Hi Junio,
On Fri, 4 Oct 2019, Junio C Hamano wrote:
> "Johannes Schindelin via GitGitGadget" <gitgitgadget@xxxxxxxxx>
> writes:
>
> > While at it, we take care of reporting overflows (which are unlikely,
> > but hey, defensive programming is good!).
> >
> > We _also_ take pains of casting the unsigned value to signed: otherwise,
> > the signed operand (i.e. the `-1`) would be cast to unsigned before
> > doing the arithmetic.
>
> These three look good and too similar to each other, which makes me
> wonder if we want to allow them simply write
>
> return insert_pos_as_negative_offset(nr);
>
> with something like
>
> static int insert_pos_as_negative_offset(uintmax_t nr)
> {
> if (INT_MAX < nr)
> die("overflow: -1 - %"PRIuMAX, nr);
> return -1 - (int)nr;
> }
>
> to avoid repetition.
I tried not to do that because there are two different data types in
play: `unsigned int` and `size_t`. But I guess by making this an
`inline` function, compilers can optimize for the common case and avoid
casting _twice_.
Will be fixed in v2,
Dscho
>
> > Helped-by: Denton Liu <liu.denton@xxxxxxxxx>
> > Signed-off-by: Johannes Schindelin <johannes.schindelin@xxxxxx>
> > ---
> > read-cache.c | 9 ++++++---
> > sha1-lookup.c | 12 +++++++++---
> > 2 files changed, 15 insertions(+), 6 deletions(-)
> >
> > diff --git a/read-cache.c b/read-cache.c
> > index c701f7f8b8..97745c2a31 100644
> > --- a/read-cache.c
> > +++ b/read-cache.c
> > @@ -1275,8 +1275,11 @@ static int add_index_entry_with_check(struct index_state *istate, struct cache_e
> > * we can avoid searching for it.
> > */
> > if (istate->cache_nr > 0 &&
> > - strcmp(ce->name, istate->cache[istate->cache_nr - 1]->name) > 0)
> > - pos = -istate->cache_nr - 1;
> > + strcmp(ce->name, istate->cache[istate->cache_nr - 1]->name) > 0) {
> > + if (istate->cache_nr > INT_MAX)
> > + die("overflow: -1 - %u", istate->cache_nr);
> > + pos = -1 - (int)istate->cache_nr;
> > + }
> > else
> > pos = index_name_stage_pos(istate, ce->name, ce_namelen(ce), ce_stage(ce));
> >
> > @@ -1894,7 +1897,7 @@ static size_t estimate_cache_size(size_t ondisk_size, unsigned int entries)
> > /*
> > * Account for potential alignment differences.
> > */
> > - per_entry += align_padding_size(sizeof(struct cache_entry), -sizeof(struct ondisk_cache_entry));
> > + per_entry += align_padding_size(per_entry, 0);
> > return ondisk_size + entries * per_entry;
> > }
> >
> > diff --git a/sha1-lookup.c b/sha1-lookup.c
> > index 796ab68da8..bb786b5420 100644
> > --- a/sha1-lookup.c
> > +++ b/sha1-lookup.c
> > @@ -69,8 +69,12 @@ int sha1_pos(const unsigned char *sha1, void *table, size_t nr,
> > miv = take2(sha1 + ofs);
> > if (miv < lov)
> > return -1;
> > - if (hiv < miv)
> > - return -1 - nr;
> > + if (hiv < miv) {
> > + if (nr > INT_MAX)
> > + die("overflow: -1 - %"PRIuMAX,
> > + (uintmax_t)nr);
> > + return -1 - (int)nr;
> > + }
> > if (lov != hiv) {
> > /*
> > * At this point miv could be equal
> > @@ -97,7 +101,9 @@ int sha1_pos(const unsigned char *sha1, void *table, size_t nr,
> > lo = mi + 1;
> > mi = lo + (hi - lo) / 2;
> > } while (lo < hi);
> > - return -lo-1;
> > + if (nr > INT_MAX)
> > + die("overflow: -1 - %"PRIuMAX, (uintmax_t)lo);
> > + return -1 - (int)lo;
> > }
> >
> > int bsearch_hash(const unsigned char *sha1, const uint32_t *fanout_nbo,
>
