Title: Potential Deadlock in _Py_qsbr_reserve · Issue #148953 · python/cpython · GitHub
Open Graph Title: Potential Deadlock in _Py_qsbr_reserve · Issue #148953 · python/cpython
X Title: Potential Deadlock in _Py_qsbr_reserve · Issue #148953 · python/cpython
Description: Crash report What happened? Hello I don't have a good unit test for it. I'm running a server under free-threading 3.14.3t build. On a larger QPS, I see a rapid unstoppable RAM increase, while CPU load is okay. Based on a dump of active t...
Open Graph Description: Crash report What happened? Hello I don't have a good unit test for it. I'm running a server under free-threading 3.14.3t build. On a larger QPS, I see a rapid unstoppable RAM increase, while CPU l...
X Description: Crash report What happened? Hello I don't have a good unit test for it. I'm running a server under free-threading 3.14.3t build. On a larger QPS, I see a rapid unstoppable RAM increase, whi...
Opengraph URL: https://github.com/python/cpython/issues/148953
X: @github
Domain: github.com
{"@context":"https://schema.org","@type":"DiscussionForumPosting","headline":"Potential Deadlock in _Py_qsbr_reserve","articleBody":"# Crash report\n\n### What happened?\n\nHello\n\nI don't have a good unit test for it. I'm running a server under free-threading 3.14.3t build. On a larger QPS, I see a rapid unstoppable RAM increase, while CPU load is okay.\n\nBased on a dump of active threads, I suspect a deadlock somewhere around `_Py_qsbr_reserve` and Stop the World:\n\nE.g.\n* This thread successfully initiated the Stop The World event and is waiting for all other threads to acknowledge and pause (park).\n * `stop_the_world` indicates it is the one trying to stop everything.\n * `PyEvent_WaitTimed` shows it is sitting there waiting for a signal that everyone has stopped.\n ```\n do_futex_wait\n __new_sem_wait_slow\n _PySemaphore_Wait\n _PyParkingLot_Park\n PyEvent_WaitTimed\n stop_the_world\n type_set_abstractmethods\n type_setattro\n PyObject_SetAttr\n _abc__abc_init\n ... etc\n ```\n* This thread also needed to stop the world (to grow an internal array) but got blocked because anotheer already held the master lock.\n * `_Py_qsbr_reserve` shows it was trying to reserve space in the memory management system.\n * `_PyMutex_LockTimed` shows it is blocked waiting for a lock inside stop_the_world. This is the lock held by Thread 1.\n ```\n do_futex_wait\n __new_sem_wait_slow\n _PySemaphore_Wait\n _PyParkingLot_Park\n _PyMutex_LockTimed\n stop_the_world\n _Py_qsbr_reserve\n PyGILState_Ensure\n ... etc\n ```\n\nI suspect a modification of `_Py_qsbr_reserve` could help, the server thread dump does not complain about waiting on Stop the World after this modification:\n```cpp\nPy_ssize_t\n_Py_qsbr_reserve(PyInterpreterState *interp)\n{\n struct _qsbr_shared *shared = &interp->qsbr;\n\n PyMutex_Lock(&shared->mutex);\n // Try allocating from our internal freelist\n struct _qsbr_thread_state *qsbr = qsbr_allocate(shared);\n\n while (qsbr == NULL) {\n // Unlock before stopping the world to avoid deadlocks.\n // If we hold shared->mutex while waiting for the world to stop,\n // we might block a thread that needs to acquire shared->mutex to park.\n PyMutex_Unlock(&shared->mutex);\n _PyEval_StopTheWorld(interp);\n PyMutex_Lock(&shared->mutex);\n\n // Try allocating again, as another thread might have grown the array\n // or freed an entry while we were waiting.\n qsbr = qsbr_allocate(shared);\n if (qsbr != NULL) {\n _PyEval_StartTheWorld(interp);\n break;\n }\n\n // Still NULL, we must grow it\n if (grow_thread_array(shared) == 0) {\n qsbr = qsbr_allocate(shared);\n } else {\n // Failed to grow array (e.g. OOM). Break to avoid infinite loop.\n _PyEval_StartTheWorld(interp);\n break;\n }\n _PyEval_StartTheWorld(interp);\n }\n\n // Return an index rather than the pointer because the array may be\n // resized and the pointer invalidated.