| Message ID | 51543E86.9000706@wut.de |
|---|---|
| State | Not Applicable |
| Headers | show |
Dear Markus Königshaus, On Thu, 28 Mar 2013 13:58:46 +0100, Markus Königshaus wrote: > I have found a bug in resolv.c in conjunction with the function > res_init: res_init sets __ res_sync to res_sync_func and then calls > __res_vinit. > -> __res_vinit calls __ open_nameservers. > In __open_nameservers __nameserver and __nameservers are set by the > contents of resolv.conf. So far so good. > -> __open_nameserver calls res_sync_func > In res_sync_func__nameservers is set to rp->nscount, what is 0. > Backin _res_vinit rp->nscount is set to __nameservers then, what is 0. > So the cat bites its own tail. > > Attached is a simple patch, that fixes the problemby setting __res_sync > after calling __res_vinit. Thanks Markus for this patch. However, this is the mailing list for Buildroot, so we're not the one responsible for taking such patches. You should instead get in touch with the uClibc community, that has its own mailing list: http://uclibc.org/lists.html. Best regards, Thomas
Autor: Markus Königshaus <M.Koenigshaus@wut.de> --- uClibc-0.9.33.2.org/libc/inet/resolv.c 2012-05-15 09:20:09.000000000 +0200 +++ uClibc-0.9.33.2/libc/inet/resolv.c 2013-03-28 11:52:55.000000000 +0100 @@ -3653,12 +3653,18 @@ */ if (!_res.id) _res.id = res_randomid(); - __res_sync = res_sync_func; + __res_sync = NULL; __UCLIBC_MUTEX_UNLOCK(__resolv_lock); __res_vinit(&_res, 1); + __UCLIBC_MUTEX_LOCK(__resolv_lock); + + __res_sync = res_sync_func; + + __UCLIBC_MUTEX_UNLOCK(__resolv_lock); + return 0; } libc_hidden_def(res_init)
Hello, I have found a bug in resolv.c in conjunction with the function res_init: res_init sets __ res_sync to res_sync_func and then calls __res_vinit. -> __res_vinit calls __ open_nameservers. In __open_nameservers __nameserver and __nameservers are set by the contents of resolv.conf. So far so good. -> __open_nameserver calls res_sync_func In res_sync_func__nameservers is set to rp->nscount, what is 0. Backin _res_vinit rp->nscount is set to __nameservers then, what is 0. So the cat bites its own tail. Attached is a simple patch, that fixes the problemby setting __res_sync after calling __res_vinit. Before this patch with busybox "nslookup www.google.de" -> nslookup: can not resolve '(null)': Name or service not known Name: www.google.de Address 1: 2a00: 1450:4016:800 :: 101f muc03s01-in-x1f.1e100.net Address 2: 173.194.35.151 muc03s01-in-f23.1e100.net Address 3: 173.194.35.159 muc03s01-in-f31.1e100.net Address 4: 173.194.35.152 muc03s01-in-f24.1e100.net After the patch "nslookup www.google.de" -> nslookup www.google.de Server: 8.8.8.8 Address 1: 8.8.8.8 google-public-dns-a.google.com Name: www.google.de Address 1: 2a00: 1450:4016:803 :: 1018 muc03s08-in-x18.1e100.net Address 2: 173.194.44.56 muc03s08-in-f24.1e100.net Address 3: 173.194.44.55 muc03s08-in-f23.1e100.net Address 4: 173.194.44.63 muc03s08-in-f31.1e100.net This patch ist tested with buildroot-02.13. Regards, Markus -- Unsere Aussagen koennen Irrtuemer und Missverstaendnisse enthalten. Bitte pruefen Sie die Aussagen fuer Ihren Fall, bevor Sie Entscheidungen auf Grundlage dieser Aussagen treffen. Wiesemann & Theis GmbH, Porschestr. 12, D-42279 Wuppertal Geschaeftsfuehrer: Dipl.-Ing. Ruediger Theis Registergericht: Amtsgericht Wuppertal, HRB 6377 Tel. +49-202/2680-0, Fax +49-202/2680-265, http://www.wut.de