From owner-freebsd-net@FreeBSD.ORG Fri Feb 11 19:57:09 2011 Return-Path: Delivered-To: freebsd-net@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id A78E2106564A for ; Fri, 11 Feb 2011 19:57:09 +0000 (UTC) (envelope-from jfvogel@gmail.com) Received: from mail-gw0-f54.google.com (mail-gw0-f54.google.com [74.125.83.54]) by mx1.freebsd.org (Postfix) with ESMTP id 5522C8FC16 for ; Fri, 11 Feb 2011 19:57:09 +0000 (UTC) Received: by gwj21 with SMTP id 21so1260417gwj.13 for ; Fri, 11 Feb 2011 11:57:08 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:in-reply-to:references:date :message-id:subject:from:to:cc:content-type; bh=u9+wCxdOFfvSdx/FU3OJGWFrewqFworiuvpI+IlfJ1o=; b=Rbnv96CAuEl7lras+f0ndLvR9nGCMnt2NWaMpgkX8Ab6L3n6rBJ2QAJuzVazXJcPGD fGkPmZbr8DdZiR0fL2+rG+cA4woDuCDg09osLpK+Li62U85NIQUoJZngjSBuPcTIFvUG PPq8gOPKPsk2uqYIzHcwrxN3UimgTGN4MyckY= DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; b=A1wadKZpyLutUDU5qYEfdUyvcR0XLTyxiTMqtHIpwg2flETuF9lnav0i+voy/Nejt2 tHOaOaAGpcNODN+8Wjti8gQ1lrohIWjaDSaLMf2jU1BPidKzsSItlOFU8mokJn1F9Hll dw/lrE54JTOhdZQgozZbiVBBov9NaD6H5PnBY= MIME-Version: 1.0 Received: by 10.147.34.2 with SMTP id m2mr1130219yaj.2.1297454228180; Fri, 11 Feb 2011 11:57:08 -0800 (PST) Received: by 10.147.167.5 with HTTP; Fri, 11 Feb 2011 11:57:08 -0800 (PST) In-Reply-To: References: <12838373-FE96-443E-8979-AF5408705BF0@freebsd.org> Date: Fri, 11 Feb 2011 11:57:08 -0800 Message-ID: From: Jack Vogel To: Karim Fodil-Lemelin Content-Type: text/plain; charset=ISO-8859-1 X-Content-Filtered-By: Mailman/MimeDel 2.1.5 Cc: freebsd-net@freebsd.org Subject: Re: igb driver RX (was TX) hangs when out of mbuf clusters X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 11 Feb 2011 19:57:09 -0000 Yes, it was the fix for this, Mike has tested and confirmed it eliminates the problem, and yes I will MFC the relevant bits back to 8 and 7 for you asap. Jack On Fri, Feb 11, 2011 at 11:53 AM, Karim Fodil-Lemelin < fodillemlinkarim@gmail.com> wrote: > Hi, > > I see a commit was made in current (r218530 | jfv | 2011-02-10 20:00:26 > -0500 (Thu, 10 Feb 2011)). Is that commit done to address this issue? > > And if so Is there any MFC planned for 7.4 for this? > > Thanks, > > Karim. > > 2011/2/9 Michael Tuexen > > On Feb 9, 2011, at 6:35 PM, Jack Vogel wrote: >> >> > OK, but the question is why does the ring get totally consumed this way, >> the >> > ring has 1024 descriptors, it seems unintuitive that that whole quantity >> can be >> > used without some being recharged. Do you see the system mbuf pool being >> > depleted at the same time? >> That was the test case I created: I set up a server accepting connections >> but not reading anything. So the driver passes the mbufs to the transport >> stack and they are not consumed. Then the problem occurs. Then I kill the >> server. Now there are mbufs available again, but the driver doesn't know. >> >> I had the impression that these were the circumstances in which the >> problem >> showed up (mbuf allocations failing). >> > >> > Since you can reproduce it, do me a favor, in rxeof, change the >> processed >> > value from 8 to 4 and then 1, effectively call refresh every descriptor, >> see if >> > that eliminates the issue. >> I will do. Need to see if I can do it remotely, since I'm not in my lab >> right now. Can do it tomorrow for sure. >> >> But I do not think that this solves the problem, since I did the things >> very slowly and you call it at least when you are leaving rxeof. >> >> Best regards >> Michael >> > >> > Thanks for your help, >> > >> > Jack >> > >> > >> > On Wed, Feb 9, 2011 at 2:36 AM, Michael Tuexen >> wrote: >> > Hi Jack, >> > >> > I could recreate the problem. When the problem occurs, we see >> > >> > rx_nxt_check = n >> > rx_nxt_refresh = n + 1 >> > >> > (This was also reported in a mail from Karim) >> > >> > This means that the *whole* receive ring has no buffers anymore. This >> can >> > occur if, for some amount of time, no clusters are available. >> > >> > Now outside of the driver, at some point of time, clusters are freed. >> > I don't think that igb_refresh_mbufs() gets called, since it only gets >> > called from igb_rxeof(), which gets called when a packet has been >> received, >> > which can not happen since the receive ring is empty. So how can the >> driver >> > know? I have no idea. Maybe we can periodically check for such an event >> > and call