簡(jiǎn)介：碩士學(xué)位論文論文作者學(xué)科專業(yè)指導(dǎo)教師培養(yǎng)單位培養(yǎng)類別完成時(shí)間沈陽師范大學(xué)學(xué)位評(píng)定委員會(huì)單位代碼單位代碼10166單位代碼單位代碼10166大學(xué)生的學(xué)習(xí)動(dòng)機(jī)與英語口語熟練水平的相關(guān)大學(xué)生的學(xué)習(xí)動(dòng)機(jī)與英語口語熟練水平的相關(guān)性研究性研究THECORRELATIONBETWEENCHINESECOLLEGETHECORRELATIONBETWEENCHINESECOLLEGESTUDENTSSTUDENTS’LEARNINGMOTIVATIONANDSPOKENLEARNINGMOTIVATIONANDSPOKENENENGLISHPROFICIENCYGLISHPROFICIENCY李睿琛課程與教學(xué)論（英語）劉淑華大學(xué)外語教學(xué)部2015年05月27日全日制ITHISTHESISWASACCEPTEDASMEETINGTHERESEARCHREQUIREMENTFORTHEMASTER’SDEGREEAPPROVEDDATENAMEOFADVISORDATENAMEOFFIRSTREADERDATENAMEOFSECONDREADERAPPROVEDDATENAMEOFDEANOFDEPARTMENTOFFOREIGNLANGUAGESDATENAMEOFCHAIROFORALDEFENSECOMMITTEE

簡(jiǎn)介：論文題目大學(xué)生心理資本問卷編制及其相關(guān)因素研究學(xué)科專業(yè)應(yīng)用心理學(xué)申請(qǐng)人周利霞導(dǎo)師樊琪教授摘要積極心理學(xué)和積極組織行為學(xué)催化了心理資本及其理論的誕生，心理資本關(guān)注人的積極心理素質(zhì)和潛能，越來越引起研究者的重視。以往對(duì)心理資本的研究主要集中在組織層面，而大學(xué)生作為企業(yè)人力資源新生力量的主要來源，更應(yīng)該受到高度關(guān)注。本研究以大學(xué)生包括本科生和研究生為研究對(duì)象，根據(jù)國(guó)內(nèi)外心理資本的相關(guān)理論，對(duì)大學(xué)生心理資本進(jìn)行了研究，編制了大學(xué)生心理資本問卷，初步分析了大學(xué)生心理資本的現(xiàn)狀和差異狀況，同時(shí)探討了大學(xué)生心理資本與應(yīng)對(duì)方式、職業(yè)決策自我效能感的關(guān)系，取得了初步的研究結(jié)果。本研究共調(diào)查了1175名在校本科生和研究生，使用SPSSLL．5和LIEREL8．5統(tǒng)計(jì)軟件進(jìn)行數(shù)據(jù)分析，研究發(fā)現(xiàn)1、大學(xué)生心理資本包括六個(gè)維度，分別為愿景、自信、合作、樂觀、韌性和感恩，所編制的大學(xué)生心理資本問卷具有良好的信度和效度，符合測(cè)量學(xué)的標(biāo)準(zhǔn)。2、所調(diào)查大學(xué)生的心理資本及其各維度均分都高于一般水平，其中感恩得分最高，樂觀得分最低，大學(xué)生心理資本的狀況良好。3、大學(xué)生心理資本及其各維度在人口學(xué)變量上存在差異，尤其是在是否擔(dān)任學(xué)生干部、家庭成員關(guān)系、人際關(guān)系、學(xué)習(xí)成績(jī)、學(xué)校類型變量上差異非常顯著。