Date: Wed, 22 Aug 2001 01:58:00 -0700 From: Terry Lambert <tlambert2@mindspring.com> To: Peter Wemm <peter@wemm.org> Cc: Mitsuru IWASAKI <iwasaki@jp.FreeBSD.org>, arch@FreeBSD.ORG, audit@FreeBSD.ORG, kumabu@t3.rim.or.jp Subject: Re: CFR: Timing to enable CR4.PGE bit Message-ID: <3B837418.2D5529E5@mindspring.com> References: <20010822082749.ED86038FD@overcee.netplex.com.au>
next in thread | previous in thread | raw e-mail | index | archive | help
Peter Wemm wrote: > Page Global Enable (bit 7 of CR4). (Introduced in the P6 family > processors.) Enables the global page feature when set; disables the > global page feature when clear. [snip] In addition, the bit must not > ^^^^^^^^^^^^^^^^ > be enabled before paging is enabled via CR0.PG. Program correctness > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > may be affected by reversing this sequence, and processor performance > will be impacted. > ---- > > Currently, we enable CR4.PGE bit in create_pagetables, then enable > CR0.PG in locore.s. This seems to violate Intel's note. Ah. It looked like an inverted patch to me... it moved the code from line 725 or so in locore.s to line 375. I think moving this to _much_ later would not be a bad idea, since the only effect will be that the PG_G bit is ignored until it's actually enabled, right? FWIW: The message you quoted was from a cross-post from -audit to -arch, which I guess originally came from -audit? So the answer is "no, I didn't see the whole thread", and I only became interested enough to read it when I saw the assemply patch fly by on -arch. So since it's moving the enable to much later in the process (but earlier in the file -- in the old days, we used to program in order 8-)), I definitely don't object, unless I can talk you guys into delaying it until the 4M page for the kernel is being enabled... > .. which is why I asked if you read the entire thread. I thought that > quote was quite clear that we were in violation. I don't subscribe to audit, and the quoted stuff didn't look that important, so I didn't see it until now... sorry. -- Terry To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-arch" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3B837418.2D5529E5>