[Intel-gfx] [PATCH] drm: Inject a cond_resched() into long drm_clflush_sg()

David Laight David.Laight at ACULAB.COM
Thu Jan 16 15:04:58 UTC 2020


From: David Laight
> Sent: 16 January 2020 14:41
> > I'll do some measurements later this afternoon.
> 
> This is an Ivy bridge cpu, so clflush (not clflushopt).
> With a cond_resched for every page I get:
> (Note these calls are every 10 seconds....)

For comparison some times booted with the original drm.ko

 1) # 3125.116 us |  drm_clflush_sg [drm]();
 0) # 3181.705 us |  drm_clflush_sg [drm]();
 1) # 3108.863 us |  drm_clflush_sg [drm]();
 1) # 3051.926 us |  drm_clflush_sg [drm]();
 2) # 3088.468 us |  drm_clflush_sg [drm]();
 2) # 3012.729 us |  drm_clflush_sg [drm]();
 2) # 3191.268 us |  drm_clflush_sg [drm]();
 3) # 3044.294 us |  drm_clflush_sg [drm]();
 0) # 3163.916 us |  drm_clflush_sg [drm]();
 2) # 3029.307 us |  drm_clflush_sg [drm]();
 2) # 3116.360 us |  drm_clflush_sg [drm]();
 2) # 3031.620 us |  drm_clflush_sg [drm]();
 0) # 3349.706 us |  drm_clflush_sg [drm]();

Probably nothing really significant.

	David

-
Registered Address Lakeside, Bramley Road, Mount Farm, Milton Keynes, MK1 1PT, UK
Registration No: 1397386 (Wales)


More information about the Intel-gfx mailing list