4、大學(xué)生心理資本及其各維度與積極應(yīng)對(duì)方式呈正相關(guān)，與消極應(yīng)對(duì)方式呈負(fù)相關(guān)。5、大學(xué)生心理資本及其各維度與職業(yè)決策自我效能感及其各維度呈顯著正相關(guān)，大學(xué)生心理資本對(duì)職業(yè)決策自我效能感存在顯著的預(yù)測(cè)作用。本研究將心理資本理論的研究擴(kuò)展到大學(xué)生群體，在全國(guó)范圍內(nèi)取樣有助于了解我國(guó)大學(xué)生的心理資本的普遍現(xiàn)狀，并根據(jù)研究結(jié)果進(jìn)行了一系列的討論，有助于提高高校和教育者對(duì)大學(xué)生心理資本的重視，推動(dòng)學(xué)校心理教育的發(fā)展，同時(shí)，本文也豐富了心理資本理論和研究，具有理論和現(xiàn)實(shí)意義。關(guān)鍵詞心理資本；大學(xué)生；應(yīng)對(duì)方式；職業(yè)決策自我效能感IIIIIIILFLLLLLRLLRIIIIIY2096030PSYCHOLOGICALCAPITALISPOSITIVE．3THECOLLEGESTUDENTS’PSYCHOLOGICALCAPITALANDITSDIMENSIONDIFFERINSOMEDEMOGRAPHICVARIABLES；ESPECIALLYTHEDIFFERENCESONWHETHERJOINEDSTUDENTS’UNION,FAMILYMEMBERS’RELATIONSHIPS，INTERPERSONALRELATIONSHIPS，ACADEMICPERFORMANCE，SCHOOLTYPESAREVERYSIGNIFICANT．4PSYCHOLOGICALCAPITALINCLUDINGITSDIMENSIONSALEPOSITIVELYCORRELATEDWITHPOSITIVECOPINGSTYLE，ANDNEGATIVELYCORRELATEDWITHTHENEGATIVECOPINGSTYLE．5THECOLLEGESTUDENTS’PSYCHOLOGICALCAPITALINCLUDINGITSDIMENSIONSANDCAREERDECISIONMAKINGSELFEFFICACYINCLUDINGITSDIMENSIONSARESIGNIFICANTLYCORRELATED；PSYCHOLOGICALCAPITALHASASIGNIFICANTPREDICTIVEEFFECTONCAREERDECISIONMAKINGSELFEFFICACY．THISSTUDYEXTENDSPSYCHOLOGICALCAPITALRESEARCHTOCOLLEGESTUDENTS，NATIONWIDESAMPLINGHELPSTOKNOWTHEGENERALLYSTATUSOFPSYCHOLOGICALCAPITALOFCHINESECOLLEGESTUDENTS，ANDASERIESOFDISCUSSIONSBASEDONTHERESULTSWOULDHELPTOENHANCETHEATTENTIONOFUNIVERSITIESANDEDUCATORSONCOLLEGESTUDENTS’PSYCHOLOGICALCAPITAL，ANDALSOIT’SHELPFULTOPROMOTETHEDEVELOPMENTOFPSYCHOLOGICALEDUCATIONINUNIVERSITIES．WHAT’SMORE，THESTUDYENRICHEDPSYCHOLOGICALCAPITAL’STHEORYANDRESEARCH，ITHASTHEORETICALANDPRACTICALSIGNIFICANCE．KEYWORDSPSYCHOLOGICALCAPITAL；COLLEGESTUDENTS；COPINGSTYLE；CAREERDECISIONMAKINGSELFEFFICACYIII

