Page MenuHomeFreeBSD

vmaffione (V Maffione)
User

Projects

User Details

User Since
Mar 10 2018, 1:54 AM (150 w, 2 d)

Recent Activity

Yesterday

vmaffione committed R10:ee0005f11f2b: netmap: simplify parameter passing (authored by vmaffione).
netmap: simplify parameter passing
Sun, Jan 24, 10:00 PM
vmaffione committed R10:f80efe5016ba: iflib: netmap: move per-packet operation out of fragments loop (authored by vmaffione).
iflib: netmap: move per-packet operation out of fragments loop
Sun, Jan 24, 9:42 PM
vmaffione committed R10:aceaccab659c: iflib: netmap: add support for NS_MOREFRAG (authored by vmaffione).
iflib: netmap: add support for NS_MOREFRAG
Sun, Jan 24, 9:22 PM
vmaffione closed D27799: Iflib changes to handle multiple netmap fragments.
Sun, Jan 24, 9:22 PM
vmaffione accepted D27799: Iflib changes to handle multiple netmap fragments.

Looks ok, thanks.

Sun, Jan 24, 9:22 PM
vmaffione committed R10:08f7954e1c9f: iflib: add assert to prevent out-of-bounds array access (authored by vmaffione).
iflib: add assert to prevent out-of-bounds array access
Sun, Jan 24, 10:09 AM
vmaffione added a comment to D28266: Fix for direct attach cable and workaround for intermittent link issue with axgbe.

It can be merged there (MFC'd) as well, after a period of a minimum of 3 days. There is no need to submit a separate patch.

Sun, Jan 24, 8:07 AM

Sat, Jan 23

vmaffione committed R10:bfd75d455719: axgbe: fix some link related issues (authored by vmaffione).
axgbe: fix some link related issues
Sat, Jan 23, 1:52 PM
vmaffione closed D28266: Fix for direct attach cable and workaround for intermittent link issue with axgbe.
Sat, Jan 23, 1:52 PM
vmaffione added a comment to D28266: Fix for direct attach cable and workaround for intermittent link issue with axgbe.

Will the fix be provided as a firmware upgrade or how?
I could commit this change if you don't have someone else around that can do that for you.

Sat, Jan 23, 10:14 AM

Thu, Jan 21

vmaffione added a comment to D28266: Fix for direct attach cable and workaround for intermittent link issue with axgbe.

Wouldn't it be easier to always perform the complete phy reset if this is a known hardware issue? Or maye perform it automatically after a normal reset does not work?
I'm afraid that adding an obscure sysctl (obscure for anyone except for the axgbe developers) would be of little use. Unaware users wouldn't have a clue about what to do...
just my 2 cents

Thu, Jan 21, 8:44 PM

Sun, Jan 17

vmaffione committed R10:45b2c3bec4f2: re: netmap: enable/disable krings on interface reinit (authored by vmaffione).
re: netmap: enable/disable krings on interface reinit
Sun, Jan 17, 1:41 PM
vmaffione committed R10:fa1a4ff944f4: netmap: vtnet: stop krings during interface reset (authored by vmaffione).
netmap: vtnet: stop krings during interface reset
Sun, Jan 17, 1:40 PM
vmaffione committed R10:9882b8313200: netmap: iflib: stop krings during interface reset (authored by vmaffione).
netmap: iflib: stop krings during interface reset
Sun, Jan 17, 1:35 PM

Sat, Jan 16

vmaffione committed R10:2968dde3de6f: axgbe: driver changes for netmap support (authored by vmaffione).
axgbe: driver changes for netmap support
Sat, Jan 16, 8:30 AM
vmaffione closed D27797: AMD 10GbE driver changes for Netmap support.
Sat, Jan 16, 8:30 AM

Fri, Jan 15

vmaffione committed R10:b791b5763f2b: netmap: pkt-gen: fix offset hex formatting (authored by vmaffione).
netmap: pkt-gen: fix offset hex formatting
Fri, Jan 15, 8:59 PM
vmaffione accepted D27797: AMD 10GbE driver changes for Netmap support.

Shall I commit this change?

