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

Re: [PATCH 2/9] commit-graph: stop using signed integers to count bloom filters

From
Junio C Hamano <gitster@pobox.com>
Date
Aug 6, 2025, 15:41 UTC
Message-ID
<xmqq4iukwj83.fsf@gitster.g>
In-Reply-To
<aJL06Nlee6CR-KK0@pks.im>
Patrick Steinhardt <ps@pks.im> writes:
Show 7 quoted lines
>> I wrote these counters in 312cff5207 (bloom: split 'get_bloom_filter()'
>> in two, 2020-09-16) and 59f0d5073f (bloom: encode out-of-bounds filters
>> as non-empty, 2020-09-17), and I don't see a compelling reason that
>> these should be unsigned.
>
> I think that is going backwards though: the question to ask is why
> should these be signed if they cannot ever be negative?

Earlier I gave an example of allowing for a "not yet counted" sentinel value for a variable or a structure member. Another example may be for a function that counts that also needs to signal an error, and as usual in any C programs, the natural way to do so for any function whose "normal" return values are non-negative integers is to signal errors with a negative value.

Note that a structure member or a variable that does not need such a "not yet counted" sentinel value (e.g., it may have a separate "counted already" member associated with it, or the nature of the thing it counts does not have such "not yet counted" state), and it is possible for such a variable to live happily with a function that can signal an error.

It means the variable that receives the counted result from such a function may be able to use only half a range of values as its type implies, if that helper function is the only source of information that is assigned to it, though.

If the counter in question never needs to store such a sentinel value itself, then I am OK for it to be unsigned, and that is exactly why I said "not always a valid excuse". But if the counter variable or structure member has to work with functions that need to return sentinel values (like platform natural int that can use the usual "negative is an error, non-negative is a normal result"), it may have less chance to trigger the -Wsign-compare irritation, if you made it also signed.

