profile
viewpoint

issue commentmkubecek/vmware-host-modules

Build fails on kernel 5.1.5 and player 15.1.0

As I said, a good start would be to describe what exactly you did, i.e. how you got the source (cloned the git repository or downloaded a tarball), which of the two method you did use (building and installing from source tree or replacing the original tarballs and running vmware-modconfig --console --install-all). And, yes, full output of the failed build would be also helpful.

Joegoen

comment created time in 18 days

push eventmkubecek/vmware-host-modules

Michal Kubecek

commit sha 4f3ac281eedd07ac1a6bd1332668815f12a72169

vmmon: mark expected switch fall through Since v5.3-rc2, kernel is built with -Wimplicit-fallthrough by default so that compiler warns about switch fall thgrough unless it is marked as expected. Add comment to mark one existing fall through as expected. Note: this function is no longer present in 15.1.0 and it is probably dead code anyway as hash2() is only called from HashFunc_HashPage() which does not seem to be called anywhere. But to keep the code as close to upstream as reasonably possible, just mark the fall through and keep both functions in place.

view details

push time in 25 days

push eventmkubecek/vmware-host-modules

Michal Kubecek

commit sha 26fb3d29e071292e95f4f4a9ac3c111f1eecbf9c

vmmon: mark expected switch fall through Since v5.3-rc2, kernel is built with -Wimplicit-fallthrough by default so that compiler warns about switch fall thgrough unless it is marked as expected. Add comment to mark one existing fall through as expected. Note: this function is no longer present in 15.1.0 and it is probably dead code anyway as hash2() is only called from HashFunc_HashPage() which does not seem to be called anywhere. But to keep the code as close to upstream as reasonably possible, just mark the fall through and keep both functions in place.

view details

push time in 25 days

push eventmkubecek/vmware-host-modules

Michal Kubecek

commit sha 6c9f69c157abdb334936e0e6417eb0e937262762

vmmon: mark expected switch fall through Since v5.3-rc2, kernel is built with -Wimplicit-fallthrough by default so that compiler warns about switch fall thgrough unless it is marked as expected. Add comment to mark one existing fall through as expected. Note: this function is no longer present in 15.1.0 and it is probably dead code anyway as hash2() is only called from HashFunc_HashPage() which does not seem to be called anywhere. But to keep the code as close to upstream as reasonably possible, just mark the fall through and keep both functions in place.

view details

push time in 25 days

push eventmkubecek/vmware-host-modules

Michal Kubecek

commit sha 39afc1e8cccecbf49fa0440ec96f70d0f6cec7a7

vmmon: mark expected switch fall through Since v5.3-rc2, kernel is built with -Wimplicit-fallthrough by default so that compiler warns about switch fall thgrough unless it is marked as expected. Add comment to mark one existing fall through as expected. Note: this function is no longer present in 15.1.0 and it is probably dead code anyway as hash2() is only called from HashFunc_HashPage() which does not seem to be called anywhere. But to keep the code as close to upstream as reasonably possible, just mark the fall through and keep both functions in place.

view details

push time in 25 days

push eventmkubecek/ethnl

Michal Kubecek

commit sha f29d4395b239c4326f7fbc32f97c734ba5f1ffc4

kernel: use a typedef for legacy string arrays - introduce ethnl_string_array_t for ETH_GSTRING_LEN sized string arrays - use the type in bitset function declarations - start exporting patches with --base to include base commit id Rebase to net-next commit 5133f36cef46.

view details

push time in 25 days

issue closedmkubecek/vmware-host-modules

Is kernel 5.2 supported?

Hi,

Thanks for your work and this wonderful patch! My "make" is failing with:

/home/pyankulov/Downloads/vmware-host-modules-w15.1.0/vmmon-only/linux/driver.c:109:19: error: initialization of ‘vm_fault_t (*)(struct vm_fault *)’ {aka ‘unsigned int (*)(struct vm_fault *)’} from incompatible pointer type ‘int (*)(struct vm_fault *)’ [-Werror=incompatible-pointer-types]
         .fault  = LinuxDriverFault
                   ^~~~~~~~~~~~~~~~
/home/pyankulov/Downloads/vmware-host-modules-w15.1.0/vmmon-only/linux/driver.c:109:19: note: (near initialization for ‘vmuser_mops.fault’)
/home/pyankulov/Downloads/vmware-host-modules-w15.1.0/vmmon-only/linux/driver.c:986:1: warning: always_inline function might not be inlinable [-Wattributes]
 LinuxDriverSyncReadTSCs(uint64 *delta) // OUT: TSC max - TSC min
 ^~~~~~~~~~~~~~~~~~~~~~~
cc1: some warnings being treated as errors
make[3]: *** [scripts/Makefile.build:279: /home/pyankulov/Downloads/vmware-host-modules-w15.1.0/vmmon-only/linux/driver.o] Error 1
make[2]: *** [Makefile:1595: _module_/home/pyankulov/Downloads/vmware-host-modules-w15.1.0/vmmon-only] Error 2
make[2]: Leaving directory '/usr/src/linux-headers-5.2.2-122.current'
make[1]: *** [Makefile:110: vmmon.ko] Error 2
make[1]: Leaving directory '/home/pyankulov/Downloads/vmware-host-modules-w15.1.0/vmmon-only'
make: *** [Makefile:21: vmmon-only] Error 2

Is kernel 5.2 supported ? I'm using Solus Linux.

closed time in 25 days

presianbg

issue commentmkubecek/vmware-host-modules

Is kernel 5.2 supported?

