ena(4) became MPSAFe last year

Add bge(4) to the list of drivers that take advantage of NET_MPSAFE

fix various typos, mainly s/trasfering/transferring/ and s/theese/these/.

doc: Update TODO.smpnet to reflect invention of ETHER_F_ALLMULTI.

fix various typos in comments and messages.

doc: more MPSAFE interfaces

Remove ixgbe because both ixg and ixv exist in the list.

update the list of ethernet drivers that have been MPSAFEified.

if_tap was made mpsafe in revision 1.102

Remove references to BRIDGE_IPF, it is now compiled in by default.

BRIDGE_IPF is MP-safe, discussed with ozaki-r@

strip(4) was removed

netsmb is no more

write down some observations about the locking wild west around if_mcast_op()

FDDI support has been removed.

Token Ring has been removed.

HIPPI has been removed.

Protecting ec_multi* of ethercom with ec_lock have been finished now.

Add aq(4) and ixl(4) to the MPSAFE component list.

Add another known issue (if_flags)

Add another known issue

Note that kernels with NET_MPSAFE won't work with non-NET_MPSAFE modules.

XXX The initial merge of the [pgoyette-compat] branch will NOT address
XXX this issue.

Drop LMC-related entry from TODO.smpnet, and note removal of LMC.

remove references to isdn

Note removal of midway and NATM, and prune the entries from TODO.smpnet.

Retire EtherIP, we have L2TP instead.

Add ipsecif(4)

Categorize non MP-safe components into protected and unprotected ones

branches: 1.20.2;
Move txintr_setup() stuff from lmc_interrupt() and do it in ifnet_start().
Now we can use bpf_mtap() in the TX path. Not tested.

branches: 1.19.2;
Remove the Econet code. It was part of acorn26, which was removed a
month ago.

Mention the limitation of enabling both NET_MPSAFE and ALTQ

Mention IPv6 address selection policy isn't MP-safe yet

Though it's not a problem until a policy is set.

Mention ec_multi*

Mention NOMPSAFE

add opencrypto(9)'s scalability comment

Write known issues on scalability

Update

ipsec(4), opencrypto(9) and vlan(4) are now MP-safe.

branches: 1.11.2; 1.11.6;
Write about bpf's statistic counters

Write about lingering obsolete variables due to kvm(3)

branches: 1.9.2;
Add ixv(4) to MP-safe component's list.

Add ixg(4)

Tidy up again

Tidy up the lists

Remove multicast stuffs from the list of Non MP-safe components

Add l2tp(4)

Add BRIDGE_IPF

Add many entries

Defer bpf_mtap in Rx interrupt context to softint

bpf_mtap of some drivers is still called in hardware interrupt context.
We want to run them in softint as well as bpf_mtap of most drivers
(see if_percpuq_softint and if_input).

To this end, bpf_mtap_softint mechanism is implemented; it defers
bpf_mtap processing to a dedicated softint for a target driver.
By using the machanism, we can move bpf_mtap processing to softint
without changing target drivers much while it adds some overhead
on CPU and memory. Once target drivers are changed to softint-based,
we should return to normal bpf_mtap.

Proposed on tech-kern and tech-net

Add bge(4) to the list of drivers that take advantage of NET_MPSAFE

fix various typos, mainly s/trasfering/transferring/ and s/theese/these/.

doc: Update TODO.smpnet to reflect invention of ETHER_F_ALLMULTI.

fix various typos in comments and messages.

doc: more MPSAFE interfaces

Remove ixgbe because both ixg and ixv exist in the list.

update the list of ethernet drivers that have been MPSAFEified.

if_tap was made mpsafe in revision 1.102

Remove references to BRIDGE_IPF, it is now compiled in by default.

BRIDGE_IPF is MP-safe, discussed with ozaki-r@

strip(4) was removed

netsmb is no more

write down some observations about the locking wild west around if_mcast_op()

FDDI support has been removed.

Token Ring has been removed.

HIPPI has been removed.

