drivers/net/mlx4/mcg.c
bcc35f7: fix warning in net/ax25/sysctl_net_ax25.c
59681c6: fix warning in net/sunrpc/xprtrdma/verbs.c
ad8d03b: fix warning in net/bluetooth/rfcomm/sock.c
320451a: fix warning in drivers/net/wireless/ray_cs.c
decc165: fix warning in net/netfilter/nf_conntrack_proto_sctp.c
b547d51: fix warning in net/netfilter/ipvs/ip_vs_ctl.c
2281d27: fix warning in drivers/net/atlx/atl2.c
b5a7c2b: fix warning in net/netfilter/nf_conntrack_proto_tcp.c
822c18d: fix warning in drivers/net/wireless/iwlwifi/iwl-scan.c
69f38f6: fix warning in drivers/net/depca.c
557073e: fix warning in drivers/net/3c523.c
d138e44: drivers/net/wireless/b43/phy_g.c: type check debug printouts
7e4f788: fix warning in drivers/net/sky2.c
deb797f: fix warning in net/sunrpc/svcauth_unix.c
923a5a9: fix warning in security/selinux/netnode.c
c62cda5: fix warning in net/packet/af_packet.c
d581867: fix warning in net/rfkill/rfkill.c
Note that the reason there's so many annotations is that this a full
range of fixes for allyes/allno/allmod and randconfig kernels on x86.
That's a lot of build configurations and a lot of warnings to
annotate.
Another one of them is a real bugfix:
742cddc: fix error path in drivers/net/mlx4/en_rx.c
these are side products of -tip's -Werror testing, i.e. not merged
anywhere in the direction of linux-next or -git (obviously).
They are done mainly so that we can notice the warnings we introduce
via any of the -tip trees - but we analyze and look at all other types
of upstream kernel warnings as well.
I'll thus keep these commits indefinitely - but if someone from the
networking side would like to work with us to pick them up and
eventually nurse them upstream, that would be great!
One thing is not really feasible: there's just too many of them for me
to spend 10 minutes on each to submit them with a proper Cc:, etc.
(such overhead mounts up quickly if you do the numbers)
So i've attached them below as a dump of patches. If someone wants to
cherry-pick bits of them, they are also available in
tip/auto-warnings-next:
git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip.git auto-warnings-next
Warning: some of them might be bad, so please double check the
analysis.
Ingo
-------------->
commit d94b43b196712688c7b018edd69a4ca1650f61e0
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 12:55:10 2008 +0100
fix warning in drivers/net/s2io.c
this warning:
drivers/net/s2io.c: In function ‘rx_intr_handler’:
drivers/net/s2io.c:7369: warning: ‘lro’ may be used uninitialized in this function
triggers because GCC does not recognize the (correct) error flow
between:
- s2io_club_tcp_session()and 'lro'
Annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/s2io.c b/drivers/net/s2io.c
index 6a1375f..a150575 100644
--- a/drivers/net/s2io.c
+++ b/drivers/net/s2io.c
@@ -7366,7 +7366,7 @@ static int rx_osm_handler(struct ring_info *ring_data, struct RxD_t * rxdp)
int ring_no = ring_data->ring_no;
u16 l3_csum, l4_csum;
unsigned long long err = rxdp->Control_1 & RXD_T_CODE;
- struct lro *lro;
+ struct lro *uninitialized_var(lro);
u8 err_mask;
skb->dev = dev;
commit 78f065dd3b7f8116dafc7eb6bf966e8276580c15
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 12:58:10 2008 +0100
fix warnings in drivers/net/niu.c
these warnings:
drivers/net/niu.c: In function ‘serdes_init_niu_1g_serdes’:
drivers/net/niu.c:451: warning: ‘sig’ may be used uninitialized in this function
drivers/net/niu.c: In function ‘serdes_init_niu_10g_serdes’:
drivers/net/niu.c:550: warning: ‘sig’ may be used uninitialized in this function
triggers because GCC does not recognize that the max_retry loop
always initializes 'sig', due to max_retry != 0.
Annotate them.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/niu.c b/drivers/net/niu.c
index afa399c..7e25142 100644
--- a/drivers/net/niu.c
+++ b/drivers/net/niu.c
@@ -448,7 +448,7 @@ static int serdes_init_niu_1g_serdes(struct niu *np)
struct niu_link_config *lp = &np->link_config;
u16 pll_cfg, pll_sts;
int max_retry = 100;
- u64 sig, mask, val;
+ u64 uninitialized_var(sig), mask, val;
u32 tx_cfg, rx_cfg;
unsigned long i;
int err;
@@ -547,7 +547,7 @@ static int serdes_init_niu_10g_serdes(struct niu *np)
struct niu_link_config *lp = &np->link_config;
u32 tx_cfg, rx_cfg, pll_cfg, pll_sts;
int max_retry = 100;
- u64 sig, mask, val;
+ u64 uninitialized_var(sig), mask, val;
unsigned long i;
int err;
commit 9cc0c65af74e02da2c3d621cb061bd61b64cc24e
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 12:54:50 2008 +0100
fix warning in drivers/net/niu.c
this warning:
drivers/net/niu.c: In function ‘esr_reset’:
drivers/net/niu.c:741: warning: ‘reset’ may be used uninitialized in this function
triggers because GCC does not recognize the (correct) error flow
between:
- esr_read_reset() and 'reset'
Annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/niu.c b/drivers/net/niu.c
index 1b6f548..afa399c 100644
--- a/drivers/net/niu.c
+++ b/drivers/net/niu.c
@@ -738,7 +738,7 @@ static int esr_write_glue0(struct niu *np, unsigned long chan, u32 val)
static int esr_reset(struct niu *np)
{
- u32 reset;
+ u32 uninitialized_var(reset);
int err;
err = mdio_write(np, np->port, NIU_ESR_DEV_ADDR,
commit 32fd3b535aa43e290c33873124f91fec6a4329f7
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 12:37:40 2008 +0100
fix warning in drivers/net/qla3xxx.c
this warning:
drivers/net/qla3xxx.c: In function ‘ql3xxx_probe’:
drivers/net/qla3xxx.c:3912: warning: ‘pci_using_dac’ may be used uninitialized in this function
triggers because GCC does not recognize the (correct) error flow
between 'pci_using_dac' and 'err'.
Annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/qla3xxx.c b/drivers/net/qla3xxx.c
index 508452c..3ae623e 100644
--- a/drivers/net/qla3xxx.c
+++ b/drivers/net/qla3xxx.c
@@ -3909,7 +3909,7 @@ static int __devinit ql3xxx_probe(struct pci_dev *pdev,
struct net_device *ndev = NULL;
struct ql3_adapter *qdev = NULL;
static int cards_found = 0;
- int pci_using_dac, err;
+ int uninitialized_var(pci_using_dac), err;
err = pci_enable_device(pdev);
if (err) {
commit ba58621694ef08b59b753bc5042b3db9472c5aff
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 12:03:21 2008 +0100
fix warning in net/sunrpc/xprtrdma/svc_rdma_transport.c
this warning:
net/sunrpc/xprtrdma/svc_rdma_transport.c: In function ‘svc_rdma_accept’:
net/sunrpc/xprtrdma/svc_rdma_transport.c:830: warning: ‘dma_mr_acc’ may be used uninitialized in this function
triggers because GCC does not recognize the (correct) flow connection
between need_dma_mr and dma_mr_acc.
Annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/net/sunrpc/xprtrdma/svc_rdma_transport.c b/net/sunrpc/xprtrdma/svc_rdma_transport.c
index 6fb493c..636095c 100644
--- a/net/sunrpc/xprtrdma/svc_rdma_transport.c
+++ b/net/sunrpc/xprtrdma/svc_rdma_transport.c
@@ -827,7 +827,7 @@ static struct svc_xprt *svc_rdma_accept(struct svc_xprt *xprt)
struct rdma_conn_param conn_param;
struct ib_qp_init_attr qp_attr;
struct ib_device_attr devattr;
- int dma_mr_acc;
+ int uninitialized_var(dma_mr_acc);
int need_dma_mr;
int ret;
int i;
commit b91eb990a0d839e810bb15048af08a69ab1b6ffe
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 12:03:12 2008 +0100
fix warning in net/sched/sch_hfsc.c
this warning:
net/sched/sch_hfsc.c: In function ‘hfsc_enqueue’:
net/sched/sch_hfsc.c:1577: warning: ‘err’ may be used uninitialized in this function
triggers because GCC does not recognize the (correct) error flow
between hfsc_classify(), 'cl' and 'err'.
Annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/net/sched/sch_hfsc.c b/net/sched/sch_hfsc.c
index c1e77da..b39e3b5 100644
--- a/net/sched/sch_hfsc.c
+++ b/net/sched/sch_hfsc.c
@@ -1574,7 +1574,7 @@ static int
hfsc_enqueue(struct sk_buff *skb, struct Qdisc *sch)
{
struct hfsc_class *cl;
- int err;
+ int uninitialized_var(err);
cl = hfsc_classify(skb, sch, &err);
if (cl == NULL) {
commit c390a3784564c1fb2183e4a31ea39a9beb47d034
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 11:31:28 2008 +0100
fix warning in net/netfilter/nf_conntrack_ftp.c
this warning:
net/netfilter/nf_conntrack_ftp.c: In function ‘help’:
net/netfilter/nf_conntrack_ftp.c:360: warning: ‘matchoff’ may be used uninitialized in this function
net/netfilter/nf_conntrack_ftp.c:360: warning: ‘matchlen’ may be used uninitialized in this function
triggers because GCC does not recognize the (correct) error flow
between find_pattern(), 'found', 'matchoff' and 'matchlen'.
Annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/net/netfilter/nf_conntrack_ftp.c b/net/netfilter/nf_conntrack_ftp.c
index 4f71071..af7ad91 100644
--- a/net/netfilter/nf_conntrack_ftp.c
+++ b/net/netfilter/nf_conntrack_ftp.c
@@ -357,7 +357,7 @@ static int help(struct sk_buff *skb,
int ret;
u32 seq;
int dir = CTINFO2DIR(ctinfo);
- unsigned int matchlen, matchoff;
+ unsigned int uninitialized_var(matchlen), uninitialized_var(matchoff);
struct nf_ct_ftp_master *ct_ftp_info = &nfct_help(ct)->help.ct_ftp_info;
struct nf_conntrack_expect *exp;
union nf_inet_addr *daddr;
@@ -592,3 +592,4 @@ static int __init nf_conntrack_ftp_init(void)
module_init(nf_conntrack_ftp_init);
module_exit(nf_conntrack_ftp_fini);
+
commit 8fecea67aa434dd1653c4a275532e29903caeb88
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 11:31:20 2008 +0100
fix warning in net/ipv6/ip6_flowlabel.c
this warning:
net/ipv6/ip6_flowlabel.c: In function ‘ipv6_flowlabel_opt’:
net/ipv6/ip6_flowlabel.c:467: warning: ‘err’ may be used uninitialized in this function
triggers because GCC does not recognize the (correct) error flow
between fl_create() and 'err'.
Annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/net/ipv6/ip6_flowlabel.c b/net/ipv6/ip6_flowlabel.c
index 37a4e77..62809ec 100644
--- a/net/ipv6/ip6_flowlabel.c
+++ b/net/ipv6/ip6_flowlabel.c
@@ -464,7 +464,7 @@ static inline void fl_link(struct ipv6_pinfo *np, struct ipv6_fl_socklist *sfl,
int ipv6_flowlabel_opt(struct sock *sk, char __user *optval, int optlen)
{
- int err;
+ int uninitialized_var(err);
struct net *net = sock_net(sk);
struct ipv6_pinfo *np = inet6_sk(sk);
struct in6_flowlabel_req freq;
commit 67230f214ec140d07fcd32535841ee6eb4437364
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 11:31:12 2008 +0100
fix warning in net/dsa/mv88e6xxx.c
this warning:
net/dsa/mv88e6xxx.c: In function ‘mv88e6xxx_poll_link’:
net/dsa/mv88e6xxx.c:361: warning: ‘port_status’ may be used uninitialized in this function
triggers because GCC does not recognize the (correct) error flow
between 'link' and 'port_status'.
Annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/net/dsa/mv88e6xxx.c b/net/dsa/mv88e6xxx.c
index aa6c609..4e4d8b5 100644
--- a/net/dsa/mv88e6xxx.c
+++ b/net/dsa/mv88e6xxx.c
@@ -358,7 +358,7 @@ void mv88e6xxx_poll_link(struct dsa_switch *ds)
for (i = 0; i < DSA_MAX_PORTS; i++) {
struct net_device *dev;
- int port_status;
+ int uninitialized_var(port_status);
int link;
int speed;
int duplex;
commit 948158e35984f86e565db303e0f026e536edca85
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 11:31:04 2008 +0100
fix warning in net/dsa/mv88e6060.c
this warning:
net/dsa/mv88e6060.c: In function ‘mv88e6060_poll_link’:
net/dsa/mv88e6060.c:225: warning: ‘port_status’ may be used uninitialized in this function
triggers because GCC does not recognize the (correct) error flow
between 'link' and 'port_status'.
Annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/net/dsa/mv88e6060.c b/net/dsa/mv88e6060.c
index 54068ef..bfb4a64 100644
--- a/net/dsa/mv88e6060.c
+++ b/net/dsa/mv88e6060.c
@@ -222,7 +222,7 @@ static void mv88e6060_poll_link(struct dsa_switch *ds)
for (i = 0; i < DSA_MAX_PORTS; i++) {
struct net_device *dev;
- int port_status;
+ int uninitialized_var(port_status);
int link;
int speed;
int duplex;
commit 8574b3b4834c740d37d6eb2d6839037779b56da4
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 11:30:57 2008 +0100
fix warning in fs/dlm/netlink.c
this warning:
fs/dlm/netlink.c: In function ‘dlm_timeout_warn’:
fs/dlm/netlink.c:131: warning: ‘send_skb’ may be used uninitialized in this function
triggers because GCC does not recognize the (correct) error flow
between prepare_data() and send_skb.
Annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/fs/dlm/netlink.c b/fs/dlm/netlink.c
index 18bda83..aa2a577 100644
--- a/fs/dlm/netlink.c
+++ b/fs/dlm/netlink.c
@@ -127,8 +127,8 @@ static void fill_data(struct dlm_lock_data *data, struct dlm_lkb *lkb)
void dlm_timeout_warn(struct dlm_lkb *lkb)
{
+ struct sk_buff *uninitialized_var(send_skb);
struct dlm_lock_data *data;
- struct sk_buff *send_skb;
size_t size;
int rv;
commit 9b9c8fde6c9d175bb7a108e3d231626984364451
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 11:30:15 2008 +0100
fix warning in drivers/net/sis900.c
this warning:
drivers/net/sis900.c: In function ‘sis900_timer’:
drivers/net/sis900.c:1280: warning: ‘speed’ may be used uninitialized in this function
triggers because GCC does not recognize the (correct) error flow
between sis900_read_mode(), 'speed' and 'duplex'.
Annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/sis900.c b/drivers/net/sis900.c
index fa3a460..8be433e 100644
--- a/drivers/net/sis900.c
+++ b/drivers/net/sis900.c
@@ -1277,7 +1277,7 @@ static void sis900_timer(unsigned long data)
u16 status;
if (!sis_priv->autong_complete){
- int speed, duplex = 0;
+ int uninitialized_var(speed), duplex = 0;
sis900_read_mode(net_dev, &speed, &duplex);
if (duplex){
commit 924a289446dadc9c3c9896c87446e0403c746537
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 10:38:24 2008 +0100
fix warning in drivers/net/wireless/iwlwifi/iwl-3945.c
this warning:
drivers/net/wireless/iwlwifi/iwl-3945.c: In function ‘iwl3945_txpower_set_from_eeprom’:
drivers/net/wireless/iwlwifi/iwl-3945.c:2182: warning: ‘power_idx’ may be used uninitialized in this function
triggers because GCC does not recognize the (correct) error flow
between iwl3945_hw_reg_get_matched_power_index() and power_idx.
Annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/wireless/iwlwifi/iwl-3945.c b/drivers/net/wireless/iwlwifi/iwl-3945.c
index 7ca5627..e9e0b1a 100644
--- a/drivers/net/wireless/iwlwifi/iwl-3945.c
+++ b/drivers/net/wireless/iwlwifi/iwl-3945.c
@@ -2179,7 +2179,7 @@ int iwl3945_txpower_set_from_eeprom(struct iwl3945_priv *priv)
/* set tx power value for all OFDM rates */
for (rate_index = 0; rate_index < IWL_OFDM_RATES;
rate_index++) {
- s32 power_idx;
+ s32 uninitialized_var(power_idx);
int rc;
/* use channel group's clip-power table,
commit 22672393d84dc3c74b7cd46eebdb44d43747318b
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 10:38:08 2008 +0100
fix warning in drivers/net/wireless/b43/main.c
this warning:
drivers/net/wireless/b43/main.c: In function ‘b43_op_config’:
drivers/net/wireless/b43/main.c:3258: warning: ‘gmode’ may be used uninitialized in this function
triggers because GCC does not recognize the (correct) error flow
between the list_for_each_entry() loop, up_dev and gmode.
Annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/wireless/b43/main.c b/drivers/net/wireless/b43/main.c
index 14c44df..bd70b27 100644
--- a/drivers/net/wireless/b43/main.c
+++ b/drivers/net/wireless/b43/main.c
@@ -3255,7 +3255,7 @@ static int b43_switch_band(struct b43_wl *wl, struct ieee80211_channel *chan)
struct b43_wldev *down_dev;
struct b43_wldev *d;
int err;
- bool gmode;
+ bool uninitialized_var(gmode);
int prev_status;
/* Find a device and PHY which supports the band. */
commit de965062d615afc885bd800976e488bd3c2d1d6f
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 10:37:58 2008 +0100
fix warning in drivers/net/wireless/ath9k/rc.c
this warning:
drivers/net/wireless/ath9k/rc.c: In function ‘ath_rc_rate_getidx’:
drivers/net/wireless/ath9k/rc.c:1033: warning: ‘nextindex’ may be used uninitialized in this function
triggers because GCC does not recognize the (correct) error flow
between ath_rc_get_nextlowervalid_txrate() and &nextindex.
Annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/wireless/ath9k/rc.c b/drivers/net/wireless/ath9k/rc.c
index cca2fc5..27b35d3 100644
--- a/drivers/net/wireless/ath9k/rc.c
+++ b/drivers/net/wireless/ath9k/rc.c
@@ -1030,7 +1030,7 @@ static u8 ath_rc_rate_getidx(struct ath_softc *sc,
u16 min_rate)
{
u32 j;
- u8 nextindex;
+ u8 uninitialized_var(nextindex);
struct ath_tx_ratectrl *rate_ctrl =
(struct ath_tx_ratectrl *)(ath_rc_priv);
commit e5e5492e7094ed4bea8daf0302f60fbb21daa4c7
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 14:52:20 2008 +0100
fix warning in drivers/net/wan/z85230.c
this warning:
drivers/net/wan/z85230.c: In function ‘z8530_interrupt’:
drivers/net/wan/z85230.c:713: warning: ‘intr’ may be used uninitialized in this function
is clearly bogus - annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/wan/z85230.c b/drivers/net/wan/z85230.c
index 5bf7e01..d7bf53a 100644
--- a/drivers/net/wan/z85230.c
+++ b/drivers/net/wan/z85230.c
@@ -710,7 +710,7 @@ EXPORT_SYMBOL(z8530_nop);
irqreturn_t z8530_interrupt(int irq, void *dev_id)
{
struct z8530_dev *dev=dev_id;
- u8 intr;
+ u8 uninitialized_var(intr);
static volatile int locker=0;
int work=0;
struct z8530_irqhandler *irqs;
commit 742cddc749294be3c0fa909e6255c5a338a1ac40
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 10:37:47 2008 +0100
fix error path in drivers/net/mlx4/en_rx.c
this warning:
drivers/net/mlx4/en_rx.c: In function ‘mlx4_en_activate_rx_rings’:
drivers/net/mlx4/en_rx.c:412: warning: ‘err’ may be used uninitialized in this function
Triggers because 'err' is uninitialized in the following input
conditions: priv->rx_ring_num is zero and mlx4_en_fill_rx_buffers()
fails.
But even if ->rx_ring_num is nonzero, 'err' will be zero if
mlx4_en_fill_rx_buffers() fails and mlx4_en_activate_rx_rings() returns
success - incorrectly.
So it's best to keep the error code uptodate on mlx4_en_fill_rx_buffers()
calls as well.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/mlx4/en_rx.c b/drivers/net/mlx4/en_rx.c
index 6232227..aed7739 100644
--- a/drivers/net/mlx4/en_rx.c
+++ b/drivers/net/mlx4/en_rx.c
@@ -443,7 +443,8 @@ int mlx4_en_activate_rx_rings(struct mlx4_en_priv *priv)
/* Fill Rx buffers */
ring->full = 0;
}
- if (mlx4_en_fill_rx_buffers(priv))
+ err = mlx4_en_fill_rx_buffers(priv);
+ if (err)
goto err_buffers;
for (ring_ind = 0; ring_ind < priv->rx_ring_num; ring_ind++) {
commit 181787d1c30af364ffd15428fe6fab7798c3f3c1
Author: Ingo Molnar <[email protected]>
Date: Tue Nov 25 10:37:35 2008 +0100
fix warning in drivers/net/cassini.c
this warning:
drivers/net/cassini.c: In function ‘cas_rx_ringN’:
drivers/net/cassini.c:2350: warning: ‘skb’ may be used uninitialized in this function
triggers because GCC does not recognize the (correct) error flow
between cas_rx_process_pkt() and 'skb'.
Annotate it.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/cassini.c b/drivers/net/cassini.c
index 86909cf..e083e1c 100644
--- a/drivers/net/cassini.c
+++ b/drivers/net/cassini.c
@@ -2347,7 +2347,7 @@ static int cas_rx_ringN(struct cas *cp, int ring, int budget)
drops = 0;
while (1) {
struct cas_rx_comp *rxc = rxcs + entry;
- struct sk_buff *skb;
+ struct sk_buff *uninitialized_var(skb);
int type, len;
u64 words[4];
int i, dring;
commit 96c79254264033f1505198d5d362b68d88d1bbaf
Author: Ingo Molnar <[email protected]>
Date: Fri Oct 17 12:41:30 2008 +0200
fix warning in net/dccp/options.c
this warning:
net/dccp/options.c: In function ‘dccp_parse_options’:
net/dccp/options.c:67: warning: ‘value’ may be used uninitialized in this function
is a bogus GCC warning. The compiler does not recognize the relation
between "value" and "mandatory" variables: the code flow can ever reach
the "out_invalid_option:" label if 'mandatory' is set to 1, and when
'mandatory' is non-zero, we'll always have 'value' initialized.
Help out the compiler by annotating the variable.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/net/dccp/options.c b/net/dccp/options.c
index 0809b63..18dcfb9 100644
--- a/net/dccp/options.c
+++ b/net/dccp/options.c
@@ -64,7 +64,7 @@ int dccp_parse_options(struct sock *sk, struct dccp_request_sock *dreq,
(dh->dccph_doff * 4);
struct dccp_options_received *opt_recv = &dp->dccps_options_received;
unsigned char opt, len;
- unsigned char *value;
+ unsigned char *uninitialized_var(value);
u32 elapsed_time;
__be32 opt_val;
int rc;
commit 193e56a10e56505dc7eac9e5632834c72b55613d
Author: Ingo Molnar <[email protected]>
Date: Mon Aug 18 16:18:34 2008 +0200
fix warning in drivers/net/mlx4/mcg.c
fix warning:
drivers/net/mlx4/mcg.c: In function ‘mlx4_multicast_attach’:
drivers/net/mlx4/mcg.c:217: warning: integer overflow in expression
there was no real danger of overflow here though.