Yes, everything up to 5.3-rc2 (or even current mainline) builds fine (and works, AFAICS). From the errors you get and from "w15.1.0" in your directory name, the problem is most likely building unpatched module source (tag w15.1.0) rather than patched one (branch workstation-15.1.0).

presianbg

comment created time in 25 days

push eventmkubecek/ethnl

Michal Kubecek

commit sha 269b6c91e6f5bc7c7be4905b6bcd36a3fe856a64

kernel: drop also ETHNL_BITSET_COMPACT flag After the previous changes to bitset handling functions, the only remaining ETHNL_BITSET_* flag is ETHNL_BITSET_COMPACT. Drop it and pass the flag as bool instead.

view details

push time in 25 days

push eventmkubecek/ethnl

Michal Kubecek

commit sha 6514f32ab99559d5dc2aac603fce6f07f9927acd

kernel: separate bitset code for u32 and unsigned long based bitmaps - drop simultaneous handling of u32 and unsigned long bitmaps - make unsigned long based functions wrappers - use simple cast on 64-bit LE and 32-bit architectures - use fixed size local variable for temporary bitmap if nbits <= 128 - only use kmalloc() for nbits > 128 on 64-bit BE - pass extack rather than genl_info for error handling Rebase to net-next commit dda91bd7abb0.

view details

Michal Kubecek

commit sha 0920f4450e00121d57c458aa7a85c57d2fe73813

kernel: use only fixed size strings internally Keeping string set code able to handle both arrays of const char pointers and fixed (ETH_GSTRING_LEN = 32) size strings makes it unnecessarily complex and harder to read. For now, use legacy fixed size strings everywhere and leave replacing them with simple const char * (everywhere) for later. - convert all new string sets to legacy format - pass array of ETH_GSTRING_LEN sized strings to all string set functions - drop ETH_SS_TYPE_LEGACY / ETH_SS_TYPE_SIMPLE types - use "strings" rather than "data" in struct strset_info - replace ethnl_str_size() with ethnl_strz_size() for fixed size strings - introduce ethnl_put_strz() helper - fix handling of non-null-terminated strings in drvinfo Rebase to net-next commit 85fd8011475e.

view details

push time in 25 days

issue commentmkubecek/vmware-host-modules

Build fails on kernel 5.1.5 and player 15.1.0

Hard to say as you only show part of the output and provide no information about what exactly you did. But this rather looks like a problem with your kernel devel packages, maybe some version mismatch.

Joegoen

comment created time in a month

issue commentmkubecek/vmware-host-modules

issue making vmmon.ko with target install for player-15.1.0

If the error message was what you reported, it means the commands of "install" target were not executed at all because one of the prerequisities (vmmon-only/vmmon.ko) was either outdated or missing. Which means the problem is not in what make install does (or rather would do) but in what should have happened before that: either make wasn't run or it failed in some way.

term1n4l

comment created time in a month

issue commentmkubecek/vmware-host-modules

issue making vmmon.ko with target install for player-15.1.0

"no rule to make target vmmon-only/vmmon.ko needed by install" sounds as if you forgot to run make before make install. But then you say you have vmmon.ko in vmmon-only... Maybe it's leftover from previous build.

I guess I would suggest to start from scratch, i.e. clean up everything with git clean -dfx and then run make (as normal user) and make install (as root).

term1n4l

comment created time in a month

issue closedmkubecek/vmware-host-modules

vmware 15 pro issue/ Fedora 30 / kernel 5.1.18-300.fc30.x86_64

Hi Sir Could you be of any help with this issue. I'm very new to Linux and follow instructions from others and yourself but now i am stuck with running my vm's any help would be greatly appreciated

Screenshot from 2019-07-21 12-27-29

Screenshot from 2019-07-21 12-30-49

closed time in a month

Saja89

issue commentmkubecek/vmware-host-modules

vmware 15 pro issue/ Fedora 30 / kernel 5.1.18-300.fc30.x86_64

You have loaded module built for different version of VMware Workstation than you are trying to run. In your case, 361 is used by 15.1.0 and 360 by 15.0.0 - 15.0.4. You need to use branch workstation-15.1.0 to build the module (and make sure to install and load it).

