On Sun, Feb 07, 2016 at 11:08:07AM +0200, Michael S. Tsirkin wrote: > On Tue, Jan 26, 2016 at 02:50:25PM +0100, Igor Mammedov wrote: > > From: Roman Kagan <rka...@virtuozzo.com> > > -static Aml *build_fdc_device_aml(void) > > +static Aml *build_fdinfo_aml(int idx, uint8_t type, uint8_t cylinders, > > + uint8_t heads, uint8_t sectors) > > acpi spec says these are WORD values. Are they really uint8_t?
Yes, see struct FDrive. > > + fdi = aml_package(0x10); > > Why 0x10 and not 16? I'm unaware of the difference... If you have any preference I'll adjust. (Originally I stuck with hex because so did iasl -d and the then hand-written ASL). > > + aml_append(fdi, aml_int(idx)); /* Drive Number */ > > + aml_append(fdi, > > + aml_int(cmos_get_fd_drive_type(type))); /* Device Type */ > > + aml_append(fdi, > > + aml_int(cylinders - 1)); /* Maximum Cylinder Number */ > > + aml_append(fdi, aml_int(sectors)); /* Maximum Sector Number */ > > + aml_append(fdi, aml_int(heads - 1)); /* Maximum Head Number */ > > Doesn't above change on media change? I guess no, because this IMHO is supposed to describe the drive properties, not the diskette properties. But I'll double-check. I'm confused about the status of this patchset: I saw you post a pull request with this series last week, and now you review it. What should I do now in response to your comments: rework and resubmit it or post incremental fixes on top of it? Thanks, Roman.