md5:
db8eb55620f886c03854a2abb2ce6c3f mcg.o.before.asm
db8eb55620f886c03854a2abb2ce6c3f mcg.o.after.asm
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/mlx4/mcg.c b/drivers/net/mlx4/mcg.c
index 592c01a..224ef0f 100644
--- a/drivers/net/mlx4/mcg.c
+++ b/drivers/net/mlx4/mcg.c
@@ -215,7 +215,7 @@ int mlx4_multicast_attach(struct mlx4_dev *dev, struct mlx4_qp *qp, u8 gid[16],
if (block_mcast_loopback)
mgm->qp[members_count++] = cpu_to_be32((qp->qpn & MGM_QPN_MASK) |
- (1 << MGM_BLCK_LB_BIT));
+ (1U << MGM_BLCK_LB_BIT));
else
mgm->qp[members_count++] = cpu_to_be32(qp->qpn & MGM_QPN_MASK);
commit bcc35f75a3b9b96c53a74176b135304fc7ffe052
Author: Ingo Molnar <[email protected]>
Date: Fri Oct 17 15:49:15 2008 +0200
fix warning in net/ax25/sysctl_net_ax25.c
fix this warning:
net/ax25/sysctl_net_ax25.c:27: warning: ‘min_ds_timeout’ defined but not used
net/ax25/sysctl_net_ax25.c:27: warning: ‘max_ds_timeout’ defined but not used
These are only used in the CONFIG_AX25_DAMA_SLAVE case.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/net/ax25/sysctl_net_ax25.c b/net/ax25/sysctl_net_ax25.c
index f288fc4..735ceef 100644
--- a/net/ax25/sysctl_net_ax25.c
+++ b/net/ax25/sysctl_net_ax25.c
@@ -24,7 +24,9 @@ static int min_idle[1], max_idle[] = {65535000};
static int min_n2[] = {1}, max_n2[] = {31};
static int min_paclen[] = {1}, max_paclen[] = {512};
static int min_proto[1], max_proto[] = { AX25_PROTO_MAX };
+#ifdef CONFIG_AX25_DAMA_SLAVE
static int min_ds_timeout[1], max_ds_timeout[] = {65535000};
+#endif
static struct ctl_table_header *ax25_table_header;
commit 59681c6c30e3f700af8d716f66e41ad18f726b80
Author: Ingo Molnar <[email protected]>
Date: Mon Oct 20 18:50:11 2008 +0200
fix warning in net/sunrpc/xprtrdma/verbs.c
fix this warning:
net/sunrpc/xprtrdma/verbs.c: In function ‘rpcrdma_conn_upcall’:
net/sunrpc/xprtrdma/verbs.c:279: warning: unused variable ‘addr’
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/net/sunrpc/xprtrdma/verbs.c b/net/sunrpc/xprtrdma/verbs.c
index a5fef5e..900e595 100644
--- a/net/sunrpc/xprtrdma/verbs.c
+++ b/net/sunrpc/xprtrdma/verbs.c
@@ -276,7 +276,9 @@ rpcrdma_conn_upcall(struct rdma_cm_id *id, struct rdma_cm_event *event)
struct rpcrdma_xprt *xprt = id->context;
struct rpcrdma_ia *ia = &xprt->rx_ia;
struct rpcrdma_ep *ep = &xprt->rx_ep;
+#ifdef RPC_DEBUG
struct sockaddr_in *addr = (struct sockaddr_in *) &ep->rep_remote_addr;
+#endif
struct ib_qp_attr attr;
struct ib_qp_init_attr iattr;
int connstate = 0;
commit ad8d03b315507a954160fcf04dc94cef9f118e77
Author: Ingo Molnar <[email protected]>
Date: Mon Oct 20 17:34:02 2008 +0200
fix warning in net/bluetooth/rfcomm/sock.c
fix this warning:
net/bluetooth/rfcomm/sock.c: In function ‘rfcomm_sock_ioctl’:
net/bluetooth/rfcomm/sock.c:795: warning: unused variable ‘sk’
perhaps BT_DEBUG() should be improved to do printf format checking
instead of the #ifdef, but that looks quite intrusive: each bluetooth
.c file undefines the macro.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/net/bluetooth/rfcomm/sock.c b/net/bluetooth/rfcomm/sock.c
index 8a972b6..bc0d4a7 100644
--- a/net/bluetooth/rfcomm/sock.c
+++ b/net/bluetooth/rfcomm/sock.c
@@ -792,7 +792,9 @@ static int rfcomm_sock_getsockopt(struct socket *sock, int level, int optname, c
static int rfcomm_sock_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)
{
+#if defined(CONFIG_BT_RFCOMM_TTY) || defined(CONFIG_BT_RFCOMM_DEBUG)
struct sock *sk = sock->sk;
+#endif
int err;
BT_DBG("sk %p cmd %x arg %lx", sk, cmd, arg);
commit 320451a3f79098ca800f2e58674074c26efee020
Author: Ingo Molnar <[email protected]>
Date: Sat Oct 18 21:33:06 2008 +0200
fix warning in drivers/net/wireless/ray_cs.c
fix this warning:
drivers/net/wireless/ray_cs.c:297: warning: ‘rcsid’ defined but not used
'rcsid' is only used in the CONFIG_PROC_FS=y case.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/wireless/ray_cs.c b/drivers/net/wireless/ray_cs.c
index 1404a57..c23a811 100644
--- a/drivers/net/wireless/ray_cs.c
+++ b/drivers/net/wireless/ray_cs.c
@@ -294,7 +294,9 @@ static char hop_pattern_length[] = { 1,
JAPAN_TEST_HOP_MOD
};
+#ifdef CONFIG_PROC_FS
static char rcsid[] = "Raylink/WebGear wireless LAN - Corey <Thomas [email protected]>";
+#endif
/*=============================================================================
ray_attach() creates an "instance" of the driver, allocating
commit decc1656e870c9e5bd94f2357b8496b8da6c2e73
Author: Ingo Molnar <[email protected]>
Date: Sat Oct 18 19:14:40 2008 +0200
fix warning in net/netfilter/nf_conntrack_proto_sctp.c
fix this warning:
net/netfilter/nf_conntrack_proto_sctp.c: In function ‘sctp_packet’:
net/netfilter/nf_conntrack_proto_sctp.c:376: warning: array subscript is above array bounds
GCC noticed that new_state could in theory be SCTP_CONNTRACK_MAX,
which is beyond the limit of the sctp_timeouts[] array.
This situation could happen if for_each_sctp_chunk() exits immediately.
That should not normally happen - but it's not obvious to me and it's
better to insert an extra check to the end of the loop for this case.
This gets rid of warning as well.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/net/netfilter/nf_conntrack_proto_sctp.c b/net/netfilter/nf_conntrack_proto_sctp.c
index ae8c260..d31ced4 100644
--- a/net/netfilter/nf_conntrack_proto_sctp.c
+++ b/net/netfilter/nf_conntrack_proto_sctp.c
@@ -373,6 +373,9 @@ static int sctp_packet(struct nf_conn *ct,
}
write_unlock_bh(&sctp_lock);
+ if (new_state == SCTP_CONNTRACK_MAX)
+ goto out;
+
nf_ct_refresh_acct(ct, ctinfo, skb, sctp_timeouts[new_state]);
if (old_state == SCTP_CONNTRACK_COOKIE_ECHOED &&
commit b547d51a26c5ed23571df50c7254439e2f5dbebe
Author: Ingo Molnar <[email protected]>
Date: Sat Oct 18 19:02:37 2008 +0200
fix warning in net/netfilter/ipvs/ip_vs_ctl.c
Impact: fix potential kernel stack information exposure to user-space.
fix this warning:
net/netfilter/ipvs/ip_vs_ctl.c: In function ‘ip_vs_genl_set_cmd’:
net/netfilter/ipvs/ip_vs_ctl.c:3034: warning: ‘t.udp_timeout’ may be used uninitialized in this function
net/netfilter/ipvs/ip_vs_ctl.c:3034: note: ‘t.udp_timeout’ was declared here
GCC seems to be correct here: in the !CONFIG_IP_VS_PROTO_UDP case
__ip_vs_get_timeouts keeps the timeout structure uninitialized. But
in a couple of places we use and copy this structure to user-space,
potentially exposing kernel-stack data to user-space.
