 |
Post Av7NmlVqbajkoVgSoK by [email protected] |
 |
More posts by [email protected] |
|
Post #Av6OqLgcTSyMfZ2Esq by [email protected] |
|
0 likes, 1 repeats |
|
|
|
I visited the TrustInSoft offices in Paris today and got a demo of their latest… |
|
|
|
Post #Av6OqLolz9Uj4qqkam by [email protected] |
|
0 likes, 0 repeats |
|
|
|
"AddressSanitizer is similarly picky about overrunning the buffer." -… |
|
|
|
Post #Av6OqLv9bQbBOdpqXQ by [email protected] |
|
0 likes, 1 repeats |
|
|
|
@regehr fake news. |
|
|
|
Post #Av6dnaqd98wE8SrUSe by [email protected] |
|
0 likes, 1 repeats |
|
|
|
@regehr joke aside the trick they play here for hashing is common, Darwin's… |
|
|
|
Post #Av6dsb4riiWOINuEgi by [email protected] |
|
0 likes, 1 repeats |
|
|
|
@regehr is it legal? no. but it's damn useful for performance :/ |
|
|
|
Post #Av780SbI7Px4wijqls by [email protected] |
|
0 likes, 1 repeats |
|
|
|
@regehr This kind of overhead trick is very common. On most hardware, traps hap… |
|
|
|
Post #Av7G7naFIkljgVGAG8 by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@david_chisnall it’s a fine idea but separately I blame C for forbidding the … |
|
|
|
Post #Av7G7ngGwLabzC4yeW by [email protected] |
|
0 likes, 1 repeats |
|
|
|
@regehr It’s very hard to do that in a portable abstract machine. This kind o… |
|
|
|
Post #Av7N4DtmDikJotdZeC by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@david_chisnall I’m saying that if we’re giving up on portability anyway, m… |
|
|
|
Post #Av7N4E09pzqm8gcfaq by [email protected] |
|
0 likes, 1 repeats |
|
|
|
@regehr I guess, as long as there’s a way of explicitly telling the programme… |
|
|
|
Post #Av7ND7a15AvcpSdaU4 by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@regehr yeah I'm kinda with the code on this one. Reading an aligned chunk … |
|
|
|
Post #Av7ND7gkg8JfALmxyy by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@harold sure, just allocate 3 extra bytes and there's no problem |
|
|
|
Post #Av7ND7nUH5hhVEwLTs by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@regehr @harold sometimes you gotta work with a string allocated by someone els… |
|
|
|
Post #Av7ND7t9w0EzmpasK0 by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@amonakov @harold sure! live dangerously!!! |
|
|
|
Post #Av7ND7zXYHLS6cZyGe by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@regehr @harold on asan/compiler side, I think it would help if along with intr… |
|
|
|
Post #Av7ND85DDBskODEV6m by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@amonakov @regehr @harold my hot take is that intrinsic loads should be blessed… |
|
|
|
Post #Av7NHwyTwN9vNvNqd6 by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@regehr those casts from byte buffer to integer buffer are also dodgy as hell… |
|
|
|
Post #Av7NHx4VZxyngcCf1U by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@mirabilos @regehr guess "every machine I've seen" also supports … |
|
|
|
Post #Av7NIBlYwgjnGCSF4i by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@regehr had a short interaction with one of their engineers at some point, seem… |
|
|
|
Post #Av7NIBsIXe7pb5bcZc by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@fay59 I mean, look at this thread. |
|
|
|
Post #Av7NJf21ceycJbrArw by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@david_chisnall @regehr perhaps the right intrinsic is a wide read with length … |
|
|
|
Post #Av7NLZFw4EDWyvGwvw by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@amonakov @harold or, even better, perhaps the folks standardizing the language… |
|
|
|
Post #Av7NLZMJgVJzIiG2sa by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@amonakov @harold similarly, it would be great if people didn't have to cho… |
|
|
|
Post #Av7NLZShImQRcVF8pE by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@regehr @harold I see they are serving some strong stuff in Paris |
|
|
|
Post #Av7NLnBqGFGGCHSsUq by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@amonakov @regehr @harold then you can’t access past the allocated length, an… |
|
|
|
Post #Av7NMij8L8lAe2Tlnk by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@steve @amonakov @harold ✅ |
