Skip to contentSkip to conversation

Remove the deprecated bluetooth subsystem

PATCH In review Thomas Huth 7 participants 23 messages Nov 20, 2019 – Feb 6, 2020 Kbuild

lore
Share & tools
lore
#0 lore raw

cover Remove the deprecated bluetooth subsystem

This patch series removes the bitrotten bluetooth subsystem. See the patch description of the third patch for the rationale.

Thomas Huth (4):
hw/arm/nseries: Replace the bluetooth chardev with a "null" chardev hw/usb: Remove the USB bluetooth dongle device
Remove the core bluetooth code
Remove libbluetooth / bluez from the CI tests

 .gitlab-ci.yml                             |    2 +-
 Makefile.objs                              |    2 -
 bt-host.c                                  |  198 --
 bt-vhci.c                                  |  167 --
 configure                                  |   31 -
 hw/Kconfig                                 |    1 -
 hw/Makefile.objs                           |    1 -
 hw/arm/nseries.c                           |   16 +-
 hw/bt/Kconfig                              |    2 -
 hw/bt/Makefile.objs                        |    3 -
 hw/bt/core.c                               |  143 --
 hw/bt/hci-csr.c                            |  512 -----
 hw/bt/hci.c                                | 2263 --------------------
 hw/bt/hid.c                                |  553 -----
 hw/bt/l2cap.c                              | 1367 ------------
 hw/bt/sdp.c                                |  989 ---------
 hw/usb/Kconfig                             |    5 -
 hw/usb/Makefile.objs                       |    1 -
 hw/usb/dev-bluetooth.c                     |  581 -----
 include/hw/bt.h                            | 2177 -------------------
 include/sysemu/bt.h                        |   20 -
 qemu-deprecated.texi                       |    7 -
 qemu-doc.texi                              |   17 -
 qemu-options.hx                            |   79 -
 tests/docker/dockerfiles/fedora.docker     |    1 -
 tests/docker/dockerfiles/ubuntu.docker     |    1 -
 tests/docker/dockerfiles/ubuntu1804.docker |    1 -
 vl.c                                       |  136 --
 28 files changed, 8 insertions(+), 9268 deletions(-)
 delete mode 100644 bt-host.c
 delete mode 100644 bt-vhci.c
 delete mode 100644 hw/bt/Kconfig
 delete mode 100644 hw/bt/Makefile.objs
 delete mode 100644 hw/bt/core.c
 delete mode 100644 hw/bt/hci-csr.c
 delete mode 100644 hw/bt/hci.c
 delete mode 100644 hw/bt/hid.c
 delete mode 100644 hw/bt/l2cap.c
 delete mode 100644 hw/bt/sdp.c
 delete mode 100644 hw/usb/dev-bluetooth.c
 delete mode 100644 include/hw/bt.h
 delete mode 100644 include/sysemu/bt.h
Signature · 1 line
2.23.0
#1 lore raw

PATCH 1/4 hw/arm/nseries: Replace the bluetooth chardev with a "null" chardev

We are going to remove the bitrotten bluetooth backend, so we can not use it in the n800 and n810 machines anymore. Replace the chardev here with a "null" chardev and stop including the "hw/bt.h" header.

I am not aware of a full working N8x0 environment that can be used to test these machines, but the Linux kernel from the Meego project that can be found on http://stskeeps.subnetmask.net/meego-n8x0/ still seems to boot as far as it did before.

---
 hw/arm/nseries.c | 16 +++++++---------
 qemu-doc.texi    |  2 --
 2 files changed, 7 insertions(+), 11 deletions(-)
hw/arm/nseries.c +7 -9
index 7e361936a9..3fd196fb30 100644
@@ -21,6 +21,7 @@
 #include "qemu/osdep.h"
 #include "qapi/error.h"
 #include "cpu.h"
+#include "chardev/char.h"
 #include "qemu/cutils.h"
 #include "qemu/bswap.h"
 #include "sysemu/reset.h"
@@ -39,7 +40,6 @@
 #include "hw/qdev-properties.h"
 #include "hw/block/flash.h"
 #include "hw/hw.h"
-#include "hw/bt.h"
 #include "hw/loader.h"
 #include "hw/sysbus.h"
 #include "qemu/log.h"
@@ -792,13 +792,11 @@ static void n8x0_cbus_setup(struct n800_s *s)
 
 static void n8x0_uart_setup(struct n800_s *s)
 {
-    Chardev *radio = uart_hci_init();
-
-    qdev_connect_gpio_out(s->mpu->gpio, N8X0_BT_RESET_GPIO,
-                    csrhci_pins_get(radio)[csrhci_pin_reset]);
-    qdev_connect_gpio_out(s->mpu->gpio, N8X0_BT_WKUP_GPIO,
-                    csrhci_pins_get(radio)[csrhci_pin_wakeup]);
-
+    Chardev *radio = qemu_chr_new("bt-dummy-uart", "null", NULL);
+    /*
+     * Note: We used to connect N8X0_BT_RESET_GPIO and N8X0_BT_WKUP_GPIO
+     * here, but this code has been removed with the bluetooth backend.
+     */
     omap_uart_attach(s->mpu->uart[BT_UART], radio);
 }
 
@@ -1137,7 +1135,7 @@ static struct omap_partition_info_s {
     { 0, 0, 0, NULL }
 };
 
-static bdaddr_t n8x0_bd_addr = {{ N8X0_BD_ADDR }};
+static uint8_t n8x0_bd_addr[6] = { N8X0_BD_ADDR };
 
 static int n8x0_atag_setup(void *p, int model)
 {
qemu-doc.texi +0 -2
index 3ddf5c0a68..ed3df01dac 100644
@@ -2308,8 +2308,6 @@ Secure Digital card connected to OMAP MMC/SD host
 @item
 Three OMAP on-chip UARTs and on-chip STI debugging console
 @item
-A Bluetooth(R) transceiver and HCI connected to an UART
-@item
 Mentor Graphics "Inventra" dual-role USB controller embedded in a TI
 TUSB6010 chip - only USB host mode is supported
 @item
Signature · 1 line
2.23.0
#3 lore raw

PATCH 2/4 hw/usb: Remove the USB bluetooth dongle device

In reply to Thomas Huth

We are going to remove the bluetooth backend, so the USB bluetooth dongle can not work anymore. It's a completely optional device, no board depends on it, so let's simply remove it now.

---
 hw/usb/Kconfig         |   5 -
 hw/usb/Makefile.objs   |   1 -
 hw/usb/dev-bluetooth.c | 581 -----------------------------------------
 qemu-doc.texi          |  15 --
 4 files changed, 602 deletions(-)
 delete mode 100644 hw/usb/dev-bluetooth.c
hw/usb/Kconfig +0 -5
index 555e09377b..5e70ed5f7b 100644
@@ -82,11 +82,6 @@ config USB_NETWORK
     default y
     depends on USB
 
-config USB_BLUETOOTH
-    bool
-    default y
-    depends on USB
-
 config USB_SMARTCARD
     bool
     default y
hw/usb/Makefile.objs +0 -1
index 303ac084a0..0ab20f9d73 100644
@@ -25,7 +25,6 @@ common-obj-$(CONFIG_USB_STORAGE_UAS)  += dev-uas.o
 common-obj-$(CONFIG_USB_AUDIO)        += dev-audio.o
 common-obj-$(CONFIG_USB_SERIAL)       += dev-serial.o
 common-obj-$(CONFIG_USB_NETWORK)      += dev-network.o
-common-obj-$(CONFIG_USB_BLUETOOTH)    += dev-bluetooth.o
 
 ifeq ($(CONFIG_USB_SMARTCARD),y)
 common-obj-y                          += dev-smartcard-reader.o
hw/usb/dev-bluetooth.c deleted +0 -581