Previous: Junio C HamanoNext: Patrick Steinhardt
Message 12 of 69 in “commit-graph: remove reliance on global state”
  1. 0/9 commit-graph: remove reliance on global statePatrick Steinhardt, Aug 4, 2025
  2. 1/9 trace2: introduce function to trace unsigned integersPatrick Steinhardt, Aug 4, 2025
  3. Taylor BlauAug 4, 2025
  4. 2/9 commit-graph: stop using signed integers to count bloom filtersPatrick Steinhardt, Aug 4, 2025
  5. Oswald BuddenhagenAug 4, 2025
  6. Patrick SteinhardtAug 4, 2025
  7. Junio C HamanoAug 4, 2025
  8. Taylor BlauAug 4, 2025
  9. Patrick SteinhardtAug 6, 2025
  10. Oswald BuddenhagenAug 6, 2025
  11. Junio C HamanoAug 6, 2025
  12. Junio C HamanoAug 6, 2025
  13. Patrick SteinhardtAug 7, 2025
  14. Junio C HamanoAug 7, 2025
  15. Patrick SteinhardtAug 11, 2025
  16. Junio C HamanoAug 5, 2025
  17. Taylor BlauAug 4, 2025
  18. 3/9 commit-graph: fix type for some write optionsPatrick Steinhardt, Aug 4, 2025
  19. Taylor BlauAug 4, 2025
  20. 4/9 commit-graph: fix sign comparison warningsPatrick Steinhardt, Aug 4, 2025
  21. Taylor BlauAug 4, 2025
  22. Patrick SteinhardtAug 6, 2025
  23. 5/9 commit-graph: stop using `the_hash_algo` via macrosPatrick Steinhardt, Aug 4, 2025
  24. Taylor BlauAug 4, 2025
  25. 6/9 commit-graph: store the hash algorithm instead of its lengthPatrick Steinhardt, Aug 4, 2025
  26. Taylor BlauAug 4, 2025
  27. 7/9 commit-graph: stop using `the_hash_algo`Patrick Steinhardt, Aug 4, 2025
  28. Taylor BlauAug 4, 2025
  29. Patrick SteinhardtAug 6, 2025
  30. 8/9 commit-graph: stop using `the_repository`Patrick Steinhardt, Aug 4, 2025
  31. Taylor BlauAug 4, 2025
  32. 9/9 commit-graph: stop passing in redundant repositoryPatrick Steinhardt, Aug 4, 2025
  33. Derrick StoleeAug 5, 2025
  34. Patrick SteinhardtAug 6, 2025
  35. 00/10 commit-graph: remove reliance on global statePatrick Steinhardt, Aug 6, 2025
  36. 01/10 trace2: introduce function to trace unsigned integersPatrick Steinhardt, Aug 6, 2025
  37. 02/10 commit-graph: stop using signed integers to count Bloom filtersPatrick Steinhardt, Aug 6, 2025
  38. 03/10 commit-graph: fix type for some write optionsPatrick Steinhardt, Aug 6, 2025
  39. Oswald BuddenhagenAug 6, 2025
  40. Junio C HamanoAug 6, 2025
  41. Patrick SteinhardtAug 7, 2025
  42. 04/10 commit-graph: fix sign comparison warningsPatrick Steinhardt, Aug 6, 2025
  43. 05/10 commit-graph: stop using `the_hash_algo` via macrosPatrick Steinhardt, Aug 6, 2025
  44. 06/10 commit-graph: store the hash algorithm instead of its lengthPatrick Steinhardt, Aug 6, 2025
  45. 07/10 commit-graph: refactor `parse_commit_graph()` to take a repositoryPatrick Steinhardt, Aug 6, 2025
  46. 08/10 commit-graph: stop using `the_hash_algo`Patrick Steinhardt, Aug 6, 2025
  47. 09/10 commit-graph: stop using `the_repository`Patrick Steinhardt, Aug 6, 2025
  48. 10/10 commit-graph: stop passing in redundant repositoryPatrick Steinhardt, Aug 6, 2025
  49. 00/10 commit-graph: remove reliance on global statePatrick Steinhardt, Aug 7, 2025
  50. 01/10 trace2: introduce function to trace unsigned integersPatrick Steinhardt, Aug 7, 2025
  51. 02/10 commit-graph: stop using signed integers to count Bloom filtersPatrick Steinhardt, Aug 7, 2025
  52. 03/10 commit-graph: fix type for some write optionsPatrick Steinhardt, Aug 7, 2025
  53. Junio C HamanoAug 7, 2025
  54. Patrick SteinhardtAug 11, 2025
  55. 04/10 commit-graph: fix sign comparison warningsPatrick Steinhardt, Aug 7, 2025
  56. 05/10 commit-graph: stop using `the_hash_algo` via macrosPatrick Steinhardt, Aug 7, 2025
  57. 06/10 commit-graph: store the hash algorithm instead of its lengthPatrick Steinhardt, Aug 7, 2025
  58. 07/10 commit-graph: refactor `parse_commit_graph()` to take a repositoryPatrick Steinhardt, Aug 7, 2025
  59. 08/10 commit-graph: stop using `the_hash_algo`Patrick Steinhardt, Aug 7, 2025
  60. 09/10 commit-graph: stop using `the_repository`Patrick Steinhardt, Aug 7, 2025
  61. 10/10 commit-graph: stop passing in redundant repositoryPatrick Steinhardt, Aug 7, 2025
  62. 0/6 commit-graph: remove reliance on global statePatrick Steinhardt, Aug 15, 2025
  63. 1/6 commit-graph: stop using `the_hash_algo` via macrosPatrick Steinhardt, Aug 15, 2025
  64. 2/6 commit-graph: store the hash algorithm instead of its lengthPatrick Steinhardt, Aug 15, 2025
  65. 3/6 commit-graph: refactor `parse_commit_graph()` to take a repositoryPatrick Steinhardt, Aug 15, 2025
  66. 4/6 commit-graph: stop using `the_hash_algo`Patrick Steinhardt, Aug 15, 2025
  67. 5/6 commit-graph: stop using `the_repository`Patrick Steinhardt, Aug 15, 2025
  68. 6/6 commit-graph: stop passing in redundant repositoryPatrick Steinhardt, Aug 15, 2025
  69. Derrick StoleeAug 15, 2025

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.