On 7/19/23 12:43, Peter Krempa wrote: > On Thu, Jun 29, 2023 at 17:13:42 +0200, Michal Privoznik wrote: >> Introduced in QEMU commit of v0.14.0-rc0~83^2~1 and not being >> able to compile the .removable attribute of the "usb-storage" >> object out, renders our corresponding capability >> QEMU_CAPS_USB_STORAGE_REMOVABLE useless. Retire it. >> >> Signed-off-by: Michal Privoznik <mprivozn@xxxxxxxxxx> >> --- >> src/qemu/qemu_capabilities.c | 9 +- >> src/qemu/qemu_capabilities.h | 2 +- >> src/qemu/qemu_command.c | 10 +- >> src/qemu/qemu_validate.c | 8 - >> .../caps_4.2.0_aarch64.replies | 195 ++++---------- >> .../caps_4.2.0_aarch64.xml | 1 - >> .../caps_4.2.0_ppc64.replies | 179 +++--------- >> .../qemucapabilitiesdata/caps_4.2.0_ppc64.xml | 1 - >> .../caps_4.2.0_x86_64.replies | 211 +++++---------- >> .../caps_4.2.0_x86_64.xml | 1 - >> .../caps_5.0.0_aarch64.replies | 211 ++++----------- >> .../caps_5.0.0_aarch64.xml | 1 - >> .../caps_5.0.0_ppc64.replies | 203 ++++---------- >> .../qemucapabilitiesdata/caps_5.0.0_ppc64.xml | 1 - >> .../caps_5.0.0_riscv64.replies | 191 +++---------- >> .../caps_5.0.0_riscv64.xml | 1 - >> .../caps_5.0.0_x86_64.replies | 227 +++++----------- >> .../caps_5.0.0_x86_64.xml | 1 - >> .../caps_5.1.0_x86_64.replies | 231 +++++----------- >> .../caps_5.1.0_x86_64.xml | 1 - >> .../caps_5.2.0_aarch64.replies | 219 ++++----------- >> .../caps_5.2.0_aarch64.xml | 1 - >> .../caps_5.2.0_ppc64.replies | 207 ++++---------- >> .../qemucapabilitiesdata/caps_5.2.0_ppc64.xml | 1 - >> .../caps_5.2.0_riscv64.replies | 195 +++----------- >> .../caps_5.2.0_riscv64.xml | 1 - >> .../caps_5.2.0_x86_64.replies | 235 +++++----------- >> .../caps_5.2.0_x86_64.xml | 1 - >> .../caps_6.0.0_aarch64.replies | 217 ++++----------- >> .../caps_6.0.0_aarch64.xml | 1 - >> .../caps_6.0.0_x86_64.replies | 233 +++++----------- >> .../caps_6.0.0_x86_64.xml | 1 - >> .../caps_6.1.0_x86_64.replies | 239 +++++----------- >> .../caps_6.1.0_x86_64.xml | 1 - >> .../caps_6.2.0_aarch64.replies | 223 ++++----------- >> .../caps_6.2.0_aarch64.xml | 1 - >> .../caps_6.2.0_ppc64.replies | 211 ++++----------- >> .../qemucapabilitiesdata/caps_6.2.0_ppc64.xml | 1 - >> .../caps_6.2.0_x86_64.replies | 243 +++++------------ >> .../caps_6.2.0_x86_64.xml | 1 - >> .../caps_7.0.0_aarch64+hvf.replies | 227 ++++------------ >> .../caps_7.0.0_aarch64+hvf.xml | 1 - >> .../caps_7.0.0_aarch64.replies | 227 ++++------------ >> .../caps_7.0.0_aarch64.xml | 1 - >> .../caps_7.0.0_ppc64.replies | 211 ++++----------- >> .../qemucapabilitiesdata/caps_7.0.0_ppc64.xml | 1 - >> .../caps_7.0.0_x86_64.replies | 243 +++++------------ >> .../caps_7.0.0_x86_64.xml | 1 - >> .../caps_7.1.0_ppc64.replies | 211 ++++----------- >> .../qemucapabilitiesdata/caps_7.1.0_ppc64.xml | 1 - >> .../caps_7.1.0_x86_64.replies | 243 +++++------------ >> .../caps_7.1.0_x86_64.xml | 1 - >> .../caps_7.2.0_ppc.replies | 211 +++------------ >> tests/qemucapabilitiesdata/caps_7.2.0_ppc.xml | 1 - >> .../caps_7.2.0_x86_64+hvf.replies | 255 +++++------------- >> .../caps_7.2.0_x86_64+hvf.xml | 1 - >> .../caps_7.2.0_x86_64.replies | 255 +++++------------- >> .../caps_7.2.0_x86_64.xml | 1 - >> .../caps_8.0.0_riscv64.replies | 207 +++----------- >> .../caps_8.0.0_riscv64.xml | 1 - >> .../caps_8.0.0_x86_64.replies | 255 +++++------------- >> .../caps_8.0.0_x86_64.xml | 1 - >> .../caps_8.1.0_x86_64.replies | 255 +++++------------- >> .../caps_8.1.0_x86_64.xml | 1 - >> 64 files changed, 1704 insertions(+), 5025 deletions(-) > > Please split this at least into a patch that retires the capability > (removes the code), you can include removal of the detection of the > capability but usually split that separately too. > > For those one/two patches you can use: > > Reviewed-by: Peter Krempa <pkrempa@xxxxxxxxxx> > > A separate patch then drops the actual probing of the usb-storage qemu > device. > > Additionally consider querying the device regardless. In case when there > will need to be another feature to be probed all of the .replies changes > will need to be re-added. > > In case of the USB device there's still possibility that some new > feature will be added. > > I've recently dropped the WWN detection from the IDE device and left in > the probing of the device itself. Whether that will get any improvement > is unlikely though. > > If you decide that USB storage is not likely get detectable features > then for the last patch you can also use: > > Reviewed-by: Peter Krempa <pkrempa@xxxxxxxxxx> > Fair enough. So let me just retire the cap and leave device probing be. We might need it at some point. Thanks! Michal
