February 2009 - Darwin-dev - lists.apple.com

iPhone 2.2.1 Source Released
by William Siegrist 19 Feb '09

19 Feb '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com http://www.opensource.apple.com/darwinsource/iPhone2.2.1/ Thank you. ---- William Siegrist Darwin Team _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… The Darwin Team has released the source for iPhone 2.2.1. You will find new versions of WebCore and JavaScriptCore at the usual location: This email sent to site_archiver(a)lists.apple.com

1 0

Are there any API's available for performing cryptographic functions in a kext?
by Tom Fortmann 19 Feb '09

19 Feb '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Thread-index: AcmSuK4nKel61JzISqeoXSiHpWb4ww== _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… This email sent to site_archiver(a)lists.apple.com

1 0

Re: broken pipe
by Greg Parker 19 Feb '09

19 Feb '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Thanks M and Shawn. I have already handler for SIGPIPE in which it is doing nothing, just ignoring. When I get SIGPIPE, I am not getting any error in write() call, does it mean my app is getting SIGPIPE due to some other thing? Shawn, I do not have any socket() in my app. I am communicating to serial port through read() and write(), after setting some parameters. -- Greg Parker gparker(a)apple.com Runtime Wrangler _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… On Feb 18, 2009, at 8:38 PM, Rakesh Singhal wrote: If you catch the signal, or ignore it with SIG_IGN, then the failed write() should return EPIPE. Try this: set a breakpoint in your SIGPIPE signal handler and run your app. When you get SIGPIPE and hit the breakpoint, look at the thread's stack trace. The failing write() should be there. This email sent to site_archiver(a)lists.apple.com

1 0

Re: broken pipe
by Rakesh Singhal 19 Feb '09

19 Feb '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:in-reply-to:references :date:message-id:subject:from:to:cc:content-type :content-transfer-encoding; bh=j+QUEn7NqzxrK/tWClDJjVZKwSH9IYzTy5Zb7bLDBj8=; b=hsmpFsNtnb32AGWK4wwaR95U7Aoj5MfbJO8qjIh9zj+2LfiGcbM1eGp65AywtDHsm+ XFjF3ux7VWohSNjnRGgUehmJfaPJl/I4sjH7E9A8c4vb/T5h1xa3V/GwEUaWLKS2NFfG uOR5AMxdgyHsHWjRwNoXl5F0Mnvat3btVQwj4= 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:content-transfer-encoding; b=Wu0qR6Kp++gjyFJh5Br3qh7tXFz4K6ACshX9uN8mCLhmcHRcWQwg5fm7+JinhBcfQZ tJw4qbI0YPN/bS6MRF+1yik0LgQx2oAOZ6ATMLSZuKl1xs9tfG3pGElzGXPN5pBfBClf OfLpwHirAeg+TdHleEKXCmS9CIa8wlwxwkpOg= Thanks M and Shawn. I have already handler for SIGPIPE in which it is doing nothing, just ignoring. When I get SIGPIPE, I am not getting any error in write() call, does it mean my app is getting SIGPIPE due to some other thing? Shawn, I do not have any socket() in my app. I am communicating to serial port through read() and write(), after setting some parameters. Regards rksinghal On Thu, Feb 19, 2009 at 5:23 AM, Shawn Erickson <shawnce(a)gmail.com> wrote: >>> On Feb 17, 2009, at 2:06 PM, Rakesh Singhal wrote: >>> >>>> Hi all, >>>> >>>> I am getting SIGPIPE (broken pipe) signal. I have implemented signal >>>> handler. >>>> >>>> I do not know, why I am getting it in my app. I do not have any >>>> sockets in my app. I am communicating with serial port and my own >>>> dynamic libraries. How can I find the cause of this signal? Please >>>> suggest. >>> >>> You are writing to descriptor which has nothing on the other side (reading). >>> You should ignore the SIGPIPE signal on all threads and check for >>> errno==EPIPE instead. > > To get EPIPE you need set socket to return such a value instead of > fire a SIGPIPE. > > "SO_NOSIGPIPE is an option that prevents SIGPIPE from being raised > when a write fails on a socket to which there is no reader; instead, > the write to the socket returns with the error EPIPE when there is no > reader." > > int sigPipe = 1; > ::setsockopt(_fd, SOL_SOCKET, SO_NOSIGPIPE, (const char*)&sigPipe, > sizeof(sigPipe)); > > Anyway how are you talking with the serial port? > > -Shawn > _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… This email sent to site_archiver(a)lists.apple.com

