git/list[1] front-page[2] threads[3] people[4] search[5] about
 

Re: [PATCH 00/38] SHA-256, part 3/3

From
brian m. carlson <sandals@crustytoothpaste.net>
Date
Jul 10, 2020, 19:55 UTC
Message-ID
<20200710195507.GK9782@camp.crustytoothpaste.net>
In-Reply-To
<68cc8fe8-3400-501a-ead5-a9c008605e74@gmail.com>
On 2020-07-10 at 15:14:37, Derrick Stolee wrote:
Show 5 quoted lines
> On 7/9/2020 10:46 PM, brian m. carlson wrote:
> > This is the final part required for the stage 4 implementation of
> > SHA-256.
> 
> WOOHOO! What a milestone!

I'm also excited about this. It's been a lot of work, but we're finally here.

> As usual, your commits are excellently organized and clear. I could
> not find any fault in any of them.
Thanks.
> The proof is really in the pudding: does this pass the test suite
> when GIT_TEST_DEFAULT_HASH=sha256? You add that as a mode to the
> CI scripts, so we will know.

I've seen several cases where we've accidentally regressed things with SHA-256, so it seemed only prudent to set up CI. I've run it locally on my system and it works for me, but we'll see how it fares on the CI system.

Show 5 quoted lines
> I made a recommendation for a different model with how to do the CI,
> but it's super minor and can be done later. Basically, if we create
> a new job for SHA-256 mode, then we can more quickly identify when
> a test failure is due to that toggle and not other optional GIT_TEST_*
> variables.

I think that's a good suggestion, and I'm familiar enough with GitHub Actions that I think I can set up an additional job. If I reroll, I'll try to squash such a change in.

> I hope to play around with SHA256-enabled repos a bit later, to see
> if I can find any issues poking around on my own. I doubt I will,
> with how thoroughly you modified the test suite.

For folks who are looking for a more convenient way to get patches, you're welcome to grab them from the transition-stage-4 branch on https://github.com/bk2204/git.git or https://git.crustytoothpaste.net/git/bmc/git.git. The GitHub URL has slightly more bandwidth and a generally better uptime, since I don't live in a datacenter.

-- 
brian m. carlson: Houston, Texas, US
Previous: Derrick StoleeNext: Junio C Hamano
Message 47 of 50 in “SHA-256, part 3/3”
  1. 00/38 SHA-256, part 3/3brian m. carlson, Jul 10, 2020
  2. 01/38 t: make test-bloom initialize repositorybrian m. carlson, Jul 10, 2020
  3. 03/38 t3305: make hash agnosticbrian m. carlson, Jul 10, 2020
  4. 02/38 t1001: use $ZERO_OIDbrian m. carlson, Jul 10, 2020
  5. 04/38 t3404: prepare 'short SHA-1 collision' tests for SHA-256brian m. carlson, Jul 10, 2020
  6. 11/38 t7063: make hash size independentbrian m. carlson, Jul 10, 2020
  7. Junio C HamanoJul 11, 2020
  8. brian m. carlsonJul 11, 2020
  9. 05/38 t6100: make hash size independentbrian m. carlson, Jul 10, 2020
  10. 07/38 t6301: make hash size independentbrian m. carlson, Jul 10, 2020
  11. 10/38 t7003: compute appropriate length constantbrian m. carlson, Jul 10, 2020
  12. 12/38 t7201: abstract away SHA-1-specific constantsbrian m. carlson, Jul 10, 2020
  13. 13/38 t7102: abstract away SHA-1-specific constantsbrian m. carlson, Jul 10, 2020
  14. 18/38 t8002: make hash size independentbrian m. carlson, Jul 10, 2020
  15. 16/38 t7506: avoid checking for SHA-1-specific constantsbrian m. carlson, Jul 10, 2020
  16. 23/38 t9301: make hash size independentbrian m. carlson, Jul 10, 2020
  17. 27/38 t5308: make test work with SHA-256brian m. carlson, Jul 10, 2020
  18. 28/38 t0410: mark test with SHA1 prerequisitebrian m. carlson, Jul 10, 2020
  19. 37/38 docs: add documentation for extensions.objectFormatbrian m. carlson, Jul 10, 2020
  20. Martin ÅgrenJul 10, 2020
  21. 33/38 t: add test_oid option to select hash algorithmbrian m. carlson, Jul 10, 2020
  22. 32/38 Enable SHA-256 support by defaultbrian m. carlson, Jul 10, 2020
  23. 36/38 ci: run tests with SHA-256brian m. carlson, Jul 10, 2020
  24. Derrick StoleeJul 10, 2020
  25. 09/38 t6501: avoid hard-coded objectsbrian m. carlson, Jul 10, 2020
  26. 08/38 t6500: specify test values for SHA-256brian m. carlson, Jul 10, 2020
  27. 06/38 t6101: make hash size independentbrian m. carlson, Jul 10, 2020
  28. 35/38 t: make SHA1 prerequisite depend on default hashbrian m. carlson, Jul 10, 2020
  29. Denton LiuJul 12, 2020
  30. Junio C HamanoJul 12, 2020
  31. 38/38 t: remove test_oid_init in testsbrian m. carlson, Jul 10, 2020
  32. 31/38 setup: add support for reading extensions.objectformatbrian m. carlson, Jul 10, 2020
  33. 25/38 t9500: ensure that algorithm info is preserved in configbrian m. carlson, Jul 10, 2020
  34. 22/38 t9300: use $ZERO_OID instead of hard-coded object IDbrian m. carlson, Jul 10, 2020
  35. 26/38 t9700: make hash size independentbrian m. carlson, Jul 10, 2020
  36. 24/38 t9350: make hash size independentbrian m. carlson, Jul 10, 2020
  37. 20/38 t8011: make hash size independentbrian m. carlson, Jul 10, 2020
  38. 30/38 builtin/verify-pack: implement an --object-format optionbrian m. carlson, Jul 10, 2020
  39. 34/38 t: allow testing different hash algorithms via environmentbrian m. carlson, Jul 10, 2020
  40. 29/38 http-fetch: set up git directory before parsing pack hashesbrian m. carlson, Jul 10, 2020
  41. 21/38 t9300: abstract away SHA-1-specific constantsbrian m. carlson, Jul 10, 2020
  42. 19/38 t8003: make hash size independentbrian m. carlson, Jul 10, 2020
  43. 14/38 t7400: make hash size independentbrian m. carlson, Jul 10, 2020
  44. 17/38 t7508: use $ZERO_OID instead of hard-coded constantbrian m. carlson, Jul 10, 2020
  45. 15/38 t7405: make hash size independentbrian m. carlson, Jul 10, 2020
  46. Derrick StoleeJul 10, 2020
  47. brian m. carlsonJul 10, 2020
  48. Junio C HamanoJul 10, 2020
  49. Junio C HamanoJul 11, 2020
  50. brian m. carlsonJul 11, 2020

Read the whole thread, see it on lore, or plain text.

$ cat FOOTERMessages come from the public archive at lore.kernel.org/git, fetched every hour. The front page is chosen and written each morning by an AI editor and can be wrong; the threads themselves are the record. About and API. For agents: an MCP server at https://gitlist.dev/mcp, and any thread, story or person page as Markdown by adding .md to its URL (or sending Accept: text/markdown). Details in /llms.txt.