While it's a minor issue that only triggers with certain config variations,
it's best to always initialize this structure in __ip_vs_get_timeouts().
(GCC will eliminate this memset)
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/net/netfilter/ipvs/ip_vs_ctl.c b/net/netfilter/ipvs/ip_vs_ctl.c
index 0302cf3..7e3a0c7 100644
--- a/net/netfilter/ipvs/ip_vs_ctl.c
+++ b/net/netfilter/ipvs/ip_vs_ctl.c
@@ -2327,6 +2327,7 @@ __ip_vs_get_dest_entries(const struct ip_vs_get_dests *get,
static inline void
__ip_vs_get_timeouts(struct ip_vs_timeout_user *u)
{
+ memset(u, 0, sizeof(*u));
#ifdef CONFIG_IP_VS_PROTO_TCP
u->tcp_timeout =
ip_vs_protocol_tcp.timeout_table[IP_VS_TCP_S_ESTABLISHED] / HZ;
commit 2281d27663a5f17689ec49e029b4962cba19d614
Author: Ingo Molnar <[email protected]>
Date: Sat Oct 18 16:27:37 2008 +0200
fix warning in drivers/net/atlx/atl2.c
fix this warning:
drivers/net/atlx/atl2.c: In function ‘atl2_request_irq’:
drivers/net/atlx/atl2.c:644: warning: unused variable ‘err’
'err' is unused in the !CONFIG_PCI_MSI case.
Instead of further increasing the #ifdeffery in this function,
restructure the code a bit and get rid of the #ifdef. This
relies on the fact that pci_enable_msi() will always fail in
the !CONFIG_PCI_MSI case.
There should be no change in driver behavior.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/atlx/atl2.c b/drivers/net/atlx/atl2.c
index 8571e8c..e02bd7e 100644
--- a/drivers/net/atlx/atl2.c
+++ b/drivers/net/atlx/atl2.c
@@ -644,7 +644,6 @@ static int atl2_request_irq(struct atl2_adapter *adapter)
int flags, err = 0;
flags = IRQF_SHARED;
-#ifdef CONFIG_PCI_MSI
adapter->have_msi = true;
err = pci_enable_msi(adapter->pdev);
if (err)
@@ -652,7 +651,6 @@ static int atl2_request_irq(struct atl2_adapter *adapter)
if (adapter->have_msi)
flags &= ~IRQF_SHARED;
-#endif
return request_irq(adapter->pdev->irq, &atl2_intr, flags, netdev->name,
netdev);
commit b5a7c2b9e1a1b4d80da5d66b330545fdeafe1dad
Author: Ingo Molnar <[email protected]>
Date: Fri Oct 17 19:48:51 2008 +0200
fix warning in net/netfilter/nf_conntrack_proto_tcp.c
fix this warning:
net/netfilter/nf_conntrack_proto_tcp.c: In function ‘tcp_in_window’:
net/netfilter/nf_conntrack_proto_tcp.c:491: warning: unused variable ‘net’
net/netfilter/nf_conntrack_proto_tcp.c: In function ‘tcp_packet’:
net/netfilter/nf_conntrack_proto_tcp.c:812: warning: unused variable ‘net’
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/include/net/netfilter/nf_conntrack_l4proto.h b/include/net/netfilter/nf_conntrack_l4proto.h
index 7f2f43c..debdaf7 100644
--- a/include/net/netfilter/nf_conntrack_l4proto.h
+++ b/include/net/netfilter/nf_conntrack_l4proto.h
@@ -129,7 +129,7 @@ extern const struct nla_policy nf_ct_port_nla_policy[];
&& net_ratelimit())
#endif
#else
-#define LOG_INVALID(net, proto) 0
+static inline int LOG_INVALID(struct net *net, int proto) { return 0; }
#endif /* CONFIG_SYSCTL */
#endif /*_NF_CONNTRACK_PROTOCOL_H*/
commit 822c18db871bc88b2a60d7a9618ec79a8b5e1ef5
Author: Ingo Molnar <[email protected]>
Date: Fri Oct 17 18:31:33 2008 +0200
fix warning in drivers/net/wireless/iwlwifi/iwl-scan.c
fix this warning:
drivers/net/wireless/iwlwifi/iwl-scan.c:92: warning: ‘iwl_escape_essid’ defined but not used
Convert the IWL debug macros to proper printf format checking.
Besides better type checking, this gets rid of the warning as well.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/wireless/iwlwifi/iwl-debug.h b/drivers/net/wireless/iwlwifi/iwl-debug.h
index e548d67..ddc1fbf 100644
--- a/drivers/net/wireless/iwlwifi/iwl-debug.h
+++ b/drivers/net/wireless/iwlwifi/iwl-debug.h
@@ -68,8 +68,10 @@ void iwl_dbgfs_unregister(struct iwl_priv *priv);
#endif
#else
-#define IWL_DEBUG(level, fmt, args...)
-#define IWL_DEBUG_LIMIT(level, fmt, args...)
+static inline void __attribute__ ((format(printf, 2, 3)))
+IWL_DEBUG(int level, const char *fmt, ...) { }
+static inline void __attribute__ ((format(printf, 2, 3)))
+IWL_DEBUG_LIMIT(int level, const char *fmt, ...) { }
#endif /* CONFIG_IWLWIFI_DEBUG */
diff --git a/drivers/net/wireless/iwlwifi/iwl-led.c b/drivers/net/wireless/iwlwifi/iwl-led.c
index 4eee1b1..f4f0664 100644
--- a/drivers/net/wireless/iwlwifi/iwl-led.c
+++ b/drivers/net/wireless/iwlwifi/iwl-led.c
@@ -43,7 +43,6 @@
#include "iwl-io.h"
#include "iwl-helpers.h"
-#ifdef CONFIG_IWLWIFI_DEBUG
static const char *led_type_str[] = {
__stringify(IWL_LED_TRG_TX),
__stringify(IWL_LED_TRG_RX),
@@ -51,7 +50,6 @@ static const char *led_type_str[] = {
__stringify(IWL_LED_TRG_RADIO),
NULL
};
-#endif /* CONFIG_IWLWIFI_DEBUG */
static const struct {
commit 69f38f617c3f832e088be9d35242d929aa623fd4
Author: Ingo Molnar <[email protected]>
Date: Wed Aug 27 19:06:48 2008 +0200
fix warning in drivers/net/depca.c
fix warning:
drivers/net/depca.c: In function ‘depca_eisa_probe’:
drivers/net/depca.c:1564: warning: ‘mem_start’ may be used uninitialized in this function
this seems to be a real bug - depca_eisa_probe() does not check
for failure. Add it, symmetric to depca_isa_probe().