|
|
|
Post #Av7NNcZUiAK7qMOrYG by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@regehr Aside from this code simply being, in general, hella sus, the worst par… |
|
|
|
Post #Av7NStPtVbwTXHQJUG by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@steve finitely-presented monoids are so simple in comparison to the contortion… |
|
|
|
Post #Av7NVQc33dx14RJjjE by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@cliffle @regehr *cough*alphasparcpowerpc, IIRC… |
|
|
|
Post #Av7NVQhiiYUJM1yGZM by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@cliffle @regehr and even on those which do, you lose performance |
|
|
|
Post #Av7NVQnONT1bdccnPU by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@cliffle @regehr *and* atomicity (possibility of load tearing) |
|
|
|
Post #Av7NVcLTR5p5SVefYm by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@slava see what we need here is some partial monoids where everything else is U… |
|
|
|
Post #Av7NYjA8XwQpTEpqBU by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@regehr oh! How long are you staying in France? |
|
|
|
Post #Av7NYjFoCqy7kpUN1c by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@MonniauxD leaving early Sunday -- I'll visit you another time! had to pick… |
|
|
|
Post #Av7Nb1hMbLNrXPTNgm by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@regehr oh I was wondering if you were staying in Paris - I’ll be traveling t… |
|
|
|
Post #Av7NfwUTOLIYpkmtRQ by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@amonakov @harold indeed |
|
|
|
Post #Av7NiC8gBL3DwhfSuO by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@regehr they're just programming on an infinite tape, like the good Turing … |
|
|
|
Post #Av7NjBCh91VZKp128W by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@regehrRaphael told me about you :) enjoy your week-end in Paris (did you buy … |
|
|
|
Post #Av7NmlVqbajkoVgSoK by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@amonakov @regehr @harold the “deliberately-partially-oob” is UB, though, a… |
|
|
|
Post #Av7NnNXFEqQwjqd7GS by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@madcoder @regehr |
|
|
|
Post #Av7NtTEWxw469GmVZg by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@glipari @MonniauxD alas, no! I’m not supposed to take it home but I will pro… |
|
|
|
Post #Av7Nz9NvfsjeKLsz5s by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@regehr @amonakov @harold If in France and have never tried one I'd suggest… |
|
|
|
Post #Av7O1JwHPDphdgzJEu by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@regehr What, you don't like libraries randomly writing past the memory all… |
|
|
|
Post #Av7O20iyKZjoKs5my0 by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@madcoder @regehr perfectly legal since it’s written in assembly instead of C |
|
|
|
Post #Av7O4VVWqo5CO3zk24 by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@regehr @harold If I ran the run-time zoo, I would insist that there is 63 sacr… |
|
|
|
Post #Av7O4X7qpoKbPCvrP6 by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@regehr @harold (note those 63 bytes before and after are once per region, not … |
|
|
|
Post #Av7OBhOt21m7IcJjbE by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@harold @regehr GCC’s builtin strlen does this sometimes, which then also tri… |
|
|
|
Post #Av7OI3o0dtykWd6MSW by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@regehr how dare asan sanitize my addressing when I merely asked it to do exact… |
|
|
|
Post #Av7OI5kXPmOGYRUcKG by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@steve clearly this calls for a type system where type checking doesn’t termi… |
|
|
|
Post #Av7ON1ohSrXVnxnGIS by [email protected] |
|
0 likes, 0 repeats |
|
|
|
@horenmar 😬 |
|
|
|
Post #AvAoshMIF9kRu79G3U by [email protected] |
|
0 likes, 1 repeats |
|
|
|
@amonakov @regehr @harold though sometimes what you actually need is more than … |
|
|
|
Post #AvApMrcEfjccP4zOWe by [email protected] |
|
0 likes, 1 repeats |
|
|
|
@amonakov @regehr @harold but I guess numbers 2 and 3 are kinda outside the sco… |