Protecting ec_multi* of ethercom with ec_lock have been finished now.

Add aq(4) and ixl(4) to the MPSAFE component list.

Add another known issue (if_flags)

Add another known issue

Note that kernels with NET_MPSAFE won't work with non-NET_MPSAFE modules.

XXX The initial merge of the [pgoyette-compat] branch will NOT address
XXX this issue.

Drop LMC-related entry from TODO.smpnet, and note removal of LMC.

remove references to isdn

Note removal of midway and NATM, and prune the entries from TODO.smpnet.

Retire EtherIP, we have L2TP instead.

Add ipsecif(4)

Categorize non MP-safe components into protected and unprotected ones

branches: 1.20.2;
Move txintr_setup() stuff from lmc_interrupt() and do it in ifnet_start().
Now we can use bpf_mtap() in the TX path. Not tested.

branches: 1.19.2;
Remove the Econet code. It was part of acorn26, which was removed a
month ago.

Mention the limitation of enabling both NET_MPSAFE and ALTQ

Mention IPv6 address selection policy isn't MP-safe yet

Though it's not a problem until a policy is set.

Mention ec_multi*

Mention NOMPSAFE

add opencrypto(9)'s scalability comment

Write known issues on scalability

Update

ipsec(4), opencrypto(9) and vlan(4) are now MP-safe.

branches: 1.11.2; 1.11.6;
Write about bpf's statistic counters

Write about lingering obsolete variables due to kvm(3)

branches: 1.9.2;
Add ixv(4) to MP-safe component's list.

Add ixg(4)

Tidy up again

Tidy up the lists

Remove multicast stuffs from the list of Non MP-safe components

Add l2tp(4)

Add BRIDGE_IPF

Add many entries

Defer bpf_mtap in Rx interrupt context to softint

bpf_mtap of some drivers is still called in hardware interrupt context.
We want to run them in softint as well as bpf_mtap of most drivers
(see if_percpuq_softint and if_input).

To this end, bpf_mtap_softint mechanism is implemented; it defers
bpf_mtap processing to a dedicated softint for a target driver.
By using the machanism, we can move bpf_mtap processing to softint
without changing target drivers much while it adds some overhead
on CPU and memory. Once target drivers are changed to softint-based,
we should return to normal bpf_mtap.

Proposed on tech-kern and tech-net

fix various typos, mainly s/trasfering/transferring/ and s/theese/these/.

doc: Update TODO.smpnet to reflect invention of ETHER_F_ALLMULTI.

fix various typos in comments and messages.

doc: more MPSAFE interfaces

Remove ixgbe because both ixg and ixv exist in the list.

update the list of ethernet drivers that have been MPSAFEified.

if_tap was made mpsafe in revision 1.102

Remove references to BRIDGE_IPF, it is now compiled in by default.

BRIDGE_IPF is MP-safe, discussed with ozaki-r@

strip(4) was removed

netsmb is no more

write down some observations about the locking wild west around if_mcast_op()

FDDI support has been removed.

Token Ring has been removed.

HIPPI has been removed.

Protecting ec_multi* of ethercom with ec_lock have been finished now.

Add aq(4) and ixl(4) to the MPSAFE component list.

Add another known issue (if_flags)

Add another known issue

Note that kernels with NET_MPSAFE won't work with non-NET_MPSAFE modules.

XXX The initial merge of the [pgoyette-compat] branch will NOT address
XXX this issue.

Drop LMC-related entry from TODO.smpnet, and note removal of LMC.

remove references to isdn

Note removal of midway and NATM, and prune the entries from TODO.smpnet.

Retire EtherIP, we have L2TP instead.

Add ipsecif(4)

Categorize non MP-safe components into protected and unprotected ones

branches: 1.20.2;
Move txintr_setup() stuff from lmc_interrupt() and do it in ifnet_start().
Now we can use bpf_mtap() in the TX path. Not tested.

branches: 1.19.2;
Remove the Econet code. It was part of acorn26, which was removed a
month ago.