581 lines not on this page · raw message

qemu-doc.texi +0 -15
index ed3df01dac..eea91a2d1e 100644
@@ -983,21 +983,6 @@ For instance, user-mode networking can be used with
 Smartcard reader device
 @item usb-audio
 USB audio device
-@item usb-bt-dongle
-Bluetooth dongle for the transport layer of HCI. It is connected to HCI
-scatternet 0 by default (corresponds to @code{-bt hci,vlan=0}).
-Note that the syntax for the @code{-device usb-bt-dongle} option is not as
-useful yet as it was with the legacy @code{-usbdevice} option. So to
-configure an USB bluetooth device, you might need to use
-"@code{-usbdevice bt}[:@var{hci-type}]" instead. This configures a
-bluetooth dongle whose type is specified in the same format as with
-the @option{-bt hci} option, @pxref{bt-hcis,,allowed HCI types}.  If
-no type is given, the HCI logic corresponds to @code{-bt hci,vlan=0}.
-This USB device implements the USB Transport Layer of HCI.  Example
-usage:
-@example
-@command{@value{qemu_system}} [...@var{OPTIONS}...] @option{-usbdevice} bt:hci,vlan=3 @option{-bt} device:keyboard,vlan=3
-@end example
 @end table
 
 @node host_usb_devices
Signature · 1 line
2.23.0
#5 lore raw

PATCH 3/4 Remove the core bluetooth code

In reply to Thomas Huth

It's been deprecated since QEMU v3.1. We've explicitly asked in the deprecation message that people should speak up on qemu-devel in case they are still actively using the bluetooth part of QEMU, but nobody ever replied that they are really still using it.

I've tried it on my own to use this bluetooth subsystem for one of my guests, but I was also not able to get it running anymore: When I was trying to pass-through a real bluetooth device, either the guest did not see the device at all, or the guest crashed.

Even worse for the emulated device: When running

qemu-system-x86_64 -bt device:keyboard

QEMU crashes once you hit a key.

So it seems like the bluetooth stack is not only neglected, it is completely bitrotten, as far as I can tell. The only attention that this code got during the past years were some CVEs that have been spotted there. So this code is a burden for the developers, without any real benefit anymore. Time to remove it.

---
 Makefile.objs        |    2 -
 bt-host.c            |  198 ----
 bt-vhci.c            |  167 ----
 configure            |   31 -
 hw/Kconfig           |    1 -
 hw/Makefile.objs     |    1 -
 hw/bt/Kconfig        |    2 -
 hw/bt/Makefile.objs  |    3 -
 hw/bt/core.c         |  143 ---
 hw/bt/hci-csr.c      |  512 ----------
 hw/bt/hci.c          | 2263 ------------------------------------------
 hw/bt/hid.c          |  553 -----------
 hw/bt/l2cap.c        | 1367 -------------------------
 hw/bt/sdp.c          |  989 ------------------
 include/hw/bt.h      | 2177 ----------------------------------------
 include/sysemu/bt.h  |   20 -
 qemu-deprecated.texi |    7 -
 qemu-options.hx      |   79 --
 vl.c                 |  136 ---
 19 files changed, 8651 deletions(-)
 delete mode 100644 bt-host.c
 delete mode 100644 bt-vhci.c
 delete mode 100644 hw/bt/Kconfig
 delete mode 100644 hw/bt/Makefile.objs
 delete mode 100644 hw/bt/core.c
 delete mode 100644 hw/bt/hci-csr.c
 delete mode 100644 hw/bt/hci.c
 delete mode 100644 hw/bt/hid.c
 delete mode 100644 hw/bt/l2cap.c
 delete mode 100644 hw/bt/sdp.c
 delete mode 100644 include/hw/bt.h
 delete mode 100644 include/sysemu/bt.h
Makefile.objs +0 -2
index 11ba1a36bd..4e4d6ddbb3 100644
@@ -65,8 +65,6 @@ common-obj-y += replay/
 
 common-obj-y += ui/
 common-obj-m += ui/
-common-obj-y += bt-host.o bt-vhci.o
-bt-host.o-cflags := $(BLUEZ_CFLAGS)
 
 common-obj-y += dma-helpers.o
 common-obj-y += vl.o
bt-host.c deleted +0 -198

198 lines not on this page · raw message

bt-vhci.c deleted +0 -167

167 lines not on this page · raw message

configure +0 -31
index 6099be1d84..ecce4ada2d 100755
@@ -349,7 +349,6 @@ unset target_list_exclude
 # Distributions want to ensure that several features are compiled in, and it
 # is impossible without a --enable-foo that exits if a feature is not found.
 
-bluez=""
 brlapi=""
 curl=""
 curses=""
@@ -1151,10 +1150,6 @@ for opt do
   ;;
   --enable-brlapi) brlapi="yes"
   ;;
-  --disable-bluez) bluez="no"
-  ;;
-  --enable-bluez) bluez="yes"
-  ;;
   --disable-kvm) kvm="no"
   ;;
   --enable-kvm) kvm="yes"
@@ -1762,7 +1757,6 @@ disabled with --disable-FEATURE, default is enabled if available:
   curl            curl connectivity
   membarrier      membarrier system call (for Linux 4.14+ or Windows)
   fdt             fdt device tree
-  bluez           bluez stack connectivity
   kvm             KVM acceleration support
   hax             HAX acceleration support
   hvf             Hypervisor.framework acceleration support
@@ -3665,26 +3659,6 @@ EOF
   fi
 fi # test "$curl"
 
-##########################################
-# bluez support probe
-if test "$bluez" != "no" ; then
-  cat > $TMPC << EOF
-#include <bluetooth/bluetooth.h>
-int main(void) { return bt_error(0); }
-EOF
-  bluez_cflags=$($pkg_config --cflags bluez 2>/dev/null)
-  bluez_libs=$($pkg_config --libs bluez 2>/dev/null)
-  if compile_prog "$bluez_cflags" "$bluez_libs" ; then
-    bluez=yes
-    libs_softmmu="$bluez_libs $libs_softmmu"
-  else
-    if test "$bluez" = "yes" ; then
-      feature_not_found "bluez" "Install bluez-libs/libbluetooth devel"
-    fi
-    bluez="no"
-  fi
-fi
-
 ##########################################
 # glib support probe
 
@@ -6509,7 +6483,6 @@ if test "$xen" = "yes" ; then
   echo "xen ctrl version  $xen_ctrl_version"
 fi
 echo "brlapi support    $brlapi"
-echo "bluez  support    $bluez"
 echo "Documentation     $docs"
 echo "PIE               $pie"
 echo "vde support       $vde"
@@ -6933,10 +6906,6 @@ if test "$brlapi" = "yes" ; then
   echo "CONFIG_BRLAPI=y" >> $config_host_mak
   echo "BRLAPI_LIBS=$brlapi_libs" >> $config_host_mak
 fi
-if test "$bluez" = "yes" ; then
-  echo "CONFIG_BLUEZ=y" >> $config_host_mak
-  echo "BLUEZ_CFLAGS=$bluez_cflags" >> $config_host_mak
-fi
 if test "$gtk" = "yes" ; then
   echo "CONFIG_GTK=m" >> $config_host_mak
   echo "GTK_CFLAGS=$gtk_cflags" >> $config_host_mak
hw/Kconfig +0 -1
index b9685b3944..ecf491bf04 100644
@@ -4,7 +4,6 @@ source acpi/Kconfig
 source adc/Kconfig
 source audio/Kconfig
 source block/Kconfig
-source bt/Kconfig
 source char/Kconfig
 source core/Kconfig
 source display/Kconfig
hw/Makefile.objs +0 -1
index 66eef20561..660e2b4373 100644
@@ -5,7 +5,6 @@ devices-dirs-y += acpi/
 devices-dirs-y += adc/
 devices-dirs-y += audio/
 devices-dirs-y += block/
-devices-dirs-y += bt/
 devices-dirs-y += char/
 devices-dirs-y += cpu/
 devices-dirs-y += display/
hw/bt/Kconfig deleted +0 -2
deleted file mode 100644
index 554a9ee75e..0000000000
@@ -1,2 +0,0 @@
-config BLUETOOTH
-    bool
hw/bt/Makefile.objs deleted +0 -3
deleted file mode 100644
index 867a7d2e8a..0000000000
@@ -1,3 +0,0 @@
-common-obj-y += core.o l2cap.o sdp.o hci.o hid.o
-common-obj-y += hci-csr.o
-
hw/bt/core.c deleted +0 -143

