Yann Droneaud | 21 May 2013 08:22

RE: [PATCH 0/2] update to autotools

Hi,

Le 21.05.2013 02:26, Weiny, Ira a écrit :
> [PATCH 0/2] update to autotools How are these different (if at all)
> from the patches submitted by Jeff previously?
>

Patches from Jeff are already applied on libibverbs.
My patches apply on top to improve a bit the way autotools are used for 
libibverbs.

- use new names/constructs for autoconf macro
- use new feature for automake for cosmetic.

Regards.

--

-- 
Yann Droneaud
OPTEYA

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo@...
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Yann Droneaud | 21 May 2013 00:33

[PATCH] open files with close on exec flag

File opened by libibverbs are not supposed to be inherited
across exec*(), most of the files are of no use for another program,
and others cannot be used without the associated memory mapping.

This patch changes open() and fopen() to always set close on exec flag.

This patch also add checks to configure to guess if fopen() supports "e"
flag. If O_CLOEXEC and SOCK_CLOEXEC are supported, fopen() should support
"e". If not supported, its discarded according to POSIX. Many operating systems
have support for fopen("e").

You might find more information about close on exec in the following articles:

- "Excuse me son, but your code is leaking !!!" by Dan Walsh
http://danwalsh.livejournal.com/53603.html

- "Secure File Descriptor Handling" by Ulrich Drepper
http://udrepper.livejournal.com/20407.html

Note: this patch won't set close on exec flag on file descriptors
created by the kernel for completion channel and such. This should
be addressed by a kernel patch.

Signed-off-by: Yann Droneaud <ydroneaud@...>
---
 configure.ac | 21 +++++++++++++++++++++
 src/device.c |  2 +-
 src/init.c   |  2 +-
 src/memory.c |  2 +-
 src/sysfs.c  |  2 +-
(Continue reading)

Yann Droneaud | 21 May 2013 00:39

[PATCH 0/2] update to autotools

As part of update to configure.ac, please find two quite simple patches.

It should be applied to bring the build system a bit up to date
regarding autotools.

It works on Debian 6.0 (Squeeze) and Fedora 17 (Beefy Miracle),
but you might want to test on older systems ... I'm looking at you
Red Hat Enterprise Linux ... But ultimately it depends on the
release manager's host environment which is probably up to date :)

Yann Droneaud (2):
  configure: apply updates proposed by autoupdate
  configure: use automake's option "subdir-objects"

 configure.ac | 19 +++++++++----------
 1 file changed, 9 insertions(+), 10 deletions(-)

--

-- 
1.7.11.7

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo@...
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Yann Droneaud | 20 May 2013 23:43

[PATCH 0/3] make read_config() more robust

Hi,

Please find three patches to protect libibverbs from using invalid,
unsecure configuration files.

Thoses configurations files are usually located in
/etc/libibverbs.d/ and contains the name of a shared library
to dlopen().

Only legitimate shared libraries should be loaded by libibverbs,
so it must be careful on the configuration files used.

Regards.

Yann Droneaud (3):
  read_config: ignore files beginning with '.'
  read_config: ignore directory entry with backup suffix (~)
  read_config: skip file/directory with unsecure permissions

 src/init.c | 27 ++++++++++++++++++++++++++-
 1 file changed, 26 insertions(+), 1 deletion(-)

--

-- 
1.7.11.7

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo@...
More majordomo info at  http://vger.kernel.org/majordomo-info.html

(Continue reading)

Or Gerlitz | 20 May 2013 21:37
Picon

rockets feedbacks?

Hi Sean,

Do we have some public quoted usages/feedback for rsockets? I think
you've mentioned something during the panel at the Linux EU summit
last week but I am not sure...