Mention the limitation of enabling both NET_MPSAFE and ALTQ

Mention IPv6 address selection policy isn't MP-safe yet

Though it's not a problem until a policy is set.

Mention ec_multi*

Mention NOMPSAFE

add opencrypto(9)'s scalability comment

Write known issues on scalability

Update

ipsec(4), opencrypto(9) and vlan(4) are now MP-safe.

branches: 1.11.2; 1.11.6;
Write about bpf's statistic counters

Write about lingering obsolete variables due to kvm(3)

branches: 1.9.2;
Add ixv(4) to MP-safe component's list.

Add ixg(4)

Tidy up again

Tidy up the lists

Remove multicast stuffs from the list of Non MP-safe components

Add l2tp(4)

Add BRIDGE_IPF

Add many entries

Defer bpf_mtap in Rx interrupt context to softint

bpf_mtap of some drivers is still called in hardware interrupt context.
We want to run them in softint as well as bpf_mtap of most drivers
(see if_percpuq_softint and if_input).

To this end, bpf_mtap_softint mechanism is implemented; it defers
bpf_mtap processing to a dedicated softint for a target driver.
By using the machanism, we can move bpf_mtap processing to softint
without changing target drivers much while it adds some overhead
on CPU and memory. Once target drivers are changed to softint-based,
we should return to normal bpf_mtap.

Proposed on tech-kern and tech-net

doc: Update TODO.smpnet to reflect invention of ETHER_F_ALLMULTI.

fix various typos in comments and messages.

doc: more MPSAFE interfaces

Remove ixgbe because both ixg and ixv exist in the list.

update the list of ethernet drivers that have been MPSAFEified.

if_tap was made mpsafe in revision 1.102

Remove references to BRIDGE_IPF, it is now compiled in by default.

BRIDGE_IPF is MP-safe, discussed with ozaki-r@

strip(4) was removed

netsmb is no more

write down some observations about the locking wild west around if_mcast_op()

FDDI support has been removed.

Token Ring has been removed.

HIPPI has been removed.

Protecting ec_multi* of ethercom with ec_lock have been finished now.

Add aq(4) and ixl(4) to the MPSAFE component list.

Add another known issue (if_flags)

Add another known issue

Note that kernels with NET_MPSAFE won't work with non-NET_MPSAFE modules.

XXX The initial merge of the [pgoyette-compat] branch will NOT address
XXX this issue.

Drop LMC-related entry from TODO.smpnet, and note removal of LMC.

remove references to isdn

Note removal of midway and NATM, and prune the entries from TODO.smpnet.

Retire EtherIP, we have L2TP instead.

Add ipsecif(4)

Categorize non MP-safe components into protected and unprotected ones

branches: 1.20.2;
Move txintr_setup() stuff from lmc_interrupt() and do it in ifnet_start().
Now we can use bpf_mtap() in the TX path. Not tested.

branches: 1.19.2;
Remove the Econet code. It was part of acorn26, which was removed a
month ago.

Mention the limitation of enabling both NET_MPSAFE and ALTQ

Mention IPv6 address selection policy isn't MP-safe yet

Though it's not a problem until a policy is set.

Mention ec_multi*

Mention NOMPSAFE

add opencrypto(9)'s scalability comment

Write known issues on scalability

Update

ipsec(4), opencrypto(9) and vlan(4) are now MP-safe.

branches: 1.11.2; 1.11.6;
Write about bpf's statistic counters

Write about lingering obsolete variables due to kvm(3)

branches: 1.9.2;
Add ixv(4) to MP-safe component's list.

Add ixg(4)

Tidy up again

Tidy up the lists

Remove multicast stuffs from the list of Non MP-safe components

Add l2tp(4)

Add BRIDGE_IPF

Add many entries

Defer bpf_mtap in Rx interrupt context to softint

bpf_mtap of some drivers is still called in hardware interrupt context.
We want to run them in softint as well as bpf_mtap of most drivers
(see if_percpuq_softint and if_input).