Fri, Jan 15, 8:48 PM

Thu, Jan 14

vmaffione accepted D27797: AMD 10GbE driver changes for Netmap support.

Looks good to me.

Thu, Jan 14, 9:40 PM

Wed, Jan 13

vmaffione requested changes to D27797: AMD 10GbE driver changes for Netmap support.

I added some comments, mostly about readability.
If you tested all the valid combinations of sph_enable and netmap/non-netmap I don't have any objection in this being merged.

Wed, Jan 13, 8:29 PM

Tue, Jan 12

vmaffione committed R10:d7493759fbfb: netmap: pkt-gen: fix offset hex formatting (authored by vmaffione).
netmap: pkt-gen: fix offset hex formatting
Tue, Jan 12, 10:06 PM

Mon, Jan 11

vmaffione committed R10:3005e10ddbfb: netmap: vtnet: fix RX initialization after netmap_reset() (authored by vmaffione).
netmap: vtnet: fix RX initialization after netmap_reset()
Mon, Jan 11, 9:43 PM

Sun, Jan 10

vmaffione committed R10:55f0ad5fdee1: netmap: restore hwofs and support it in iflib (authored by vmaffione).
netmap: restore hwofs and support it in iflib
Sun, Jan 10, 10:57 PM
vmaffione committed R10:54bbcca4f979: re: netmap: enable/disable krings on interface reinit (authored by vmaffione).
re: netmap: enable/disable krings on interface reinit
Sun, Jan 10, 3:13 PM
vmaffione committed R10:8aa8484cbf06: iflib: fix build failure in case DEV_NETMAP is not defined (authored by vmaffione).
iflib: fix build failure in case DEV_NETMAP is not defined
Sun, Jan 10, 2:46 PM
vmaffione added a comment to D27797: AMD 10GbE driver changes for Netmap support.

Il giorno dom 10 gen 2021 alle ore 13:20 rajesh1.kumar_amd.com (Rajesh
Kumar) <phabric-noreply@freebsd.org> ha scritto:

Sun, Jan 10, 2:30 PM
vmaffione committed R10:bb714db6d395: netmap: vtnet: enable/disable krings on any interface reinit (authored by vmaffione).
netmap: vtnet: enable/disable krings on any interface reinit
Sun, Jan 10, 2:11 PM
vmaffione committed R10:4ba9ad0dc316: iflib: add assert to prevent out-of-bounds array access (authored by vmaffione).
iflib: add assert to prevent out-of-bounds array access
Sun, Jan 10, 2:00 PM
vmaffione committed R10:3d65fd97e85a: netmap: iflib: enable/disable krings on any interface reinit (authored by vmaffione).
netmap: iflib: enable/disable krings on any interface reinit
Sun, Jan 10, 12:06 PM
vmaffione committed R10:af352cf1397f: netmap: bridge: fix NS_MOREFRAG support (authored by vmaffione).
netmap: bridge: fix NS_MOREFRAG support
Sun, Jan 10, 9:11 AM

Sat, Jan 9

vmaffione committed R10:9ac59d42c0b4: netmap: vtnet: stop krings during interface reset (authored by vmaffione).
netmap: vtnet: stop krings during interface reset
Sat, Jan 9, 10:57 PM
vmaffione committed R10:7ba6ecf216fb: netmap: refactor netmap_reset (authored by vmaffione).
netmap: refactor netmap_reset
Sat, Jan 9, 10:18 PM
vmaffione committed R10:3189ba61673a: netmap: iflib: fix asserts in netmap_fl_refill() (authored by vmaffione).
netmap: iflib: fix asserts in netmap_fl_refill()
Sat, Jan 9, 9:43 PM
vmaffione committed R10:1d238b07d5d4: netmap: iflib: stop krings during interface reset (authored by vmaffione).
netmap: iflib: stop krings during interface reset
Sat, Jan 9, 9:02 PM

Fri, Jan 8

vmaffione added a comment to D27799: Iflib changes to handle multiple netmap fragments.

Thanks to both for the clarification.

Fri, Jan 8, 8:58 PM
vmaffione resigned from D27799: Iflib changes to handle multiple netmap fragments.
Fri, Jan 8, 8:56 PM
vmaffione added a comment to D27877: Support for NS_MOREFRAG in Netmap bridge utility.

