From: Max Gurtovoy <maxg@mellanox.com>
To: Greg KH <gregkh@linuxfoundation.org>
Cc: matanb@mellanox.com, leon@leon.nu, sagi@grimberg.me,
linux-rdma@vger.kernel.org, stable@vger.kernel.org,
robert@leblancnet.us
Subject: Re: [PATCH] IB/core: Fix bit curruption in ib_device_cap_flags structure
Date: Thu, 2 Jun 2016 19:41:10 +0300 [thread overview]
Message-ID: <575061A6.1060302@mellanox.com> (raw)
In-Reply-To: <20160602162426.GC26699@kroah.com>
On 6/2/2016 7:24 PM, Greg KH wrote:
> On Thu, Jun 02, 2016 at 12:28:05PM +0300, Max Gurtovoy wrote:
>> ib_device_cap_flags 64-bit expansion caused caps overlapping
>> and made consumers read wrong device capabilities. For example
>> IB_DEVICE_SG_GAPS_REG was falsely read by the iser driver causing
>> it to use a non-existing capability. This happened because signed
>> int becomes sign extended when converted it to u64. Fix this by
>> casting IB_DEVICE_ON_DEMAND_PAGING enumeration to ULL.
>>
>> Fixes: fb532d6a79b9 ('IB/{core, ulp} Support above 32 possible device capability flags')
>> Reported-by: Robert LeBlanc <robert@leblancnet.us>
>> Cc: Stable <stable@vger.kernel.org> #[v4.6+]
>> Acked-by: Sagi Grimberg <sagi@grimberg.me>
>> Signed-off-by: Max Gurtovoy <maxg@mellanox.com>
>> Signed-off-by: Matan Barak <matanb@mellanox.com>
>> ---
>> include/rdma/ib_verbs.h | 2 +-
>> 1 files changed, 1 insertions(+), 1 deletions(-)
>>
>> diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h
>> index 432bed5..c97357b 100644
>> --- a/include/rdma/ib_verbs.h
>> +++ b/include/rdma/ib_verbs.h
>> @@ -217,7 +217,7 @@ enum ib_device_cap_flags {
>> IB_DEVICE_CROSS_CHANNEL = (1 << 27),
>> IB_DEVICE_MANAGED_FLOW_STEERING = (1 << 29),
>> IB_DEVICE_SIGNATURE_HANDOVER = (1 << 30),
>> - IB_DEVICE_ON_DEMAND_PAGING = (1 << 31),
>> + IB_DEVICE_ON_DEMAND_PAGING = (1ULL << 31),
>> IB_DEVICE_SG_GAPS_REG = (1ULL << 32),
>> IB_DEVICE_VIRTUAL_FUNCTION = ((u64)1 << 33),
>> IB_DEVICE_RAW_SCATTER_FCS = ((u64)1 << 34),
>
> Why not just use the BIT() and BIT_ULL() macros instead of "open coding"
> these?
Good idea. do you think it will be a good idea to set all the elements
in this ib_device_cap_flags enum to BIT_ULL() in this patch ? to have 1
style instead of many ?
>
> thanks,
>
> greg k-h
>
WARNING: multiple messages have this Message-ID (diff)
From: Max Gurtovoy <maxg@mellanox.com>
To: Greg KH <gregkh@linuxfoundation.org>
Cc: <matanb@mellanox.com>, <leon@leon.nu>, <sagi@grimberg.me>,
<linux-rdma@vger.kernel.org>, <stable@vger.kernel.org>,
<robert@leblancnet.us>
Subject: Re: [PATCH] IB/core: Fix bit curruption in ib_device_cap_flags structure
Date: Thu, 2 Jun 2016 19:41:10 +0300 [thread overview]
Message-ID: <575061A6.1060302@mellanox.com> (raw)
In-Reply-To: <20160602162426.GC26699@kroah.com>
On 6/2/2016 7:24 PM, Greg KH wrote:
> On Thu, Jun 02, 2016 at 12:28:05PM +0300, Max Gurtovoy wrote:
>> ib_device_cap_flags 64-bit expansion caused caps overlapping
>> and made consumers read wrong device capabilities. For example
>> IB_DEVICE_SG_GAPS_REG was falsely read by the iser driver causing
>> it to use a non-existing capability. This happened because signed
>> int becomes sign extended when converted it to u64. Fix this by
>> casting IB_DEVICE_ON_DEMAND_PAGING enumeration to ULL.
