Re: OpenSSI preview on Debian Lenny (on x86_64)
Roger Tsang <[email protected]> Sat, 17 Apr 2010 18:03:47 -0400
| Newsgroups | gmane.linux.cluster.ssic.devel |
|---|---|
| Message-ID | <[email protected]> |
--===============2260041280928682437== Content-Type: multipart/alternative; boundary=000325556512f10493048475e69d --000325556512f10493048475e69d Content-Type: text/plain; charset=ISO-8859-1 Also support for 4-level page table is not yet there in vproc/as_xscribe.c. It currently skips the PUD table. That is probably why process migration isn't working properly in x86_64. I'm adding code to traverse the PUD table as we speak. On Sat, Apr 17, 2010 at 4:29 PM, Roger Tsang <[email protected]> wrote: > Further it appears as_do_pg() is just a dupe of Linux install_arg_page(). > To avoid this duplication it is better to modify as_do_pg() to call > something like ssi_install_arg_page() which is install_arg_page() without > force_sig() in error path. > > On Sat, Apr 17, 2010 at 2:36 PM, John Hughes <[email protected]> wrote: > >> John Hughes wrote: >> >> John Hughes wrote: >>> >>>> >>>> pud = pud_offset(pgd, apip->api_addr); >>>> >>> It seems than when we get here "pgd_none (pgd)" is true, so maybe we >>> should be doing a pud_alloc instead of just "pud = pud_offset(...)". >>> >>> But it doesn't seem to work, that just makes us hang. >>> >> Because I screwed up. >> >> Yes, we should be doing pud_alloc, not just pud_offset. >> >> This patch makes migration work. >> >> Well, almost. The migrated process seems not to be able to access its >> terminal. >> >> >> >> >> >> ------------------------------------------------------------------------------ >> Download Intel® Parallel Studio Eval >> Try the new software tools for yourself. Speed compiling, find bugs >> proactively, and fine-tune applications for parallel performance. >> See why Intel Parallel Studio got high marks during beta. >> http://p.sf.net/sfu/intel-sw-dev >> _______________________________________________ >> ssic-linux-devel mailing list >> [email protected] >> https://lists.sourceforge.net/lists/listinfo/ssic-linux-devel >> >> > --000325556512f10493048475e69d Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Also support for 4-level page table is not yet there in vproc/as_xscribe.c.= It currently skips the PUD table.=A0 That is probably why process migratio= n isn't working properly in x86_64.=A0 I'm adding code to traverse = the PUD table as we speak.<br> <br><div class=3D"gmail_quote">On Sat, Apr 17, 2010 at 4:29 PM, Roger Tsang= <span dir=3D"ltr"><<a href=3D"mailto:[email protected]">roger.tsang= @gmail.com</a>></span> wrote:<br><blockquote class=3D"gmail_quote" style= =3D"border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; p= adding-left: 1ex;"> Further it appears as_do_pg() is just a dupe of Linux install_arg_page().= =A0 To avoid this duplication it is better to modify as_do_pg() to call somethi= ng like ssi_install_arg_page() which is install_arg_page() without force_si= g() in error path.<br><br><div class=3D"gmail_quote"><div><div></div><div c= lass=3D"h5"> On Sat, Apr 17, 2010 at 2:36 PM, John Hughes <span dir=3D"ltr"><<a href= =3D"mailto:[email protected]" target=3D"_blank">[email protected]</a>></span> = wrote:<br> </div></div><blockquote class=3D"gmail_quote" style=3D"border-left: 1px sol= id rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><div>= <div></div><div class=3D"h5">John Hughes wrote:<div><br> <blockquote class=3D"gmail_quote" style=3D"border-left: 1px solid rgb(204, = 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"> John Hughes wrote:<br> <blockquote class=3D"gmail_quote" style=3D"border-left: 1px solid rgb(204, = 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"> <br> =A0 =A0 =A0 pud =3D pud_offset(pgd, apip->api_addr);<br> </blockquote> It seems than when we get here "pgd_none (pgd)" is true, so maybe= we should be doing a pud_alloc instead of just "pud =3D pud_offset(..= .)".<br> <br> But it doesn't seem to work, that just makes us hang.<br> </blockquote></div> Because I screwed up.<br> <br> Yes, we should be doing pud_alloc, not just pud_offset.<br> <br> This patch makes migration work.<br> <br> Well, almost. =A0The migrated process seems not to be able to access its te= rminal.<br> <br> <br> <br> <br></div></div><div class=3D"im">-----------------------------------------= -------------------------------------<br> Download Intel&#174; Parallel Studio Eval<br> Try the new software tools for yourself. Speed compiling, find bugs<br> proactively, and fine-tune applications for parallel performance.<br> See why Intel Parallel Studio got high marks during beta.<br> <a href=3D"http://p.sf.net/sfu/intel-sw-dev" target=3D"_blank">http://p.sf.= net/sfu/intel-sw-dev</a><br>_______________________________________________= <br> ssic-linux-devel mailing list<br> <a href=3D"mailto:[email protected]" target=3D"_blank"= >[email protected]</a><br> <a href=3D"https://lists.sourceforge.net/lists/listinfo/ssic-linux-devel" t= arget=3D"_blank">https://lists.sourceforge.net/lists/listinfo/ssic-linux-de= vel</a><br> <br></div></blockquote></div><br> </blockquote></div><br> --000325556512f10493048475e69d-- --===============2260041280928682437== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline ------------------------------------------------------------------------------ Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev --===============2260041280928682437== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ ssic-linux-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/ssic-linux-devel --===============2260041280928682437==--