Unless they're going to add a proper warning for the condition to gcc 'today' it won't, really.
Sure there are enough developers to go over the kernel to make sure such errors haven't been missed elsewhere, but all it takes is one to miss it and it's still there. Then there's all the other software compiled by gcc..
I'm not entirely sure how it can lead to an exploit (short of remapping page zero, which requires root privileges so doesn't really count) but since it has it's going to need a proper fix.
No. That doesn't fix the problem. All it does is stop the broken optimisation (why the *hell* did someone at gcc think such a thing should be default anyway?)
You need an -ferror-on-bogus-null-pointer-checks parameter so that the code can be fixed.
It's an easy error to make. It's the compilers job to warn you.. in this case not only did it fail to throw a warning it also made the problem worse by 'optimising' it.
gcc is doing precisely the wrong thing - optimising out the if on the theory that the app would have crashed if it was null.
What it *should* do is throw a warning (even an error, given the clear intent of the code) pointing out that the variable is dereferensed before it is tested.
This kind of error being missed by gcc is going to affect a *lot* of code - it's really not that uncommon a coding error, and is easy to do.
Windows has a special allocation function that returns 36bit pointers, but you still have to map them into the 32bit address space to use it - an app on 32bit windows can't address more than about 2.5gb simultaneously (much less, once fragmentation is taken into account).
In this day and age it's a silly argument anyway.. desktop processors are generally 64bit capable, 64bit versions are available of all your favourite OS.. why stick to 32bit if your memory requirements are that large?
It's a safe bet that they'll extend copyright again just as Mickey Mouse is looking like public domain.
You guys in the US won't have a public domain to speak of in a few years.. it'll all be owned by the great grandchildren of once famous authors - the new ruling elite.
Not always. In the UK they recalled perfectly working Pace Satellite boxes (only think wrong was there was a nonzero chance of the cables getting trapped, so they needed cable ties adding) and replaced them with secondhand Thomsons (3 year old plus) with dodgy power supplies, or if you were *really* lucky Amstrads that were, well, Amstrad.
Smart people told Sky to F Off and kept their nice boxes.
Pretty much the same as happens with bittorrent now.
A certain percentage of files will be fine. A certain percentage will be served at dialup speeds (boy is that one fun! Yes I'm really prepared to wait 4 days to download that 50mb file). A certain percentage will be incomplete and never finish A certain percentage will have no seeds online at all
There is no inherent reliability in such a system.. reliability improves with popularity, so if you want something that's 'cool' you'll probably find it in the first bracket, otherwise it's hit and miss.
They could probably get you for contributory infringement - after all, you just became a node on a network called 'the pirate bay'. Also you're getting paid (albeit paid in kind rather than cash)... and as this is a 'legit' company they'll have to keep accurate records of all this - just waiting for the subpoena.
If you're getting money, almost certainly (unless you have a business account) - resale is one of the first things that a TOS prohibits.
Whether they consider 'free subs' as resale depends on the ISP/Lawyer concerned.
Whether merely running internet services is against TOS depends on the ISP - certainly on some of the cheaper contracts this is prohibited but it's far from universal.
A place I go to uses a mixture of last.fm and Spotify to provide their piped music.. this isn't unusual any more - I've been in several pubs and been surprised to hear the spotify premium advert suddenly play...
This is the future - buying music is going the way of the dodo.
Way to set an impossible barrier. Firefox isn't a microsoft product so won't pull updates from WSUS. Duh. Anyway, you're talking about companies who are still running IE6 so what do they care about updates?
If you want to wrap the firefox installer in an MSI, knock yourself out. Takes about 5 minutes. The requirement for that is *way* overstated.. We spent months writing a full fledged MSI installer. Number of customers who actually use the feature to push it out automatically? Zero. They stick it on the intranet and tell people to double click on it, or the admin runs around and installs it.
You don't set what sites can be reached in the browser. That's a retarded way to do it. You do it on the gateway.
50% of consumers using IE6? Not on this planet. 90% of users at work sounds dodgy as well - most of the places I've seen are at least on IE7 even firefox is fairly well represented. Haven't seen an IE6 for ages, although I believe it's still required by some of the more neanderthal companies.
That's normally between 24 hours and 7 days, so it's irrelevant to the discussion.
The article you mention (which btw. seems to be a wikipedia invention) would be done with custom DNS anyway, otherwise it's easily blocked by the ISP setting its cache to ignore a TTL less than a couple of hours (as most do.. hell, my even my home DNS does that).
POP isn't encrypted. There's pop3s but I've never seen an ISP implement it - ISPs know they their punters don't know about encryption so never bother with it. The passwords are even sent plaintext, most of the time.