1 0

Re: broken pipe
by Shawn Erickson 18 Feb '09

18 Feb '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:in-reply-to:references :date:message-id:subject:from:to:cc:content-type :content-transfer-encoding; bh=0Z8YohWQVVpnFmUjIP9M/Oq5Ca3cyWy0OEGlAzqlPOQ=; b=Ym/FGqDBj5VpZX1fNhUlHG+7A51GI/ULRxAixwItzPolGc21bb/CnL1OQxCisF16Al lN/xktU2YkqceAk5ZQp1tMdpmfOEZc25/hG7av1hF1RcD9in+Iv/99+X+tAvQn72DrW9 P4N/L5q6UwJJrolWLyfKO9BaDBmZks15fIfzs= 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:content-transfer-encoding; b=H+nM1m6h9ic+zKG3PVtd5yWHJJGK+mF7mLiG+wGKFbEYtAH8rMsbf78mItrqgLmCJC hyT80O3cGmZ8ysrpWCd41f/detM2YM/05TYDyclutgBSo+zbKOEBGl1OoWa8+UO5q1OI /zaN1N+NpHMcFmLUUucuRf/T2FTHtiSIUO9EA= >> On Feb 17, 2009, at 2:06 PM, Rakesh Singhal wrote: >> >>> Hi all, >>> >>> I am getting SIGPIPE (broken pipe) signal. I have implemented signal >>> handler. >>> >>> I do not know, why I am getting it in my app. I do not have any >>> sockets in my app. I am communicating with serial port and my own >>> dynamic libraries. How can I find the cause of this signal? Please >>> suggest. >> >> You are writing to descriptor which has nothing on the other side (reading). >> You should ignore the SIGPIPE signal on all threads and check for >> errno==EPIPE instead. To get EPIPE you need set socket to return such a value instead of fire a SIGPIPE. "SO_NOSIGPIPE is an option that prevents SIGPIPE from being raised when a write fails on a socket to which there is no reader; instead, the write to the socket returns with the error EPIPE when there is no reader." int sigPipe = 1; ::setsockopt(_fd, SOL_SOCKET, SO_NOSIGPIPE, (const char*)&sigPipe, sizeof(sigPipe)); Anyway how are you talking with the serial port? -Shawn _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… This email sent to site_archiver(a)lists.apple.com

1 0

Re: Secure IPC question
by Tim Murison 18 Feb '09

18 Feb '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Thread-index: AcmSEgW7OjypHpeSHEOiJVctXZeh4AAC4B5s Thread-topic: Secure IPC question User-agent: Microsoft-Entourage/12.15.0.081119 > pid_t pid; > mach_port_t port; > name_t name = "com.mycompany.myportname"; > > bootstrap_look_up(bootstrap_port, name, &port); > pid_for_task(port, &pid); > printf("the server pid is %d\n", pid); > > Any ideas? After examining the launchd source, it looks like I can do this with the launch.h API, but it would be very ugly (and brittle?). Are there any better ways? *********************************************************************** This e-mail and its attachments are confidential, legally privileged, may be subject to copyright and sent solely for the attention of the addressee(s). Any unauthorized use or disclosure is prohibited. Statements and opinions expressed in this e-mail may not represent those of Radialpoint. Le contenu de ce courriel est confidentiel, privilégié et peut être soumis à des droits d'auteur. Il est envoyé à l'intention exclusive de son ou de ses destinataires. Il est interdit de l'utiliser ou de le divulguer sans autorisation. Les opinions exprimées dans le présent courriel peuvent diverger de celles de Radialpoint. _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… This email sent to site_archiver(a)lists.apple.com

1 0

Secure IPC question
by Tim Murison 18 Feb '09

18 Feb '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Thread-index: AcmSEgW7OjypHpeSHEOiJVctXZeh4A== Thread-topic: Secure IPC question User-agent: Microsoft-Entourage/12.15.0.081119 Hi Guys, I'm trying to implement a secure IPC system for my client/server software. The IPC mechanism is mach ports. The security goal for the IPC system is each end of the communication establishing the validity of the other. The server establishes the validity of a client using the mach_msg_audit_trailer_t structure to retrieve the PID of the sender and validate its code signature. What I want to do with the client is establish the validity of the server on startup. So far, my code reads: pid_t pid; mach_port_t port; name_t name = "com.mycompany.myportname"; bootstrap_look_up(bootstrap_port, name, &port); pid_for_task(port, &pid); printf("the server pid is %d\n", pid); I had rather hoped that this would work, but after some research I realize that it makes no sense. I'm beginning to think there is no way to do what I want. Any ideas? Thanks in advance, -Tim *********************************************************************** This e-mail and its attachments are confidential, legally privileged, may be subject to copyright and sent solely for the attention of the addressee(s). Any unauthorized use or disclosure is prohibited. Statements and opinions expressed in this e-mail may not represent those of Radialpoint. Le contenu de ce courriel est confidentiel, privilégié et peut être soumis à des droits d'auteur. Il est envoyé à l'intention exclusive de son ou de ses destinataires. Il est interdit de l'utiliser ou de le divulguer sans autorisation. Les opinions exprimées dans le présent courriel peuvent diverger de celles de Radialpoint. _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… This email sent to site_archiver(a)lists.apple.com