簡(jiǎn)介：學(xué)號(hào)2007020164研究生姓名李莉聯(lián)系電話15866760373EMAIL598797173QQCOM所在院系心理學(xué)院

簡(jiǎn)介：大學(xué)生呼吸肌耐力與肺活量、最大攝氧量的相關(guān)研究論文作者簽名曼紐莖曼指導(dǎo)教師簽名論文評(píng)閱人1評(píng)閱人2評(píng)閱人3評(píng)閱人4評(píng)閱人5答辯委員會(huì)主席委員1委員2委員3委員4委員5答辯日期瑚厶∈邁致謝時(shí)光如梭，轉(zhuǎn)眼間兩年的碩士學(xué)習(xí)生活即將過去，在這段難忘的日子里，曾得到許多老師、同學(xué)、朋友及家人的關(guān)心幫助，作者謹(jǐn)以本文獻(xiàn)給他們，以此表達(dá)深深地祝福和誠(chéng)摯的謝意。首先感謝我的導(dǎo)師徐玉明教授。徐老師知識(shí)淵博、學(xué)術(shù)嚴(yán)謹(jǐn)。他忘我的工作精神、誠(chéng)摯的處事態(tài)度、風(fēng)趣幽默的語言、細(xì)致入微的關(guān)懷給我留下了深刻的印象，使我受益匪淺。從論文開題、指標(biāo)選擇、測(cè)試設(shè)計(jì)到最后的撰寫論文，徐老師都給與了極大的幫助和關(guān)懷，在此致以衷心的敬意和深深的謝意感謝我的同F(xiàn)J師兄弟陳玉平、羅宇婷、鄭冬霞、張亞新、喬暢等，感謝你們對(duì)我學(xué)7工作上的幫助，正是有了你們的幫助和鼓勵(lì)，我才能順利完成實(shí)驗(yàn)，在此對(duì)你們表示深深的謝意。感謝同班好友王亞圣、劉晴、李征等，在我碩士期間陪我度過的許多美好時(shí)光，每當(dāng)想起這些，我的心里充滿溫暖和熱情，在這里祝你們前程似錦，幸福美好。感謝我的家人，尤其是我的父親和母親二十多年來給與我的大力支持和無私的愛最后，向在百忙中抽出時(shí)間參與我論文答辯和評(píng)審的各位專家學(xué)者表示由衷的感謝

簡(jiǎn)介：分類號(hào)密級(jí)無UDC單位代碼10118山西師范大學(xué)研究生碩士學(xué)位論文大學(xué)生體育鍛煉與情緒管理能力的相關(guān)性研究屈振榮屈振榮指導(dǎo)教師張海潮教授山西師范大學(xué)申請(qǐng)學(xué)位級(jí)別教育學(xué)碩士專業(yè)名稱體育教育訓(xùn)練學(xué)論文提交日期2015年4月7日論文答辯日期2015年5月24日學(xué)位授予單位山西師范大學(xué)學(xué)位授予日期2015年6月4日答辯委員會(huì)主席邱招義教授評(píng)閱人張廣林教授崔浩瀾教授2015年5月20日中文摘要I論文題目大學(xué)生體育鍛煉與情緒管理能力的相關(guān)性研究專業(yè)體育教育訓(xùn)練學(xué)碩士生屈振榮簽名指導(dǎo)教師張海潮簽名摘要生活滿意感和生命意義感是衡量大千世界中個(gè)體生活質(zhì)量的綜合性心理指標(biāo)，其中對(duì)生活滿意感和生命意義感評(píng)定的重要因素就是對(duì)情緒管理能力的控制和調(diào)節(jié)。因而，提高生活滿意感和生命意義感的總體水平，加強(qiáng)情緒管理的控制，對(duì)于提高人們的生活質(zhì)量就顯得尤為重要。