RE: git ftbfs on s390x (test failures)

Hi,

I was able to narrow down the problem and I do have an idea for a fix.

diff --git a/builtin/backfill.c b/builtin/backfill.c
index 33e1ea2f84..18f9701487 100644
--- a/builtin/backfill.c
+++ b/builtin/backfill.c
@@ -123,7 +123,7 @@ int cmd_backfill(int argc, const char **argv, const char *prefix, struct reposit
                .sparse = 0,
        };
        struct option options[] = {
-               OPT_INTEGER(0, "min-batch-size", &ctx.min_batch_size,
+               OPT_MAGNITUDE(0, "min-batch-size", &ctx.min_batch_size,
                            N_("Minimum number of objects to request at a time")),
                OPT_BOOL(0, "sparse", &ctx.sparse,
                         N_("Restrict the missing objects to the current sparse-checkout")),


This is passing all of the test cases. But with this change a size_t pointer is being cast to an unsigned long pointer.
This could potentially lead to some issue in some architecture.
If I can verify that the min_batch_size variable (which is a size_t var) is not taking a value larger than an int could store, then we can change its data type from size_t to int. So, I was checking if there is any evidence that indicates a potential range for min_batch_size.
I was also planning to ping with the upstream community for clarification.

Thanks,
Pranav


________________________________
From: Pranav P <[email protected]>
Sent: Monday, April 21, 2025 10:08 AM
To: Bastian Blank <[email protected]>; debian-s390 <[email protected]>; [email protected] <[email protected]>; elbrus <[email protected]>
Subject: Re: [EXTERNAL] Re: git ftbfs on s390x (test failures)

Hi,

I was able to reproduce the issue.
I will update all of my findings in the mailing list.

Thanks,
Pranav
________________________________
From: Bastian Blank <[email protected]>
Sent: Thursday, April 17, 2025 4:45 PM
To: debian-s390 <[email protected]>; [email protected] <[email protected]>
Subject: [EXTERNAL] Re: git ftbfs on s390x (test failures)

On Thu, Apr 17, 2025 at 12:46:08PM +0200, Paul Gevers wrote:
> > git ftbfs on s390x (test failures), there's no good summary on the build
> > log:

The failing test is:

| expecting success of 5620.4 'do partial clone 2, backfill min batch size':
|        git clone --no-checkout --filter=blob:none      \
|                --single-branch --branch=main            \
|                "file://$(pwd)/srv.bare" backfill2 &&
|
|        GIT_TRACE2_EVENT="$(pwd)/batch-trace" git \
|                -C backfill2 backfill --min-batch-size=20 &&
|
|        # Batches were used
|        test_trace2_data promisor fetch_count 20 <batch-trace >matches &&
|        test_line_count = 2 matches &&
|        test_trace2_data promisor fetch_count 8 <batch-trace &&
|
|        # No more missing objects!
|        git -C backfill2 rev-list --quiet --objects --missing=print HEAD >revs2 &&
|        test_line_count = 0 revs2
|
| Cloning into 'backfill2'...
| not ok 4 - do partial clone 2, backfill min batch size

>From the test log it's not remotely possible to see what is going on.

Bastian

--
I'm a soldier, not a diplomat.  I can only tell the truth.
                -- Kirk, "Errand of Mercy", stardate 3198.9