If you think it's OK to hijack DNS think about what happens if you mistype an email address, or what happens when your configured NTP server goes offline.
Well 3x faster than BT would have to be an LLU Operator, because everything else is all BT (openreach) infrastructure & the only difference between ISPs that use that is how good they are at kicking BT up the arse when things go wrong*.
That limits it in practical terms to Be and Sky (O2 are Be resellers). None of these have a 1 month rolling contract with no cancellation fee. Both of them are IWF encumbered, which is a negative mark against them.
* In that respect I recommend A&A who even have a bot that reopens trouble tickets every time BT close them prematurely.. it's quite fun to watch.
That's expensive IMO. I pay £5 a month (about 6 euros)... it's capable of 7.2Mbps but you really only get about 6-6.5 at best in cities, and a lot lower out in the sticks.
Damn thing is a godsend. Wifi is too expensive - expect to pay £10 per *hour* in the average hotel, and of course doesn't work when moving.
I've been in plenty of offices that are 100% wireless, with a netgear in the corner serving their network. Not all companies are IT and need servers etc. and the average email/browsing/bespoke app stuff needs very little bandwidth to work well.
As far as your 3G comments go... have you been asleep or are you just American? We have cities here delivering 15mbps over 3G and even from my office I can stream a solid 6mbps. The backhauls on some of these towers have *huge* amounts of bandwidth & the slowdowns you used to get are becoming a lot rarer.
You can drive it as an import but can't register it in the UK, so it'll have foreign plates. Insurance would be insane on something like a Tesla, given that there are approximately zero garages that know how to fix them and no parts availablility, plus with that 0-60 it's going to be classed as sports.
Slashdot Mirror
Unless they're going to add a proper warning for the condition to gcc 'today' it won't, really.
Sure there are enough developers to go over the kernel to make sure such errors haven't been missed elsewhere, but all it takes is one to miss it and it's still there. Then there's all the other software compiled by gcc..
I'm not entirely sure how it can lead to an exploit (short of remapping page zero, which requires root privileges so doesn't really count) but since it has it's going to need a proper fix.
No. That doesn't fix the problem. All it does is stop the broken optimisation (why the *hell* did someone at gcc think such a thing should be default anyway?)
You need an -ferror-on-bogus-null-pointer-checks parameter so that the code can be fixed.
It's an easy error to make. It's the compilers job to warn you.. in this case not only did it fail to throw a warning it also made the problem worse by 'optimising' it.
gcc is definitely doing the wrong thing here.
Given the code:
a = foo->bar
if(foo) something()
gcc is doing precisely the wrong thing - optimising out the if on the theory that the app would have crashed if it was null.
What it *should* do is throw a warning (even an error, given the clear intent of the code) pointing out that the variable is dereferensed before it is tested.
This kind of error being missed by gcc is going to affect a *lot* of code - it's really not that uncommon a coding error, and is easy to do.
Windows has a special allocation function that returns 36bit pointers, but you still have to map them into the 32bit address space to use it - an app on 32bit windows can't address more than about 2.5gb simultaneously (much less, once fragmentation is taken into account).
In this day and age it's a silly argument anyway.. desktop processors are generally 64bit capable, 64bit versions are available of all your favourite OS.. why stick to 32bit if your memory requirements are that large?
Funny thing is, although it's repeated about 3 times a day here it's one programme I can keep watching and finding funny.
OTOH I can watch new simpsons episodes and find them so dull I switch off. They cancelled the wrong show.
I agree though - voice actors change all the time and we don't notice unless someone tells us.
They turned him into a brainwashed paedophile and then had him die by falling off a cliff and get eaten by rabid dogs.
Somewhat harsh, IMO.
It's a safe bet that they'll extend copyright again just as Mickey Mouse is looking like public domain.
You guys in the US won't have a public domain to speak of in a few years.. it'll all be owned by the great grandchildren of once famous authors - the new ruling elite.
Not always. In the UK they recalled perfectly working Pace Satellite boxes (only think wrong was there was a nonzero chance of the cables getting trapped, so they needed cable ties adding) and replaced them with secondhand Thomsons (3 year old plus) with dodgy power supplies, or if you were *really* lucky Amstrads that were, well, Amstrad.
Smart people told Sky to F Off and kept their nice boxes.
Ping time, probably.
Pretty much the same as happens with bittorrent now.
A certain percentage of files will be fine.
A certain percentage will be served at dialup speeds (boy is that one fun! Yes I'm really prepared to wait 4 days to download that 50mb file).
