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

Re: Unaligned accesses in sha1dc

From
Andreas Schwab <schwab@suse.de>
Date
Jun 1, 2017, 09:32 UTC
Message-ID
<mvmr2z4t5ns.fsf@suse.de>
In-Reply-To
<20170601091856.r3oaddwzsndniyfa@genre.crustytoothpaste.net>
On Jun 01 2017, "brian m. carlson" <sandals@crustytoothpaste.net> wrote:
Show 23 quoted lines
> On Thu, Jun 01, 2017 at 10:28:52AM +0200, Andreas Schwab wrote:
>> The sh1dc implementation is making unaligned accesses, which will crash
>> on some architectures, others have to emulate them in software.
>> 
>> Breakpoint 4, sha1_compression_states (ihv=0x600ffffffffe7010, 
>>     m=<optimized out>, W=0x600ffffffffe70a8, states=0x600ffffffffe7328)
>>     at sha1dc/sha1.c:398
>> 398             SHA1COMPRESS_FULL_ROUND2_STEP(e, a, b, c, d, W, 21, temp);
>> (gdb) n
>> 403             SHA1COMPRESS_FULL_ROUND2_STEP(d, e, a, b, c, W, 22, temp);
>> (gdb) 
>> 408             SHA1COMPRESS_FULL_ROUND2_STEP(c, d, e, a, b, W, 23, temp);
>> (gdb) 
>> 413             SHA1COMPRESS_FULL_ROUND2_STEP(b, c, d, e, a, W, 24, temp);
>> (gdb) 
>> 418             SHA1COMPRESS_FULL_ROUND2_STEP(a, b, c, d, e, W, 25, temp);
>> (gdb) 
>> 291             SHA1COMPRESS_FULL_ROUND1_STEP_LOAD(a, b, c, d, e, m, W, 0, temp);
>> (gdb) 
>> git(21728): unaligned access to 0x600000000009f8d5, ip=0x40000000003336d0
>> 423             SHA1COMPRESS_FULL_ROUND2_STEP(e, a, b, c, d, W, 26, temp);
>
> What architecture are you seeing this on?
It doesn't matter.
Andreas.
-- 
Andreas Schwab, SUSE Labs, schwab@suse.de
GPG Key fingerprint = 0196 BAD8 1CE9 1970 F4BE  1748 E4D4 88E3 0EEA B9D7
"And now for something completely different."
Previous: brian m. carlsonNext: Lars Schneider
Message 6 of 28 in “Unaligned accesses in sha1dc”
  1. Andreas SchwabJun 1, 2017
  2. Junio C HamanoJun 1, 2017
  3. Andreas SchwabJun 1, 2017
  4. Junio C HamanoJun 1, 2017
  5. brian m. carlsonJun 1, 2017
  6. Andreas SchwabJun 1, 2017
  7. Lars SchneiderJun 1, 2017
  8. Junio C HamanoJun 1, 2017
  9. Andreas SchwabJun 1, 2017
  10. Martin ÅgrenJun 1, 2017
  11. Ævar Arnfjörð BjarmasonJun 1, 2017
  12. Martin ÅgrenJun 1, 2017
  13. Martin ÅgrenJun 1, 2017
  14. Junio C HamanoJun 2, 2017
  15. Ævar Arnfjörð BjarmasonJun 2, 2017
  16. Martin ÅgrenJun 2, 2017
  17. Ævar Arnfjörð BjarmasonJun 2, 2017
  18. Martin ÅgrenJun 2, 2017
  19. Ævar Arnfjörð BjarmasonJun 2, 2017
  20. demerphqJun 2, 2017
  21. demerphqJun 2, 2017
  22. Ævar Arnfjörð BjarmasonJun 2, 2017
  23. Linus TorvaldsJun 2, 2017
  24. Junio C HamanoJun 3, 2017
  25. Liam R. HowlettJun 2, 2017
  26. Ævar Arnfjörð BjarmasonJun 2, 2017
  27. Junio C HamanoJun 3, 2017
  28. Andreas SchwabJun 1, 2017

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.