\n Py_ssize_t index = -1;\n if (qsbr != NULL) {\n index = (struct _qsbr_pad *)qsbr - shared->array;\n }\n PyMutex_Unlock(&shared->mutex);\n return index;\n}\n```\n\nSimilar issues in the past:\n* https://github.com/python/cpython/issues/129732\n* https://github.com/python/cpython/issues/144513\n\n### CPython versions tested on:\n\n3.14\n\n### Operating systems tested on:\n\nLinux\n\n### Output from running 'python -VV' on the command line:\n\n3.14.3 (free-threading)\n\n\u003c!-- gh-linked-prs -->\n### Linked PRs\n* gh-149100\n\u003c!-- /gh-linked-prs -->\n","author":{"url":"https://github.com/AnnaAr321","@type":"Person","name":"AnnaAr321"},"datePublished":"2026-04-24T11:53:47Z","url":"https://github.com/python/cpython/issues/148953"}
| react-profiling | 0 |
| react-import-map | react,react-dom,react-dom/client,react-dom/profiling,react-is,react/compiler-runtime,react/jsx-dev-runtime,react/jsx-runtime,scheduler@b8488ee2c322 |
| route-pattern | /_view_fragments/issues/show/:user_id/:repository/:id/issue_layout(.:format) |
| route-controller | voltron_issues_fragments |
| route-action | issue_layout |
| fetch-nonce | v2:0692197a-a284-dfeb-1fe2-3f4327098c2a |
| current-catalog-service-hash | 81bb79d38c15960b92d99bca9288a9108c7a47b18f2423d0f6438c5b7bcd2114 |
| request-id | C648:3E418A:2008A6F:29F2C28:6B02367F |
| html-safe-nonce | c154641608c9c4d33ffdf05cc27924dfdefea83011c24fce76a5c5f483b4acb8 |
| visitor-payload | eyJyZWZlcnJlciI6IiIsInJlcXVlc3RfaWQiOiJDNjQ4OjNFNDE4QToyMDA4QTZGOjI5RjJDMjg6NkIwMjM2N0YiLCJ2aXNpdG9yX2lkIjoiNzU5MDU0Mjc2Mzk3Mjc2OTQwNyIsInJlZ2lvbl9lZGdlIjoiaWFkIiwicmVnaW9uX3JlbmRlciI6ImlhZCJ9 |
| visitor-hmac | 968a874ca971a001692307b751695d4a883af7317b7e5d3d511340841c6ddc36 |
| hovercard-subject-tag | issue:4322705188 |
| github-keyboard-shortcuts | repository,issues,copilot |
| google-site-verification | Apib7-x98H0j5cPqHWwSMm6dNU4GmODRoqxLiDzdx9I |
| octolytics-url | https://collector.github.com/github/collect |
| analytics-location | / |
| fb:app_id | 1401488693436528 |
| apple-itunes-app | app-id=1477376905, app-argument=https://github.com/_view_fragments/issues/show/python/cpython/148953/issue_layout |
| twitter:image | https://opengraph.githubassets.com/1e34322bf505d2569201c51c020e256a6bb50177cc80f1a7a1b33c2210c12de0/python/cpython/issues/148953 |
| twitter:card | summary_large_image |
| og:image | https://opengraph.githubassets.com/1e34322bf505d2569201c51c020e256a6bb50177cc80f1a7a1b33c2210c12de0/python/cpython/issues/148953 |
| og:image:alt | Crash report What happened? Hello I don't have a good unit test for it. I'm running a server under free-threading 3.14.3t build. On a larger QPS, I see a rapid unstoppable RAM increase, while CPU l... |
| og:image:width | 1200 |
| og:image:height | 600 |
| og:site_name | GitHub |
| og:type | object |
| og:author:username | AnnaAr321 |
| hostname | github.com |
| expected-hostname | github.com |
| None | 87e774ee7c6b3bee0490394e4d3e57ec9dbc956f74b87ea685dc8996c1676456 |
| turbo-cache-control | no-preview |
| go-import | github.com/python/cpython git https://github.com/python/cpython.git |
| octolytics-dimension-user_id | 1525981 |
| octolytics-dimension-user_login | python |
| octolytics-dimension-repository_id | 81598961 |
| octolytics-dimension-repository_nwo | python/cpython |
| octolytics-dimension-repository_public | true |
| octolytics-dimension-repository_is_fork | false |
| octolytics-dimension-repository_network_root_id | 81598961 |
| octolytics-dimension-repository_network_root_nwo | python/cpython |
| turbo-body-classes | logged-out env-production page-responsive |
| disable-turbo | false |
| browser-stats-url | https://api.github.com/_private/browser/stats |
| browser-errors-url | https://api.github.com/_private/browser/errors |
| release | 423dbb54c5e7bfecc838eb387c1435eb3d0e760d |
| ui-target | full |
| theme-color | #1e2327 |
| color-scheme | light dark |
| react-app-name | issues |
Links:
Viewport: width=device-width