To this end, bpf_mtap_softint mechanism is implemented; it defers
bpf_mtap processing to a dedicated softint for a target driver.
By using the machanism, we can move bpf_mtap processing to softint
without changing target drivers much while it adds some overhead
on CPU and memory. Once target drivers are changed to softint-based,
we should return to normal bpf_mtap.

Proposed on tech-kern and tech-net

fix various typos in comments and messages.

doc: more MPSAFE interfaces

Remove ixgbe because both ixg and ixv exist in the list.

update the list of ethernet drivers that have been MPSAFEified.

if_tap was made mpsafe in revision 1.102

Remove references to BRIDGE_IPF, it is now compiled in by default.

BRIDGE_IPF is MP-safe, discussed with ozaki-r@

strip(4) was removed

netsmb is no more

write down some observations about the locking wild west around if_mcast_op()

FDDI support has been removed.

Token Ring has been removed.

HIPPI has been removed.

Protecting ec_multi* of ethercom with ec_lock have been finished now.

Add aq(4) and ixl(4) to the MPSAFE component list.

Add another known issue (if_flags)

Add another known issue

Note that kernels with NET_MPSAFE won't work with non-NET_MPSAFE modules.

XXX The initial merge of the [pgoyette-compat] branch will NOT address
XXX this issue.

Drop LMC-related entry from TODO.smpnet, and note removal of LMC.

remove references to isdn

Note removal of midway and NATM, and prune the entries from TODO.smpnet.

Retire EtherIP, we have L2TP instead.

Add ipsecif(4)

Categorize non MP-safe components into protected and unprotected ones

branches: 1.20.2;
Move txintr_setup() stuff from lmc_interrupt() and do it in ifnet_start().
Now we can use bpf_mtap() in the TX path. Not tested.

branches: 1.19.2;
Remove the Econet code. It was part of acorn26, which was removed a
month ago.

Mention the limitation of enabling both NET_MPSAFE and ALTQ

Mention IPv6 address selection policy isn't MP-safe yet

Though it's not a problem until a policy is set.

Mention ec_multi*

Mention NOMPSAFE

add opencrypto(9)'s scalability comment

Write known issues on scalability

Update

ipsec(4), opencrypto(9) and vlan(4) are now MP-safe.

branches: 1.11.2; 1.11.6;
Write about bpf's statistic counters

Write about lingering obsolete variables due to kvm(3)

branches: 1.9.2;
Add ixv(4) to MP-safe component's list.

Add ixg(4)

Tidy up again

Tidy up the lists

Remove multicast stuffs from the list of Non MP-safe components

Add l2tp(4)

Add BRIDGE_IPF

Add many entries

Defer bpf_mtap in Rx interrupt context to softint

bpf_mtap of some drivers is still called in hardware interrupt context.
We want to run them in softint as well as bpf_mtap of most drivers
(see if_percpuq_softint and if_input).

To this end, bpf_mtap_softint mechanism is implemented; it defers
bpf_mtap processing to a dedicated softint for a target driver.
By using the machanism, we can move bpf_mtap processing to softint
without changing target drivers much while it adds some overhead
on CPU and memory. Once target drivers are changed to softint-based,
we should return to normal bpf_mtap.

Proposed on tech-kern and tech-net

doc: more MPSAFE interfaces

Remove ixgbe because both ixg and ixv exist in the list.

update the list of ethernet drivers that have been MPSAFEified.

if_tap was made mpsafe in revision 1.102

Remove references to BRIDGE_IPF, it is now compiled in by default.

BRIDGE_IPF is MP-safe, discussed with ozaki-r@

strip(4) was removed

netsmb is no more

write down some observations about the locking wild west around if_mcast_op()

FDDI support has been removed.

Token Ring has been removed.

HIPPI has been removed.

Protecting ec_multi* of ethercom with ec_lock have been finished now.

Add aq(4) and ixl(4) to the MPSAFE component list.

