| Message ID | E1QpTpl-00026F-D6@rvml.edm.orcorp.ca (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
Hi Rolf, On 17:29 Fri 05 Aug , Rolf Manderscheid wrote: > > The code treats them as 16-bit fields (by using htons), make > the structure match. This also avoids the 32 bits of padding > before the length field. > There are more places where lids defined as integer. Why did you choose to change only these two? > Signed-off-by: Rolf Manderscheid <rvm@obsidianresearch.com> > --- > include/ibsim.h | 4 ++-- > 1 files changed, 2 insertions(+), 2 deletions(-) > > diff --git a/include/ibsim.h b/include/ibsim.h > index 15fc37c..b6b08bd 100644 > --- a/include/ibsim.h > +++ b/include/ibsim.h > @@ -62,8 +62,8 @@ struct sim_port { > #define SIM_CTL_MAX_DATA 64 > > struct sim_request { > - uint32_t dlid; > - uint32_t slid; > + uint16_t dlid; > + uint16_t slid; > uint32_t dqp; > uint32_t sqp; > uint32_t status; > -- > 1.7.1 > > > -- > To unsubscribe from this list: send the line "unsubscribe linux-rdma" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html
Hi Alex, > There are more places where lids defined as integer. Why did you choose to > change only these two? This structure defines what request packets look like on the wire to the simulator. The motivation for the change is a simulator client for python-rdma, and this issue shows up when creating the structure format strings for the request packets. Rolf -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Wed, Aug 10, 2011 at 11:18:33AM -0600, Rolf Manderscheid wrote: > Hi Alex, > >There are more places where lids defined as integer. Why did you choose to > >change only these two? > > This structure defines what request packets look like on the wire to > the simulator. The motivation for the change is a simulator client > for python-rdma, and this issue shows up when creating the structure > format strings for the request packets. Also that uint32_t x = htons(y) Simply doesn't work when you mix big and little endian. So, the choices are: switch the type to uint16, or switch to htonl. This is the reason only this struct was changed - because it is the only place where htons is being applied incorrectly, while finding all the wrong htons is a bit harder. python cares because it cannot easily construct something so malformed :) Jason -- To unsubscribe from this list: send the line "unsubscribe linux-rdma" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/include/ibsim.h b/include/ibsim.h index 15fc37c..b6b08bd 100644 --- a/include/ibsim.h +++ b/include/ibsim.h @@ -62,8 +62,8 @@ struct sim_port { #define SIM_CTL_MAX_DATA 64 struct sim_request { - uint32_t dlid; - uint32_t slid; + uint16_t dlid; + uint16_t slid; uint32_t dqp; uint32_t sqp; uint32_t status;
The code treats them as 16-bit fields (by using htons), make the structure match. This also avoids the 32 bits of padding before the length field. Signed-off-by: Rolf Manderscheid <rvm@obsidianresearch.com> --- include/ibsim.h | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-)