Message ID | 20190621132324.2165-1-mreitz@redhat.com (mailing list archive) |
---|---|
State | New, archived |
Headers | show |
On Fri, 21 Jun 2019 at 14:23, Max Reitz <mreitz@redhat.com> wrote: > > The following changes since commit 33d609990621dea6c7d056c86f707b8811320ac1: > > Merge remote-tracking branch 'remotes/kevin/tags/for-upstream' into staging (2019-06-18 17:00:52 +0100) > > are available in the Git repository at: > > https://github.com/XanClic/qemu.git tags/pull-block-2019-06-21 > > for you to fetch changes up to e2a76186f7948b8b75d1b2b52638de7c2f7f7472: > > iotests: Fix 205 for concurrent runs (2019-06-21 14:40:28 +0200) > > ---------------------------------------------------------------- > Block patches: > - The SSH block driver now uses libssh instead of libssh2 > - The VMDK block driver gets read-only support for the seSparse > subformat > - Various fixes > Hi; this failed to build on my s390 box: /home/linux1/qemu/block/ssh.c: In function ‘check_host_key_knownhosts’: /home/linux1/qemu/block/ssh.c:367:27: error: implicit declaration of function ‘ssh_get_fingerprint_hash’ [-Werror=implicit-function-declaration] fingerprint = ssh_get_fingerprint_hash(SSH_PUBLICKEY_HASH_SHA1, ^ /home/linux1/qemu/block/ssh.c:367:13: error: nested extern declaration of ‘ssh_get_fingerprint_hash’ [-Werror=nested-externs] fingerprint = ssh_get_fingerprint_hash(SSH_PUBLICKEY_HASH_SHA1, ^ /home/linux1/qemu/block/ssh.c:367:25: error: assignment makes pointer from integer without a cast [-Werror=int-conversion] fingerprint = ssh_get_fingerprint_hash(SSH_PUBLICKEY_HASH_SHA1, ^ It looks like that function was introduced in libssh 0.8.3, and this box has 0.6.3. (configure has correctly not defined HAVE_LIBSSH_0_8 but this usage is inside a bit of code that's compiled even when that is not defined.) thanks -- PMM
On 23.06.19 19:18, Peter Maydell wrote: > On Fri, 21 Jun 2019 at 14:23, Max Reitz <mreitz@redhat.com> wrote: >> >> The following changes since commit 33d609990621dea6c7d056c86f707b8811320ac1: >> >> Merge remote-tracking branch 'remotes/kevin/tags/for-upstream' into staging (2019-06-18 17:00:52 +0100) >> >> are available in the Git repository at: >> >> https://github.com/XanClic/qemu.git tags/pull-block-2019-06-21 >> >> for you to fetch changes up to e2a76186f7948b8b75d1b2b52638de7c2f7f7472: >> >> iotests: Fix 205 for concurrent runs (2019-06-21 14:40:28 +0200) >> >> ---------------------------------------------------------------- >> Block patches: >> - The SSH block driver now uses libssh instead of libssh2 >> - The VMDK block driver gets read-only support for the seSparse >> subformat >> - Various fixes >> > > Hi; this failed to build on my s390 box: > > /home/linux1/qemu/block/ssh.c: In function ‘check_host_key_knownhosts’: > /home/linux1/qemu/block/ssh.c:367:27: error: implicit declaration of > function ‘ssh_get_fingerprint_hash’ > [-Werror=implicit-function-declaration] > fingerprint = ssh_get_fingerprint_hash(SSH_PUBLICKEY_HASH_SHA1, > ^ > /home/linux1/qemu/block/ssh.c:367:13: error: nested extern declaration > of ‘ssh_get_fingerprint_hash’ [-Werror=nested-externs] > fingerprint = ssh_get_fingerprint_hash(SSH_PUBLICKEY_HASH_SHA1, > ^ > /home/linux1/qemu/block/ssh.c:367:25: error: assignment makes pointer > from integer without a cast [-Werror=int-conversion] > fingerprint = ssh_get_fingerprint_hash(SSH_PUBLICKEY_HASH_SHA1, > ^ > > It looks like that function was introduced in libssh 0.8.3, and this box > has 0.6.3. (configure has correctly not defined HAVE_LIBSSH_0_8 > but this usage is inside a bit of code that's compiled even when > that is not defined.) Pino, would you be OK with dropping that piece of code for pre-0.8 and just replacing it with the else-error_setg()? Max