Add another known issue (if_flags)

Add another known issue

Note that kernels with NET_MPSAFE won't work with non-NET_MPSAFE modules.

XXX The initial merge of the [pgoyette-compat] branch will NOT address
XXX this issue.

Drop LMC-related entry from TODO.smpnet, and note removal of LMC.

remove references to isdn

Note removal of midway and NATM, and prune the entries from TODO.smpnet.

Retire EtherIP, we have L2TP instead.

Add ipsecif(4)

Categorize non MP-safe components into protected and unprotected ones

branches: 1.20.2;
Move txintr_setup() stuff from lmc_interrupt() and do it in ifnet_start().
Now we can use bpf_mtap() in the TX path. Not tested.

branches: 1.19.2;
Remove the Econet code. It was part of acorn26, which was removed a
month ago.

Mention the limitation of enabling both NET_MPSAFE and ALTQ

Mention IPv6 address selection policy isn't MP-safe yet

Though it's not a problem until a policy is set.

Mention ec_multi*

Mention NOMPSAFE

add opencrypto(9)'s scalability comment

Write known issues on scalability

Update

ipsec(4), opencrypto(9) and vlan(4) are now MP-safe.

branches: 1.11.2; 1.11.6;
Write about bpf's statistic counters

Write about lingering obsolete variables due to kvm(3)

branches: 1.9.2;
Add ixv(4) to MP-safe component's list.

Add ixg(4)

Tidy up again

Tidy up the lists

Remove multicast stuffs from the list of Non MP-safe components

Add l2tp(4)

Add BRIDGE_IPF

Add many entries

Defer bpf_mtap in Rx interrupt context to softint

bpf_mtap of some drivers is still called in hardware interrupt context.
We want to run them in softint as well as bpf_mtap of most drivers
(see if_percpuq_softint and if_input).

To this end, bpf_mtap_softint mechanism is implemented; it defers
bpf_mtap processing to a dedicated softint for a target driver.
By using the machanism, we can move bpf_mtap processing to softint
without changing target drivers much while it adds some overhead
on CPU and memory. Once target drivers are changed to softint-based,
we should return to normal bpf_mtap.

Proposed on tech-kern and tech-net

Remove ixgbe because both ixg and ixv exist in the list.

update the list of ethernet drivers that have been MPSAFEified.

if_tap was made mpsafe in revision 1.102

Remove references to BRIDGE_IPF, it is now compiled in by default.

BRIDGE_IPF is MP-safe, discussed with ozaki-r@

strip(4) was removed

netsmb is no more

write down some observations about the locking wild west around if_mcast_op()

FDDI support has been removed.

Token Ring has been removed.

HIPPI has been removed.

Protecting ec_multi* of ethercom with ec_lock have been finished now.

Add aq(4) and ixl(4) to the MPSAFE component list.

Add another known issue (if_flags)

Add another known issue

Note that kernels with NET_MPSAFE won't work with non-NET_MPSAFE modules.

XXX The initial merge of the [pgoyette-compat] branch will NOT address
XXX this issue.

Drop LMC-related entry from TODO.smpnet, and note removal of LMC.

remove references to isdn

Note removal of midway and NATM, and prune the entries from TODO.smpnet.

Retire EtherIP, we have L2TP instead.

Add ipsecif(4)

Categorize non MP-safe components into protected and unprotected ones

branches: 1.20.2;
Move txintr_setup() stuff from lmc_interrupt() and do it in ifnet_start().
Now we can use bpf_mtap() in the TX path. Not tested.

branches: 1.19.2;
Remove the Econet code. It was part of acorn26, which was removed a
month ago.

Mention the limitation of enabling both NET_MPSAFE and ALTQ

Mention IPv6 address selection policy isn't MP-safe yet

Though it's not a problem until a policy is set.