Saja89

comment created time in a month

push eventmkubecek/vmware-host-modules

Michal Kubecek

commit sha 5f884777a35ba42b80f9202a3d23fd46a29eaf9b

vmmon: fix HostIF_SetFastClockRate() not to use force_sig() Commit 3cf5d076fb4d ("signal: Remove task parameter from force_sig") in v5.3-rc1 drops second argument of force_sig(); before that, all callers which passed something else than current task as second argument were fixed, mostly to use send_sig() instead. The situation in HostIF_SetFastClockRate() is the same as e.g. in bpfilter call fixed by commit 1dfd1711de29 ("signal/bpfilter: Fix bpfilter_kernl to use send_sig not force_sig"): locking in force_sig_info() cannot handle task exiting and using force_sig() for SIGKILL is pointless anyway as this signal cannot be blocked. As send_sig() is present with unchanged signature since the pre-git era, we can use send_sig() unconditionally, regardless of kernel version.

view details

push time in a month

push eventmkubecek/vmware-host-modules

Michal Kubecek

commit sha 98008ff6adec118962df8ac26f814b55c3861efb

vmmon: fix HostIF_SetFastClockRate() not to use force_sig() Commit 3cf5d076fb4d ("signal: Remove task parameter from force_sig") in v5.3-rc1 drops second argument of force_sig(); before that, all callers which passed something else than current task as second argument were fixed, mostly to use send_sig() instead. The situation in HostIF_SetFastClockRate() is the same as e.g. in bpfilter call fixed by commit 1dfd1711de29 ("signal/bpfilter: Fix bpfilter_kernl to use send_sig not force_sig"): locking in force_sig_info() cannot handle task exiting and using force_sig() for SIGKILL is pointless anyway as this signal cannot be blocked. As send_sig() is present with unchanged signature since the pre-git era, we can use send_sig() unconditionally, regardless of kernel version.

view details

push time in a month

push eventmkubecek/vmware-host-modules

Michal Kubecek

commit sha 4f88bce6846f4b0b148236b86b4468a73da0bf74

vmmon: fix HostIF_SetFastClockRate() not to use force_sig() Commit 3cf5d076fb4d ("signal: Remove task parameter from force_sig") in v5.3-rc1 drops second argument of force_sig(); before that, all callers which passed something else than current task as second argument were fixed, mostly to use send_sig() instead. The situation in HostIF_SetFastClockRate() is the same as e.g. in bpfilter call fixed by commit 1dfd1711de29 ("signal/bpfilter: Fix bpfilter_kernl to use send_sig not force_sig"): locking in force_sig_info() cannot handle task exiting and using force_sig() for SIGKILL is pointless anyway as this signal cannot be blocked. As send_sig() is present with unchanged signature since the pre-git era, we can use send_sig() unconditionally, regardless of kernel version.

view details

push time in a month

push eventmkubecek/vmware-host-modules

Michal Kubecek

commit sha c37ec19b19f949d55866ac688b77d81887326043

vmmon: fix HostIF_SetFastClockRate() not to use force_sig() Commit 3cf5d076fb4d ("signal: Remove task parameter from force_sig") in v5.3-rc1 drops second argument of force_sig(); before that, all callers which passed something else than current task as second argument were fixed, mostly to use send_sig() instead. The situation in HostIF_SetFastClockRate() is the same as e.g. in bpfilter call fixed by commit 1dfd1711de29 ("signal/bpfilter: Fix bpfilter_kernl to use send_sig not force_sig"): locking in force_sig_info() cannot handle task exiting and using force_sig() for SIGKILL is pointless anyway as this signal cannot be blocked. As send_sig() is present with unchanged signature since the pre-git era, we can use send_sig() unconditionally, regardless of kernel version.

view details

push time in a month

push eventmkubecek/vmware-host-modules

Michal Kubecek

commit sha 5ecf2faaa61d089c9b5715489e14b27742316f1e

vmmon: fix HostIF_SetFastClockRate() not to use force_sig() Commit 3cf5d076fb4d ("signal: Remove task parameter from force_sig") in v5.3-rc1 drops second argument of force_sig(); before that, all callers which passed something else than current task as second argument were fixed, mostly to use send_sig() instead. The situation in HostIF_SetFastClockRate() is the same as e.g. in bpfilter call fixed by commit 1dfd1711de29 ("signal/bpfilter: Fix bpfilter_kernl to use send_sig not force_sig"): locking in force_sig_info() cannot handle task exiting and using force_sig() for SIGKILL is pointless anyway as this signal cannot be blocked. As send_sig() is present with unchanged signature since the pre-git era, we can use send_sig() unconditionally, regardless of kernel version.