當(dāng)代大學(xué)生具有心理變化大，心理素質(zhì)不穩(wěn)定，可塑性較強(qiáng)的心理特征，受社會(huì)、周圍環(huán)境等各方面影響和自身心理特點(diǎn)的制約，易產(chǎn)生多重負(fù)面心理體驗(yàn)，如空虛、寂寞、迷茫等，處于心理亞健康狀態(tài)，為各種心理問題埋下隱患。因此，重視當(dāng)代大學(xué)生的生活滿意感、生命意義感，提高大學(xué)生的情緒管理能力，通過體育鍛煉的方式達(dá)到消除各種消極的心理影響的效果，是預(yù)防大學(xué)生心理疾病，優(yōu)化大學(xué)生心理健康的需要。科學(xué)的體育鍛煉有利于調(diào)節(jié)不良情緒，增進(jìn)積極情感，促進(jìn)人的身心健康。但是當(dāng)代大學(xué)生對(duì)體育鍛煉強(qiáng)身健心功效的認(rèn)識(shí)程度并不深刻，其治療心理亞健康，預(yù)防心理疾病的心理學(xué)功效并未得到廣泛認(rèn)識(shí)和重視，相關(guān)研究論證也十分欠缺。因此，本論文所要研究的核心內(nèi)容是通過探討大學(xué)生參與體育鍛煉與情緒管理能力之間的相互關(guān)系，為改善大學(xué)生的生活質(zhì)量，提高大學(xué)生的情緒管理能力，優(yōu)化大學(xué)生的心理健康提供理論參考，為普通高校制定大學(xué)生情緒管理能力和心理健康教育方案提供有益的借鑒，為體育鍛煉的強(qiáng)身健心功效提供實(shí)證研究資料。采用大學(xué)生情緒管理能力問卷、體育活動(dòng)等級(jí)量表和體育鍛煉情況調(diào)查表。問卷信、效度較高，對(duì)山西省五所普通高等學(xué)校的1000名在校大學(xué)生進(jìn)行調(diào)查，運(yùn)用SPSS170數(shù)據(jù)統(tǒng)計(jì)軟件進(jìn)行數(shù)據(jù)分析。結(jié)論如下（1）大學(xué)生參加體育鍛煉的意識(shí)較好，但達(dá)到標(biāo)準(zhǔn)體育鍛煉人口的人數(shù)較少，即參加體育鍛煉的運(yùn)動(dòng)量不太理想。（2）體育鍛煉的時(shí)間、強(qiáng)度、頻率與情緒管理能力均存在顯著相關(guān)性，且體育鍛煉強(qiáng)度和頻率與情緒管理能力的相關(guān)系數(shù)較高?！娟P(guān)鍵詞】大學(xué)生體育鍛煉情緒管理能力相關(guān)性研究【論文類型】關(guān)系型研究

簡(jiǎn)介：AI遼SEARCHONTHEI迮LATIONBETWEENBODYIMAGEANDWEIGHTLOSSBEHAVIOROFCOLLEGESTUDENTSADISSERTATIONSUBMITTEDTOSOUTHEASTUNIVERSITYFORTHEACADEMICDEGREEOFMASTEROFEDUCATIONBYZHANGYANSUPERVISEDBYSUPERVISEDBYPROF．HELUNSCHOOLOFHUMANITIESSOUTHEASTUNIVERSITYDECEMBER2009摘要摘要所謂身體意象，是指對(duì)自己身體的意象，是關(guān)于自己的外貌、身體型象在他人面前是怎樣的一種想象，也包括別人是如何看待自己的身體功能方面的意象。也就是說，意象一方面是自己對(duì)自己身體的想象，另一方面，是別人對(duì)自己身體的想象。大學(xué)生正處于青春期，是自我概念發(fā)展的關(guān)鍵期，正面臨著自我同一性與發(fā)展危機(jī)，在這一階段，個(gè)體開始關(guān)注自我，其自我意識(shí)迅速發(fā)展并逐步形成。