Messages in this thread | | | Date | Mon, 27 Jul 1998 19:26:19 +0200 (MET DST) | From | Magnus Sjoegren <> | Subject | Re: VM: Found a writable swap-cached page! |
| |
Article: <35BCB546.21EBF0A0@transmeta.com> Date: Mon, 27 Jul 1998 10:13:42 -0700 Author: Bill Hawes <whawes@transmeta.com>
> > Jul 27 17:32:31 atlas kernel: page=c0234ac0@00065200, found=c0234ac0, > > count=3 > > Jul 27 17:32:31 atlas kernel: VM: Found a writable swap-cached page! > > Jul 27 17:32:31 atlas kernel: pte 871047, vma flags 000000fb, page > > flags 0000028c, count 2 > > Jul 27 17:32:31 atlas kernel: page=c023f530@00065800, found=c023f530, > > count=3 > > Hmm, very interesting. This time the pte present bit is set, so the page is > really there. The pte flags say dirty/user/rw/present, and the vma flags say > shared/write/read. Looks like it really shouldn't be swap-cached. > > So the writable swap-cache problem appears to be back in a different form. Can > you reproduce this easily? We need to figure out what sequence of events leads > up to this ... >
Sure,
1: Start X and netscape. 2: Switch to console and start any svgalib app, lets say zgv. View a nice picture. 3: Switch back to netscape and view a webpage. Move around the page as usual. Wait for syslog to fill.
That's all it takes on my machine (486 DX/2 66, 16MB). I first saw this while playing with some svgalib code, and I think any svgalib app will do. Tested with squake, zgv and the examples that come with svgalib. __ Magnus Sjögren <dat98msj@student3.lu.se>
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html
| |