1 0

Re: Set process' working set size explictly
by Maxim Zhuravlev 18 Feb '09

18 Feb '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:in-reply-to:references :date:message-id:subject:from:to:cc:content-type :content-transfer-encoding; bh=L7MQKD0pEGmp7urImlDjMiFgaadwtIKOYKZrUCcClpg=; b=gxjZZbmHN8pn8Ak+FueQ7RZZ0hMyqyXxcieMvpVzc5lA17BiFw7jlWSZDlXUxVAPfH PV6iGT6crXvnNHNgPK58aqlqUtjLoBym4+took+IRmR1u9XZxwgDTDnKETWI2XVrMtzi kIZFxF7WTe9wLCPiGnCjOVDd2ZuUJWOmIxDBQ= 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:content-transfer-encoding; b=ulZgoaJGTYoFWiBIRYubRV8cWPb+GHewEXkAlxHfS3MybX+2QX+YzPb+jL9y/akKUd 0+oddeHzU8wzjSN522WHRpI1X47/tV/IFa1hvbu86IRAKsf6G2jLZaN2IfL2GLXov9Lk yjvA785YjhCE54Ah23Ab05rAnhDd7geJitdZ8= 2009/2/18 Michael Smith <drivers(a)mu.org>: > > By constraining the amount of virtual memory your process uses. > > Terry's trying to avoid explaining in detail the differences between > Darwin's approach to resource allocation (squeaky wheel) and Windows' (no > you can't have a pony) - the latter is probably where you're coming from. > > I would be happy to go into more detail, if you'd like. > > = Mike > > Sure, I'd like. Thanks in advance, - Maxim. _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… This email sent to site_archiver(a)lists.apple.com

1 0

Re: Set process' working set size explictly
by Maxim Zhuravlev 18 Feb '09

18 Feb '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:in-reply-to:references :date:message-id:subject:from:to:cc:content-type :content-transfer-encoding; bh=G2sE6zamK3iB7vat6HttTVpLM+3J2iUjvTt24X35MLg=; b=bZruohO9kVqyl3Qjig2035vI4LOxnPHOJUJC1bV2DAQz9Ynnl3Bt+M4DuafaUj5yF8 tiNLhaWEe8rd8X1mX3wXZYgFciOMmoel0ILRQXRM0ShCRlnVszqW8bt6/cvYDJV0XhIV HQiZ+yE8u39nXe9bfB7X0eoAqvRXvhzua0rVM= 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:content-transfer-encoding; b=P/Ry2irUoZ+rRsJt4rliZJb1TDGcOta5vm1+0/QEKyDhg8l36qW0a3SfAIeJrVjpg3 jeVfS8xyC8IuH7HX5UYEVq9b88UWQVJClCOJ58J3JFNF4RtJ4R0JxO6rlWi7oGP7JM7I dtEIvXERkhgHaRhVn03sr2MVB8L8szFbrBwKI= 2009/2/18 Terry Lambert <tlambert(a)apple.com>: > On Feb 18, 2009, at 6:02 AM, Maxim Zhuravlev wrote: >> >> Hi, all. >> >> Is there any (kernel- or user-mode) way to set a process' working set >> size explicitly in runtime? > > setrlimit(2) will set the resource limit, but it's purely voluntary. Good > code like gcc will look at this limit and voluntarily not exceed it. By what means do I *voluntary* not exceed it? I mean, given this, I can't ensure, how much *real* memory does my process consume. Then how do I not exceed it, if virtual memory consumed exceeds required working set? Thanks, - Maxim. _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… This email sent to site_archiver(a)lists.apple.com

1 0

Query on CrashReporter and atos
by Arnab Ganguly 18 Feb '09

18 Feb '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:date:message-id:subject :from:to:content-type; bh=iyWI+nX7sRqN1Ttl8NYBRxGSank78a/Qd+3xlBK4Kwo=; b=WE+HLK/RCNJvArX2i4GXbws9WnY2mVhyD+g/vusQ+DGyBuyqLGjU/YvDDpUMEcHA64 2XWSve8S0Hrq14a5WIm9Rh4tUCXwHhyN+uaGFGwAa7D07C27Pp6nYGbsVJeY+Vk0EvQS dMV5x0AHTbMQwHK8TQl4q0Finfv2Z2utI3stE= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:date:message-id:subject:from:to:content-type; b=JRHfa3Pkddlsev5Zn3JdLSdXr8AwL0DPTqb9znA2fk/bSf2hCNUF+WYaaqNJyMxVv5 futVlB1X3tKv1tZBi/0q8MIs5GFX4dHgexb+Sj51wwXzE9P3x4/C53IRwwFXNeVv+KL0 n8NxoAfbYCdcCqXsp35JEUPhUTHPMmPnA4KAM= _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… This email sent to site_archiver(a)lists.apple.com

1 0