igb_refresh_mbufs(). >> > >> > Does this make sense to you? >> > >> > Best regards >> > Michael >> > >> > >> > On Feb 9, 2011, at 8:32 AM, Jack Vogel wrote: >> > >> > > Hmmm, well so much for that theory :) >> > > >> > > Jack >> > > >> > > >> > > On Tue, Feb 8, 2011 at 4:06 PM, Karim Fodil-Lemelin < >> fodillemlinkarim@gmail.com> wrote: >> > > >> > > >> > > 2011/2/8 Jack Vogel >> > > >> > > >> > > I have been following this, and thinking about it. I still am working >> from a theoretical >> > > standpoint, but based on a patch I got quite a long time back and >> never quite groked, >> > > I believe now that I might have a solution. >> > > >> > > The original PR and patch was kern/150516 from Beezar Liu, I was >> never quite comfortable >> > > with the code changes, nor convinced that it was a real issue and not >> a misunderstanding. >> > > However I think now that this very report might be behind what we are >> seeing today. I have >> > > a slightly different approach to solving it, of course it remains to >> be seen if it handles it >> > > properly. >> > > >> > > Please try the patch I've attached, I'm open to further correction or >> polishing of the >> > > changes. And thanks to Beezar for his original report and changes, >> this is not for em, >> > > but if this eliminates the problem its clearly needed in all drivers. >> > > >> > > Jack >> > > >> > > >> > > Hi Jack, >> > > >> > > Thanks for your help. I tried your patch and it didn't work so I added >> a couple of printf to see if the added code was getting hit: >> > > >> > > --- a/freebsd/sys/dev/e1000/if_igb.c >> > > --More--(byte 1253)+++ b/freebsd/sys/dev/e1000/if_igb.c >> > > @@ -612,7 +612,7 @@ igb_attach(device_t dev) >> > > device_get_nameunit(dev)); >> > > >> > > INIT_DEBUGOUT("igb_attach: end"); >> > > - >> > > + printf("this driver has a patch from Jack Vogel\n"); >> > > return (0); >> > > >> > > err_late: >> > > @@ -4131,6 +4131,7 @@ igb_rxeof(struct igb_queue *que, int count, int >> *done) >> > > struct mbuf *sendmp, *mh, *mp; >> > > struct igb_rx_buf *rxbuf; >> > > u16 hlen, plen, hdr, vtag; >> > > + int commit; >> > > bool eop = FALSE; >> > > >> > > cur = &rxr->rx_base[i]; >> > > @@ -4255,10 +4256,23 @@ next_desc: >> > > bus_dmamap_sync(rxr->rxdma.dma_tag, >> rxr->rxdma.dma_map, >> > > BUS_DMASYNC_PREREAD | BUS_DMASYNC_PREWRITE); >> > > >> > > + commit = i; /* capture the old index */ >> > > + >> > > /* Advance our pointers to the next descriptor. */ >> > > if (++i == adapter->num_rx_desc) >> > > i = 0; >> > > /* >> > > + ** Sanity test for ring full, if this >> > > + ** happens we need to refresh immediately >> > > + ** or refresh may deadlock. >> > > + */ >> > > + if (i == rxr->next_to_refresh) { >> > > + igb_refresh_mbufs(rxr, commit); >> > > + printf("igb_refresh_mbufs called with commit >> %d\n", commit); >> > > + processed = 0; >> > > + } >> > > + >> > > + /* >> > > ** Send to the stack or LRO >> > > */ >> > > if (sendmp != NULL) { >> > > >> > > Here is the results: >> > > >> > > # dmesg | grep Vogel >> > > this driver has a patch from Jack Vogel >> > > this driver has a patch from Jack Vogel >> > > >> > > # netstat -m >> > > 60453/52707/113160 mbufs in use (current/cache/total) >> > > 48416/51584/100000/100000 mbuf clusters in use >> (current/cache/total/max) >> > > 2894/690 mbuf+clusters out of packet secondary zone in use >> (current/cache) >> > > 11946/854/12800/12800 4k (page size) jumbo clusters in use >> (current/cache/total/max) >> > > 0/0/0/6400 9k jumbo clusters in use (current/cache/total/max) >> > > 0/0/0/3200 16k jumbo clusters in use (current/cache/total/max) >> > > 164834K/119760K/284595K bytes allocated to network >> (current/cache/total) >> > > 0/339/0 requests for mbufs denied (mbufs/clusters/mbuf+clusters) >> > > 0/0/0 requests for jumbo clusters denied (4k/9k/16k) >> > > 0/4/6656 sfbufs in use (current/peak/max) >> > > 0 requests for sfbufs denied >> > > 0 requests for sfbufs delayed >> > > 0 requests for I/O initiated by sendfile >> > > 0 calls to protocol drain routines >> > > # dmesg | grep commit >> > > >> > > At this point RX has hung. >> > > >> > > Somehow the check (i == rxr->next_to_refresh) is never true in this >> case. Also, I did read kern/150516 and couldn't wrap my head around the >> patch for the em driver that Beezar Liu suggested. >> > > >> > > Regards, >> > > >> > > Karim. >> > > >> > > >> > >> > >> >> >