>>
>> Fixes: fb532d6a79b9 ('IB/{core, ulp} Support above 32 possible device capability flags')
>> Reported-by: Robert LeBlanc <robert@leblancnet.us>
>> Cc: Stable <stable@vger.kernel.org> #[v4.6+]
>> Acked-by: Sagi Grimberg <sagi@grimberg.me>
>> Signed-off-by: Max Gurtovoy <maxg@mellanox.com>
>> Signed-off-by: Matan Barak <matanb@mellanox.com>
>> ---
>> include/rdma/ib_verbs.h | 2 +-
>> 1 files changed, 1 insertions(+), 1 deletions(-)
>>
>> diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h
>> index 432bed5..c97357b 100644
>> --- a/include/rdma/ib_verbs.h
>> +++ b/include/rdma/ib_verbs.h
>> @@ -217,7 +217,7 @@ enum ib_device_cap_flags {
>> IB_DEVICE_CROSS_CHANNEL = (1 << 27),
>> IB_DEVICE_MANAGED_FLOW_STEERING = (1 << 29),
>> IB_DEVICE_SIGNATURE_HANDOVER = (1 << 30),
>> - IB_DEVICE_ON_DEMAND_PAGING = (1 << 31),
>> + IB_DEVICE_ON_DEMAND_PAGING = (1ULL << 31),
>> IB_DEVICE_SG_GAPS_REG = (1ULL << 32),
>> IB_DEVICE_VIRTUAL_FUNCTION = ((u64)1 << 33),
>> IB_DEVICE_RAW_SCATTER_FCS = ((u64)1 << 34),
>
> Why not just use the BIT() and BIT_ULL() macros instead of "open coding"
> these?
Good idea. do you think it will be a good idea to set all the elements
in this ib_device_cap_flags enum to BIT_ULL() in this patch ? to have 1
style instead of many ?
>
> thanks,
>
> greg k-h
>
next prev parent reply other threads:[~2016-06-02 16:41 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-06-02 9:28 [PATCH] IB/core: Fix bit curruption in ib_device_cap_flags structure Max Gurtovoy
2016-06-02 9:28 ` Max Gurtovoy
2016-06-02 11:33 ` Christoph Hellwig
[not found] ` <20160602113353.GA18494-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>
2016-06-02 11:48 ` Max Gurtovoy
2016-06-02 11:48 ` Max Gurtovoy
2016-06-02 11:56 ` Sagi Grimberg
2016-06-02 16:51 ` Jason Gunthorpe
[not found] ` <1464859685-18666-1-git-send-email-maxg-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
2016-06-02 16:24 ` Greg KH
2016-06-02 16:24 ` Greg KH
2016-06-02 16:41 ` Max Gurtovoy [this message]
2016-06-02 16:41 ` Max Gurtovoy
[not found] ` <575061A6.1060302-VPRAkNaXOzVWk0Htik3J/w@public.gmane.org>
2016-06-02 16:52 ` Greg KH
2016-06-02 16:52 ` Greg KH
[not found] ` <20160602162426.GC26699-U8xfFu+wG4EAvxtiuMwx3w@public.gmane.org>
2016-06-02 16:41 ` Bart Van Assche
2016-06-02 16:41 ` Bart Van Assche
[not found] ` <c72fac21-0a3d-fc40-ec82-e156f29c2fab-XdAiOPVOjttBDgjK7y7TUQ@public.gmane.org>
2016-06-02 16:52 ` Greg KH
2016-06-02 16:52 ` Greg KH
2016-06-02 16:48 ` Christoph Lameter
2016-06-02 16:48 ` Christoph Lameter
2016-06-03 10:47 ` Leon Romanovsky
2016-06-03 10:47 ` Leon Romanovsky
2016-06-02 16:52 ` Jason Gunthorpe
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=575061A6.1060302@mellanox.com \
--to=maxg@mellanox.com \
--cc=gregkh@linuxfoundation.org \
--cc=leon@leon.nu \
--cc=linux-rdma@vger.kernel.org \
--cc=matanb@mellanox.com \
--cc=robert@leblancnet.us \
--cc=sagi@grimberg.me \
--cc=stable@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.