Il giorno ven 8 gen 2021 alle ore 10:37 rajesh1.kumar_amd.com (Rajesh
Kumar) <phabric-noreply@freebsd.org> ha scritto:

Fri, Jan 8, 8:56 PM
vmaffione added a comment to D27797: AMD 10GbE driver changes for Netmap support.

Are both netmap_test and sph_enable actually necessary? Would it make sense to have only sph_enable, and set isc_nfl = 1 when sph_enable = 0 and isc_nfl = 2 when sph_enable = 1?
If not, I think netmap_test should be called something like single_free_list

I agree, both sysctl variables now depend on each other. I will create a change and submit it for review here once I find the time.

@stephan.dewt_yahoo.co.uk, I have already made the changes for this and about to submit it.

@vmaffione, I feel it's better to retain both the variables, since it will give some flexibility to the user to try disabling split header even with non-netmap cases if needed. But as you suggested, I have modified the "netmap_test" variable to "single_fl".

But if you people see no usage for the same, we shall remov it.

Fri, Jan 8, 4:47 PM

Thu, Jan 7

vmaffione added a comment to D27797: AMD 10GbE driver changes for Netmap support.

Are both netmap_test and sph_enable actually necessary? Would it make sense to have only sph_enable, and set isc_nfl = 1 when sph_enable = 0 and isc_nfl = 2 when sph_enable = 1?
If not, I think netmap_test should be called something like single_free_list

Thu, Jan 7, 9:04 AM
vmaffione added a comment to D27799: Iflib changes to handle multiple netmap fragments.

Il giorno mer 6 gen 2021 alle ore 20:16 rajesh1.kumar_amd.com (Rajesh
Kumar) <phabric-noreply@freebsd.org> ha scritto:

Thu, Jan 7, 8:25 AM
vmaffione added a comment to D27877: Support for NS_MOREFRAG in Netmap bridge utility.

Il giorno mer 6 gen 2021 alle ore 20:01 rajesh1.kumar_amd.com (Rajesh
Kumar) <phabric-noreply@freebsd.org> ha scritto:

Thu, Jan 7, 7:17 AM
vmaffione closed D27980: netmap: bridge: fix bridge support for NS_MOREFRAG.
Thu, Jan 7, 7:07 AM
vmaffione committed R10:163f4f15733f: netmap: bridge: fix NS_MOREFRAG support (authored by vmaffione).
netmap: bridge: fix NS_MOREFRAG support
Thu, Jan 7, 7:07 AM

Wed, Jan 6

vmaffione added a comment to D27980: netmap: bridge: fix bridge support for NS_MOREFRAG.

@rajesh1.kumar_amd.com
Thanks for testing, I'll push this right away.

Wed, Jan 6, 8:15 PM

Tue, Jan 5

vmaffione added a comment to D27877: Support for NS_MOREFRAG in Netmap bridge utility.

Good catch, thanks.
I think the existing line 113 (copying the NS_MOREFRAG flag) was never tested, because it copies the flag the other way around.

Tue, Jan 5, 5:15 PM
vmaffione requested review of D27980: netmap: bridge: fix bridge support for NS_MOREFRAG.
Tue, Jan 5, 5:09 PM
vmaffione added a comment to D27799: Iflib changes to handle multiple netmap fragments.

Il giorno mar 5 gen 2021 alle ore 07:55 rajesh1.kumar_amd.com (Rajesh
Kumar) <phabric-noreply@freebsd.org> ha scritto:

Tue, Jan 5, 4:34 PM

Sat, Jan 2

vmaffione accepted D27799: Iflib changes to handle multiple netmap fragments.

Thanks.
I tested it on my side on em(3) and vmx(3) in a VM to look for regressions.
What kind of tests have you done (I guess you are using axgbe)?
Have you tried the pkt-gen -F and -M options on the transmitter side (pkt-gen supports multifrag), and see if a pkt-gen receiver receives the correct traffic?
Do you want me to commit this change?

Sat, Jan 2, 10:32 AM

Tue, Dec 29

vmaffione accepted D27814: [vale] Fix valectl to compile on a 32-bit platform.

