On 25.01.2019 16:48, Burakov, Anatoly wrote: > On 25-Jan-19 9:53 AM, David Marchand wrote: >> >> >> On Fri, Jan 25, 2019 at 9:06 AM Ilya Maximets <i.maxim...@samsung.com >> <mailto:i.maxim...@samsung.com>> wrote: >> >> On 25.01.2019 10:55, Ilya Maximets wrote: >> > Dynamic memory mode allowes zero socket-mem because all the >> > required memory could be allocated on demand. >> > >> > Fixes: 339c2244b4f1 ("eal: fix parsing zero socket memory and >> limits") >> > Cc: sta...@dpdk.org <mailto:sta...@dpdk.org> >> > >> >> Reported-by: Shuai Zhu <shuaix....@intel.com >> <mailto:shuaix....@intel.com>> >> >> > Signed-off-by: Ilya Maximets <i.maxim...@samsung.com >> <mailto:i.maxim...@samsung.com>> >> >> > --- >> > test/test/test_eal_flags.c | 6 +++--- >> > 1 file changed, 3 insertions(+), 3 deletions(-) >> > >> > diff --git a/test/test/test_eal_flags.c b/test/test/test_eal_flags.c >> > index e3a60c7ae..81e345b87 100644 >> > --- a/test/test/test_eal_flags.c >> > +++ b/test/test/test_eal_flags.c >> > @@ -1158,7 +1158,7 @@ test_memory_flags(void) >> > const char *argv1[] = {prgname, "-c", "10", "-n", "2", >> > "--file-prefix=" memtest, "-m", >> DEFAULT_MEM_SIZE}; >> > >> > - /* invalid (zero) --socket-mem flag */ >> > + /* valid (zero) --socket-mem flag */ >> > const char *argv2[] = {prgname, "-c", "10", "-n", "2", >> > "--file-prefix=" memtest, >> "--socket-mem=0,0,0,0"}; >> > >> > @@ -1256,8 +1256,8 @@ test_memory_flags(void) >> > printf("Error - process failed with valid -m flag!\n"); >> > return -1; >> > } >> > - if (launch_proc(argv2) == 0) { >> > - printf("Error - process run ok with invalid (zero) >> --socket-mem!\n"); >> > + if (launch_proc(argv2) != 0) { >> > + printf("Error - process failed with valid (zero) >> --socket-mem!\n"); >> > return -1; >> > } >> > >> > >> >> >> Reviewed-by: David Marchand <david.march...@redhat.com >> <mailto:david.march...@redhat.com>> >> >> >> -- >> David Marchand > > Now that i think of it, maybe it's not that simple. > > --socket-mem/-m flag with zero is still an invalid value *if* --legacy-mem is > involved. However, it is a valid value in non-legacy mode. > > So maybe the test should reflect this, and the previous fix should have > instead added a check for legacy mode rather than disabling the zero check > outright. >
I don't think that it's a big deal, because "--socket-mem=0 --legacy-mem" quickly fails with clear: EAL: WARNING: Master core has no memory on local socket! IMHO, It's actually more informative than previous: EAL: invalid parameters for --socket-limit I agree that we could add a test for a legacy-mem cases, but that's a bit different task.