A certain percentage will be incomplete and never finish
A certain percentage will have no seeds online at all
There is no inherent reliability in such a system.. reliability improves with popularity, so if you want something that's 'cool' you'll probably find it in the first bracket, otherwise it's hit and miss.
Actually, they wouldn't need to do that. A letter to your ISP saying you're involved in commercial piracy and your internet connection is toast.
They could probably get you for contributory infringement - after all, you just became a node on a network called 'the pirate bay'. Also you're getting paid (albeit paid in kind rather than cash)... and as this is a 'legit' company they'll have to keep accurate records of all this - just waiting for the subpoena.
If you're getting money, almost certainly (unless you have a business account) - resale is one of the first things that a TOS prohibits.
Whether they consider 'free subs' as resale depends on the ISP/Lawyer concerned.
Whether merely running internet services is against TOS depends on the ISP - certainly on some of the cheaper contracts this is prohibited but it's far from universal.
A place I go to uses a mixture of last.fm and Spotify to provide their piped music.. this isn't unusual any more - I've been in several pubs and been surprised to hear the spotify premium advert suddenly play...
This is the future - buying music is going the way of the dodo.
Way to set an impossible barrier. Firefox isn't a microsoft product so won't pull updates from WSUS. Duh. Anyway, you're talking about companies who are still running IE6 so what do they care about updates?
If you want to wrap the firefox installer in an MSI, knock yourself out. Takes about 5 minutes. The requirement for that is *way* overstated.. We spent months writing a full fledged MSI installer. Number of customers who actually use the feature to push it out automatically? Zero. They stick it on the intranet and tell people to double click on it, or the admin runs around and installs it.
You don't set what sites can be reached in the browser. That's a retarded way to do it. You do it on the gateway.
50% of consumers using IE6? Not on this planet. 90% of users at work sounds dodgy as well - most of the places I've seen are at least on IE7 even firefox is fairly well represented. Haven't seen an IE6 for ages, although I believe it's still required by some of the more neanderthal companies.
That's normally between 24 hours and 7 days, so it's irrelevant to the discussion.
The article you mention (which btw. seems to be a wikipedia invention) would be done with custom DNS anyway, otherwise it's easily blocked by the ISP setting its cache to ignore a TTL less than a couple of hours (as most do.. hell, my even my home DNS does that).
When opendns started it was precisely that - an open DNS system which even had its own set of free TLDs to play with.
Then they smelled money. And the rest is history.
Use the anycast DNS at 4.2.2.1, 4.2.2.2, etc. Run by Level3 who have plenty of money anyway and don't need to nickel and dime DNS for it.
How, exactly? DNS happens once. If the PC queries for a valid address it will get it, whether it's in a botnet or not.
The ISP can block attacks by measuring traffic across their routers.. DNS is a stupid place to do it.
POP isn't encrypted. There's pop3s but I've never seen an ISP implement it - ISPs know they their punters don't know about encryption so never bother with it. The passwords are even sent plaintext, most of the time.
If you think it's OK to hijack DNS think about what happens if you mistype an email address, or what happens when your configured NTP server goes offline.
Bit of guesswork on who it is...
Well 3x faster than BT would have to be an LLU Operator, because everything else is all BT (openreach) infrastructure & the only difference between ISPs that use that is how good they are at kicking BT up the arse when things go wrong*.
That limits it in practical terms to Be and Sky (O2 are Be resellers). None of these have a 1 month rolling contract with no cancellation fee. Both of them are IWF encumbered, which is a negative mark against them.
* In that respect I recommend A&A who even have a bot that reopens trouble tickets every time BT close them prematurely.. it's quite fun to watch.
That's expensive IMO. I pay £5 a month (about 6 euros)... it's capable of 7.2Mbps but you really only get about 6-6.5 at best in cities, and a lot lower out in the sticks.
Damn thing is a godsend. Wifi is too expensive - expect to pay £10 per *hour* in the average hotel, and of course doesn't work when moving.
I've been in plenty of offices that are 100% wireless, with a netgear in the corner serving their network. Not all companies are IT and need servers etc. and the average email/browsing/bespoke app stuff needs very little bandwidth to work well.
As far as your 3G comments go... have you been asleep or are you just American? We have cities here delivering 15mbps over 3G and even from my office I can stream a solid 6mbps. The backhauls on some of these towers have *huge* amounts of bandwidth & the slowdowns you used to get are becoming a lot rarer.
You can drive it as an import but can't register it in the UK, so it'll have foreign plates. Insurance would be insane on something like a Tesla, given that there are approximately zero garages that know how to fix them and no parts availablility, plus with that 0-60 it's going to be classed as sports.