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/depca.c b/drivers/net/depca.c
index ace39ec..f00bd64 100644
--- a/drivers/net/depca.c
+++ b/drivers/net/depca.c
@@ -1558,6 +1558,7 @@ static int __init depca_isa_probe (struct platform_device *device)
#ifdef CONFIG_EISA
static int __init depca_eisa_probe (struct device *device)
{
+ enum depca_type adapter = unknown;
struct eisa_device *edev;
struct net_device *dev;
struct depca_private *lp;
@@ -1576,7 +1577,11 @@ static int __init depca_eisa_probe (struct device *device)
* the EISA configuration structures (yet... :-), just rely on
* the ISA probing to sort it out... */
- depca_shmem_probe (&mem_start);
+ adapter = depca_shmem_probe (&mem_start);
+ if (adapter == unknown) {
+ status = -ENODEV;
+ goto out_free;
+ }
dev->base_addr = ioaddr;
dev->irq = irq;
commit 557073e3dac7b8479df69c140a5fd2b8f4a9cfad
Author: Ingo Molnar <[email protected]>
Date: Wed Aug 27 19:05:19 2008 +0200
fix warning in drivers/net/3c523.c
fix warning:
drivers/net/3c523.c:582: warning: ‘cleanup_card’ defined but not used
No code changed:
md5:
ebe4a1b27d3f21b0b12a78c58463b0d7 3c523.o.before.asm
ebe4a1b27d3f21b0b12a78c58463b0d7 3c523.o.after.asm
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/3c523.c b/drivers/net/3c523.c
index e2ce41d..7b9d427 100644
--- a/drivers/net/3c523.c
+++ b/drivers/net/3c523.c
@@ -578,13 +578,13 @@ err_out:
return retval;
}
+#ifdef MODULE
static void cleanup_card(struct net_device *dev)
{
mca_set_adapter_procfn(((struct priv *) (dev->priv))->slot, NULL, NULL);
release_region(dev->base_addr, ELMC_IO_EXTENT);
}
-
-#ifndef MODULE
+#else
struct net_device * __init elmc_probe(int unit)
{
struct net_device *dev = alloc_etherdev(sizeof(struct priv));
commit d138e44e2f8d0f26e04b0386d328d9cc7e33d82e
Author: Ingo Molnar <[email protected]>
Date: Fri Oct 17 14:30:37 2008 +0200
drivers/net/wireless/b43/phy_g.c: type check debug printouts
this warning:
drivers/net/wireless/b43/phy_g.c: In function ‘b43_gphy_op_recalc_txpower’:
drivers/net/wireless/b43/phy_g.c:3191: warning: unused variable ‘dbm’
is caused because b43dbg() is a macro, hence in the !B43_DEBUG
case the compiler does not know that the variables are used.
Convert it to a proper inline instead. This also improves type checking
in the !B43_DEBUG case.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/wireless/b43/b43.h b/drivers/net/wireless/b43/b43.h
index 427b820..ac55b62 100644
--- a/drivers/net/wireless/b43/b43.h
+++ b/drivers/net/wireless/b43/b43.h
@@ -853,7 +853,8 @@ void b43warn(struct b43_wl *wl, const char *fmt, ...)
void b43dbg(struct b43_wl *wl, const char *fmt, ...)
__attribute__ ((format(printf, 2, 3)));
#else /* DEBUG */
-# define b43dbg(wl, fmt...) do { /* nothing */ } while (0)
+static inline void __attribute__ ((format(printf, 2, 3)))
+b43dbg(struct b43_wl *wl, const char *fmt, ...) { }
#endif /* DEBUG */
/* A WARN_ON variant that vanishes when b43 debugging is disabled.
commit 7e4f7889aca92f2cf95ba31e005fc1368c1cbc5a
Author: Ingo Molnar <[email protected]>
Date: Fri Oct 17 17:47:04 2008 +0200
fix warning in drivers/net/sky2.c
fix this warning:
drivers/net/sky2.c: In function ‘sky2_mhz’:
drivers/net/sky2.c:2753: warning: control reaches end of non-void function
BUG() can be NOP on !CONFIG_BUG.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/drivers/net/sky2.c b/drivers/net/sky2.c
index 3813d15..85fdacc 100644
--- a/drivers/net/sky2.c
+++ b/drivers/net/sky2.c
@@ -2748,7 +2748,7 @@ static u32 sky2_mhz(const struct sky2_hw *hw)
return 156;
default:
- BUG();
+ panic("sky2_mhz: unknown chip id!");
}
}
commit deb797f96d920083ba8a0bcfa6f57b19fc593571
Author: Ingo Molnar <[email protected]>
Date: Fri Oct 17 17:34:27 2008 +0200
fix warning in net/sunrpc/svcauth_unix.c
fix this warning:
net/sunrpc/svcauth_unix.c: In function ‘svcauth_unix_set_client’:
net/sunrpc/svcauth_unix.c:681: warning: ‘sin6’ may be used uninitialized in this function
caused by BUG() being a NOP on !CONFIG_BUG.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/net/sunrpc/svcauth_unix.c b/net/sunrpc/svcauth_unix.c
index f24800f..200ed93 100644
--- a/net/sunrpc/svcauth_unix.c
+++ b/net/sunrpc/svcauth_unix.c
@@ -692,7 +692,7 @@ svcauth_unix_set_client(struct svc_rqst *rqstp)
sin6 = svc_addr_in6(rqstp);
break;
default:
- BUG();
+ panic("svcauth_unix_set_client: bad address family!");
}
rqstp->rq_client = NULL;
@@ -873,3 +873,4 @@ struct auth_ops svcauth_unix = {
.set_client = svcauth_unix_set_client,
};
+
commit 923a5a960577dcc5189b6113300a122c0212b292
Author: Ingo Molnar <[email protected]>
Date: Fri Oct 17 16:55:34 2008 +0200
fix warning in security/selinux/netnode.c
fix this warning:
security/selinux/netnode.c: In function ‘sel_netnode_find’:
security/selinux/netnode.c:131: warning: ‘idx’ may be used uninitialized in this function
BUG() can be a NOP in the !CONFIG_BUG case. Return NULL in that case,
instead of running in an undefined way.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/security/selinux/netnode.c b/security/selinux/netnode.c
index 7100072..a307122 100644
--- a/security/selinux/netnode.c
+++ b/security/selinux/netnode.c
@@ -140,6 +140,7 @@ static struct sel_netnode *sel_netnode_find(const void *addr, u16 family)
break;
default:
BUG();
+ return NULL;
}
list_for_each_entry_rcu(node, &sel_netnode_hash[idx].list, list)
commit c62cda5dfaf7630edefe097a3e23af7732060b08
Author: Ingo Molnar <[email protected]>
Date: Fri Oct 17 16:26:02 2008 +0200
fix warning in net/packet/af_packet.c
fix this warning:
net/packet/af_packet.c: In function ‘tpacket_rcv’:
net/packet/af_packet.c:604: warning: ‘hdrlen’ may be used uninitialized in this function
This is because in the !CONFIG_BUG case BUG() is a NOP.