view details

push time in a month

push eventmkubecek/vmware-host-modules

Michal Kubecek

commit sha d20641f619bd3766cf5b50b2c750700eab189bd9

vmmon: fix HostIF_SetFastClockRate() not to use force_sig() Commit 3cf5d076fb4d ("signal: Remove task parameter from force_sig") in v5.3-rc1 drops second argument of force_sig(); before that, all callers which passed something else than current task as second argument were fixed, mostly to use send_sig() instead. The situation in HostIF_SetFastClockRate() is the same as e.g. in bpfilter call fixed by commit 1dfd1711de29 ("signal/bpfilter: Fix bpfilter_kernl to use send_sig not force_sig"): locking in force_sig_info() cannot handle task exiting and using force_sig() for SIGKILL is pointless anyway as this signal cannot be blocked. As send_sig() is present with unchanged signature since the pre-git era, we can use send_sig() unconditionally, regardless of kernel version.

view details

push time in a month

issue closedmkubecek/vmware-host-modules

make error generated/autoconf.h: No such file or directory

I have encountered an error when making. My system is Manjaro.

Here are the details.

Using standalone build system. In file included from ./autoconf/geninclude.c:20:0: ./include/compat_autoconf.h:38:13: fatal error: generated/autoconf.h: No such file or directory

include <generated/autoconf.h>

         ^~~~~~~~~~~~~~~~~~~~~~

compilation terminated. In file included from .././autoconf/geninclude.c:20:0: .././include/compat_autoconf.h:38:13: fatal error: generated/autoconf.h: No such file or directory

include <generated/autoconf.h>

         ^~~~~~~~~~~~~~~~~~~~~~

compilation terminated. Dependencies for hostif.c cc1: error: code model kernel does not support PIC mode make[1]: *** [.././Makefile.normal:132: hostif.d] Error 1 make: *** [Makefile.normal:143: deps] Error 2

closed time in a month

CoCoThink

issue commentmkubecek/vmware-host-modules

make error generated/autoconf.h: No such file or directory

You don't need kernel sources, you need kernel-default-devel (and kernel-devel but that would be installed as a dependency). Also make sure you clean up leftovers from previous failed builds before trying again.

CoCoThink

comment created time in a month

issue commentmkubecek/vmware-host-modules

asm error building 15.1.0 for kernel 5.1.14

That could happen if you accidentally run make as root once; it would create files owned by root which make executed as normal user cannot overwrite. Running make clean (or git clean -dfx) as root should get rid of them.

hmnd

comment created time in a month

push eventmkubecek/ethnl

Michal Kubecek

commit sha ce11a322acba3a1668ebd72a2f08eb37994696a4

kernel: rename request flag constants - replace ETHTOOL_RF_ prefix with ETHTOOL_[GR]FLAG_ - rename ETHTOOL_RF_COMPACT to ETHTOOL_GFLAG_COMPACT_BITSETS - rename ETHTOOL_RF_REPLY to ETHTOOL_GFLAG_OMIT_REPLY (inverted logic!) - rename ETHTOOL_RF_STRSET_COUNTS to ETHTOOL_RFLAG_COUNTS_ONLY - fix missing notifications on netdev feature changes Rebase to net-next commit 1ff2f0fa450e.

view details

Michal Kubecek

commit sha 5de79c0ddfe718b1dca4a882418617b29d4d7756

ethtool: rename request flag constants - sync UAPI header with kernel - mirror the ETHTOOL_RF_* -> ETHTOOL_[GR]FLAG_ rename in kernel UAPI - request all optional SET/ACT replies again