143 lines not on this page · raw message

hw/bt/hci-csr.c deleted +0 -512

512 lines not on this page · raw message

hw/bt/hci.c deleted +0 -2263

1,886 lines not on this page · raw message

hw/bt/hid.c deleted +0 -553

553 lines not on this page · raw message

hw/bt/l2cap.c deleted +0 -1367

1,367 lines not on this page · raw message

hw/bt/sdp.c deleted +0 -989

989 lines not on this page · raw message

include/hw/bt.h deleted +0 -2177

2,177 lines not on this page · raw message

include/sysemu/bt.h deleted +0 -20
deleted file mode 100644
index 2fd8c0f14b..0000000000

Lines of this file are not shown. View the raw message.

qemu-deprecated.texi +0 -7
index 4b4b7425ac..cba73eed07 100644

Lines of this file are not shown. View the raw message.

qemu-options.hx +0 -79

79 lines not on this page · raw message

vl.c +0 -136

136 lines not on this page · raw message

The rest of this diff is not shown. View the raw message.

Signature · 1 line
2.23.0

Part of the diff is not shown here. View the raw message.

#21 lore raw

PATCH 4/4 Remove libbluetooth / bluez from the CI tests

In reply to Thomas Huth

Since the bluetooth code has been removed, we don't need to test with this library anymore.

---
 .gitlab-ci.yml                             | 2 +-
 tests/docker/dockerfiles/fedora.docker     | 1 -
 tests/docker/dockerfiles/ubuntu.docker     | 1 -
 tests/docker/dockerfiles/ubuntu1804.docker | 1 -
 4 files changed, 1 insertion(+), 4 deletions(-)
.gitlab-ci.yml +1 -1
index be57c6a454..a28e389794 100644
@@ -36,7 +36,7 @@ build-disabled:
 
 build-tcg-disabled:
  script:
- - apt-get install -y -qq clang libgtk-3-dev libbluetooth-dev libusb-dev
+ - apt-get install -y -qq clang libgtk-3-dev libusb-dev
  - ./configure --cc=clang --enable-werror --disable-tcg --audio-drv-list=""
  - make -j2
  - make check-unit
tests/docker/dockerfiles/fedora.docker +0 -1
index 4ddc7dd112..f143b26cf3 100644
@@ -2,7 +2,6 @@ FROM fedora:30
 ENV PACKAGES \
     bc \
     bison \
-    bluez-libs-devel \
     brlapi-devel \
     bzip2 \
     bzip2-devel \
tests/docker/dockerfiles/ubuntu.docker +0 -1
index f486492224..1b3b2693d2 100644
@@ -19,7 +19,6 @@ ENV PACKAGES flex bison \
     glusterfs-common \
     libaio-dev \
     libattr1-dev \
-    libbluetooth-dev \
     libbrlapi-dev \
     libbz2-dev \
     libcacard-dev \
tests/docker/dockerfiles/ubuntu1804.docker +0 -1
index 3cc4f492c4..21a5f2c45c 100644
@@ -8,7 +8,6 @@ ENV PACKAGES flex bison \
     glusterfs-common \
     libaio-dev \
     libattr1-dev \
-    libbluetooth-dev \
     libbrlapi-dev \
     libbz2-dev \
     libcacard-dev \
Signature · 1 line
2.23.0
Reviewed-by on v1 2/4
#4 lore raw

Re: [libvirt] [PATCH for-5.0 2/4] hw/usb: Remove the USB bluetooth dongle device

In reply to Thomas Huth
Quoted 4 lines · Thomas Huth

We are going to remove the bluetooth backend, so the USB bluetooth dongle can not work anymore. It's a completely optional device, no board depends on it, so let's simply remove it now.

Quoted 9 lines

---

hw/usb/Kconfig         |   5 -
hw/usb/Makefile.objs   |   1 -
hw/usb/dev-bluetooth.c | 581 -----------------------------------------
qemu-doc.texi          |  15 --
4 files changed, 602 deletions(-)

delete mode 100644 hw/usb/dev-bluetooth.c

Jano

Reviewed-by on v1 3/4
#6 lore raw

Re: [libvirt] [PATCH for-5.0 3/4] Remove the core bluetooth code

In reply to Thomas Huth
Quoted 36 lines · Thomas Huth

It's been deprecated since QEMU v3.1. We've explicitly asked in the deprecation message that people should speak up on qemu-devel in case they are still actively using the bluetooth part of QEMU, but nobody ever replied that they are really still using it.

I've tried it on my own to use this bluetooth subsystem for one of my guests, but I was also not able to get it running anymore: When I was trying to pass-through a real bluetooth device, either the guest did not see the device at all, or the guest crashed.

Even worse for the emulated device: When running

qemu-system-x86_64 -bt device:keyboard

QEMU crashes once you hit a key.

So it seems like the bluetooth stack is not only neglected, it is completely bitrotten, as far as I can tell. The only attention that this code got during the past years were some CVEs that have been spotted there. So this code is a burden for the developers, without any real benefit anymore. Time to remove it.

---

Makefile.objs        |    2 -
bt-host.c            |  198 ----
bt-vhci.c            |  167 ----
configure            |   31 -
hw/Kconfig           |    1 -
hw/Makefile.objs     |    1 -
hw/bt/Kconfig        |    2 -
hw/bt/Makefile.objs  |    3 -
hw/bt/core.c         |  143 ---
hw/bt/hci-csr.c      |  512 ----------
hw/bt/hci.c          | 2263 ------------------------------------------
hw/bt/hid.c          |  553 -----------
Quoted 8 lines
hw/bt/l2cap.c        | 1367 -------------------------
hw/bt/sdp.c          |  989 ------------------
include/hw/bt.h      | 2177 ----------------------------------------
include/sysemu/bt.h  |   20 -
qemu-deprecated.texi |    7 -
qemu-options.hx      |   79 --
vl.c                 |  136 ---
19 files changed, 8651 deletions(-)

Jano

Reviewed-by on v1 1/4
#2 lore raw

Re: [PATCH for-5.0 1/4] hw/arm/nseries: Replace the bluetooth chardev with a "null" chardev

In reply to Thomas Huth
Quoted 5 lines · Thomas Huth

We are going to remove the bitrotten bluetooth backend, so we can not use it in the n800 and n810 machines anymore. Replace the chardev here with a "null" chardev and stop including the "hw/bt.h" header.

Quoted 7 lines

I am not aware of a full working N8x0 environment that can be used to test these machines, but the Linux kernel from the Meego project that can be found on http://stskeeps.subnetmask.net/meego-n8x0/ still seems to boot as far as it did before.

---

thanks
-- PMM

#7 lore raw

Re: [PATCH for-5.0 3/4] Remove the core bluetooth code

In reply to Thomas Huth
Quoted 48 lines · Thomas Huth

It's been deprecated since QEMU v3.1. We've explicitly asked in the deprecation message that people should speak up on qemu-devel in case they are still actively using the bluetooth part of QEMU, but nobody ever replied that they are really still using it.

I've tried it on my own to use this bluetooth subsystem for one of my guests, but I was also not able to get it running anymore: When I was trying to pass-through a real bluetooth device, either the guest did not see the device at all, or the guest crashed.

Even worse for the emulated device: When running

qemu-system-x86_64 -bt device:keyboard

QEMU crashes once you hit a key.

So it seems like the bluetooth stack is not only neglected, it is completely bitrotten, as far as I can tell. The only attention that this code got during the past years were some CVEs that have been spotted there. So this code is a burden for the developers, without any real benefit anymore. Time to remove it.