LGTM
Thanks!

Tue, Dec 29, 8:58 PM

Mon, Dec 28

vmaffione requested changes to D27799: Iflib changes to handle multiple netmap fragments.
Mon, Dec 28, 9:36 PM

Dec 3 2020

vmaffione added a comment to D27381: [bhyve] virtio-net: Do not allow receiving packets until features have been negotiated..

I'd say let's go step by step.
First we should merge this patch, which makes sense as is (unless @grehan has some more observations)?
Then we can talk about the other issue in a separate review.

Dec 3 2020, 9:43 PM

Nov 29 2020

vmaffione accepted D27381: [bhyve] virtio-net: Do not allow receiving packets until features have been negotiated..
Nov 29 2020, 7:54 PM

Nov 28 2020

vmaffione added a comment to D27381: [bhyve] virtio-net: Do not allow receiving packets until features have been negotiated..

I agree with @grehan , but see my comments.

Nov 28 2020, 11:35 AM

Nov 26 2020

vmaffione accepted D27382: [ng_socket] Don't take the SOCKBUF_LOCK() twice in the RX data path..

Not familiar with the code, but it looks ok to me.

Nov 26 2020, 6:31 PM

Nov 25 2020

vmaffione committed rS368037: MFC r367936.
MFC r367936
Nov 25 2020, 9:25 PM
vmaffione committed rS368036: MFC r367935.
MFC r367935
Nov 25 2020, 9:25 PM
vmaffione committed rS368035: MFC r367934.
MFC r367934
Nov 25 2020, 9:24 PM
vmaffione committed rS368034: MFC r367933.
MFC r367933
Nov 25 2020, 9:23 PM
vmaffione committed rS368033: MFC r367932.
MFC r367932
Nov 25 2020, 9:22 PM
vmaffione committed rS368032: MFC r367931.
MFC r367931
Nov 25 2020, 9:21 PM

Nov 24 2020

vmaffione committed rS367996: MFC r367920, r367921.
MFC r367920, r367921
Nov 24 2020, 7:55 PM

Nov 22 2020

vmaffione committed rS367936: netmap: bridge: improve readability.
netmap: bridge: improve readability
Nov 22 2020, 1:40 PM
vmaffione committed rS367935: netmap: valectl: switch to libnetmap.
netmap: valectl: switch to libnetmap
Nov 22 2020, 10:03 AM
vmaffione committed rS367934: netmap: nmreplay: switch to libnetmap.
netmap: nmreplay: switch to libnetmap
Nov 22 2020, 9:38 AM
vmaffione committed rS367933: netmap: lb: switch to libnetmap.
netmap: lb: switch to libnetmap
Nov 22 2020, 9:29 AM
vmaffione committed rS367932: netmap: pkt-gen: switch to libnetmap.
netmap: pkt-gen: switch to libnetmap
Nov 22 2020, 9:20 AM
vmaffione committed rS367931: netmap: bridge: switch to libnetmap.
netmap: bridge: switch to libnetmap
Nov 22 2020, 9:10 AM

Nov 21 2020

vmaffione committed rS367921: netmap: bridge, pkt-gen: fix man page typo.
netmap: bridge, pkt-gen: fix man page typo
Nov 21 2020, 6:47 PM
vmaffione committed rS367920: netmap: bridge: update man page with more information.
netmap: bridge: update man page with more information
Nov 21 2020, 6:21 PM

Nov 11 2020

vmaffione committed rS367599: MFC r367093, r367117.
MFC r367093, r367117
Nov 11 2020, 9:27 PM

Oct 28 2020

vmaffione committed rS367117: iflib: fix typo bug introduced by r367093.
iflib: fix typo bug introduced by r367093
Oct 28 2020, 9:07 PM

Oct 27 2020

vmaffione committed rS367093: iflib: add per-tx-queue netmap timer.
iflib: add per-tx-queue netmap timer
Oct 27 2020, 9:53 PM

Oct 25 2020

vmaffione committed rS367036: MFC r366952.
MFC r366952
Oct 25 2020, 7:48 AM

Oct 22 2020

vmaffione committed rS366952: netmap: fix mutex double unlock bug.
netmap: fix mutex double unlock bug
Oct 22 2020, 8:21 PM
vmaffione added a comment to D26896: ixl: honor IPI_TX_INTR.

