| History log of /src/sys/net/bpfjit.c |
| Revision | | Date | Author | Comments |
| 1.48 |
| 01-Feb-2020 |
riastradh | Fix wrong memory order and switch bpf to atomic_load/store_*.
|
| 1.47 |
| 20-Jan-2019 |
alnsn | branches: 1.47.6;
Add missing include.
|
| 1.46 |
| 29-Jul-2016 |
alnsn | branches: 1.46.14; 1.46.16;
Don't trigger BJ_ASSERT(false) on invalid BPF_Jxxx opcode in jmp_to_op().
This change helps survive AFL fuzzing without calling bpf_validate() first.
Also change alu_to_op() function to have a similar interface.
|
| 1.45 |
| 29-May-2016 |
alnsn | branches: 1.45.2;
Adapt to the new version of sljit@r313.
|
| 1.44 |
| 29-Dec-2015 |
alnsn | Replace the nsaveds() function with #define NSAVEDS 3. No functional change.
Patch from Michael McConville.
|
| 1.43 |
| 14-Feb-2015 |
alnsn | Copyright year.
|
| 1.42 |
| 14-Feb-2015 |
alnsn | In some implementations pc->k is signed. Cast it to uint32_t before comparing.
|
| 1.41 |
| 14-Feb-2015 |
alnsn | Properly track initialisation of registers for BPF_JMP instructions.
|
| 1.40 |
| 13-Feb-2015 |
alnsn | Don't emit wrapped-around reads. They're dead code but dead code elimination
logic isn't smart enough to figure it out.
Found by afl fuzzer http://lcamtuf.coredump.cx/afl/.
|
| 1.39 |
| 12-Feb-2015 |
alnsn | Fix bugs found by afl fuzzer http://lcamtuf.coredump.cx/afl/.
|
| 1.38 |
| 15-Jan-2015 |
christos | rename variable to avoid conflict with "div"
|
| 1.37 |
| 08-Dec-2014 |
justin | Help gcc by initialising variable
|
| 1.36 |
| 20-Nov-2014 |
alnsn | branches: 1.36.2;
Implement BPF_MOD.
|
| 1.35 |
| 20-Nov-2014 |
alnsn | Implement BPF_ALU+BPF_MOD-BPF_K when pc->k is a power of 2. Get rid of divt
and divw arguments in emit_moddiv(), they're accessible via the pc argument.
|
| 1.34 |
| 20-Nov-2014 |
alnsn | Follow argument convension of other emit_xxx() functions.
|
| 1.33 |
| 19-Nov-2014 |
christos | Add BPF_MOD/BPF_XOR (untested, needs work)
|
| 1.32 |
| 26-Jul-2014 |
alnsn | branches: 1.32.2;
Don't use saved EREG registers because sljit 0.91 can generate
bogus code on amd64. The A and X registers are saved on the stack.
The most recent version of sljit fixes bogus code generation but
it's not backward compatible with sljit 0.91.
|
| 1.31 |
| 24-Jul-2014 |
alnsn | For P[X+0] load, don't emit wrap around check and copy X intead of emitting X+0.
|
| 1.30 |
| 22-Jul-2014 |
alnsn | Two tweaks: don't use a temporary register to dereference the err agrument
after xcall and don't generate ((tmp1 & 0xf) << 2) twice in emit_msh().
|
| 1.29 |
| 22-Jul-2014 |
alnsn | Don't use scratch registers for X and to restore A after BPF_COPX call.
|
| 1.28 |
| 13-Jul-2014 |
alnsn | Refactor BPF_COPX code. New version doesn't load buf and buflen after copx call.
|
| 1.27 |
| 13-Jul-2014 |
alnsn | Don't use BJ_TMP2REG for 32bit packet reads. Assign this register to (buf+X)
in BPF_LD+BPF_IND and save one instruction.
|
| 1.26 |
| 12-Jul-2014 |
alnsn | emit_xcall: check overflow by comparing X with (UINT32_MAX - pk->k), restore
the A register after checking that xcall succeeded.
|
| 1.25 |
| 12-Jul-2014 |
alnsn | Initialise status to avoid -Wuninitialized warning.
|
| 1.24 |
| 12-Jul-2014 |
alnsn | Some small changes: add missing error checks; move sjump initialisation away
from optimize(); +BJ_HINT_PKT, -BJ_HINT_IND; tweak comments.
|
| 1.23 |
| 11-Jul-2014 |
alnsn | Handle overflow in BPF_LD+BPF_IND for mbuf chains and make two minor changes:
move sljit_emit_return() to generate_insn_code() and use a different register
for checking errors after xcall.
|
| 1.22 |
| 08-Jul-2014 |
alnsn | Most filter programs in the kernel need 3 scratch registers.
|
| 1.21 |
| 05-Jul-2014 |
alnsn | Review some SLJIT_MOV instructions with respect to width.
|
| 1.20 |
| 04-Jul-2014 |
alnsn | Add optimization hints. They replace nscratches and ncopfuncs and improve
readability.
|
| 1.19 |
| 01-Jul-2014 |
alnsn | Move the main loop in bpfjit_generate_code() to a new function and make few
small changes.
|
| 1.18 |
| 25-Jun-2014 |
alnsn | Default initialize external memwords.