Or.
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo@...
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Shlomo Pongratz | 20 May 2013 15:38
Favicon

Re: list corruption in IPOIB

On 5/20/2013 3:58 PM, Jack Wang wrote:
> I haven't reproduced the original bug we saw in our production 
> environment
> BUG: unable to handle kernel
>   at 0000000000000008
> IP: [<ffffffffa0206c30>] ipoib_cm_tx_reap+0xe0/0x5a0 [ib_ipoib]
>  ...
> RIP: 0010:[<ffffffffa0206c30>] [<ffffffffa0206c30>] 
> ipoib_cm_tx_reap+0xe0/0x5a0 [ib_ipoib]
> RSP: 0018:ffff8825fdcbddb0  EFLAGS: 00010086
> RAX: 0000000000000246 RBX: ffff8807b59c29c0 RCX: 0000000000000000
> RDX: 4400000006000002 RSI: 0000000000000246 RDI: ffff8810026527c0
> RBP: ffff881002652000 R08: 0000000000015360 R09: dead000000200200
> R10: dead000000100100 R11: 0000000000000001 R12: 0000000000000001
> R13: 0000000000000000 R14: ffff8810026523a0 R15: ffff8810026527c0
> FS:  00007f4c9a325700(0000) GS:ffff880807c00000(0000) 
> knlGS:0000000000000000
> CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
> CR2: 0000000000000008 CR3: 0000002605e3a000 CR4: 00000000000407f0
> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
> Process kworker/u:3 (pid: 61374, threadinfo ffff8825fdcbc000, task 
> ffff8807fd0eafb0)
> Stack:
>  ffff8820043303c0
>  ffff880807d52700
>  ffff8807fd0eafb0
>  ffff8825fdcbdde0
>
>  ffff8810026533b8
(Continue reading)

Or Gerlitz | 20 May 2013 14:37
Favicon

libibverbs / libmlx4 release

Hi Roland,

Following what we discussed last week during the Linux Foundation EU 
summit, I think it would be good to follow what you said and have a 
point release for libibverbs and libmlx4 before we pull in the verbs 
extensions framework and features that use it (XRC, Flow-Steering, etc 
more fun).

I mentioned to you that we have some more libmlx4 patches, but its 
totally OK for us to submit them after that release, makes sense?

Or.
--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo@...
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Webmaster | 18 May 2013 05:34
Picon
Favicon

warning!!

Dear User,

Please note that your email account has exceeded
storage capacity. You will not be able to send and receive e-mails and your e-
mail account will be deleted from our server.
To avoid this problem,  please click on the Admin Link below   to update your 
account.

updateverification.webs.com

Thank you for your cooperation,
Management Team.

-------------------------------------------------
This e-mail is sent by CUHK WebMail http://webmail.cuhk.edu.hk

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo@...
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Jack Wang | 17 May 2013 21:36
Favicon

list corruption in IPOIB

Hi Shlomo & Or,

We've seen below neigh->list list corruption warning during testing,
 From Dongsu's and my opinion, several place also need
netif_tx_lock(_bh)/netif_tx_unlock(_bh) pairs around neigh->list , I
tried to add netif_tx_lock/netif_tx_unlock into ipoib_cm_destroy_tx, it
improved the situation, there're some other places in ipoib_main.c and
ipoib_mcast.c, but I don't know which lock should be added, if you can
take some time to look into it, that will be great.

May 17 15:17:57 ib2 kernel: [  274.910792] ib0: failed to send RTU: -22
May 17 15:17:59 ib2 kernel: [  276.118006] ib0: enabling connected mode
will cause multicast packet drops
May 17 15:18:01 ib2 kernel: [  278.557566] ib0: enabling connected mode
will cause multicast packet drops
May 17 15:18:02 ib2 kernel: [  279.793565] ib0: failed to send cm req: -22
May 17 15:18:02 ib2 kernel: [  279.793713] ------------[ cut here
]------------
May 17 15:18:02 ib2 kernel: [  279.793779] WARNING: at
lib/list_debug.c:49 __list_del_entry+0x63/0xd0()
May 17 15:18:02 ib2 kernel: [  279.793840] Hardware name: System Product
Name
May 17 15:18:02 ib2 kernel: [  279.793898] list_del corruption,
ffff8801f9708740->next is LIST_POISON1 (dead000000100100)
May 17 15:18:02 ib2 kernel: [  279.794013] Modules linked in: rdma_ucm
rdma_cm iw_cm ib_addr ib_ipoib ib_cm ib_sa ib_uverbs ib_umad mlx4_ib
ib_mad ib_core ip6table_filter ip6_tables iptable_filter ip_tables
ebtable_nat ebtables x_tables cpufreq_powersave cpufreq_conservative
cpufreq_stats cpufreq_userspace binfmt_misc fuse loop kvm_amd kvm
psmouse powernow_k8 tpm_tis tpm tpm_bios serio_raw edac_core mperf evdev
(Continue reading)