---
 Makefile.objs        |    2 -
 bt-host.c            |  198 ----
 bt-vhci.c            |  167 ----
 configure            |   31 -
 hw/Kconfig           |    1 -
 hw/Makefile.objs     |    1 -
 hw/bt/Kconfig        |    2 -
 hw/bt/Makefile.objs  |    3 -
 hw/bt/core.c         |  143 ---
 hw/bt/hci-csr.c      |  512 ----------
 hw/bt/hci.c          | 2263 ------------------------------------------
 hw/bt/hid.c          |  553 -----------
 hw/bt/l2cap.c        | 1367 -------------------------
 hw/bt/sdp.c          |  989 ------------------
 include/hw/bt.h      | 2177 ----------------------------------------
 include/sysemu/bt.h  |   20 -
 qemu-deprecated.texi |    7 -
 qemu-options.hx      |   79 --
 vl.c                 |  136 ---
 19 files changed, 8651 deletions(-)
 delete mode 100644 bt-host.c
 delete mode 100644 bt-vhci.c
 delete mode 100644 hw/bt/Kconfig
 delete mode 100644 hw/bt/Makefile.objs
Quoted 9 lines
 delete mode 100644 hw/bt/core.c
 delete mode 100644 hw/bt/hci-csr.c
 delete mode 100644 hw/bt/hci.c
 delete mode 100644 hw/bt/hid.c
 delete mode 100644 hw/bt/l2cap.c
 delete mode 100644 hw/bt/sdp.c
 delete mode 100644 include/hw/bt.h
 delete mode 100644 include/sysemu/bt.h

[...]> diff --git a/configure b/configure

Quoted 12 lines

index 6099be1d84..ecce4ada2d 100755

configure +0 -1
@@ -349,7 +349,6 @@ unset target_list_exclude
 # Distributions want to ensure that several features are compiled in, and it
 # is impossible without a --enable-foo that exits if a feature is not found.
 
-bluez=""
 brlapi=""
 curl=""
 curses=""

configure for opt do · line 1157

Quoted 7 lines
configure +0 -4
@@ -1151,10 +1150,6 @@ for opt do
   ;;
   --enable-brlapi) brlapi="yes"
   ;;
-  --disable-bluez) bluez="no"
-  ;;
-  --enable-bluez) bluez="yes"
-  ;;

Now than I'm bisecting over this commit, I realize removing this option was not a good idea, we should have done like commit cb6414dfec8 or 315d3184525:

@@ -886,10 +885,6 @@ for opt do

  -  --disable-uuid) uuid="no"
  -  ;;
  -  --enable-uuid) uuid="yes"
  -  ;;

...
+ --enable-uuid|--disable-uuid)
+ echo "$0: $opt is obsolete, UUID support is always built" >&2
+ ;;

Quoted 57 lines
configure +0 -26
   --disable-kvm) kvm="no"
   ;;
   --enable-kvm) kvm="yes"
@@ -1762,7 +1757,6 @@ disabled with --disable-FEATURE, default is enabled if available:
   curl            curl connectivity
   membarrier      membarrier system call (for Linux 4.14+ or Windows)
   fdt             fdt device tree
-  bluez           bluez stack connectivity
   kvm             KVM acceleration support
   hax             HAX acceleration support
   hvf             Hypervisor.framework acceleration support
@@ -3665,26 +3659,6 @@ EOF
   fi
 fi # test "$curl"
 
-##########################################
-# bluez support probe
-if test "$bluez" != "no" ; then
-  cat > $TMPC << EOF
-#include <bluetooth/bluetooth.h>
-int main(void) { return bt_error(0); }
-EOF
-  bluez_cflags=$($pkg_config --cflags bluez 2>/dev/null)
-  bluez_libs=$($pkg_config --libs bluez 2>/dev/null)
-  if compile_prog "$bluez_cflags" "$bluez_libs" ; then
-    bluez=yes
-    libs_softmmu="$bluez_libs $libs_softmmu"
-  else
-    if test "$bluez" = "yes" ; then
-      feature_not_found "bluez" "Install bluez-libs/libbluetooth devel"
-    fi
-    bluez="no"
-  fi
-fi
-
 ##########################################
 # glib support probe
 
@@ -6509,7 +6483,6 @@ if test "$xen" = "yes" ; then
   echo "xen ctrl version  $xen_ctrl_version"
 fi
 echo "brlapi support    $brlapi"
-echo "bluez  support    $bluez"
 echo "Documentation     $docs"
 echo "PIE               $pie"
 echo "vde support       $vde"
@@ -6933,10 +6906,6 @@ if test "$brlapi" = "yes" ; then
   echo "CONFIG_BRLAPI=y" >> $config_host_mak
   echo "BRLAPI_LIBS=$brlapi_libs" >> $config_host_mak
 fi
-if test "$bluez" = "yes" ; then
-  echo "CONFIG_BLUEZ=y" >> $config_host_mak
-  echo "BLUEZ_CFLAGS=$bluez_cflags" >> $config_host_mak
-fi
 if test "$gtk" = "yes" ; then
   echo "CONFIG_GTK=m" >> $config_host_mak
   echo "GTK_CFLAGS=$gtk_cflags" >> $config_host_mak

[...]

#8 lore raw

Re: [PATCH for-5.0 3/4] Remove the core bluetooth code

Quoted 84 lines · Philippe Mathieu-Daudé
Quoted 48 lines · Thomas Huth

It's been deprecated since QEMU v3.1. We've explicitly asked in the deprecation message that people should speak up on qemu-devel in case they are still actively using the bluetooth part of QEMU, but nobody ever replied that they are really still using it.

I've tried it on my own to use this bluetooth subsystem for one of my guests, but I was also not able to get it running anymore: When I was trying to pass-through a real bluetooth device, either the guest did not see the device at all, or the guest crashed.

Even worse for the emulated device: When running

qemu-system-x86_64 -bt device:keyboard

QEMU crashes once you hit a key.

So it seems like the bluetooth stack is not only neglected, it is completely bitrotten, as far as I can tell. The only attention that this code got during the past years were some CVEs that have been spotted there. So this code is a burden for the developers, without any real benefit anymore. Time to remove it.

---
 Makefile.objs        |    2 -
 bt-host.c            |  198 ----
 bt-vhci.c            |  167 ----
 configure            |   31 -
 hw/Kconfig           |    1 -
 hw/Makefile.objs     |    1 -
 hw/bt/Kconfig        |    2 -
 hw/bt/Makefile.objs  |    3 -
 hw/bt/core.c         |  143 ---
 hw/bt/hci-csr.c      |  512 ----------
 hw/bt/hci.c          | 2263 ------------------------------------------
 hw/bt/hid.c          |  553 -----------
 hw/bt/l2cap.c        | 1367 -------------------------
 hw/bt/sdp.c          |  989 ------------------
 include/hw/bt.h      | 2177 ----------------------------------------
 include/sysemu/bt.h  |   20 -
 qemu-deprecated.texi |    7 -
 qemu-options.hx      |   79 --
 vl.c                 |  136 ---
 19 files changed, 8651 deletions(-)
 delete mode 100644 bt-host.c
 delete mode 100644 bt-vhci.c
 delete mode 100644 hw/bt/Kconfig
 delete mode 100644 hw/bt/Makefile.objs
Quoted 9 lines
 delete mode 100644 hw/bt/core.c
 delete mode 100644 hw/bt/hci-csr.c
 delete mode 100644 hw/bt/hci.c
 delete mode 100644 hw/bt/hid.c
 delete mode 100644 hw/bt/l2cap.c
 delete mode 100644 hw/bt/sdp.c
 delete mode 100644 include/hw/bt.h
 delete mode 100644 include/sysemu/bt.h

[...]> diff --git a/configure b/configure

Quoted 12 lines

index 6099be1d84..ecce4ada2d 100755

configure +0 -1
@@ -349,7 +349,6 @@ unset target_list_exclude
 # Distributions want to ensure that several features are compiled in, and it
 # is impossible without a --enable-foo that exits if a feature is not found.
 