Mention ec_multi*

Mention NOMPSAFE

add opencrypto(9)'s scalability comment

Write known issues on scalability

Update

ipsec(4), opencrypto(9) and vlan(4) are now MP-safe.

branches: 1.11.2; 1.11.6;
Write about bpf's statistic counters

Write about lingering obsolete variables due to kvm(3)

branches: 1.9.2;
Add ixv(4) to MP-safe component's list.

Add ixg(4)

Tidy up again

Tidy up the lists

Remove multicast stuffs from the list of Non MP-safe components

Add l2tp(4)

Add BRIDGE_IPF

Add many entries

Defer bpf_mtap in Rx interrupt context to softint

bpf_mtap of some drivers is still called in hardware interrupt context.
We want to run them in softint as well as bpf_mtap of most drivers
(see if_percpuq_softint and if_input).

To this end, bpf_mtap_softint mechanism is implemented; it defers
bpf_mtap processing to a dedicated softint for a target driver.
By using the machanism, we can move bpf_mtap processing to softint
without changing target drivers much while it adds some overhead
on CPU and memory. Once target drivers are changed to softint-based,
we should return to normal bpf_mtap.

Proposed on tech-kern and tech-net

if_tap was made mpsafe in revision 1.102

Remove references to BRIDGE_IPF, it is now compiled in by default.

BRIDGE_IPF is MP-safe, discussed with ozaki-r@

strip(4) was removed

netsmb is no more

write down some observations about the locking wild west around if_mcast_op()

FDDI support has been removed.

Token Ring has been removed.

HIPPI has been removed.

Protecting ec_multi* of ethercom with ec_lock have been finished now.

Add aq(4) and ixl(4) to the MPSAFE component list.

Add another known issue (if_flags)

Add another known issue

Note that kernels with NET_MPSAFE won't work with non-NET_MPSAFE modules.

XXX The initial merge of the [pgoyette-compat] branch will NOT address
XXX this issue.

Drop LMC-related entry from TODO.smpnet, and note removal of LMC.

remove references to isdn

Note removal of midway and NATM, and prune the entries from TODO.smpnet.

Retire EtherIP, we have L2TP instead.

Add ipsecif(4)

Categorize non MP-safe components into protected and unprotected ones

branches: 1.20.2;
Move txintr_setup() stuff from lmc_interrupt() and do it in ifnet_start().
Now we can use bpf_mtap() in the TX path. Not tested.

branches: 1.19.2;
Remove the Econet code. It was part of acorn26, which was removed a
month ago.

Mention the limitation of enabling both NET_MPSAFE and ALTQ

Mention IPv6 address selection policy isn't MP-safe yet

Though it's not a problem until a policy is set.

Mention ec_multi*

Mention NOMPSAFE

add opencrypto(9)'s scalability comment

Write known issues on scalability

Update

ipsec(4), opencrypto(9) and vlan(4) are now MP-safe.

branches: 1.11.2; 1.11.6;
Write about bpf's statistic counters

Write about lingering obsolete variables due to kvm(3)

branches: 1.9.2;
Add ixv(4) to MP-safe component's list.

Add ixg(4)

Tidy up again

Tidy up the lists

Remove multicast stuffs from the list of Non MP-safe components

Add l2tp(4)

Add BRIDGE_IPF

Add many entries

Defer bpf_mtap in Rx interrupt context to softint

bpf_mtap of some drivers is still called in hardware interrupt context.
We want to run them in softint as well as bpf_mtap of most drivers
(see if_percpuq_softint and if_input).

To this end, bpf_mtap_softint mechanism is implemented; it defers
bpf_mtap processing to a dedicated softint for a target driver.
By using the machanism, we can move bpf_mtap processing to softint
without changing target drivers much while it adds some overhead
on CPU and memory. Once target drivers are changed to softint-based,
we should return to normal bpf_mtap.

Proposed on tech-kern and tech-net

Remove references to BRIDGE_IPF, it is now compiled in by default.

BRIDGE_IPF is MP-safe, discussed with ozaki-r@

strip(4) was removed

netsmb is no more

write down some observations about the locking wild west around if_mcast_op()

FDDI support has been removed.