這一時(shí)期急劇的身體成熟，使得他們更加注意關(guān)注自己的身體，身體滿意度逐漸下降，減肥行為也隨之增加，他們成為追求苗條體型、追求時(shí)尚的主要人群之一。傳統(tǒng)觀念的人們，大多數(shù)會(huì)認(rèn)為肥胖者們才會(huì)去減肥，殊不知在如今大學(xué)生中存在很大～部分群體，盡管在別人的眼里他們的體重正常或偏瘦，但他們?nèi)匀贿€存在減肥行為，而這些與自身體重、體型根本無關(guān)。因此，本文從大學(xué)生著手，探討他們的減肥行為，力圖探索這一群體的身體意象與減肥行為的關(guān)系，分析其減肥行為背后的心理動(dòng)因。本研究采用采取整群抽樣的方法，各年級(jí)抽取兩個(gè)班的學(xué)生進(jìn)行檢查。以STORZ．GREENE體型圖、多維自我體象關(guān)系問卷MBSRQ和自編減肥行為問卷為工具對(duì)766名大學(xué)生進(jìn)行了調(diào)查，得出結(jié)論L、大多數(shù)學(xué)生對(duì)于他們的身體意象是滿意，只有少數(shù)學(xué)生不滿意，甚至存在負(fù)性扭曲現(xiàn)象。大學(xué)生身體意象在性別、年級(jí)、專業(yè)、生源及BML分組方面存在差異。2、總體看來，大部分學(xué)生傾向于選擇中間范圍的體重作為自己實(shí)際體重和理想體重。這些認(rèn)知在性別、年級(jí)、專業(yè)、生源及BML分組方面存在差異。3、大學(xué)生存在減肥行為，女生表現(xiàn)出更高的減肥意向與關(guān)注；在減肥群體中，大多是正常體重的群體，而且很多都是心理認(rèn)同的肥胖，和生理指數(shù)關(guān)系不大。4、大學(xué)生身體意象與減肥行為存在顯著負(fù)相關(guān)，越是低程度的身體意象，減肥意向和減肥關(guān)注程度越高。5、大學(xué)生對(duì)自己身體的不滿意和對(duì)體重的認(rèn)知錯(cuò)覺是導(dǎo)致大學(xué)生減肥行為的重要原因之關(guān)鍵詞；大學(xué)生身體意象減肥行為

簡(jiǎn)介：碩士學(xué)位論文大一學(xué)生網(wǎng)絡(luò)成癮與人格特質(zhì)、孤獨(dú)、自尊及學(xué)校適應(yīng)性的相關(guān)性研究ONRELEVANCEOFFIRSTYEARCOLLEGESTUDENTS’INTERNETADDICTIONTOPERSONALTRAITS，LONELINESS，SELFESTEEMANDCOLLEGEADJUSTMENT作者齊曉穎指導(dǎo)教師學(xué)科專業(yè)研究方向?qū)W位類型姜英杰副教授’發(fā)展與教育心理學(xué)網(wǎng)絡(luò)成癮‘高校教師東北師范大學(xué)學(xué)位評(píng)定委員會(huì)2010年5月獨(dú)創(chuàng)性聲明本人聲明所呈交的學(xué)位論文是本人在導(dǎo)師指導(dǎo)下進(jìn)行的研究工作及取得的研究成果。據(jù)我所知，除了文中特LLIJL以標(biāo)注和致謝的地方外，論文中不包含其他人已經(jīng)發(fā)表或撰寫過的研究成果，也不包含為獲得東北師范大學(xué)或其他教育機(jī)構(gòu)的學(xué)位或證書而使用過的材料。與我一同工作的同志對(duì)本研究所做的任何貢獻(xiàn)均已在論文中作了明確的說明并表示謝意?！