-bluez=""
 brlapi=""
 curl=""
 curses=""

configure for opt do · line 1157

Quoted 7 lines
configure +0 -4
@@ -1151,10 +1150,6 @@ for opt do
   ;;
   --enable-brlapi) brlapi="yes"
   ;;
-  --disable-bluez) bluez="no"
-  ;;
-  --enable-bluez) bluez="yes"
-  ;;

Now than I'm bisecting over this commit, I realize removing this option was not a good idea, we should have done like commit cb6414dfec8 or 315d3184525:

@@ -886,10 +885,6 @@ for opt do

Quoted 8 lines
  -  --disable-uuid) uuid="no"
  -  ;;
  -  --enable-uuid) uuid="yes"
  -  ;;

...
+ --enable-uuid|--disable-uuid)
+ echo "$0: $opt is obsolete, UUID support is always built" >&2
+ ;;

Looks trivial ... so if it bugs you, just send a patch?

Thomas

#9 lore raw

Re: [PATCH for-5.0 3/4] Remove the core bluetooth code

Quoted 94 lines · Thomas Huth
Quoted 84 lines · Philippe Mathieu-Daudé
Quoted 48 lines · Thomas Huth

It's been deprecated since QEMU v3.1. We've explicitly asked in the deprecation message that people should speak up on qemu-devel in case they are still actively using the bluetooth part of QEMU, but nobody ever replied that they are really still using it.

I've tried it on my own to use this bluetooth subsystem for one of my guests, but I was also not able to get it running anymore: When I was trying to pass-through a real bluetooth device, either the guest did not see the device at all, or the guest crashed.

Even worse for the emulated device: When running

qemu-system-x86_64 -bt device:keyboard

QEMU crashes once you hit a key.

So it seems like the bluetooth stack is not only neglected, it is completely bitrotten, as far as I can tell. The only attention that this code got during the past years were some CVEs that have been spotted there. So this code is a burden for the developers, without any real benefit anymore. Time to remove it.

---
 Makefile.objs        |    2 -
 bt-host.c            |  198 ----
 bt-vhci.c            |  167 ----
 configure            |   31 -
 hw/Kconfig           |    1 -
 hw/Makefile.objs     |    1 -
 hw/bt/Kconfig        |    2 -
 hw/bt/Makefile.objs  |    3 -
 hw/bt/core.c         |  143 ---
 hw/bt/hci-csr.c      |  512 ----------
 hw/bt/hci.c          | 2263 ------------------------------------------
 hw/bt/hid.c          |  553 -----------
 hw/bt/l2cap.c        | 1367 -------------------------
 hw/bt/sdp.c          |  989 ------------------
 include/hw/bt.h      | 2177 ----------------------------------------
 include/sysemu/bt.h  |   20 -
 qemu-deprecated.texi |    7 -
 qemu-options.hx      |   79 --
 vl.c                 |  136 ---
 19 files changed, 8651 deletions(-)
 delete mode 100644 bt-host.c
 delete mode 100644 bt-vhci.c
 delete mode 100644 hw/bt/Kconfig
 delete mode 100644 hw/bt/Makefile.objs
Quoted 9 lines
 delete mode 100644 hw/bt/core.c
 delete mode 100644 hw/bt/hci-csr.c
 delete mode 100644 hw/bt/hci.c
 delete mode 100644 hw/bt/hid.c
 delete mode 100644 hw/bt/l2cap.c
 delete mode 100644 hw/bt/sdp.c
 delete mode 100644 include/hw/bt.h
 delete mode 100644 include/sysemu/bt.h

[...]> diff --git a/configure b/configure

Quoted 12 lines

index 6099be1d84..ecce4ada2d 100755

configure +0 -1
@@ -349,7 +349,6 @@ unset target_list_exclude
 # Distributions want to ensure that several features are compiled in, and it
 # is impossible without a --enable-foo that exits if a feature is not found.
 
-bluez=""
 brlapi=""
 curl=""
 curses=""

configure for opt do · line 1157

Quoted 7 lines
configure +0 -4
@@ -1151,10 +1150,6 @@ for opt do
   ;;
   --enable-brlapi) brlapi="yes"
   ;;
-  --disable-bluez) bluez="no"
-  ;;
-  --enable-bluez) bluez="yes"
-  ;;

Now than I'm bisecting over this commit, I realize removing this option was not a good idea, we should have done like commit cb6414dfec8 or 315d3184525:

@@ -886,10 +885,6 @@ for opt do

Quoted 8 lines
  -  --disable-uuid) uuid="no"
  -  ;;
  -  --enable-uuid) uuid="yes"
  -  ;;

...
+ --enable-uuid|--disable-uuid)
+ echo "$0: $opt is obsolete, UUID support is always built" >&2
+ ;;

Quoted 1 line

Looks trivial ... so if it bugs you, just send a patch?

I thought about it but this won't fix much, it is too late now.

I simply wanted to share this bugged me so we try to avoid doing the same mistake again.

#10 lore raw

Re: [PATCH for-5.0 3/4] Remove the core bluetooth code

Quoted 95 lines · Philippe Mathieu-Daudé
Quoted 94 lines · Thomas Huth
Quoted 92 lines · Philippe Mathieu-Daudé
Quoted 48 lines · Thomas Huth

It's been deprecated since QEMU v3.1. We've explicitly asked in the deprecation message that people should speak up on qemu-devel in case they are still actively using the bluetooth part of QEMU, but nobody ever replied that they are really still using it.

I've tried it on my own to use this bluetooth subsystem for one of my guests, but I was also not able to get it running anymore: When I was trying to pass-through a real bluetooth device, either the guest did not see the device at all, or the guest crashed.

Even worse for the emulated device: When running

qemu-system-x86_64 -bt device:keyboard

QEMU crashes once you hit a key.

So it seems like the bluetooth stack is not only neglected, it is completely bitrotten, as far as I can tell. The only attention that this code got during the past years were some CVEs that have been spotted there. So this code is a burden for the developers, without any real benefit anymore. Time to remove it.

---
 Makefile.objs        |    2 -
 bt-host.c            |  198 ----
 bt-vhci.c            |  167 ----
 configure            |   31 -
 hw/Kconfig           |    1 -
 hw/Makefile.objs     |    1 -
 hw/bt/Kconfig        |    2 -
 hw/bt/Makefile.objs  |    3 -
 hw/bt/core.c         |  143 ---
 hw/bt/hci-csr.c      |  512 ----------
 hw/bt/hci.c          | 2263 ------------------------------------------
 hw/bt/hid.c          |  553 -----------
 hw/bt/l2cap.c        | 1367 -------------------------
 hw/bt/sdp.c          |  989 ------------------
 include/hw/bt.h      | 2177 ----------------------------------------
 include/sysemu/bt.h  |   20 -
 qemu-deprecated.texi |    7 -
 qemu-options.hx      |   79 --
 vl.c                 |  136 ---
 19 files changed, 8651 deletions(-)
 delete mode 100644 bt-host.c
 delete mode 100644 bt-vhci.c
 delete mode 100644 hw/bt/Kconfig
 delete mode 100644 hw/bt/Makefile.objs
Quoted 9 lines
 delete mode 100644 hw/bt/core.c
 delete mode 100644 hw/bt/hci-csr.c
 delete mode 100644 hw/bt/hci.c
 delete mode 100644 hw/bt/hid.c
 delete mode 100644 hw/bt/l2cap.c
 delete mode 100644 hw/bt/sdp.c
 delete mode 100644 include/hw/bt.h
 delete mode 100644 include/sysemu/bt.h

[...]> diff --git a/configure b/configure

Quoted 12 lines

index 6099be1d84..ecce4ada2d 100755

configure +0 -1
@@ -349,7 +349,6 @@ unset target_list_exclude
 # Distributions want to ensure that several features are compiled in, and it
 # is impossible without a --enable-foo that exits if a feature is not found.
 
