<div>I updated the HCA "InfiniBand: Mellanox Technologies: Unknown device 634a (rev a0)" to the latest firmware and issue remains. "fw_ver" is now 2.6.000.</div>
<div> </div>
<div>Any ideas on why the time out is occuring in the function?</div>
<div><br> </div>
<div class="pre"><a class="linenr" id="l203" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l203">203</a> static int mlx4_cmd_poll(struct mlx4_dev *dev, u64 in_param, u64 *out_param,</div>
<div class="pre"><a class="linenr" id="l204" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l204">204</a> int out_is_imm, u32 in_modifier, u8 op_modifier,</div>
<div class="pre"><a class="linenr" id="l205" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l205">205</a> u16 op, unsigned long timeout)</div>
<div class="pre"><a class="linenr" id="l206" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l206">206</a> {</div>
<div class="pre"><a class="linenr" id="l207" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l207">207</a> struct mlx4_priv *priv = mlx4_priv(dev);</div>
<div class="pre"><a class="linenr" id="l208" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l208">208</a> void __iomem *hcr = priv->cmd.hcr;</div>
<div class="pre"><a class="linenr" id="l209" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l209">209</a> int err = 0;</div>
<div class="pre"><a class="linenr" id="l210" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l210">210</a> unsigned long end;</div>
<div class="pre"><a class="linenr" id="l211" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l211">211</a> </div>
<div class="pre"><a class="linenr" id="l212" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l212">212</a> down(&priv->cmd.poll_sem);</div>
<div class="pre"><a class="linenr" id="l213" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l213">213</a> </div>
<div class="pre"><a class="linenr" id="l214" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l214">214</a> err = mlx4_cmd_post(dev, in_param, out_param ? *out_param : 0,</div>
<div class="pre"><a class="linenr" id="l215" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l215">215</a> in_modifier, op_modifier, op, CMD_POLL_TOKEN, 0);</div>
<div class="pre"><a class="linenr" id="l216" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l216">216</a> if (err)</div>
<div class="pre"><a class="linenr" id="l217" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l217">217</a> goto out;</div>
<div class="pre"><a class="linenr" id="l218" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l218">218</a> </div>
<div class="pre"><a class="linenr" id="l219" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l219">219</a> end = msecs_to_jiffies(timeout) + jiffies;</div>
<div class="pre"><a class="linenr" id="l220" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l220">220</a> while (cmd_pending(dev) && time_before(jiffies, end))</div>
<div class="pre"><a class="linenr" id="l221" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l221">221</a> cond_resched();</div>
<div class="pre"><a class="linenr" id="l222" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l222">222</a> </div>
<div class="pre"><a class="linenr" id="l223" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l223">223</a> if (cmd_pending(dev)) {</div>
<div class="pre"><a class="linenr" id="l224" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l224">224</a> err = -<span class="__mozilla-findbar-search" style="PADDING-RIGHT: 0pt; DISPLAY: inline; PADDING-LEFT: 0pt; PADDING-BOTTOM: 0pt; COLOR: black; PADDING-TOP: 0pt; BACKGROUND-COLOR: yellow">ETIMEDOUT</span>;</div>
<div class="pre"><a class="linenr" id="l225" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l225">225</a> goto out;</div>
<div class="pre"><a class="linenr" id="l226" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l226">226</a> }</div>
<div class="pre"><a class="linenr" id="l227" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l227">227</a> </div>
<div class="pre"><a class="linenr" id="l228" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l228">228</a> if (out_is_imm)</div>
<div class="pre"><a class="linenr" id="l229" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l229">229</a> *out_param =</div>
<div class="pre"><a class="linenr" id="l230" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l230">230</a> (u64) be32_to_cpu((__force __be32)</div>
<div class="pre"><a class="linenr" id="l231" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l231">231</a> __raw_readl(hcr + HCR_OUT_PARAM_OFFSET)) << 32 |</div>
<div class="pre"><a class="linenr" id="l232" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l232">232</a> (u64) be32_to_cpu((__force __be32)</div>
<div class="pre"><a class="linenr" id="l233" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l233">233</a> __raw_readl(hcr + HCR_OUT_PARAM_OFFSET + 4));</div>
<div class="pre"><a class="linenr" id="l234" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l234">234</a> </div>
<div class="pre"><a class="linenr" id="l235" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l235">235</a> err = mlx4_status_to_errno(be32_to_cpu((__force __be32)</div>
<div class="pre"><a class="linenr" id="l236" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l236">236</a> __raw_readl(hcr + HCR_STATUS_OFFSET)) >> 24);</div>
<div class="pre"><a class="linenr" id="l237" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l237">237</a> </div>
<div class="pre"><a class="linenr" id="l238" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l238">238</a> out:</div>
<div class="pre"><a class="linenr" id="l239" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l239">239</a> up(&priv->cmd.poll_sem);</div>
<div class="pre"><a class="linenr" id="l240" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l240">240</a> return err;</div>
<div class="pre"><a class="linenr" id="l241" href="http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.24.y.git;a=blob;f=drivers/net/mlx4/cmd.c;h=db49051b97b168b7a004e52c865bef94311ce56a;hb=master#l241">241</a> }</div>
<br><br>
<div class="gmail_quote">On Sat, Mar 7, 2009 at 12:55 AM, Dotan Barak <span dir="ltr"><<a href="mailto:dotanba@gmail.com">dotanba@gmail.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid">Please make sure that you have the latest firmware for the HCA that has the failure.<br><font color="#888888"><br>
Dotan</font>
<div>
<div></div>
<div class="h5"><br><br>Phillip Wilson wrote:<br>
<blockquote class="gmail_quote" style="PADDING-LEFT: 1ex; MARGIN: 0px 0px 0px 0.8ex; BORDER-LEFT: #ccc 1px solid">This is related to the thread I stared on Feb 24<br><br>Re: [ofa-general] ***SPAM*** Mellanox ibv_reg_mr (memory region) function call fails under load when using the mlx4 driver<br>
So far I have modified the "num_mtt" to 1 << 21, as Dotan suggested, but I will try some more values this weekend.<br> I think the code for this is in ../drivers/net/mlx4/cmd.c mlx4_cmd( ... )<br> -ETIMEOUT > System Information:<br>
><br>><br>><br>> The system has 4GB of memory.<br>><br>><br>><br>> uname -a<br>><br>> Linux (none) 2.6.24.02.02.08 #21 SMP Thu Feb 19 11:04:35 PST 2009 ia64<br>> unknown<br>><br>><br>
><br>> OFED 1.2.5<br>><br>><br>><br>> lspci -d 15b3:<br>><br>><br>><br>> 0000:10:00.0 InfiniBand: Mellanox Technologies MT25208 InfiniHost III Ex<br>> (Tavor compatibility mode) (rev 20)<br>
><br>> 0000:c3:00.0 InfiniBand: Mellanox Technologies: Unknown device 634a (rev a0)<br>><br>><br>><br>> lspci -d 15b3: -n<br>><br>> 0000:10:00.0 0c06: 15b3:6278 (rev 20)<br>><br>> 0000:c3:00.0 0c06: 15b3:634a (rev a0)<br>
><br>><br>><br>> ibv_devinfo -v<br>><br>> hca_id: mlx4_0<br>><br>> fw_ver: 2.5.000<br>><br>><br>><br>> hca_id: mthca0<br>><br>> fw_ver: 4.8.930<br>
</blockquote><br></div></div></blockquote></div><br>