Token Ring has been removed.

HIPPI has been removed.

Protecting ec_multi* of ethercom with ec_lock have been finished now.

Add aq(4) and ixl(4) to the MPSAFE component list.

Add another known issue (if_flags)

Add another known issue

Note that kernels with NET_MPSAFE won't work with non-NET_MPSAFE modules.

XXX The initial merge of the [pgoyette-compat] branch will NOT address
XXX this issue.

Drop LMC-related entry from TODO.smpnet, and note removal of LMC.

remove references to isdn

Note removal of midway and NATM, and prune the entries from TODO.smpnet.

Retire EtherIP, we have L2TP instead.

Add ipsecif(4)

Categorize non MP-safe components into protected and unprotected ones

branches: 1.20.2;
Move txintr_setup() stuff from lmc_interrupt() and do it in ifnet_start().
Now we can use bpf_mtap() in the TX path. Not tested.

branches: 1.19.2;
Remove the Econet code. It was part of acorn26, which was removed a
month ago.

Mention the limitation of enabling both NET_MPSAFE and ALTQ

Mention IPv6 address selection policy isn't MP-safe yet

Though it's not a problem until a policy is set.

Mention ec_multi*

Mention NOMPSAFE

add opencrypto(9)'s scalability comment

Write known issues on scalability

Update

ipsec(4), opencrypto(9) and vlan(4) are now MP-safe.

branches: 1.11.2; 1.11.6;
Write about bpf's statistic counters

Write about lingering obsolete variables due to kvm(3)

branches: 1.9.2;
Add ixv(4) to MP-safe component's list.

Add ixg(4)

Tidy up again

Tidy up the lists

Remove multicast stuffs from the list of Non MP-safe components

Add l2tp(4)

Add BRIDGE_IPF

Add many entries

Defer bpf_mtap in Rx interrupt context to softint

bpf_mtap of some drivers is still called in hardware interrupt context.
We want to run them in softint as well as bpf_mtap of most drivers
(see if_percpuq_softint and if_input).

To this end, bpf_mtap_softint mechanism is implemented; it defers
bpf_mtap processing to a dedicated softint for a target driver.
By using the machanism, we can move bpf_mtap processing to softint
without changing target drivers much while it adds some overhead
on CPU and memory. Once target drivers are changed to softint-based,
we should return to normal bpf_mtap.

Proposed on tech-kern and tech-net

strip(4) was removed

netsmb is no more

write down some observations about the locking wild west around if_mcast_op()

FDDI support has been removed.

Token Ring has been removed.

HIPPI has been removed.

Protecting ec_multi* of ethercom with ec_lock have been finished now.

Add aq(4) and ixl(4) to the MPSAFE component list.

Add another known issue (if_flags)

Add another known issue

Note that kernels with NET_MPSAFE won't work with non-NET_MPSAFE modules.

XXX The initial merge of the [pgoyette-compat] branch will NOT address
XXX this issue.

Drop LMC-related entry from TODO.smpnet, and note removal of LMC.

remove references to isdn

Note removal of midway and NATM, and prune the entries from TODO.smpnet.

Retire EtherIP, we have L2TP instead.

Add ipsecif(4)

Categorize non MP-safe components into protected and unprotected ones

branches: 1.20.2;
Move txintr_setup() stuff from lmc_interrupt() and do it in ifnet_start().
Now we can use bpf_mtap() in the TX path. Not tested.

branches: 1.19.2;
Remove the Econet code. It was part of acorn26, which was removed a
month ago.

Mention the limitation of enabling both NET_MPSAFE and ALTQ

Mention IPv6 address selection policy isn't MP-safe yet

Though it's not a problem until a policy is set.

Mention ec_multi*

Mention NOMPSAFE

add opencrypto(9)'s scalability comment

Write known issues on scalability

Update

ipsec(4), opencrypto(9) and vlan(4) are now MP-safe.

branches: 1.11.2; 1.11.6;
Write about bpf's statistic counters

Write about lingering obsolete variables due to kvm(3)

branches: 1.9.2;
Add ixv(4) to MP-safe component's list.

