Re: [GSOC Patch v2 3/4] commit-graph: use generation directly when writing commit-graph

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Mon, Jun 08, 2020 at 06:31:49PM +0200, Jakub Narębski wrote:
> Abhishek Kumar <abhishekkumar8222@xxxxxxxxx> writes:
>
> > commit_graph_generation() returns GENERATION_NUMBER_INFINITY if the
> > graph position for commit is COMMIT_NOT_FROM_GRAPH.
> >
> > While this is true when reading from a commit graph, no graph positions
> > are associated with a commit when writing a commit graph. Therefore, the
> > helper incorrectly returns GENERATION_NUMBER_INFINITY despite having a
> > finite generation number.
> >
> > Let's fix this by using generation number directly when writing a commit
> > graph.
>
> I think that to avoid having non-working patch (which can cause problems
> when bisecting), it would be a better idea to switch the order of
> patches 2 and 3.  This way we won't have incorrect behaviour.

Yup... agreed (with the additional caveat that the unused function in
the first patch also be moved into this new--larger--patch).

> >
> > Signed-off-by: Abhishek Kumar <abhishekkumar8222@xxxxxxxxx>
> > ---
> >  commit-graph.c | 13 ++++++++-----
> >  1 file changed, 8 insertions(+), 5 deletions(-)
> >
> > diff --git a/commit-graph.c b/commit-graph.c
> > index f7cca4def4..0dc79e7c90 100644
> > --- a/commit-graph.c
> > +++ b/commit-graph.c
> > @@ -1070,7 +1070,7 @@ static void write_graph_chunk_data(struct hashfile *f, int hash_len,
> >  		else
> >  			packedDate[0] = 0;
> >
> > -		packedDate[0] |= htonl(commit_graph_generation((*list)) << 2);
> > +		packedDate[0] |= htonl(commit_graph_data_at(*list)->generation << 2);
> >
>
> All right.
>
> >  		packedDate[1] = htonl((*list)->date);
> >  		hashwrite(f, packedDate, 8);
> > @@ -1301,9 +1301,11 @@ static void compute_generation_numbers(struct write_commit_graph_context *ctx)
> >  					_("Computing commit graph generation numbers"),
> >  					ctx->commits.nr);
> >  	for (i = 0; i < ctx->commits.nr; i++) {
> > +		uint32_t generation = commit_graph_data_at(ctx->commits.list[i])->generation;
> > +
> >  		display_progress(ctx->progress, i + 1);
> > -		if (commit_graph_generation(ctx->commits.list[i]) != GENERATION_NUMBER_INFINITY &&
> > -		    commit_graph_generation(ctx->commits.list[i]) != GENERATION_NUMBER_ZERO)
> > +		if (generation != GENERATION_NUMBER_INFINITY &&
> > +		    generation != GENERATION_NUMBER_ZERO)
> >  			continue;
> >
>
> All right; this also introduces local variable to avoid accessing the
> slab twice^W four times...
>
> >  		commit_list_insert(ctx->commits.list[i], &list);
> > @@ -1314,8 +1316,9 @@ static void compute_generation_numbers(struct write_commit_graph_context *ctx)
> >  			uint32_t max_generation = 0;
> >
> >  			for (parent = current->parents; parent; parent = parent->next) {
> > -				if (commit_graph_generation(parent->item) == GENERATION_NUMBER_INFINITY ||
> > -				    commit_graph_generation(parent->item) == GENERATION_NUMBER_ZERO) {
> > +
> > +				if (generation == GENERATION_NUMBER_INFINITY ||
> > +				    generation == GENERATION_NUMBER_ZERO) {
> >  					all_parents_computed = 0;
> >  					commit_list_insert(parent->item, &list);
> >  					break;
>
> ... which is then used here.
>
> Best,
> --
> Jakub Narębski
Thanks,
Taylor



[Index of Archives]     [Linux Kernel Development]     [Gcc Help]     [IETF Annouce]     [DCCP]     [Netdev]     [Networking]     [Security]     [V4L]     [Bugtraq]     [Yosemite]     [MIPS Linux]     [ARM Linux]     [Linux Security]     [Linux RAID]     [Linux SCSI]     [Fedora Users]

  Powered by Linux