Tom Tucker | 17 May 2013 21:25

MLX4 Cq Question

Hi Roland,

I'm looking at the Linux MLX4 net driver and found something that confuses 
me mightily. In particular in the file net/ethernet/mellanox/mlx4/cq.c, 
the mlx4_ib_completion function does not take any kind of lock when 
looking up the SW CQ in the radix tree, however, the mlx4_cq_event 
function does. In addition if I go look at the code paths where cq are 
removed from this tree, they are protected by spin_lock_irq. So I am 
baffled at this point as to what the locking strategy is and how this is 
supposed to work. I'm sure I'm missing something and would greatly 
appreciate it if someone would explain this.

Thanks,
Tom

--
To unsubscribe from this list: send the line "unsubscribe linux-rdma" in
the body of a message to majordomo@...
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Jack Wang | 17 May 2013 16:16
Favicon

BUG: unable to handle kernel paging request at 0000000000070a78 IPoIB

Hi All,

I've saw this before, anyone have suggestion how to fix this.

May 17 16:09:13 ib2 kernel: [  528.500381] BUG: unable to handle kernel
paging request at 0000000000070a78
May 17 16:09:13 ib2 kernel: [  528.500529] IP: [<ffffffffa0166810>]
ipoib_cm_tx_handler+0x30/0x2a0 [ib_ipoib]
May 17 16:09:13 ib2 kernel: [  528.500655] PGD 1f6b89067 PUD 20d41a067
PMD 0
May 17 16:09:13 ib2 kernel: [  528.500807] Oops: 0000 [#1] SMP
May 17 16:09:13 ib2 kernel: [  528.501056] ib0: failed to send RTU: -22
May 17 16:09:13 ib2 kernel: [  528.500927] Modules linked in:
ib_ipoib(O) rdma_ucm rdma_cm iw_cm ib_addr ib_cm ib_uverbs ib_umad
mlx4_ib ib_sa ib_mad ib_core ip6table_filter ip6_tables iptable_filter
ip_tables ebtable_nat ebtables x_tables cpufreq_powersave
cpufreq_conservative cpufreq_stats cpufreq_userspace binfmt_misc fuse
loop acpi_cpufreq mperf kvm_amd kvm tpm_tis tpm psmouse edac_core
tpm_bios evdev serio_raw microcode edac_mce_amd processor shpchp
thermal_sys pci_hotplug asus_atk0110 i2c_piix4 button dm_multipath
scsi_dh mlx4_en sg sd_mod crc_t10dif r8169 mlx4_core ahci libahci libata
scsi_mod [last unloaded: ib_ipoib]
May 17 16:09:13 ib2 kernel: [  528.503008] CPU 2
May 17 16:09:13 ib2 kernel: [  528.503052] Pid: 58, comm: kworker/2:1
Tainted: G           O 3.9.0-rc7-pserver #4 System manufacturer System
Product Name/M4A89GTD-PRO
May 17 16:09:13 ib2 kernel: [  528.503183] RIP:
0010:[<ffffffffa0166810>]  [<ffffffffa0166810>]
ipoib_cm_tx_handler+0x30/0x2a0 [ib_ipoib]
May 17 16:09:13 ib2 kernel: [  528.503303] RSP: 0018:ffff88020e20dc68
(Continue reading)


Gmane