view details

push time in a month

create barnchmkubecek/vmware-host-modules

branch : tmp/workstation-12.5.9-k5.3

created branch time in a month

create barnchmkubecek/vmware-host-modules

branch : tmp/workstation-14.1.7-k5.3

created branch time in a month

delete branch mkubecek/vmware-host-modules

delete branch : workstation-15.0.3-test

delete time in 2 months

create barnchmkubecek/vmware-host-modules

branch : tmp/workstation-15.1.0-k5.3

created branch time in 2 months

push eventmkubecek/vmware-host-modules

Michal Kubecek

commit sha 00b8e2f065c121cad22b3d575c833f795ea5db8f

mark branch player-14.1.5 as retired

view details

push time in 2 months

push eventmkubecek/vmware-host-modules

Michal Kubecek

commit sha 155ee015393358ac410d82794628817d3a54097f

mark branch player-15.0.2 as retired

view details

push time in 2 months

push eventmkubecek/vmware-host-modules

Michal Kubecek

commit sha 36880c2364774b3b8b64f9dd31352feb3e681fbb

mark branch workstation-15.0.2 as retired

view details

push time in 2 months

push eventmkubecek/vmware-host-modules

Michal Kubecek

commit sha 67d1de7be00700eaf88d2d28468b7e733507ffb8

mark branch player-15.0.4 as retired

view details

push time in 2 months

push eventmkubecek/vmware-host-modules

Michal Kubecek

commit sha a718c005c48bb24084c8a916cfd62852f9534479

mark branch player-14.1.6 as retired

view details

push time in 2 months

push eventmkubecek/vmware-host-modules

Michal Kubecek

commit sha b513c2469eb2c7da7ea10e15969ed66658fcfb05

mark branch workstation-15.0.4 as retired

view details

push time in 2 months

push eventmkubecek/vmware-host-modules

Michal Kubecek

commit sha c103d150d8e37678f21fd620431ff990bc88afa6

mark branch player-15.0.3 as retired

view details

push time in 2 months

push eventmkubecek/vmware-host-modules

Michal Kubecek

commit sha cb3a140945be2414d4d80cc06ef14e52448aeb79

mark branch workstation-14.1.5 as retired

view details

push time in 2 months

push eventmkubecek/vmware-host-modules

Michal Kubecek

commit sha cde047d4d7778e5062cc8e2f69243098371d29e7

mark branch workstation-14.1.6 as retired

view details

push time in 2 months

push eventmkubecek/vmware-host-modules

Michal Kubecek

commit sha f2b5b2a7158bbe91065984038f5c7243ea714fdc

mark branch workstation-15.0.3 as retired

view details

push time in 2 months

push eventmkubecek/ethnl

Michal Kubecek

commit sha 7d253a64d88fe8a6c8aa649589e87cc978ee53fa

kernel: redesign ethnl_update_* helpers (v6 review) - update helpers now get pointer to mod as a parameter - they set it to true if data is modified, leave it as it was if not - ethnl_update_bitset() and ethnl_update_bitset32() return error code - more consistent return value handling in ethnl_set_settings() Rebase to net-next commit a51df9f8da43.

view details

push time in 2 months

push eventmkubecek/ethnl

Michal Kubecek

commit sha f742b9d547c611ee557a4a81527ac79c0c34eab3

kernel: small changes from v6 review - minor documentation improvements - rename ethnl_parse_header() parameter nest to header - drop cleanup mention in ethnl_fill_reply_header() description - fix ethnl_reply_init() error path - make ethnl_update_binary() parameter dst void pointer Rebase to net-next commit dbcdb61aaf1a.

view details

push time in 2 months

created tagmkubecek/ethnl

tagv6

netlink interface for ethtool - work in progress

created time in 2 months

push eventmkubecek/ethnl

Michal Kubecek

commit sha 5d582d55d910647933cb39d7232f2f0202ac79be

kernel: one more spatch warning Not sure why spatch doesn't like this assignment, gcc doesn't seem to have a problem with it. Make it an explicit cast to avoid kbuild complaints.

view details

push time in 2 months

push eventmkubecek/ethnl

Michal Kubecek

commit sha 915c33708440b2f69484e2d314b3cea07e2297c1

kernel: use netlink_callback::ctx rather than args To store parameters to keep through iteration over device list in dumps, use a structure laid over the newly introduced netlink_callback::ctx rather than the old array of longs (netlink_callback::args). Rebase on net-next commit 337d1ccb3dc9.

view details

Michal Kubecek

commit sha 9c57aa89bfe80ac49f0f5c4419f81af6ba70e6e9

kernel: final pre-v6 touches - add Acked-by/Reviewed-by from v5 review - unsplit some GENL_SET_ERR_MSG() lines

view details

push time in 2 months

push eventmkubecek/ethnl

Michal Kubecek

commit sha d97fcaeafc65302187db51d602b5b5cefbc8690b

address checkpatch issues - minor coding style fixes - two typos in commit messages

view details

push time in 2 months

push eventmkubecek/ethnl

Michal Kubecek

commit sha 1d30a17d6422ff3de610a3836e802a5ccb823a85

kernel: address sparse warnings Functions ethnl_get_doit() and ethnl_get_dumpit() and array get_requests are only used in netlink.c so make them static.

view details

push time in 2 months

push eventmkubecek/ethnl

Michal Kubecek

commit sha 3500d42cef5d62230fc86b5f57d57852b77ecb35

kernel: update cover letter for v6 submission - update paragraph about message types to reflect recent changes - add v5 -> v6 changes section - drop ToDo section - minor editing

view details

push time in 2 months

push eventmkubecek/ethnl

Michal Kubecek

commit sha cdee1bc604a825fb4f2e7231c84eda573edb9883

kernel: reorder patches to prepare for v6 None of the patches to be submitted in v6 of first part uses the nla_put_bitfield32() helper so move the patch adding it just before the first patch using it. In the end, we may use arbitrary size bitmaps even for the attributes where NLA_BITFIELD32 is used now so that the helper wouldn't be used at all.

view details

push time in 2 months

push eventmkubecek/ethnl

Michal Kubecek

commit sha f0bb99eb0fa7fc11c72d96b2b51c82a6c70002ea

kernel: improve commit messages - more descriptive commit messages for commits to be submitted in v6 - minor whitespace cleanup in documentation - drop leftover paragraph from older version of documentation Rebase to net-next commit 954a5a029472.