There is no other sane way to continue but to panic the system
unconditionally, regardless of the CONFIG_BUG setting.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index c718e7e..1478fb8 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -709,7 +709,7 @@ static int tpacket_rcv(struct sk_buff *skb, struct net_device *dev, struct packe
hdrlen = sizeof(*h.h2);
break;
default:
- BUG();
+ panic("AF_PACKET: bad tp->version");
}
sll = h.raw + TPACKET_ALIGN(hdrlen);
commit d581867031ebe3c6f4e018ea32ed8b9ee5141ead
Author: Ingo Molnar <[email protected]>
Date: Fri Oct 17 16:15:59 2008 +0200
fix warning in net/rfkill/rfkill.c
fix this warning:
net/rfkill/rfkill.c: In function ‘rfkill_get_type_str’:
net/rfkill/rfkill.c:393: warning: control reaches end of non-void function
in the embedded !CONFIG_BUG case BUG() is a NOP.
Return NULL to solve this (unlikely) scenario. All callers feed this
pointer into sprintf %s so NULL should be fine.
Signed-off-by: Ingo Molnar <[email protected]>
diff --git a/net/rfkill/rfkill.c b/net/rfkill/rfkill.c
index f949a48..eecb9a8 100644
--- a/net/rfkill/rfkill.c
+++ b/net/rfkill/rfkill.c
@@ -389,6 +389,7 @@ static const char *rfkill_get_type_str(enum rfkill_type type)
return "wwan";
default:
BUG();
+ return NULL;
}
}
I already applied these patches, see my reply in that thread.
I was saying to submit these (individually, please!) next time
with netdev CC:'d.
Posting a ton all at once is pointless, they all get logged to
patchwork as one big blog and I cannot annotate them individually
and give you proper feedback.
* David Miller <[email protected]> wrote:
> > fix warning in drivers/net/sky2.c
> > fix warning in net/sunrpc/svcauth_unix.c
> > fix warning in security/selinux/netnode.c
> > fix warning in net/packet/af_packet.c
> > fix warning in net/rfkill/rfkill.c
>
> Not applied. Changing this from a BUG() to a panic() is wrong.
> BUG() kills the current execution context, and allows the system to
> potentially continue. Panic() takes down the entire machine.
Yeah, although in practice a BUG() in some of those places will take
down the whole machine:
> > fix warning in net/packet/af_packet.c
That's packet receive path, holding spinlocks or in other atomic
context (irq/softirq)? A BUG() will take down the machine as we will
crash and try to schedule in atomic context.
> The !CONFIG_BUG BUG() definition should be marked in such a way that
> gcc thinks it is noreturn. Otherwise BUG() isn't really BUG()
> anymore. The other option is to never allow BUG to be disabled or
> have it always evaluate to a simple bug trap.
Yeah, but note that that's exactly the current behavior (and intent)
of !CONFIG_BUG: it is not doing anything (it's just an empty macro
which returns!), and we get those build warnings because the control
flow becomes undefined. We cannot mark it noreturn because it does
return.
Ingo
Hi Linus:
At this late stage there is just one fix for a regression from
2.6.27. We used to allow algorithms to be built-in with the new
testing framework built as a module. This meant that the built-in
algorithms may not get tested properly if modules can't be loaded
when the built-in algorithms are registered. It also seems to
trigger a modprobe loop on one distro. This configuration is now
forbidden.
Please pull from
git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6.git
or
master.kernel.org:/pub/scm/linux/kernel/git/herbert/crypto-2.6.git
Herbert Xu (1):
crypto: api - Disallow cryptomgr as a module if algorithms are built-in
crypto/Kconfig | 36 ++++++++++++++++++++++++++++++++----
crypto/Makefile | 18 +++++++++---------
2 files changed, 41 insertions(+), 13 deletions(-)
Thanks,
--
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
Hi Linus:
A regression has been reported where the new algorithm testing
infrastructure may cause the optimised versions of AES to fail
when it's built into the kernel (as opposed to as a module).
This is due to its use of shared AES tables that may not have
been initialised yet. This has been fixed by pregenerating
the tables instead of having them computed at boot/run-time.
Please pull from
git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6.git
or
master.kernel.org:/pub/scm/linux/kernel/git/herbert/crypto-2.6.git
Herbert Xu (1):
crypto: aes - Precompute tables
crypto/aes_generic.c | 1145 ++++++++++++++++++++++++++++++++++++++++++++++----
include/crypto/aes.h | 8 +-
2 files changed, 1059 insertions(+), 94 deletions(-)
Thanks,
--
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
On Thu, 25 Dec 2008, Herbert Xu wrote:
>
> A regression has been reported where the new algorithm testing
> infrastructure may cause the optimised versions of AES to fail
> when it's built into the kernel (as opposed to as a module).
Can't we just get rid of that stupid testing infrastructure, or make it a
module-only thing? I refuse to pull a 1000+ line patch for something like
this, especially this late. It's insane.
Linus
On Wed, Dec 24, 2008 at 02:32:28PM -0800, Linus Torvalds wrote:
>
> Can't we just get rid of that stupid testing infrastructure, or make it a
> module-only thing? I refuse to pull a 1000+ line patch for something like
> this, especially this late. It's insane.
Well the testing infrastructure simply flagged a problem that
was always there. Previously if something used AES early enough
it would silently give the wrong result instead. Of course this
is fairly unlikely since the number of AES users is limited.
Now that the testing infrastructure is available, we could just
do nothing for 2.6.28 since the C AES version will always be there
and work correctly.
Thanks,
--
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <[email protected]>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt