Roland Dreier [EMAIL PROTECTED] writes:
Definitely makes sense. I queued the following version for 2.6.20,
which gets the max CDB size directly from struct srp_cmd. Does this
look OK to you?
Looks good to me. I guess there's no way to sneak it into 2.6.19
anymore, right? ;)
Thanks,
Arne
While importing these patches, I got several Space in indent is
followed by a tab. errors. For example, the line
+ __constant_cpu_to_be16(ETH_P_8021Q))) {
I did test my patches with git-apply --whitespace=error-all,
before mailing them but did not see any error.
I am
Hi,
I just installed Debian Etch RC1 on a Dell PowerEdge 1950.
I have written an application that performs a RDMA read and write on a server,
and it works on Suse 9.3 with OFED 1.0 installed.
Now, when I start that application on Debian, I got the following error:
-
libibverbs: Fatal:
Yesterday I complained about a gen2 x86 client RDMA read not working on
a gen1 PowerPC server.
Today I got my hardware guy measuring the PCI-Express interface on the
PowerPC gen1 server.
It seems as if the RDMA read is at least started. So all ideas about
endianess problems
To explain the
Dotan and all others.
I could isolate the problem with hardware measurement a little more
today
I just opened a new thread on the problem naming:
How to increase the waiting period/timeout for the completion of an
RDMA operation?
Thanks for the support so far.
Thomas
Umm.. what's the point of the previous patch that changed spin_lock to
spin_lock_nested? All it did was hide this bug.
Also BTW your subject lines are off -- these are not RDMA/iwcm
patches.
- R.
___
openib-general mailing list
+static u32 next_random(u32 rand)
+{
+u32 y, ylast;
+
+y = rand;
+ylast = y;
+y = (y * 69069) 0x;
+y = (y 0x8000) + (ylast 0x7fff);
+if ((y 1))
+y = ylast ^ (y 1) ^ (2567483615UL);
+else
+y =
On Thu, 2006-11-16 at 20:45 -0800, Roland Dreier wrote:
+struct t3_send_wr {
+ struct fw_riwrh wrh;/* 0 */
+ union t3_wrid wrid; /* 1 */
+
+ enum t3_rdma_opcode rdmaop:8;
+ u32 reserved:24;/* 2 */
Does this do the right thing wrt endianness? I'd be more
Looks good to me. I guess there's no way to sneak it into 2.6.19
anymore, right? ;)
I thought about it but I couldn't really justify it to myself...
___
openib-general mailing list
openib-general@openib.org
libibverbs: Fatal: couldn't read uverbs ABI version.
Do you have the ib_uverbs kernel module installed?
To load it automatically on boot you can add a line
ib_uverbs
to your /etc/modules directory.
BTW out of curiousity are you using the standard Debian kernel and
libibverbs/libmthca
How to increase the waiting period/timeout for the completion of an
RDMA operation?
This is the ACK timeout I think. What is your gen2 client setting for
the local ACK timeout when connecting a QP?
- R.
___
openib-general mailing list
On Fri, 2006-11-17 at 08:54 -0800, Roland Dreier wrote:
+static u32 next_random(u32 rand)
+{
+ u32 y, ylast;
+
+ y = rand;
+ ylast = y;
+ y = (y * 69069) 0x;
+ y = (y 0x8000) + (ylast 0x7fff);
+ if ((y 1))
+ y = ylast ^ (y 1)
Roland,
actually I'm answering from home. So I can't tell.
Can you give me the exact attribute name so that I can modify it tomorrow?
Is this a parameter that the CM I'm using modifies/overwrites?
Thomas
-Ursprüngliche Nachricht-
Von: Roland Dreier [mailto:[EMAIL PROTECTED]
Gesendet:
Steve Wise wrote:
+static inline void *vzmalloc(int size)
+{
+ void *p = vmalloc(size);
+ memset(p, 0, size);
+ return p;
+}
This isn't checking the return value from vmalloc.
Also, we could do with a generic vzalloc and vcalloc, just as we now
have kzalloc and kcalloc.
On Fri, 2006-11-17 at 09:53 -0800, Bryan O'Sullivan wrote:
Steve Wise wrote:
+static inline void *vzmalloc(int size)
+{
+ void *p = vmalloc(size);
+ memset(p, 0, size);
+ return p;
+}
This isn't checking the return value from vmalloc.
Oops...
Also, we could do with a
Steve Wise wrote:
+static void release_tid(struct t3cdev *tdev, u32 hwtid, struct sk_buff *skb)
+{
+ struct cpl_tid_release *req;
+
+ skb = get_skb(skb, sizeof *req, GFP_KERNEL);
+ if (!skb) {
+ return;
+ }
Style micronit: no curlies for single-statement
Steve Wise wrote:
T3 WQE and CQE structures, defines, etc...
I notice that none of the fields in these structs seem to be
endianness-annotated, but that there's a lot of cpu_to_be64 and so on
being used to frob values into them. Please make sure that the driver
passes a sparse check, which
On Fri, 2006-11-17 at 10:07 -0800, Bryan O'Sullivan wrote:
Steve Wise wrote:
+static void release_tid(struct t3cdev *tdev, u32 hwtid, struct sk_buff
*skb)
+{
+ struct cpl_tid_release *req;
+
+ skb = get_skb(skb, sizeof *req, GFP_KERNEL);
+ if (!skb) {
+ return;
On Fri, 2006-11-17 at 10:19 -0800, Bryan O'Sullivan wrote:
Steve Wise wrote:
T3 WQE and CQE structures, defines, etc...
I notice that none of the fields in these structs seem to be
endianness-annotated, but that there's a lot of cpu_to_be64 and so on
being used to frob values into them.
Steve Wise wrote:
It passes sparse with only a few warnings about calling memset() with a
size 10.
You need to pass in CF=-D__CHECK_ENDIAN__ too, on the kernel build
command line. Otherwise, the endianness annotations aren't turned on in
the kernel headers, and you get this nice false
I think we can use random32() or get_random_bytes(). I need to
re-review how this algorithm works. Its randomizing the stag IDs so
they are not predictable.
I assume based on the algorithm you have now that they don't need to
be cryptographically unpredictable. So random32() would
I think the attribute to look at is struct ibv_qp_attr.timeout. I'm
not really sure if the CM sets it or not.
- R.
___
openib-general mailing list
openib-general@openib.org
http://openib.org/mailman/listinfo/openib-general
To unsubscribe, please
I did test my patches with git-apply --whitespace=error-all,
before mailing them but did not see any error.
I am using GIT 1.3, may be I should upgrade ?
Yes, git 1.3 is not that new. I am using git 1.4.3.2, and I think the
relevant code you are missing is in the changeset
commit
The Chelsio driver is hogging lots of memory right now for mapping
PDIDs, QPIDs, CQIDs, and STAG IDs back to their respective kernel
structures. This is done via an array of pointers, indexed by the ID.
The critical performance mapping is finding a QP struct from the QPID in
the poll path.
I think the attribute to look at is struct ibv_qp_attr.timeout. I'm
not really sure if the CM sets it or not.
- R.
The CM sets this attribute and it called local_ack_timeout in the CM
structures.
Dotan
___
openib-general mailing list
25 matches
Mail list logo