On 12/22/2017 10:48 AM, Bjørn Mork wrote: > Julian Labus <jul...@labus-online.de> writes: > >> diff --git a/package/utils/usbmode/Makefile b/package/utils/usbmode/Makefile >> index 7f0195fdce..1b1ef147b6 100644 >> --- a/package/utils/usbmode/Makefile >> +++ b/package/utils/usbmode/Makefile >> @@ -40,8 +40,13 @@ endef >> define Build/Prepare >> $(Build/Prepare/Default) >> tar xvfj $(DL_DIR)/$(PKG_DATA_FILENAME) -C $(PKG_BUILD_DIR) >> - rm -f \ >> - >> $(PKG_BUILD_DIR)/$(PKG_DATA_PATH)/usb_modeswitch.d/05c6:1000:sVe=GT >> + #remove devices with unsupported modes >> + for filevar in $(PKG_BUILD_DIR)/$(PKG_DATA_PATH)/usb_modeswitch.d/* ; \ >> + do \ >> + if grep -q -E '(Quanta|Option|Blackberry|Pantech)Mode' >> "$$$$filevar" ; then \ >> + rm "$$$$filevar" ; \ >> + fi \ >> + done >> cp ./data/* $(PKG_BUILD_DIR)/$(PKG_DATA_PATH)/usb_modeswitch.d/ >> #in order to keep the Lede GIT repo free of filenames with colons, >> #we name the files xxxx-yyyy > > > Sorry for the extremely late comment on this. I missed the discussion > when it happened, but was made aware of it now because users with Option > modems complain that the modems have stopped working with LEDE. > > This patch is papering over a regression introduced by updating the data > package. I strongly object to that sort of bug "fixing". Option modems > used to work with the old data package, as it included the explicit > message for each Option device entry. Adding new modems does not > justify dropping support for older modems. > > Please fix this properly instead if you want to update to the config > data.
Hi Bjørn, I didn't recognize that new modes where introduced for existing devices. To me it looked like they where introduced to support previously unsupported devices. Knowing that, I wouldn't have submitted the patch. I saw that you already updated usbmode to support the missing modes. Once the package is updated accordingly my "patch" can be reverted. Kind regards, Julian Kind regards, Julian _______________________________________________ Lede-dev mailing list Lede-dev@lists.infradead.org http://lists.infradead.org/mailman/listinfo/lede-dev
