From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757413AbZHZNwD (ORCPT ); Wed, 26 Aug 2009 09:52:03 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1757376AbZHZNwD (ORCPT ); Wed, 26 Aug 2009 09:52:03 -0400 Received: from mail-px0-f174.google.com ([209.85.216.174]:45643 "EHLO mail-px0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756078AbZHZNwC (ORCPT ); Wed, 26 Aug 2009 09:52:02 -0400 X-Greylist: delayed 460 seconds by postgrey-1.27 at vger.kernel.org; Wed, 26 Aug 2009 09:52:02 EDT DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:date:message-id:subject:from:to:content-type :content-transfer-encoding; b=riyaAi3fscOdNe106aQI9aM5irs/YfRqD3Q4NNWHFfRWw4UuwGo9xW6Ww5oGG5e8Tt kypcpicC//XxUNBftbMSLoi8+r8Ktiu9sl4gF383CKhcEx4e7+Lp5zzOKDS4uiQvI9HZ JW2bkF+sPTGxuCUgbzg4onPFl1CK8qJQk66Fw= MIME-Version: 1.0 Date: Wed, 26 Aug 2009 19:14:23 +0530 Message-ID: Subject: sk_stream_wait_memory: Query From: Ravish Tayal To: linux-arm@vger.kernel.org, Kernel development list Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi, In " tcp_sendmsg" function in 2 situation, tcp code attempts to jump to "wait_for_memory: " case when 1. it fails to allocate to the new skb [sk_stream_alloc_pskb failure] 2. it fails to allocate new cache page. and invokes the sk_stream_wait_memory function. inside which it invokes the sk_wait_event(sk, ¤t_timeo, sk_stream_memory_free(sk) && vm_wait); in case if sk_stream_wait_memory is invoked due to skb allocation failure, it would busy loop and would not relinquish the control. [For non-preemptive kernel] Is it fair expectation, or even this case it should let it current task get scheduled and re try after timer expiry. regards Ravish Tayal in the case if linux kernel is configured as non-premeptive