Package: dpdk / 16.11.11-1+deb9u2
Metadata
| Package | Version | Patches format |
|---|---|---|
| dpdk | 16.11.11-1+deb9u2 | 3.0 (quilt) |
Patch series
view the series file| Patch | File delta | Description |
|---|---|---|
| fix vhost user socket permission.patch | (download) |
doc/guides/testpmd_app_ug/run_app.rst |
19 19 + 0 - 0 ! |
eal: provide option to set vhost_user socket owner/permissions The API doesn't hold a way to specify a owner/permission set for vhost_user created sockets. Projects consuming DPDK started to do 'their own workarounds' like openvswitch https://patchwork.ozlabs.org/patch/559043/ https://patchwork.ozlabs.org/patch/559045/ But for this specific example they are blocked/stalled behind a bigger rework (https://patchwork.ozlabs.org/patch/604898/). We need something now for existing code linking against DPDK. That implies to avoid changing API/ABI. So I created a DPDK EAL commandline option based ideas in the former patches. Fixes LP: #1546565 *Update* - with the split libs it now nees to be listed in lib/librte_eal/linuxapp/eal/rte_eal_version.map to work on link steps - please note that upstream gravitates towards not extending but creating a new the API in DPDK as long term solution (will take a while) - also as listed before most affected projects seem to create their own workaround. So over time we have to check when we can drop it at the price of a config transition - likely OVS 2.6 won't need it anymore. *Update* - the handling and lifecycle of this changed in Openvswitch 2.7 so we can no more use internal_config. - Also the upstreaming was aborted as that now clearly goes towards client mode vhost sockets for this (and other issues). - But until that is fully working we have to carry the workaround. - Updated to work with Openvswitch 2.7 (and backward compatible to 2.6) |
| fix power default config.patch | (download) |
config/defconfig_ppc_64-power8-linuxapp-gcc |
1 0 + 1 - 0 ! |
config: fix power8 default conf Latest dpdk FTBFS for power8 default config. This (re-)enables a config symbol as recommended by IBM (gowrishankar.m@in.ibm.com) Signed-off-by: Christian Ehrhardt <christian.ehrhardt@canonical.com> |
| 0001 vhost reject invalid log base mmap offset.patch | (download) |
lib/librte_vhost/vhost_user.c |
9 9 + 0 - 0 ! |
vhost: reject invalid log base mmap offset If the log base mmap_offset is larger than mmap_size then it points outside the mmap region. We must not write to memory outside the mmap region, so validate mmap_offset in vhost_user_set_log_base(). Upstream: yes |
| 0002 vhost check log mmap offset and size overflow.patch | (download) |
lib/librte_vhost/vhost_user.c |
6 3 + 3 - 0 ! |
vhost: check log mmap offset and size overflow vhost_user_set_log_base() is a message handler that is called to handle the VHOST_USER_SET_LOG_BASE message. Its payload contains a 64 bit size and offset. Both are added up and used as a size when calling mmap(). There is no integer overflow check. If an integer overflow occurs a smaller memory map would be created than requested. Since the returned mapping is mapped as writable and used for logging, a memory corruption could occur. This issue has been assigned CVE-2020-10722 Upstream: yes |