畬W(xué)位論文作者簽名學(xué)位論文版權(quán)使用授權(quán)書本學(xué)位論文作者完全了解東北師范大學(xué)有關(guān)保留、使用學(xué)位論文的規(guī)定，即東北師范大學(xué)有權(quán)保留并向國(guó)家有關(guān)部門或機(jī)構(gòu)送交學(xué)位論文的復(fù)印件和磁盤，允許論文被查閱和借閱。本人授權(quán)東北師范大學(xué)可以將學(xué)位論文的全部或部分內(nèi)容編入有關(guān)數(shù)據(jù)庫(kù)進(jìn)行檢索，可以采用影印、縮印或其它復(fù)制手段保存、匯編學(xué)位論文。保密的學(xué)位論文在解密后適用本授權(quán)書學(xué)位論文作者簽名壺隆題指導(dǎo)教師簽名日期日期學(xué)位論文作者畢業(yè)后去向電話郵編13944691028132101

簡(jiǎn)介：四川師范大學(xué)學(xué)位論文獨(dú)創(chuàng)性聲明本人聲明所呈交學(xué)位論文太眾佳媒佳直信星懟丕回眭別太堂生佳直叢翅髭響的扭差比較巫究二叢四叫垣范太堂趁倒，是本人在導(dǎo)師龐面主教援指導(dǎo)下，獨(dú)立進(jìn)行研究工作所取得的成果。除文中已經(jīng)注明引用的內(nèi)容外，本論文不含任何其他個(gè)人或集體己經(jīng)發(fā)表或撰寫過的作品或成果。對(duì)本文的研究做出重要貢獻(xiàn)的個(gè)人和集體，均已在文中以明確方式標(biāo)明。本聲明的法律結(jié)果由本人承擔(dān)。本人承諾己提交的學(xué)位論文電子版與論文紙本的內(nèi)容一致。如因不符而引起的學(xué)術(shù)聲譽(yù)上的損失由本人自負(fù)。學(xué)位論文作者稀，『13芬及簽字日期2．O／B年6月多’日四川師范大學(xué)學(xué)位論文版權(quán)使用授權(quán)書本人同意所撰寫學(xué)位論文的使用授權(quán)遵照學(xué)校的管理規(guī)定學(xué)校作為申請(qǐng)學(xué)位的條件之一，學(xué)位論文著作權(quán)擁有者須授權(quán)所在大學(xué)擁有學(xué)位論文的部分使用權(quán)，即1己獲學(xué)位的研究生必須按學(xué)校規(guī)定提交印刷版和電子版學(xué)位論文，可以將學(xué)位論文的全部或部分內(nèi)容編入有關(guān)數(shù)據(jù)庫(kù)供檢索；2為教學(xué)、科研和學(xué)術(shù)交流目的，學(xué)?？梢詫⒐_的學(xué)位論文或解密后的學(xué)位論文作為資料在圖書館、資料室等場(chǎng)所或在有關(guān)網(wǎng)絡(luò)上供閱讀、瀏覽。本人授權(quán)萬方數(shù)據(jù)電子出版社將本學(xué)位論文收錄到中國(guó)學(xué)位論文全文數(shù)據(jù)庫(kù)，并通過網(wǎng)絡(luò)向社會(huì)公眾提供信息服務(wù)。同意按相關(guān)規(guī)定享受相關(guān)權(quán)益。保密的學(xué)位論文在解密后適用本授權(quán)書學(xué)位論文作者簽名私LF私簽字日期加佑年多月≯≥日翩魏寵簽字日期勃鰣年莎月≥JJL日萬方數(shù)據(jù)四川師范大學(xué)碩士學(xué)位論文II萬方數(shù)據(jù)

簡(jiǎn)介：??????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????????“?ABC1DEF??BG???HIJKLMN?