Add ixg(4)

Tidy up again

Tidy up the lists

Remove multicast stuffs from the list of Non MP-safe components

Add l2tp(4)

Add BRIDGE_IPF

Add many entries

Defer bpf_mtap in Rx interrupt context to softint

bpf_mtap of some drivers is still called in hardware interrupt context.
We want to run them in softint as well as bpf_mtap of most drivers
(see if_percpuq_softint and if_input).

To this end, bpf_mtap_softint mechanism is implemented; it defers
bpf_mtap processing to a dedicated softint for a target driver.
By using the machanism, we can move bpf_mtap processing to softint
without changing target drivers much while it adds some overhead
on CPU and memory. Once target drivers are changed to softint-based,
we should return to normal bpf_mtap.

Proposed on tech-kern and tech-net

netsmb is no more

write down some observations about the locking wild west around if_mcast_op()

FDDI support has been removed.

Token Ring has been removed.

HIPPI has been removed.

Protecting ec_multi* of ethercom with ec_lock have been finished now.

Add aq(4) and ixl(4) to the MPSAFE component list.

Add another known issue (if_flags)

Add another known issue

Note that kernels with NET_MPSAFE won't work with non-NET_MPSAFE modules.

XXX The initial merge of the [pgoyette-compat] branch will NOT address
XXX this issue.

Drop LMC-related entry from TODO.smpnet, and note removal of LMC.

remove references to isdn

Note removal of midway and NATM, and prune the entries from TODO.smpnet.

Retire EtherIP, we have L2TP instead.

Add ipsecif(4)

Categorize non MP-safe components into protected and unprotected ones

branches: 1.20.2;
Move txintr_setup() stuff from lmc_interrupt() and do it in ifnet_start().
Now we can use bpf_mtap() in the TX path. Not tested.

branches: 1.19.2;
Remove the Econet code. It was part of acorn26, which was removed a
month ago.

Mention the limitation of enabling both NET_MPSAFE and ALTQ

Mention IPv6 address selection policy isn't MP-safe yet

Though it's not a problem until a policy is set.

Mention ec_multi*

Mention NOMPSAFE

add opencrypto(9)'s scalability comment

Write known issues on scalability

Update

ipsec(4), opencrypto(9) and vlan(4) are now MP-safe.

branches: 1.11.2; 1.11.6;
Write about bpf's statistic counters

Write about lingering obsolete variables due to kvm(3)

branches: 1.9.2;
Add ixv(4) to MP-safe component's list.

Add ixg(4)

Tidy up again

Tidy up the lists

Remove multicast stuffs from the list of Non MP-safe components

Add l2tp(4)

Add BRIDGE_IPF

Add many entries

Defer bpf_mtap in Rx interrupt context to softint

bpf_mtap of some drivers is still called in hardware interrupt context.
We want to run them in softint as well as bpf_mtap of most drivers
(see if_percpuq_softint and if_input).

To this end, bpf_mtap_softint mechanism is implemented; it defers
bpf_mtap processing to a dedicated softint for a target driver.
By using the machanism, we can move bpf_mtap processing to softint
without changing target drivers much while it adds some overhead
on CPU and memory. Once target drivers are changed to softint-based,
we should return to normal bpf_mtap.

Proposed on tech-kern and tech-net

write down some observations about the locking wild west around if_mcast_op()

FDDI support has been removed.

Token Ring has been removed.

HIPPI has been removed.

Protecting ec_multi* of ethercom with ec_lock have been finished now.

Add aq(4) and ixl(4) to the MPSAFE component list.

Add another known issue (if_flags)

Add another known issue

Note that kernels with NET_MPSAFE won't work with non-NET_MPSAFE modules.

XXX The initial merge of the [pgoyette-compat] branch will NOT address
XXX this issue.

Drop LMC-related entry from TODO.smpnet, and note removal of LMC.