Although the change seems legit, it has been reported to cause hangs with netmap pkt-gen.
See:
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=248652 (from comment 29)
Any idea why this may be the case?
In stable/11, the same logic (two report flags set for every ring) does not cause hangs.

Oct 22 2020, 8:11 PM

Oct 21 2020

vmaffione requested review of D26896: ixl: honor IPI_TX_INTR.
Oct 21 2020, 8:14 PM

Oct 6 2020

vmaffione committed rS366499: MFC r366394.
MFC r366394
Oct 6 2020, 7:15 PM
vmaffione committed rS366498: MFC r366393.
MFC r366393
Oct 6 2020, 7:14 PM
vmaffione committed rS366497: MFC r366389.
MFC r366389
Oct 6 2020, 7:13 PM
vmaffione committed rS366496: MFC r366388.
MFC r366388
Oct 6 2020, 7:12 PM
vmaffione committed rS366495: MFC r366387.
MFC r366387
Oct 6 2020, 7:11 PM

Oct 3 2020

vmaffione committed rS366394: netmap: tools: extend CFLAGS after including bsd.prog.mk.
netmap: tools: extend CFLAGS after including bsd.prog.mk
Oct 3 2020, 1:27 PM
vmaffione committed rS366393: netmap: tools: fix several compiler warnings.
netmap: tools: fix several compiler warnings
Oct 3 2020, 1:20 PM
vmaffione committed rS366389: netmap: minor documentation fix.
netmap: minor documentation fix
Oct 3 2020, 9:36 AM
vmaffione committed rS366388: netmap: fix constness warnings generated by "-Wcast-qual".
netmap: fix constness warnings generated by "-Wcast-qual"
Oct 3 2020, 9:33 AM
vmaffione committed rS366387: netmap: pkt-gen: minor corrections to documentation.
netmap: pkt-gen: minor corrections to documentation
Oct 3 2020, 9:23 AM

Sep 24 2020

vmaffione committed rS366130: MFC r366011.
MFC r366011
Sep 24 2020, 8:01 PM

Sep 22 2020

vmaffione committed rS366014: MFC r365061.
MFC r365061
Sep 22 2020, 9:44 PM
vmaffione committed rS366013: MFC r364936, r365023.
MFC r364936, r365023
Sep 22 2020, 9:16 PM
vmaffione committed rS366011: libnetmap: fix cast from uint64_t to void*.
libnetmap: fix cast from uint64_t to void*
Sep 22 2020, 8:20 PM

Sep 1 2020

vmaffione committed rS365066: r364770.
r364770
Sep 1 2020, 8:58 PM
vmaffione committed rS365065: MFC r364655.
MFC r364655
Sep 1 2020, 8:50 PM
vmaffione committed rS365064: MFC r364165.
MFC r364165
Sep 1 2020, 8:48 PM
vmaffione committed rS365063: MFC r364164.
MFC r364164
Sep 1 2020, 8:45 PM
vmaffione committed rS365061: iflib: leave only 1 receive descriptor unused.
iflib: leave only 1 receive descriptor unused
Sep 1 2020, 8:42 PM
vmaffione closed D26191: iflib: leave only 1 receive descriptor unused.
Sep 1 2020, 8:42 PM
vmaffione committed rS365023: lib: libnetmap: add missing copyright headers.
lib: libnetmap: add missing copyright headers
Sep 1 2020, 6:06 AM

Aug 31 2020

vmaffione updated the diff for D26191: iflib: leave only 1 receive descriptor unused.

Load the correct patch.

Aug 31 2020, 8:33 PM
vmaffione added inline comments to D26191: iflib: leave only 1 receive descriptor unused.
Aug 31 2020, 8:27 PM
vmaffione updated the diff for D26191: iflib: leave only 1 receive descriptor unused.

Added assert in iflib_fl_refill().
Improved comment.
Removed assert() at first refill.

Aug 31 2020, 8:27 PM