view details

push time in 2 months

push eventmkubecek/ethnl

Michal Kubecek

commit sha 5b6d12461a8833a8eb81b95756b429ab85f421de

kernel: reorganize the patches to reduce "defined but not used" warnings - split ethnl_std_notify and related out of generic GET handling patch - put both patches just before the functions from them are first used - move all SETTINGS_NTF code from "GET" patch to "SET" one - move all PARAMS_NTF code from "GET" patch to "SET" one - improve comments documenting get_request_ops and related data structure - no functional or code changes in final tree (only comments) Rebase on net-next commit 5cdda5f1d6ad.

view details

push time in 2 months

push eventmkubecek/ethnl

Michal Kubecek

commit sha 9594306a234891c9c09fc820bcf1b726d2e6c500

kernel: coding style cleanup (v5 review) - use nla_put_failure label where appropriate - avoid using ret variable if it's always -EMSGSIZE - drop always fulfilled conditions - keep include of <net/xdp_sock.h> in ioctl.c (needed after rebase) Rebase to net-next commit ee7dd7733b20.

view details

push time in 2 months

push eventmkubecek/ethnl

Michal Kubecek

commit sha e6a606b81e7f8a3dc678160de35e647320c058d3

kernel: use unified request/reply header - unified request and reply header in all messages - simplified and unified header processing - one more nesting level for string sets - minor documentation improvements - fix indirection table fill with "equal N" Rebase to net-next commit 8b89d8dad5df.

view details

Michal Kubecek

commit sha 88608c346ec35da42a0353777a95eb34f03481cf

ethtool: unified request/reply header - sync UAPI with kernel series - unified request/reply header, generate and process - one more nesting level for string sets - reorganize early patches to make each commit build - fix crash when rx hash options are missing (monitor)

view details

push time in 2 months

issue closedmkubecek/vmware-host-modules

can't compile on Fedora 30 after update to kernel 5.1.12-300.fc30.x86_64

Version mismatch: module vmmon-only/vmmon.ko 5.1.11-300.fc30.x86_64, kernel 5.1.12-300.fc30.x86_64 Version mismatch: module vmnet-only/vmnet.ko 5.1.11-300.fc30.x86_64, kernel 5.1.12-300.fc30.x86_64

closed time in 2 months

shuifeng1988

issue commentmkubecek/vmware-host-modules

can't compile on Fedora 30 after update to kernel 5.1.12-300.fc30.x86_64

This happens when you run make install with different value of VM_UNAME than you used for make. (If VM_UNAME is not set, it defaults to the output of uname -r.)

shuifeng1988

comment created time in 2 months

issue commentmkubecek/vmware-host-modules

compile vmmon-only failed on workstation-15.1.0 branch, kernel: 5.0.13-200.fc29.x86_64 fedora 30.

Some kernel devel package is missing. Please check the documentation of your distribution to see what you need to build out-of-tree kernel modules.

xg6402

comment created time in 2 months

issue closedmkubecek/vmware-host-modules

compile vmmon-only failed on workstation-15.1.0 branch, kernel: 5.0.13-200.fc29.x86_64 fedora 30.

./include/compat_autoconf.h:38:13: fatal error: generated/autoconf.h: No such file or directory
38 | # include <generated/autoconf.h>
| ^~~~~~~~~~~~~~~~~~~~~~
compilation terminated.

closed time in 2 months

xg6402

issue commentmkubecek/vmware-host-modules

compile vmmon-only failed on workstation-15.1.0 branch, kernel: 5.0.13-200.fc29.x86_64 fedora 30.

You don't have necessary kernel devel packages of your distribution.

xg6402

comment created time in 2 months

push eventmkubecek/ethnl

Michal Kubecek

commit sha 99d84ce0228d60154786578aefea092ffc668d66

kernel: reorganize message type constants (v5 review) - use two separate enums for kernel and userspace messages - distinguish kernel replies and notifications from userspace SET requests - use ETHTOOL_MSG_ prefix rather than ETHNL_CMD_ - drop no longer valid "notification type is get request type + 1" logic - improve documentation on message type naming scheme - mark the point to add new constants in UAPI header file Rebase to net-next commit f464100f5766.

view details

Michal Kubecek

commit sha d122ef4ca5814aa310c08207fe3ec6fc87310026

ethtool: sync with kernel series - sync UAPI with kernel series (message type rename) - use new message type constants in ethtool netlink code - update message type output in --debug - drop handling of INFO notification (there isn't any) Known issue: some patches use functions declared/defined in later patches so that the series is not bisectable at the moment. Will fix.

view details

push time in 2 months

push eventmkubecek/ethnl

Michal Kubecek

commit sha 262f3b8ed57f82ba4f3f8fe72d8f34884e3e1958

kernel: reorder patches for v6 submission - drop EVENT notifications, rtnetlink is to be used to track device list - move GET_INFO commits out of the part to be submitted until it's clear what to keep in DRVINFO and what to use devlink (and if there will be "generic devlink" for devices without actual devlink support Rebase to net-next commit 877cd9ffbc9c.

