Re: [dpdk-stable] [PATCH] app/test-compress-perf: fix reliance on integer endianness

[Shally Verma <shallyv@marvell.com>]

> -----Original Message-----
> From: Trahe, Fiona <fiona.trahe@intel.com>
> Sent: Wednesday, May 29, 2019 7:17 PM
> To: Jozwiak, TomaszX <tomaszx.jozwiak@intel.com>; Shally Verma
> <shallyv@marvell.com>; dev@dpdk.org; stable@dpdk.org
> Cc: Trybula, ArturX <arturx.trybula@intel.com>; Trahe, Fiona
> <fiona.trahe@intel.com>
> Subject: [EXT] RE: [PATCH] app/test-compress-perf: fix reliance on integer
> endianness
> 
> External Email
> 
> ----------------------------------------------------------------------
> Hi Shally, Tomasz,
> 
> > > > > > -----Original Message-----
> > > > > > From: Jozwiak, TomaszX
> > > > > > Sent: Monday, May 20, 2019 2:26 PM
> > > > > > To: dev@dpdk.org; Trahe, Fiona <fiona.trahe@intel.com>;
> > > > > > Jozwiak, TomaszX <tomaszx.jozwiak@intel.com>;
> > > > > > shallyv@marvell.com; stable@dpdk.org
> > > > > > Subject: [PATCH] app/test-compress-perf: fix reliance on
> > > > > > integer endianness
> > > > > >
> > > > > > This patch fixes coverity issue:
> > > > > > Reliance on integer endianness (INCOMPATIBLE_CAST) in
> > > > > parse_window_sz
> > > > > > function.
> > > > > >
> > > > > > Coverity issue: 328524
> > > > > > Fixes: e0b6287c035d ("app/compress-perf: add parser")
> > > > > > Cc: stable@dpdk.org
> > > > > >
> > > > > > Signed-off-by: Tomasz Jozwiak <tomaszx.jozwiak@intel.com>
> > > > > > ---
> > > > > >  app/test-compress-perf/comp_perf_options_parse.c | 4 +++-
> > > > > >  1 file changed, 3 insertions(+), 1 deletion(-)
> > > > > >
> > > > > > diff --git a/app/test-compress-perf/comp_perf_options_parse.c
> > > > > > b/app/test-compress- perf/comp_perf_options_parse.c index
> > > > > > 2fb6fb4..56ca580 100644
> > > > > > --- a/app/test-compress-perf/comp_perf_options_parse.c
> > > > > > +++ b/app/test-compress-perf/comp_perf_options_parse.c
> > > > > > @@ -364,13 +364,15 @@ parse_max_num_sgl_segs(struct
> > > > > comp_test_data
> > > > > > *test_data, const char *arg)  static int
> > > > > > parse_window_sz(struct comp_test_data *test_data, const char *arg)
> {
> > > > > > -	int ret = parse_uint16_t((uint16_t *)&test_data->window_sz,
> arg);
> > > > > > +	uint16_t tmp;
> > > > > > +	int ret = parse_uint16_t(&tmp, arg);
> > > > > >
> > > > > >  	if (ret) {
> > > > > >  		RTE_LOG(ERR, USER1, "Failed to parse window
> size\n");
> > > > > >  		return -1;
> > > > > >  	}
> > > > > >
> > > > > > +	test_data->window_sz = (int)tmp;
> > > > > >  	return 0;
> > > > > >  }
> > > > > [Fiona] I expect this fixes this coverity issue - but will it
> > > > > result in another
> > > one?
> > > > > window_sz on the xform is uint8_t - so this int will get
> > > > > truncated later, and there's no cast done at that point.
> > > > > Would it be better to add a new parse_uint8_t fn and change
> > > > > test-data-
> > > > > >window_sz to a unit8_t?
> > > > > Or add that cast?
> > > > [Tomek] I measn it's ok. There's a check inside
> > > > comp_perf_check_capabilities function.
> > > > If the value from test_data->window_sz > cap->window_size we have
> > > > a
> > > fail.
> > > > Also during parsing there's a check is value from command line
> > > > between
> > > > 0 and UINT16_MAX, so in my opinion all cases are tested. The point
> > > > is there's only one place where we're parsing uint8_t value.
> > > > parse_uint8_t function will be especially for that.
> > > [Shally] What is window_sz in test data ?is it base 2 log of (actual
> > > window
> > > length) or actual window length in bytes? lib spec mention this as
> > > struct rte_param_log2_range, so If test window size is actual window
> > > length in bytes then I assume test perf should check for
> > > test_data->window_sz > 2 pow cap->window_size but that doesn't look like
> the case.
> > > So if it is log value, then coding wise typecasting here doesn't look right.
> > > Though it add need for extra function to parse_uint8, but that looks
> > > like cleaner approach to use.
> > [Tomek] I mean it's log 2  (please take a look at help usage function
> > in comp_perf_options_parse.c:37
> >
> > " --window-sz N: base two log value of compression window size\n"
> > 		"		(e.g.: 15 => 32k, default: max supported by
> PMD)\n"
> >
> > I mean it's ok, still. We have many types in command line and can be
> > much more in the future. The idea is to parse them into a sort of
> > common range value first ( it should be max range for all digital
> > command line options - in our case there's uint16 or uint32_t) even if
> > it's shorter like uint8_t or etc. We store these values in
> > comp_test_data structure first. Next we check the ranges each of them.
> > In case of window_sz this test is in comp_perf_check_capabilities
> > function. That approach reduce a  set of parsing functions we needed.
> > Of course we can create separate parsing function for each of command
> > line type value, but is this really needed ? :D Please let me know
> > your thoughts - if this new parsing function will clear the code -
> > I'll add this in v2
> [Fiona] ok, I reviewed again and see I'd misunderstood.
> The param being parsed is intentionally not being stored in test_data struct as
> uint8_t, but as an int because it uses -1 as a default value. And there are range
> checks on the input, so an invalid value will never be passed to the PMD.
> So I'm ok with the fix as is - it resolves the coverity issues reported on the param
> parsing.
> 
> However there's a second issue, which coverity is likely to throw up after the
> above fix is applied - when the test_data value is later passed to the PMD, it
> should have a cast from int to unit8_t.
> But that's a separate issue, not referred to by this coverity report, so we'll send a
> separate patch for it.
> @Shally, are you ok with this?
[Shally] No hard choices here. But why we cant app use 0 as default value instead of -1? It would save us an additional typecast when passing to PMD

Thanks
Shally