-bluez=""
 brlapi=""
 curl=""
 curses=""

configure for opt do · line 1157

Quoted 7 lines
configure +0 -4
@@ -1151,10 +1150,6 @@ for opt do
   ;;
   --enable-brlapi) brlapi="yes"
   ;;
-  --disable-bluez) bluez="no"
-  ;;
-  --enable-bluez) bluez="yes"
-  ;;

Now than I'm bisecting over this commit, I realize removing this option was not a good idea, we should have done like commit cb6414dfec8 or 315d3184525:

@@ -886,10 +885,6 @@ for opt do

  -  --disable-uuid) uuid="no"
  -  ;;
  -  --enable-uuid) uuid="yes"
  -  ;;

...
+ --enable-uuid|--disable-uuid)
+ echo "$0: $opt is obsolete, UUID support is always built" >&2
+ ;;

Quoted 6 lines
Quoted 1 line

Looks trivial ... so if it bugs you, just send a patch?

I thought about it but this won't fix much, it is too late now.

I simply wanted to share this bugged me so we try to avoid doing the same mistake again.

Makes sense. I'll try to remember it for the future.

Thomas

#11 lore raw

Re: [PATCH for-5.0 3/4] Remove the core bluetooth code

In reply to Philippe Mathieu-Daudé
Quoted 1 line · Philippe Mathieu-Daudé

I thought about it but this won't fix much, it is too late now.

It would help distros who are already disabling bluez, since they wouldn't incur a compile-time error. Not a big deal though.

Paolo

#12 lore raw

Re: [PATCH for-5.0 3/4] Remove the core bluetooth code

In reply to Philippe Mathieu-Daudé
Quoted 96 lines · Philippe Mathieu-Daudé
Quoted 94 lines · Thomas Huth
Quoted 92 lines · Philippe Mathieu-Daudé
Quoted 48 lines · Thomas Huth

It's been deprecated since QEMU v3.1. We've explicitly asked in the deprecation message that people should speak up on qemu-devel in case they are still actively using the bluetooth part of QEMU, but nobody ever replied that they are really still using it.

I've tried it on my own to use this bluetooth subsystem for one of my guests, but I was also not able to get it running anymore: When I was trying to pass-through a real bluetooth device, either the guest did not see the device at all, or the guest crashed.

Even worse for the emulated device: When running

qemu-system-x86_64 -bt device:keyboard

QEMU crashes once you hit a key.

So it seems like the bluetooth stack is not only neglected, it is completely bitrotten, as far as I can tell. The only attention that this code got during the past years were some CVEs that have been spotted there. So this code is a burden for the developers, without any real benefit anymore. Time to remove it.

---
 Makefile.objs        |    2 -
 bt-host.c            |  198 ----
 bt-vhci.c            |  167 ----
 configure            |   31 -
 hw/Kconfig           |    1 -
 hw/Makefile.objs     |    1 -
 hw/bt/Kconfig        |    2 -
 hw/bt/Makefile.objs  |    3 -
 hw/bt/core.c         |  143 ---
 hw/bt/hci-csr.c      |  512 ----------
 hw/bt/hci.c          | 2263 ------------------------------------------
 hw/bt/hid.c          |  553 -----------
 hw/bt/l2cap.c        | 1367 -------------------------
 hw/bt/sdp.c          |  989 ------------------
 include/hw/bt.h      | 2177 ----------------------------------------
 include/sysemu/bt.h  |   20 -
 qemu-deprecated.texi |    7 -
 qemu-options.hx      |   79 --
 vl.c                 |  136 ---
 19 files changed, 8651 deletions(-)
 delete mode 100644 bt-host.c
 delete mode 100644 bt-vhci.c
 delete mode 100644 hw/bt/Kconfig
 delete mode 100644 hw/bt/Makefile.objs
Quoted 9 lines
 delete mode 100644 hw/bt/core.c
 delete mode 100644 hw/bt/hci-csr.c
 delete mode 100644 hw/bt/hci.c
 delete mode 100644 hw/bt/hid.c
 delete mode 100644 hw/bt/l2cap.c
 delete mode 100644 hw/bt/sdp.c
 delete mode 100644 include/hw/bt.h
 delete mode 100644 include/sysemu/bt.h

[...]> diff --git a/configure b/configure

Quoted 12 lines

index 6099be1d84..ecce4ada2d 100755

configure +0 -1
@@ -349,7 +349,6 @@ unset target_list_exclude
 # Distributions want to ensure that several features are compiled in, and it
 # is impossible without a --enable-foo that exits if a feature is not found.
 
-bluez=""
 brlapi=""
 curl=""
 curses=""

configure for opt do · line 1157

Quoted 7 lines
configure +0 -4
@@ -1151,10 +1150,6 @@ for opt do
   ;;
   --enable-brlapi) brlapi="yes"
   ;;
-  --disable-bluez) bluez="no"
-  ;;
-  --enable-bluez) bluez="yes"
-  ;;

Now than I'm bisecting over this commit, I realize removing this option was not a good idea, we should have done like commit cb6414dfec8 or 315d3184525:

@@ -886,10 +885,6 @@ for opt do

  -  --disable-uuid) uuid="no"
  -  ;;
  -  --enable-uuid) uuid="yes"
  -  ;;

...
+ --enable-uuid|--disable-uuid)
+ echo "$0: $opt is obsolete, UUID support is always built" >&2
+ ;;

Quoted 7 lines
Quoted 1 line

Looks trivial ... so if it bugs you, just send a patch?

I thought about it but this won't fix much, it is too late now.

I simply wanted to share this bugged me so we try to avoid doing the same mistake again.

I vote for addition of a change similar to what Philippe described.

Furthermore, it looks to me the correct way would be to now do full deprecation of --enable-bluez and --disable-bluez. This means adding this to documentation (not related to bluetooth devices support), not only a change in "configure". This also means that only after two next full cycles these options could be removed.

True, this could have been done together with bluetooth devices support deprecation (and in that case we could have deleted these options right away), but it wasn't. Users don't have a crystal ball to know that we assumed that --enable-bluez and --disable-bluez were part of bluetooth devices support, and could rightfully complain about an abrupt elimination of a compile time option.

This is my understanding, at least.

What do you think?

Aleksandar

#13 lore raw

Re: [PATCH for-5.0 3/4] Remove the core bluetooth code

Quoted 4 lines · Aleksandar Markovic
Quoted 3 lines · Philippe Mathieu-Daudé
Quoted 1 line · Thomas Huth

On 01/02/2020 17.09, Philippe Mathieu-Daudé wrote:

[...]

Quoted 36 lines
Quoted 36 lines
Quoted 35 lines
Quoted 25 lines
Quoted 12 lines

index 6099be1d84..ecce4ada2d 100755

configure +0 -1
@@ -349,7 +349,6 @@ unset target_list_exclude
 # Distributions want to ensure that several features are compiled in, and it
 # is impossible without a --enable-foo that exits if a feature is not found.
 
-bluez=""
 brlapi=""
 curl=""
 curses=""

configure for opt do · line 1157

Quoted 7 lines
configure +0 -4
@@ -1151,10 +1150,6 @@ for opt do
   ;;
   --enable-brlapi) brlapi="yes"
   ;;
-  --disable-bluez) bluez="no"
-  ;;
-  --enable-bluez) bluez="yes"
-  ;;

Now than I'm bisecting over this commit, I realize removing this option was not a good idea, we should have done like commit cb6414dfec8 or 315d3184525:

@@ -886,10 +885,6 @@ for opt do

Quoted 8 lines
  -  --disable-uuid) uuid="no"
  -  ;;
  -  --enable-uuid) uuid="yes"
  -  ;;

...
+ --enable-uuid|--disable-uuid)
+ echo "$0: $opt is obsolete, UUID support is always built" >&2
+ ;;

Looks trivial ... so if it bugs you, just send a patch?

Quoted 7 lines
Quoted 5 lines

I thought about it but this won't fix much, it is too late now.

I simply wanted to share this bugged me so we try to avoid doing the same mistake again.

I vote for addition of a change similar to what Philippe described.

Feel free to send a patch if it bugs you.

Quoted 6 lines

Furthermore, it looks to me the correct way would be to now do full deprecation of --enable-bluez and --disable-bluez. This means adding this to documentation (not related to bluetooth devices support), not only a change in "configure". This also means that only after two next full cycles these options could be removed.

