{"id":30993,"date":"2006-06-02T10:00:00","date_gmt":"2006-06-02T10:00:00","guid":{"rendered":"https:\/\/blogs.msdn.microsoft.com\/oldnewthing\/2006\/06\/02\/what-does-the-cs_classdc-class-style-do\/"},"modified":"2006-06-02T10:00:00","modified_gmt":"2006-06-02T10:00:00","slug":"what-does-the-cs_classdc-class-style-do","status":"publish","type":"post","link":"https:\/\/devblogs.microsoft.com\/oldnewthing\/20060602-00\/?p=30993","title":{"rendered":"What does the CS_CLASSDC class style do?"},"content":{"rendered":"

\nLast time, I talked about the historical background for\nthe CS_OWNDC<\/code> class style and why it starts out\nsounding like a good idea but when you think about it some more\nturns out to be an awful idea.\n<\/p>\n

\nThe CS_CLASSDC<\/code> class style is the same thing, but worse,\nfor it takes all the problems of CS_OWNDC<\/code> and magnifies them.\nRecall that the CS_OWNDC<\/code> class style instructs the window\nmanager to create a DC for the window and use that single DC in response\nto calls to BeginPaint<\/code> and GetDC<\/code>.\nThe CS_CLASSDC<\/code> takes this one step further and\ncreates a DC for all the windows of that class<\/strong>.\nSo that problem I showed last time with a function that thought\nit had two different DCs for a window can now happen even\nacross windows.\nYou think you have one DC for one window and another DC for another\nwindow, but in fact they are the same!\n<\/p>\n

\nWhat makes this even worse is that two threads can both be using\nthe same DC at the same time.\nThere is nothing in GDI to forbid it;\nit’s simply a race to see which thread’s changes prevail:\n“Last writer wins.”\nImagine two threads that happen each to have a CS_CLASSDC<\/code>\nwindow from the same window class,\nand suppose both windows need to be repainted.\nEach window gets a WM_PAINT<\/code> message, and the two\nthreads both go into their painting code.\nBut what these threads don’t know is that they are operating on\nthe same DC<\/strong>.\n<\/p>\n\n\n\n\n\n\n\n\n
Thread A<\/th>\nThread B<\/th>\n<\/tr>\n
HDC hdc = BeginPaint(hwnd, &ps);<\/td>\n<\/tr>\n
<\/td>\nHDC hdc = BeginPaint(hwnd, &ps);<\/td>\n<\/tr>\n
SetTextColor(hdc, red);<\/td>\n<\/tr>\n
<\/td>\nSetTextColor(hdc, blue);<\/td>\n<\/tr>\n
DrawText(hdc, …);<\/td>\n<\/tr>\n
<\/td>\nDrawText(hdc, …);<\/td>\n<\/tr>\n<\/table>\n

\nThe code running in Thread A\nfully expected the text to be in red since it set the text color\nto red and then drew text.\nHow was it to know that just at that moment,\nThread B went and changed it to blue?\n<\/p>\n

\nThis is the sort of race condition bug that you’ll probably never\nbe able to study under controlled conditions.\nYou’ll just get bug reports from customers saying that\nmaybe once a month, an item comes out the wrong color, and maybe you’ll\nsee it yourself once in a while, but it will never happen when\nyou have debugger breakpoints set.\nEven if you add additional diagnostic code, all you’ll see is this:\n<\/p>\n

\n...\nSetTextColor(hdc, red);\nASSERT(GetTextColor(hdc) == red); \/\/ assertion fires!<\/font>\nDrawText(hdc, ...);\n<\/pre>\n
\nGreat, the assertion fired.\nThe color you just set isn’t there.\nNow what are you going to do?\nMaybe you’ll just say “Stupid buggy Windows” and change your\ncode to<\/p>\n
\n\/\/ Stupid buggy Windows. For some reason,\n\/\/ about once a month, the SetTextColor doesn't\n\/\/ work and we have to call it twice.\ndo {\n SetTextColor(hdc, red);\n} while (GetTextColor(hdc) != red);\nDrawText(hdc, ...);\n<\/pre>\n
\nAnd even that doesn’t fix the problem, because\nThread B might have changed the color to blue\nafter the GetTextColor<\/code> and the call to DrawText<\/code>.\nNow, it’s only once every six months that the item comes out the\nwrong color.\n<\/p>\n
\nYou swear at Microsoft and vow to develop Mac software from now on.\n<\/p>\n
\nOkay, so now I hope I’ve convinced you that CS_CLASSDC<\/code> is\na horrifically bad idea.\nBut if it’s so fundamentally flawed,\nwhy does it exist in the first place?\n<\/p>\n
\nBecause 16-bit Windows is co-operatively multi-tasked.\nIn the 16-bit world, you don’t have to worry about another thread\nsneaking in and messing with your DC because, as I already noted,\n\nthe fact that you were running meant that nobody else was running<\/a>.\nThis whole multi-threaded disaster scenario could not occur,\nso CS_CLASSDC<\/code> is only slightly wackier than\nCS_OWNDC<\/code>.\nThe introduction of pre-emptive multi-tasking with multiple threads\nin a single process is what\ntook us into the world of “this has no chance of ever working\nproperly”.\nThe class style exists so people who used it in 16-bit code\ncan port to Win32 (as long as they promise to remain a single-threaded\napplication),\nbut no modern software should use it.<\/p>\n","protected":false},"excerpt":{"rendered":"
Last time, I talked about the historical background for the CS_OWNDC class style and why it starts out sounding like a good idea but when you think about it some more turns out to be an awful idea. The CS_CLASSDC class style is the same thing, but worse, for it takes all the problems of […]<\/p>\n","protected":false},"author":1069,"featured_media":111744,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[1],"tags":[25,2],"class_list":["post-30993","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-oldnewthing","tag-code","tag-history"],"acf":[],"blog_post_summary":"
Last time, I talked about the historical background for the CS_OWNDC class style and why it starts out sounding like a good idea but when you think about it some more turns out to be an awful idea. The CS_CLASSDC class style is the same thing, but worse, for it takes all the problems of […]<\/p>\n","_links":{"self":[{"href":"https:\/\/devblogs.microsoft.com\/oldnewthing\/wp-json\/wp\/v2\/posts\/30993","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/devblogs.microsoft.com\/oldnewthing\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/devblogs.microsoft.com\/oldnewthing\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/devblogs.microsoft.com\/oldnewthing\/wp-json\/wp\/v2\/users\/1069"}],"replies":[{"embeddable":true,"href":"https:\/\/devblogs.microsoft.com\/oldnewthing\/wp-json\/wp\/v2\/comments?post=30993"}],"version-history":[{"count":0,"href":"https:\/\/devblogs.microsoft.com\/oldnewthing\/wp-json\/wp\/v2\/posts\/30993\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/devblogs.microsoft.com\/oldnewthing\/wp-json\/wp\/v2\/media\/111744"}],"wp:attachment":[{"href":"https:\/\/devblogs.microsoft.com\/oldnewthing\/wp-json\/wp\/v2\/media?parent=30993"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/devblogs.microsoft.com\/oldnewthing\/wp-json\/wp\/v2\/categories?post=30993"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/devblogs.microsoft.com\/oldnewthing\/wp-json\/wp\/v2\/tags?post=30993"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}