Messages in this thread | | | Date | Fri, 11 Apr 2008 22:18:17 -0700 | From | Andrew Morton <> | Subject | Re: [PATCH 1/2] Driver for Freescale 8610 and 5121 DIU |
| |
On Fri, 11 Apr 2008 23:45:35 +0200 Jiri Slaby <jirislaby@gmail.com> wrote:
> On 03/12/2008 11:20 PM, Jiri Slaby wrote: > > On 03/12/2008 10:43 PM, York Sun wrote: > >> +static int fsl_diu_open(struct fb_info *info, int user) > >> +{ > >> + struct mfb_info *mfbi = info->par; > >> + int res = 0; > >> + > >> + spin_lock(&diu_lock); > >> + mfbi->count++; > >> + if (mfbi->count == 1) { > >> + DPRINTK("open plane index %d\n", mfbi->index); > >> + fsl_diu_check_var(&info->var, info); > >> + fsl_diu_set_par(info); > > > > Please retest your code (at least) with sleep-inside spinlock debug > > option. If I see correctly you call GFP_KERNEL allocation somewhere > > deeper in this function, which might sleep. > > > > ... > > ping. > > Seeing this in -mm yet. Are those comments all wrong? Are you working on it? >
(top-posting repaired)
Thanks. I've made a note that this patch has outstanding issues. Usually this means that I'll defer merging it until they have been addressed: either by fixing them or by successfully arguing against the objections.
| |