Quoted 6 lines

True, this could have been done together with bluetooth devices support deprecation (and in that case we could have deleted these options right away), but it wasn't. Users don't have a crystal ball to know that we assumed that --enable-bluez and --disable-bluez were part of bluetooth devices support, and could rightfully complain about an abrupt elimination of a compile time option.

I disagree. If you don't know that "bluez" is about bluetooth, then you should likely not use these options anyway.

Thomas

#15 lore raw

Re: [PATCH for-5.0 3/4] Remove the core bluetooth code

In reply to Aleksandar Markovic
Quoted 41 lines · Aleksandar Markovic
Quoted 24 lines · Philippe Mathieu-Daudé
Quoted 24 lines
Quoted 23 lines

configure for opt do · line 1157

Quoted 9 lines
configure +0 -4
@@ -1151,10 +1150,6 @@ for opt do
   ;;
   --enable-brlapi) brlapi="yes"
   ;;
-  --disable-bluez) bluez="no"
-  ;;
-  --enable-bluez) bluez="yes"
-  ;;

Now than I'm bisecting over this commit, I realize removing this option was not a good idea, we should have done like commit cb6414dfec8 or 315d3184525:

@@ -886,10 +885,6 @@ for opt do

  -  --disable-uuid) uuid="no"
  -  ;;
  -  --enable-uuid) uuid="yes"
  -  ;;

...
+ --enable-uuid|--disable-uuid)
+ echo "$0: $opt is obsolete, UUID support is always built" >&2
+ ;;

Quoted 7 lines
Quoted 1 line

Looks trivial ... so if it bugs you, just send a patch?

I thought about it but this won't fix much, it is too late now.

I simply wanted to share this bugged me so we try to avoid doing the same mistake again.

I vote for addition of a change similar to what Philippe described.

Furthermore, it looks to me the correct way would be to now do full deprecation of --enable-bluez and --disable-bluez. This means adding this to documentation (not related to bluetooth devices support), not only a change in "configure". This also means that only after two next full cycles these options could be removed.

Quoted 6 lines

True, this could have been done together with bluetooth devices support deprecation (and in that case we could have deleted these options right away), but it wasn't. Users don't have a crystal ball to know that we assumed that --enable-bluez and --disable-bluez were part of bluetooth devices support, and could rightfully complain about an abrupt elimination of a compile time option.

The deprecation policy is primarily intended for notifying of changes to QEMU's stable interfaces ( CLI, HMP, QMP) which affect behaviour and usage of QEMU at runtime & are liable to break apps managing QEMU.

Changes to build time options have no strong reason to be subjected to the deprecation process. If we remove an option at build time the effect is noticed immediately and the solution is straightforward (stop passing the option). We have added / removed configure options at will with little negative feedback historically. We'll have far biggest changes coming to the build system in future too, with the introduction of meson.

I don't think we want to constrain & complicate our work in modernizing the build system by declaring that any changes to it must go through deprecation.

Regards,
Daniel

Signature · 3 lines
|: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org         -o-            https://fstop138.berrange.com :|
|: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|
#14 lore raw

Re: [PATCH for-5.0 3/4] Remove the core bluetooth code

In reply to Thomas Huth
Quoted 44 lines · Thomas Huth
Quoted 4 lines · Aleksandar Markovic
Quoted 3 lines · Philippe Mathieu-Daudé
Quoted 1 line · Thomas Huth

On 01/02/2020 17.09, Philippe Mathieu-Daudé wrote:

[...]

Quoted 38 lines
Quoted 38 lines
Quoted 34 lines
Quoted 25 lines
Quoted 12 lines

index 6099be1d84..ecce4ada2d 100755

configure +0 -1
@@ -349,7 +349,6 @@ unset target_list_exclude
 # Distributions want to ensure that several features are compiled in, and it
 # is impossible without a --enable-foo that exits if a feature is not found.
 
-bluez=""
 brlapi=""
 curl=""
 curses=""

configure for opt do · line 1157

Quoted 7 lines
configure +0 -4
@@ -1151,10 +1150,6 @@ for opt do
   ;;
   --enable-brlapi) brlapi="yes"
   ;;
-  --disable-bluez) bluez="no"
-  ;;
-  --enable-bluez) bluez="yes"
-  ;;

Now than I'm bisecting over this commit, I realize removing this option was not a good idea, we should have done like commit cb6414dfec8 or 315d3184525:

@@ -886,10 +885,6 @@ for opt do

Quoted 8 lines
  -  --disable-uuid) uuid="no"
  -  ;;
  -  --enable-uuid) uuid="yes"
  -  ;;

...
+ --enable-uuid|--disable-uuid)
+ echo "$0: $opt is obsolete, UUID support is always built" >&2
+ ;;

Quoted 1 line

Looks trivial ... so if it bugs you, just send a patch?

I thought about it but this won't fix much, it is too late now.

Quoted 7 lines
Quoted 5 lines
Quoted 3 lines

I simply wanted to share this bugged me so we try to avoid doing the same mistake again.

I vote for addition of a change similar to what Philippe described.

Feel free to send a patch if it bugs you.

I don't think we should do this (nor the original uuid compat hack above). We don't want to set an expectation that configure arguments are considered "stable" interface to QEMU requiring back compat. This will constrain our future build system work as we move to introduce meson initially and then gradually elminate more & more of configure until it can be deleted entirely.

Certainly we should document configure arg removals in the release notes, but we don't need to go beyond that IMHO.

Regards,
Daniel

Signature · 3 lines
|: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org         -o-            https://fstop138.berrange.com :|
|: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|
#16 lore raw

Re: [PATCH for-5.0 3/4] Remove the core bluetooth code

In reply to Daniel P. Berrangé
Quoted 7 lines

The deprecation policy is primarily intended for notifying of changes to QEMU's stable interfaces ( CLI, HMP, QMP) which affect behaviour and usage of QEMU at runtime & are liable to break apps managing QEMU.

Changes to build time options have no strong reason to be subjected to the deprecation process.

This sounds reasonable to me.

But: Should our deprecation policy be clearer on what is subject to our deprecation procedure, and what is not?

Regards,
Aleksandar

Quoted 17 lines

If we remove an option at build time the effect
is noticed immediately and the solution is straightforward (stop passing the option). We have added / removed configure options at will with little negative feedback historically. We'll have far biggest changes coming to the build system in future too, with the introduction of meson.

I don't think we want to constrain & complicate our work in modernizing the build system by declaring that any changes to it must go through deprecation.

Regards,
Daniel
--

|: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org         -o-            https://fstop138.berrange.com :|
|: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|
#17 lore raw

Re: [PATCH for-5.0 3/4] Remove the core bluetooth code

Quoted 9 lines · Aleksandar Markovic
Quoted 8 lines

The deprecation policy is primarily intended for notifying of changes to QEMU's stable interfaces ( CLI, HMP, QMP) which affect behaviour and usage of QEMU at runtime & are liable to break apps managing QEMU.

Changes to build time options have no strong reason to be subjected to the deprecation process.

Quoted 8 lines

This sounds reasonable to me.

But: Should our deprecation policy be clearer on what is subject to our deprecation procedure, and what is not?

Regards,
Aleksandar

And, there is even a section devoted to the build system in
https://qemu.weilnetz.de/doc/qemu-doc.html#Deprecated-features :

B.10 Build system

B.10.1 Python 2 support (since 4.1.0)

In the future, QEMU will require Python 3 to be available at build
time. Support for Python 2 in scripts shipped with QEMU is deprecated.

The wording above comes from the file in our source tree: qemu-deprecated.texi .

Shouldn't the cases like "--enable-blues" be documented here? I think the deprecation of bluetooth devices support should have included a short section on involved configure options here.

Why is Python 2 support different to "--enable-blues" support? There should not be some unknown and arbitrary criterium what do and what doesn't come in this subsection.

Or, if our build system is not subject to our deprecation policy, than what is the entire subsection "B.10 Build system" doing here, in section "Deprecated features"?