This change doesn't affect performance of valid bpf kernel programs
because bpf_filter_ext() checks that all memwords are initialized
explicitly.
|
| 1.17 |
| 25-Jun-2014 |
alnsn | New jitcode takes two arguments.
|
| 1.16 |
| 25-Jun-2014 |
alnsn | Use SLJIT_MOV_P to copy extmem pointer.
|
| 1.15 |
| 25-Jun-2014 |
rmind | bpfjit_generate_code: emit the instruction correctly.
|
| 1.14 |
| 24-Jun-2014 |
rmind | - Improve the comments in bpf.h and KNF a little.
- Rename bpf_ctx_t member noinit to preinited (reflects the meaning better).
|
| 1.13 |
| 24-Jun-2014 |
alnsn | Implement copfuncs and external memory in bpfjit.
|
| 1.12 |
| 17-Jun-2014 |
alnsn | Update code to the latest sljit version.
|
| 1.11 |
| 23-May-2014 |
alnsn | Enable ABC optimization when one branch returns 0.
|
| 1.10 |
| 23-May-2014 |
alnsn | Loads at offsets UINT32_MAX or greater are unreachable.
|
| 1.9 |
| 23-May-2014 |
alnsn | Implement unconditional jump to "return 0" for loads at UINT32_MAX+1 or greater.
|
| 1.8 |
| 22-May-2014 |
alnsn | Some small changes in preparation for a bigger change.
- typedef for ABC variables and MAX_ABC_LENGTH constant,
- cast pc->k to uint32_t in more places,
- whitespaces.
|
| 1.7 |
| 15-May-2014 |
alnsn | Refactor bpfjit code.
- Implement Array Bounds Check Elimination for packet bytes.
- Track initialization of registers and memwords.
- Remove "bj_" prefix from struct members.
- Shorten "BPFJIT_" prefix to "BJ_".
- Other small improvements.
|
| 1.6 |
| 15-Dec-2013 |
pooka | branches: 1.6.2;
XXXgcc Wuninitialized kludge
|
| 1.5 |
| 15-Nov-2013 |
rmind | Fix the bpfjit build.
|
| 1.4 |
| 15-Nov-2013 |
rmind | - Add bpf_args_t and convert bpf_filter_ext() to use it. This allows the
caller to initialise (and re-use) the memory store.
- Add bpf_jit_generate() and bpf_jit_freecode() wrappers.
|
| 1.3 |
| 20-Sep-2013 |
rmind | bpfjit: replace malloc with kmem, KNF a little, add RCS ID.
|
| 1.2 |
| 10-Nov-2012 |
alnsn | branches: 1.2.2; 1.2.4;
Add RCSID and fix -Wsign-compare warnings.
|
| 1.1 |
| 27-Oct-2012 |
alnsn | branches: 1.1.2;
Add bpfjit and enable it for amd64.
|
| 1.1.2.4 |
| 22-May-2014 |
yamt | sync with head.
for a reference, the tree before this commit was tagged
as yamt-pagecache-tag8.
this commit was splitted into small chunks to avoid
a limitation of cvs. ("Protocol error: too many arguments")
|
| 1.1.2.3 |
| 16-Jan-2013 |
yamt | sync with (a bit old) head
|
| 1.1.2.2 |
| 30-Oct-2012 |
yamt | sync with head
|
| 1.1.2.1 |
| 27-Oct-2012 |
yamt | file bpfjit.c was added on branch yamt-pagecache on 2012-10-30 17:22:42 +0000
|
| 1.2.4.1 |
| 18-May-2014 |
rmind | sync with head
|
| 1.2.2.4 |
| 03-Dec-2017 |
jdolecek | update from HEAD
|
| 1.2.2.3 |
| 20-Aug-2014 |
tls | Rebase to HEAD as of a few days ago.
|
| 1.2.2.2 |
| 20-Nov-2012 |
tls | Resync to 2012-11-19 00:00:00 UTC
|
| 1.2.2.1 |
| 10-Nov-2012 |
tls | file bpfjit.c was added on branch tls-maxphys on 2012-11-20 03:02:46 +0000
|
| 1.6.2.1 |
| 10-Aug-2014 |
tls | Rebase.
|
| 1.32.2.1 |
| 16-Feb-2015 |
martin | Pull up following revision(s) (requested by alnsn in ticket #519):
sys/net/bpfjit.c: revision 1.39-1.41
Fix bugs found by afl fuzzer http://lcamtuf.coredump.cx/afl/.
-
Don't emit wrapped-around reads. They're dead code but dead code elimination
logic isn't smart enough to figure it out.
-
Properly track initialisation of registers for BPF_JMP instructions.
|
| 1.36.2.4 |
| 05-Oct-2016 |
skrll | Sync with HEAD
|
| 1.36.2.3 |
| 09-Jul-2016 |
skrll | Sync with HEAD
|
| 1.36.2.2 |
| 19-Mar-2016 |
skrll | Sync with HEAD
|
| 1.36.2.1 |
| 06-Apr-2015 |
skrll | Sync with HEAD
|
| 1.45.2.1 |
| 06-Aug-2016 |
pgoyette | Sync with HEAD
|
| 1.46.16.2 |
| 08-Apr-2020 |
martin | Merge changes from current as of 20200406
|
| 1.46.16.1 |
| 10-Jun-2019 |
christos | Sync with HEAD
|
| 1.46.14.1 |
| 26-Jan-2019 |
pgoyette | Sync with HEAD
|
| 1.47.6.1 |
| 29-Feb-2020 |
ad | Sync with head.
|
