Date: Wed, 26 Sep 2007 20:36:44 GMT From: Michael W Lucas <mwlucas@blackhelicopters.org> To: freebsd-gnats-submit@FreeBSD.org Subject: ports/116677: [patch] flow-tools on amd64 is broken Message-ID: <200709262036.l8QKaiIq081564@www.freebsd.org> Resent-Message-ID: <200709262040.l8QKe2p0022827@freefall.freebsd.org>
next in thread | raw e-mail | index | archive | help
>Number: 116677 >Category: ports >Synopsis: [patch] flow-tools on amd64 is broken >Confidential: no >Severity: non-critical >Priority: low >Responsible: freebsd-ports-bugs >State: open >Quarter: >Keywords: >Date-Required: >Class: change-request >Submitter-Id: current-users >Arrival-Date: Wed Sep 26 20:40:02 GMT 2007 >Closed-Date: >Last-Modified: >Originator: Michael W Lucas >Release: 7.0-current >Organization: none >Environment: FreeBSD aubsr019.us.add 7.0-CURRENT FreeBSD 7.0-CURRENT #0: Thu Jun 14 15:36:22 EDT 2007 system_mwl@aubsr019.us.add:/usr/obj/usr/src/sys/GENERIC amd64 >Description: flow-tools on amd64 platforms is known to be broken. See http://mailman.splintered.net/pipermail/flow-tools/2004-December/002501.html and http://www.mail-archive.com/flow-tools@list.splintered.net/msg00471.html -- or, if you prefer, see how many mailing list postings link to these two postings as the solution. More discussion of the problem from our own Bill F. at http://www.mail-archive.com/flow-tools@list.splintered.net/msg00470.html Maybe "flow-tools on amd64 is broken" is too strong; it does compile, after all. But the results are not useful... >How-To-Repeat: Run flow-cat flowfile-XXXX | flow-print -f 1 on any amd64 platform. Look at the dates. They are garbage. >Fix: One of the attached files is a new patch. The other is an updated version of an existing patch file. Patch attached with submission follows: --- lib/ftfile.c.orig Wed Sep 26 16:21:38 2007 +++ lib/ftfile.c Wed Sep 26 16:20:49 2007 @@ -311,7 +311,7 @@ int ftfile_expire (struct ftfile_entries *fte, int doit, int curbytes) { u_int i; - struct ftfile_entry *n1; + struct ftfile_entry *n1, *n2; u_int64 bytes; /* @@ -323,15 +323,20 @@ bytes = 0; if (fte->max_files && (fte->num_files > fte->max_files)) { + n2 = NULL; FT_TAILQ_FOREACH(n1, &fte->head, chain) { + if (n2 != NULL) { + ftfile_entry_free(n2); + n2 = NULL; + } fterr_info("remove/1 %s", n1->name); bytes += n1->size; ++i; if (doit) { + n2 = n1; FT_TAILQ_REMOVE(&fte->head, n1, chain); if (unlink(n1->name) == -1) fterr_warn("unlink(%s)", n1->name); - ftfile_entry_free(n1); } /* doit */ if ((fte->num_files - i) <= fte->max_files) break; @@ -340,6 +345,10 @@ fte->num_files -= i; fte->num_bytes -= bytes; } /* doit */ + if (n2 != NULL) { + ftfile_entry_free(n2); + n2 = NULL; + } } /* if */ if (debug) @@ -354,15 +363,20 @@ */ if (fte->max_bytes && (fte->num_bytes+curbytes > fte->max_bytes)) { + n2 = NULL; FT_TAILQ_FOREACH(n1, &fte->head, chain) { + if (n2 != NULL) { + ftfile_entry_free(n2); + n2 = NULL; + } fterr_info("remove/2 %s", n1->name); bytes += n1->size; ++i; if (doit) { + n2 = n1; FT_TAILQ_REMOVE(&fte->head, n1, chain); if (unlink(n1->name) == -1) fterr_warn("unlink(%s)", n1->name); - ftfile_entry_free(n1); } /* doit */ if ((fte->num_bytes+curbytes - bytes) <= fte->max_bytes) break; @@ -371,6 +385,10 @@ fte->num_files -= i; fte->num_bytes -= bytes; } /* doit */ + if (n2 != NULL) { + ftfile_entry_free(n2); + n2 = NULL; + } } /* if */ if (debug) @@ -499,7 +517,7 @@ * returns -1 on error * */ -int ftfile_mkpath(u_int32 ftime, int nest) +int ftfile_mkpath(time_t ftime, int nest) { struct tm *tm; char buf[32]; @@ -762,13 +780,19 @@ { struct ftfile_entry *n1, *n2; + n2 = NULL; FT_TAILQ_FOREACH(n1, &fte->head, chain) { + if (n2 != NULL) { + ftfile_entry_free(n2); + n2 = NULL; + } FT_TAILQ_REMOVE(&fte->head, n1, chain); n2 = n1; - n1 = FT_TAILQ_NEXT(n1, chain); + } + + if (n2 != NULL) { ftfile_entry_free(n2); - if (!n1) - break; + n2 = NULL; } } /* ftfile_free */ >Release-Note: >Audit-Trail: >Unformatted:
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200709262036.l8QKaiIq081564>