Quoted 17 lines
Quoted 17 lines

If we remove an option at build time the effect
is noticed immediately and the solution is straightforward (stop passing the option). We have added / removed configure options at will with little negative feedback historically. We'll have far biggest changes coming to the build system in future too, with the introduction of meson.

I don't think we want to constrain & complicate our work in modernizing the build system by declaring that any changes to it must go through deprecation.

Regards,
Daniel
--

|: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org         -o-            https://fstop138.berrange.com :|
|: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|
#19 lore raw

Re: [PATCH for-5.0 3/4] Remove the core bluetooth code

In reply to Daniel P. Berrangé
Quoted 54 lines · Daniel P. Berrangé
Quoted 41 lines · Aleksandar Markovic
Quoted 24 lines · Philippe Mathieu-Daudé
Quoted 24 lines
Quoted 23 lines

configure for opt do · line 1157

Quoted 9 lines
configure +0 -4
@@ -1151,10 +1150,6 @@ for opt do
   ;;
   --enable-brlapi) brlapi="yes"
   ;;
-  --disable-bluez) bluez="no"
-  ;;
-  --enable-bluez) bluez="yes"
-  ;;

Now than I'm bisecting over this commit, I realize removing this option was not a good idea, we should have done like commit cb6414dfec8 or 315d3184525:

@@ -886,10 +885,6 @@ for opt do

  -  --disable-uuid) uuid="no"
  -  ;;
  -  --enable-uuid) uuid="yes"
  -  ;;

...
+ --enable-uuid|--disable-uuid)
+ echo "$0: $opt is obsolete, UUID support is always built" >&2
+ ;;

Quoted 7 lines
Quoted 1 line

Looks trivial ... so if it bugs you, just send a patch?

I thought about it but this won't fix much, it is too late now.

I simply wanted to share this bugged me so we try to avoid doing the same mistake again.

I vote for addition of a change similar to what Philippe described.

Furthermore, it looks to me the correct way would be to now do full deprecation of --enable-bluez and --disable-bluez. This means adding this to documentation (not related to bluetooth devices support), not only a change in "configure". This also means that only after two next full cycles these options could be removed.

Quoted 6 lines

True, this could have been done together with bluetooth devices support deprecation (and in that case we could have deleted these options right away), but it wasn't. Users don't have a crystal ball to know that we assumed that --enable-bluez and --disable-bluez were part of bluetooth devices support, and could rightfully complain about an abrupt elimination of a compile time option.

The deprecation policy is primarily intended for notifying of changes to QEMU's stable interfaces ( CLI, HMP, QMP) which affect behaviour and usage of QEMU at runtime & are liable to break apps managing QEMU.

Quoted 6 lines

Changes to build time options have no strong reason to be subjected to the deprecation process. If we remove an option at build time the effect is noticed immediately and the solution is straightforward (stop passing the option). We have added / removed configure options at will with little negative feedback historically. We'll have far biggest changes coming to the build system in future too, with the introduction of meson.

I understand your point when looking forward (when distribution upgrade, it is easy to adapt).
However this is still an issue when looking backward when running bisection to find regressions.

Quoted 7 lines

I don't think we want to constrain & complicate our work in modernizing the build system by declaring that any changes to it must go through deprecation.

Regards,
Daniel

#18 lore raw

Re: [PATCH for-5.0 3/4] Remove the core bluetooth code

In reply to Aleksandar Markovic
Quoted 36 lines · Aleksandar Markovic
Quoted 9 lines · Aleksandar Markovic
Quoted 8 lines

The deprecation policy is primarily intended for notifying of changes to QEMU's stable interfaces ( CLI, HMP, QMP) which affect behaviour and usage of QEMU at runtime & are liable to break apps managing QEMU.

Changes to build time options have no strong reason to be subjected to the deprecation process.

Quoted 8 lines

This sounds reasonable to me.

But: Should our deprecation policy be clearer on what is subject to our deprecation procedure, and what is not?

Regards,
Aleksandar

And, there is even a section devoted to the build system in
https://qemu.weilnetz.de/doc/qemu-doc.html#Deprecated-features :

B.10 Build system

B.10.1 Python 2 support (since 4.1.0)

In the future, QEMU will require Python 3 to be available at build
time. Support for Python 2 in scripts shipped with QEMU is deprecated.

The wording above comes from the file in our source tree: qemu-deprecated.texi .

Shouldn't the cases like "--enable-blues" be documented here? I think the deprecation of bluetooth devices support should have included a short section on involved configure options here.

Quoted 7 lines

Why is Python 2 support different to "--enable-blues" support? There should not be some unknown and arbitrary criterium what do and what doesn't come in this subsection.

Or, if our build system is not subject to our deprecation policy, than what is the entire subsection "B.10 Build system" doing here, in section "Deprecated features"?

See here:

https://lists.gnu.org/archive/html/qemu-devel/2020-01/msg02552.html

i.e. the section was here rather by "accident". It will be removed once my "qemu-deprecated: Remove text about Python 2" patch hits the master branch.

Thomas

#20 lore raw

Re: [PATCH for-5.0 3/4] Remove the core bluetooth code

In reply to Philippe Mathieu-Daudé
Quoted 62 lines · Philippe Mathieu-Daudé
Quoted 54 lines · Daniel P. Berrangé
Quoted 41 lines · Aleksandar Markovic
Quoted 24 lines · Philippe Mathieu-Daudé
Quoted 24 lines
Quoted 23 lines

Review for opt do · line 1157

Quoted 9 lines
(quoted diff) +0 -4
@@ -1151,10 +1150,6 @@ for opt do
   ;;
   --enable-brlapi) brlapi="yes"
   ;;
-  --disable-bluez) bluez="no"
-  ;;
-  --enable-bluez) bluez="yes"
-  ;;

Now than I'm bisecting over this commit, I realize removing this option was not a good idea, we should have done like commit cb6414dfec8 or 315d3184525:

@@ -886,10 +885,6 @@ for opt do

  -  --disable-uuid) uuid="no"
  -  ;;
  -  --enable-uuid) uuid="yes"
  -  ;;

...
+ --enable-uuid|--disable-uuid)
+ echo "$0: $opt is obsolete, UUID support is always built" >&2
+ ;;

Quoted 7 lines
Quoted 1 line

Looks trivial ... so if it bugs you, just send a patch?

I thought about it but this won't fix much, it is too late now.

I simply wanted to share this bugged me so we try to avoid doing the same mistake again.

I vote for addition of a change similar to what Philippe described.

Furthermore, it looks to me the correct way would be to now do full deprecation of --enable-bluez and --disable-bluez. This means adding this to documentation (not related to bluetooth devices support), not only a change in "configure". This also means that only after two next full cycles these options could be removed.

Quoted 6 lines

True, this could have been done together with bluetooth devices support deprecation (and in that case we could have deleted these options right away), but it wasn't. Users don't have a crystal ball to know that we assumed that --enable-bluez and --disable-bluez were part of bluetooth devices support, and could rightfully complain about an abrupt elimination of a compile time option.

The deprecation policy is primarily intended for notifying of changes to QEMU's stable interfaces ( CLI, HMP, QMP) which affect behaviour and usage of QEMU at runtime & are liable to break apps managing QEMU.

Quoted 6 lines

Changes to build time options have no strong reason to be subjected to the deprecation process. If we remove an option at build time the effect is noticed immediately and the solution is straightforward (stop passing the option). We have added / removed configure options at will with little negative feedback historically. We'll have far biggest changes coming to the build system in future too, with the introduction of meson.

Quoted 4 lines

I understand your point when looking forward (when distribution upgrade, it is easy to adapt).
However this is still an issue when looking backward when running bisection to find regressions.

That's only a problem if you were actively setting this particular command line argument. Most developers are just fine with configure auto-detecting stuff. The only time I've ever needed to try to explicitly disable this is in downstream RHEL branches of QEMU

Regards,
Daniel

Signature · 3 lines
|: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
|: https://libvirt.org         -o-            https://fstop138.berrange.com :|
|: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|