view details

Michal Kubecek

commit sha c9301d1899b16dabad68da454e373159bdb2b72b

ethtool: update to reflect recent kernel changes - sync UAPI with net-next - sync UAPI with kernel ethnl series - drop processing of EVENT notifications Rebase to master commit 2bce6d9cb08e.

view details

push time in 2 months

push eventmkubecek/ethnl

Michal Kubecek

commit sha 154db214ae31ea0bf3058944d00e65cfd5d5dbf1

kernel: update to current net-next - drop netlink patches which are in net-next now - add support for {100,1000}baseT1 link modes - stricter check that link_mode_names[] is up-to-date - add check that link_mode_params[] is up-to-date - do not pass IFLA_PERM_ADDRESS if address is all zeros Rebase to net-next commit a842fe1425cb.

view details

Michal Kubecek

commit sha 55f5be1d68bd2273912d2ab61db9ce04551511e7

kernel: function and variable renames (addressing v5 review) - rename more identifiers (including static) to have unified prefixes - drop warn_partial_info() and few other not so much used helpers - proper cleanup when handling generic dump requests - better description of ethnl_update_*() helpers - add more function description comments

view details

Michal Kubecek

commit sha 35c76c217990f0aa1bf8dfbf081dc66056f25343

ethtool: sync with recent kernel changes - update (non-ethnl) UAPI with kernel net-next - show bad attribute offset in error/warning messsages if provided Rebase to master commit cabc523cd2c3 (v5.1).

