search for: send_fast_zero

...st-eflags.sh b/tests/test-eflags.sh index f5cd43ed..9b3a6a3a 100755 --- a/tests/test-eflags.sh +++ b/tests/test-eflags.sh @@ -68,6 +68,7 @@ SEND_DF=$(( 1 << 7 )) CAN_MULTI_CONN=$(( 1 << 8 )) SEND_RESIZE=$(( 1 << 9 )) SEND_CACHE=$(( 1 << 10 )) +SEND_FAST_ZERO=$(( 1 << 11 )) do_nbdkit () { @@ -133,8 +134,8 @@ EOF #---------------------------------------------------------------------- # can_write=true # -# NBD_FLAG_SEND_WRITE_ZEROES is set on writable connections -# even when can_zero returns false, because nbdkit reckons it +# NBD_FLAG_SEN...

This is a cover letter to a series of patches being proposed in tandem to four different projects: - nbd: Document a new NBD_CMD_FLAG_FAST_ZERO command flag - qemu: Implement the flag for both clients and server - libnbd: Implement the flag for clients - nbdkit: Implement the flag for servers, including the nbd passthrough client If you want to test the patches together, I've pushed a

...# modification, are permitted provided that the following conditions are @@ -50,7 +50,7 @@ if ! qemu-nbd --help | grep -sq -- --list; then exit 77 fi -files="eflags.out" +files="eflags.out eflags.err" late_args= rm -f $files cleanup_fn rm -f $files @@ -72,12 +72,29 @@ SEND_FAST_ZERO=$(( 1 << 11 )) do_nbdkit () { - nbdkit -v -U - "$@" sh - $late_args --run 'qemu-nbd --list -k $unixsocket' | - grep -E "flags: 0x" | grep -Eoi '0x[a-f0-9]+' > eflags.out + # Prepend a check for internal caching to the script on stdin....

When I added .can_FOO caching in 1.16, I missed the case that the sh plugin itself was calling .can_flush twice in some situations (in order to default .can_fua). Then right after, I regressed it to call .can_zero twice (in order to default .can_fast_zero). I also missed that .thread_model could use better caching, because at the time, I did not add testsuite coverage. Fix that now. Eric Blake