On Monday, 24 June 2019 14:20:11 CEST Max Reitz wrote: > On 23.06.19 19:18, Peter Maydell wrote: > > On Fri, 21 Jun 2019 at 14:23, Max Reitz <mreitz@redhat.com> wrote: > >> > >> The following changes since commit 33d609990621dea6c7d056c86f707b8811320ac1: > >> > >> Merge remote-tracking branch 'remotes/kevin/tags/for-upstream' into staging (2019-06-18 17:00:52 +0100) > >> > >> are available in the Git repository at: > >> > >> https://github.com/XanClic/qemu.git tags/pull-block-2019-06-21 > >> > >> for you to fetch changes up to e2a76186f7948b8b75d1b2b52638de7c2f7f7472: > >> > >> iotests: Fix 205 for concurrent runs (2019-06-21 14:40:28 +0200) > >> > >> ---------------------------------------------------------------- > >> Block patches: > >> - The SSH block driver now uses libssh instead of libssh2 > >> - The VMDK block driver gets read-only support for the seSparse > >> subformat > >> - Various fixes > >> > > > > Hi; this failed to build on my s390 box: > > > > /home/linux1/qemu/block/ssh.c: In function ‘check_host_key_knownhosts’: > > /home/linux1/qemu/block/ssh.c:367:27: error: implicit declaration of > > function ‘ssh_get_fingerprint_hash’ > > [-Werror=implicit-function-declaration] > > fingerprint = ssh_get_fingerprint_hash(SSH_PUBLICKEY_HASH_SHA1, > > ^ > > /home/linux1/qemu/block/ssh.c:367:13: error: nested extern declaration > > of ‘ssh_get_fingerprint_hash’ [-Werror=nested-externs] > > fingerprint = ssh_get_fingerprint_hash(SSH_PUBLICKEY_HASH_SHA1, > > ^ > > /home/linux1/qemu/block/ssh.c:367:25: error: assignment makes pointer > > from integer without a cast [-Werror=int-conversion] > > fingerprint = ssh_get_fingerprint_hash(SSH_PUBLICKEY_HASH_SHA1, > > ^ > > > > It looks like that function was introduced in libssh 0.8.3, and this box > > has 0.6.3. (configure has correctly not defined HAVE_LIBSSH_0_8 > > but this usage is inside a bit of code that's compiled even when > > that is not defined.) Oops, sorry, I did not test the latest versions with that old libssh. > Pino, would you be OK with dropping that piece of code for pre-0.8 and > just replacing it with the else-error_setg()? Some the variables in check_host_key_knownhosts must be moved within the HAVE_LIBSSH_0_8 block now; attached fixup patch, please squash with my patch (I can submit a v12, if needed/wanted).
On 24.06.19 15:15, Pino Toscano wrote: > On Monday, 24 June 2019 14:20:11 CEST Max Reitz wrote: >> On 23.06.19 19:18, Peter Maydell wrote: >>> On Fri, 21 Jun 2019 at 14:23, Max Reitz <mreitz@redhat.com> wrote: >>>> >>>> The following changes since commit 33d609990621dea6c7d056c86f707b8811320ac1: >>>> >>>> Merge remote-tracking branch 'remotes/kevin/tags/for-upstream' into staging (2019-06-18 17:00:52 +0100) >>>> >>>> are available in the Git repository at: >>>> >>>> https://github.com/XanClic/qemu.git tags/pull-block-2019-06-21 >>>> >>>> for you to fetch changes up to e2a76186f7948b8b75d1b2b52638de7c2f7f7472: >>>> >>>> iotests: Fix 205 for concurrent runs (2019-06-21 14:40:28 +0200) >>>> >>>> ---------------------------------------------------------------- >>>> Block patches: >>>> - The SSH block driver now uses libssh instead of libssh2 >>>> - The VMDK block driver gets read-only support for the seSparse >>>> subformat >>>> - Various fixes >>>> >>> >>> Hi; this failed to build on my s390 box: >>> >>> /home/linux1/qemu/block/ssh.c: In function ‘check_host_key_knownhosts’: >>> /home/linux1/qemu/block/ssh.c:367:27: error: implicit declaration of >>> function ‘ssh_get_fingerprint_hash’ >>> [-Werror=implicit-function-declaration] >>> fingerprint = ssh_get_fingerprint_hash(SSH_PUBLICKEY_HASH_SHA1, >>> ^ >>> /home/linux1/qemu/block/ssh.c:367:13: error: nested extern declaration >>> of ‘ssh_get_fingerprint_hash’ [-Werror=nested-externs] >>> fingerprint = ssh_get_fingerprint_hash(SSH_PUBLICKEY_HASH_SHA1, >>> ^ >>> /home/linux1/qemu/block/ssh.c:367:25: error: assignment makes pointer >>> from integer without a cast [-Werror=int-conversion] >>> fingerprint = ssh_get_fingerprint_hash(SSH_PUBLICKEY_HASH_SHA1, >>> ^ >>> >>> It looks like that function was introduced in libssh 0.8.3, and this box >>> has 0.6.3. (configure has correctly not defined HAVE_LIBSSH_0_8 >>> but this usage is inside a bit of code that's compiled even when >>> that is not defined.) > > Oops, sorry, I did not test the latest versions with that old libssh. > >> Pino, would you be OK with dropping that piece of code for pre-0.8 and >> just replacing it with the else-error_setg()? > > Some the variables in check_host_key_knownhosts must be moved within > the HAVE_LIBSSH_0_8 block now; attached fixup patch, please squash with > my patch (I can submit a v12, if needed/wanted). Thanks, I’ve squashed it in and sent a v2 for this pull request. Max