view details

push time in 2 months

issue closedmkubecek/vmware-host-modules

cannot build on fedora 30 / kernel 5.1.7-300.fc30.x86_64

$ make

make -C vmmon-only 
make[1]: Entering directory '/home/me/Desktop/vmware-host-modules-w15.1.0/vmmon-only'
Using kernel build system.
make -C /lib/modules/5.1.7-300.fc30.x86_64/build/include/.. SUBDIRS=$PWD SRCROOT=$PWD/. \
  MODULEBUILDDIR= modules
make[2]: Entering directory '/usr/src/kernels/5.1.7-300.fc30.x86_64'
Makefile:205: ================= WARNING ================
Makefile:206: 'SUBDIRS' will be removed after Linux 5.3
Makefile:207: Please use 'M=' or 'KBUILD_EXTMOD' instead
Makefile:208: ==========================================
  CC [M]  /home/me/Desktop/vmware-host-modules-w15.1.0/vmmon-only/linux/driver.o
In file included from /home/me/Desktop/vmware-host-modules-w15.1.0/vmmon-only/./common/vmx86.h:32,
                 from /home/me/Desktop/vmware-host-modules-w15.1.0/vmmon-only/linux/driver.h:32,
                 from /home/me/Desktop/vmware-host-modules-w15.1.0/vmmon-only/linux/driver.c:47:
/home/me/Desktop/vmware-host-modules-w15.1.0/vmmon-only/./include/x86msr.h:132: warning: "MSR_MISC_FEATURES_ENABLES" redefined
  132 | #define MSR_MISC_FEATURES_ENABLES            0x140
      | 
In file included from ./arch/x86/include/asm/nospec-branch.h:11,
                 from ./arch/x86/include/asm/paravirt_types.h:46,
                 from ./arch/x86/include/asm/ptrace.h:94,
                 from ./arch/x86/include/asm/math_emu.h:5,
                 from ./arch/x86/include/asm/processor.h:12,
                 from ./arch/x86/include/asm/cpufeature.h:5,
                 from ./arch/x86/include/asm/thread_info.h:53,
                 from ./include/linux/thread_info.h:38,
                 from ./arch/x86/include/asm/preempt.h:7,
                 from ./include/linux/preempt.h:78,
                 from ./include/linux/spinlock.h:51,
                 from ./include/linux/wait.h:9,
                 from ./include/linux/wait_bit.h:8,
                 from ./include/linux/fs.h:6,
                 from ./include/linux/highmem.h:5,
                 from /home/me/Desktop/vmware-host-modules-w15.1.0/vmmon-only/linux/driver.c:25:
./arch/x86/include/asm/msr-index.h:668: note: this is the location of the previous definition
  668 | #define MSR_MISC_FEATURES_ENABLES 0x00000140
      | 
In file included from /home/me/Desktop/vmware-host-modules-w15.1.0/vmmon-only/./common/vmx86.h:32,
                 from /home/me/Desktop/vmware-host-modules-w15.1.0/vmmon-only/linux/driver.h:32,
                 from /home/me/Desktop/vmware-host-modules-w15.1.0/vmmon-only/linux/driver.c:47:
/home/me/Desktop/vmware-host-modules-w15.1.0/vmmon-only/./include/x86msr.h:468: warning: "MSR_K7_HWCR_SMMLOCK" redefined
  468 | #define MSR_K7_HWCR_SMMLOCK        0x00000001ULL // Lock SMM environment
      | 
In file included from ./arch/x86/include/asm/nospec-branch.h:11,
                 from ./arch/x86/include/asm/paravirt_types.h:46,
                 from ./arch/x86/include/asm/ptrace.h:94,
                 from ./arch/x86/include/asm/math_emu.h:5,
                 from ./arch/x86/include/asm/processor.h:12,
                 from ./arch/x86/include/asm/cpufeature.h:5,
                 from ./arch/x86/include/asm/thread_info.h:53,
                 from ./include/linux/thread_info.h:38,
                 from ./arch/x86/include/asm/preempt.h:7,
                 from ./include/linux/preempt.h:78,
                 from ./include/linux/spinlock.h:51,
                 from ./include/linux/wait.h:9,
                 from ./include/linux/wait_bit.h:8,
                 from ./include/linux/fs.h:6,
                 from ./include/linux/highmem.h:5,
                 from /home/me/Desktop/vmware-host-modules-w15.1.0/vmmon-only/linux/driver.c:25:
./arch/x86/include/asm/msr-index.h:477: note: this is the location of the previous definition
  477 | #define MSR_K7_HWCR_SMMLOCK  BIT_ULL(MSR_K7_HWCR_SMMLOCK_BIT)
      | 
In file included from /home/me/Desktop/vmware-host-modules-w15.1.0/vmmon-only/./include/vm_asm_x86.h:41,
                 from /home/me/Desktop/vmware-host-modules-w15.1.0/vmmon-only/./include/vm_asm_x86_64.h:40,
                 from /home/me/Desktop/vmware-host-modules-w15.1.0/vmmon-only/./include/vm_asm.h:46,
                 from /home/me/Desktop/vmware-host-modules-w15.1.0/vmmon-only/linux/driver.c:49:
/home/me/Desktop/vmware-host-modules-w15.1.0/vmmon-only/./include/x86_basic_defs.h:78: warning: "CR3_PCID_MASK" redefined
   78 | #define CR3_PCID_MASK  0xFFF
      | 
In file included from ./arch/x86/include/asm/processor.h:5,
                 from ./arch/x86/include/asm/cpufeature.h:5,
                 from ./arch/x86/include/asm/thread_info.h:53,
                 from ./include/linux/thread_info.h:38,
                 from ./arch/x86/include/asm/preempt.h:7,
                 from ./include/linux/preempt.h:78,
                 from ./include/linux/spinlock.h:51,
                 from ./include/linux/wait.h:9,
                 from ./include/linux/wait_bit.h:8,
                 from ./include/linux/fs.h:6,
                 from ./include/linux/highmem.h:5,
                 from /home/me/Desktop/vmware-host-modules-w15.1.0/vmmon-only/linux/driver.c:25:
./arch/x86/include/asm/processor-flags.h:39: note: this is the location of the previous definition
   39 | #define CR3_PCID_MASK 0xFFFull
      | 
/home/me/Desktop/vmware-host-modules-w15.1.0/vmmon-only/linux/driver.c:109:19: error: initialization of ‘vm_fault_t (*)(struct vm_fault *)’ {aka ‘unsigned int (*)(struct vm_fault *)’} from incompatible pointer type ‘int (*)(struct vm_fault *)’ [-Werror=incompatible-pointer-types]
  109 |         .fault  = LinuxDriverFault
      |                   ^~~~~~~~~~~~~~~~
/home/me/Desktop/vmware-host-modules-w15.1.0/vmmon-only/linux/driver.c:109:19: note: (near initialization for ‘vmuser_mops.fault’)
/home/me/Desktop/vmware-host-modules-w15.1.0/vmmon-only/linux/driver.c:986:1: warning: always_inline function might not be inlinable [-Wattributes]
  986 | LinuxDriverSyncReadTSCs(uint64 *delta) // OUT: TSC max - TSC min
      | ^~~~~~~~~~~~~~~~~~~~~~~
cc1: some warnings being treated as errors
make[3]: *** [scripts/Makefile.build:276: /home/me/Desktop/vmware-host-modules-w15.1.0/vmmon-only/linux/driver.o] Error 1
make[2]: *** [Makefile:1575: _module_/home/me/Desktop/vmware-host-modules-w15.1.0/vmmon-only] Error 2
make[2]: Leaving directory '/usr/src/kernels/5.1.7-300.fc30.x86_64'
make[1]: *** [Makefile:110: vmmon.ko] Error 2
make[1]: Leaving directory '/home/me/Desktop/vmware-host-modules-w15.1.0/vmmon-only'
make: *** [Makefile:21: vmmon-only] Error 2

closed time in 2 months

alsoGAMER

issue commentmkubecek/vmware-host-modules

cannot build on fedora 30 / kernel 5.1.7-300.fc30.x86_64

You are compiling unpatched 15.1.0 source (tag w15.1.0?), not patched one (branch workstation-15.1.0).

alsoGAMER

comment created time in 2 months

more