Re: port-amd64/54988: possible memory leaks/swap problems

[mlh%goathill.org@localhost (MLH)]

The following reply was made to PR port-amd64/54988; it has been noted by GNATS.

From: mlh%goathill.org@localhost (MLH)
To: gnats-bugs%netbsd.org@localhost
Cc: ad%netbsd.org@localhost, gnats-admin%netbsd.org@localhost, netbsd-bugs%netbsd.org@localhost, 
 mlh%goathill.org@localhost
Subject: Re: port-amd64/54988: possible memory leaks/swap problems
Date: Thu, 26 Mar 2020 21:42:48 -0400 (EDT)

 David Holland wrote:
 > The following reply was made to PR port-amd64/54988; it has been noted by GNATS.
 > 
 > From: David Holland <dholland-bugs%netbsd.org@localhost>
 > To: gnats-bugs%netbsd.org@localhost
 > Cc: 
 > Subject: Re: port-amd64/54988: possible memory leaks/swap problems
 > Date: Fri, 27 Mar 2020 00:13:56 +0000
 > 
 >  On Tue, Mar 24, 2020 at 09:25:02PM +0000, Andrew Doran wrote:
 >   >> kmem-00192          256     4088775       0         182  4088593       15     255538          0  255538   4096 255538     0    inf    0 0x10000 100.0%
 >   >  
 >   > That's about a gigabyte of kernel memory leaked.  I have a radeon card here.
 >   > I'll try to reproduce it.
 >  
 >  I have been seeing kmem-192 leaks on a machine with a radeon as well,
 >  but nowhere as severe as described in this PR (takes weeks to burn
 >  through a couple gigs).
 
 It can take mine as little as about four hours to burn through 4G.
 The last time I checked, X didn't have to be running for it to
 happen, but I haven't tried that in about a week.
 
 [     1.027936] radeon0 at pci1 dev 0 function 0: ATI Technologies Radeon HD 6450 (rev. 0x00)
 [     8.893648] radeon0: info: VRAM: 2048M 0x0000000000000000 - 0x000000007FFFFFFF (2048M used)
 [     8.893648] radeon0: info: GTT: 1024M 0x0000000080000000 - 0x00000000BFFFFFFF
 [     8.893648] kern info: [drm] radeon: 2048M of VRAM memory ready
 [     8.893648] kern info: [drm] radeon: 1024M of GTT memory ready.
 [     8.959977] kern info: [drm] radeon: dpm initialized
 [     9.009997] radeon0: info: WB enabled
 [     9.009997] radeon0: info: fence driver on ring 0 use gpu addr 0x0000000080000c00 and cpu addr 0x0xffffbfcb7bd1fc00
 [     9.009997] radeon0: info: fence driver on ring 3 use gpu addr 0x0000000080000c0c and cpu addr 0x0xffffbfcb7bd1fc0c
 [     9.020001] radeon0: info: fence driver on ring 5 use gpu addr 0x0000000000072118 and cpu addr 0x0xffffc300780b2118
 [     9.020001] radeon0: info: radeon: MSI limited to 32-bit
 [     9.020001] radeon0: info: radeon: using MSI.
 [     9.020001] radeon0: interrupting at msi3 vec 0 (radeon0)
 [     9.020001] kern info: [drm] radeon: irq initialized.
 [    10.030413] radeondrmkmsfb0 at radeon0
 [    10.030413] radeondrmkmsfb0: framebuffer at 0xffffc300786e3000, size 1920x1200, depth 32, stride 7680
 [    10.650666] wsdisplay0 at radeondrmkmsfb0 kbdmux 1: console (default, vt100 emulation), using wskbd0
 [    68.046323] kern error: [drm:(/usr/src/sys/external/bsd/drm2/dist/drm/radeon/radeon_btc_dpm.c:2319)btc_dpm_set_power_state] *ERROR* rv770_restrict_performance_levels_before_switch failed