簡(jiǎn)介：獨(dú)創(chuàng)性聲明本人聲明所呈交的學(xué)位論文是本人在導(dǎo)師指導(dǎo)下進(jìn)行的研究工作及取得的研究成果。論文中除了特別加以標(biāo)注和致謝的地方外，不包含其他人已經(jīng)發(fā)表或撰寫過的研究成果，也不包含為獲得其他教育機(jī)構(gòu)的學(xué)位或證書而使用過的材料。與我一同工作的同志對(duì)本研究所做的任何貢獻(xiàn)均己在論文中作了明確的說明并表示謝意。學(xué)位論文作者簽名．I琵釋簽字日期護(hù)易年鄉(xiāng)月弓P日學(xué)位論文版權(quán)使用授權(quán)書本學(xué)位論文作者完全了解J匕塞跡獨(dú)醫(yī)堂隨有關(guān)保存、使用學(xué)位論文的管理辦法。有權(quán)保留并向國(guó)家有關(guān)部門或機(jī)構(gòu)送交論文的復(fù)印件和磁盤，允許論文被查閱和借閱。本人授權(quán)JE塞垃獨(dú)醫(yī)堂醫(yī)可以將學(xué)位論文的全部或部分內(nèi)容編入有關(guān)數(shù)據(jù)庫(kù)進(jìn)行檢索，可以采用影印、縮印或掃描等復(fù)制手段保存、匯編學(xué)位論文。保密的學(xué)位論文在解密后適用本授權(quán)書學(xué)位論文作者虢琵智導(dǎo)師虢簽字日期2．O16年鄉(xiāng)月；P日簽字日期砂F6年學(xué)位論文作者畢業(yè)后去向工作單位通訊地址電話郵編．B1寧欠≥月字口日平會(huì)對(duì)論文質(zhì)量產(chǎn)生顯著影響，因此認(rèn)為增加與發(fā)達(dá)國(guó)家的合著是提升合著論文質(zhì)量的途徑之一；5基金資助情況會(huì)對(duì)論文質(zhì)量產(chǎn)生顯著正向影響，提示可以通過科研項(xiàng)目之間的學(xué)術(shù)交流與合作，提高科研產(chǎn)出的質(zhì)量；6相對(duì)于傳統(tǒng)統(tǒng)計(jì)方法而言，結(jié)構(gòu)方程模型考慮了多個(gè)因素的協(xié)同作用，同時(shí)考慮了因變量的協(xié)同效應(yīng)，避免了犯二類錯(cuò)誤，評(píng)價(jià)結(jié)果更為客觀。同時(shí)本研究也存在一定的不足，研究將在多個(gè)醫(yī)學(xué)研究領(lǐng)域進(jìn)行實(shí)證分析，確保研究結(jié)果在醫(yī)學(xué)領(lǐng)域的適用性，同時(shí)將進(jìn)一步通過專家咨詢，確定各指標(biāo)對(duì)論文質(zhì)量的影響系數(shù)。關(guān)鍵詞國(guó)際合著；論文質(zhì)量；相關(guān)分析；結(jié)構(gòu)方程

簡(jiǎn)介：STABLEDETERMINISTICMULTITHREADINGTHROUGHSCHEDULEMEMOIZATIONHEMINGCUI,JINGYUEWU,CHIACHETSAI,JUNFENGYANG{HEMING,JINGYUE,CT2459,JUNFENG}CSCOLUMBIAEDUCOMPUTERSCIENCEDEPARTMENTCOLUMBIAUNIVERSITYNEWYORK,NY10027ABSTRACTADETERMINISTICMULTITHREADINGDMTSYSTEMELIMINATESNONDETERMINISMINTHREADSCHEDULING,SIMPLIFYINGTHEDEVELOPMENTOFMULTITHREADEDPROGRAMSHOWEVER,EXISTINGDMTSYSTEMSAREUNSTABLETHEYMAYFORCEAPROGRAMTOADVENTUREINTOVASTLYDIFFERENTSCHEDULESEVENFORSLIGHTLYDIFFERENTINPUTSOREXECUTIONENVIRONMENTS,DEFEATINGMANYBENEFITSOFDETERMINISMMOREOVER,FEWEXISTINGDMTSYSTEMSWORKWITHSERVERPROGRAMSWHOSEINPUTSARRIVECONTINUOUSLYANDNONDETERMINISTICALLYTERNISASTABLEDMTSYSTEMTHEKEYNOVELTYINTERNISTHEIDEAOFSCHEDULEMEMOIZATIONTHATMEMOIZESPASTWORKINGSCHEDULESANDREUSESTHEMONFUTUREINPUTS,MAKINGPROGRAMBEHAVIORSSTABLEACROSSDIFFERENTINPUTSASECONDNOVELTYINTERNISTHEIDEAOFWINDOWINGTHATEXTENDSSCHEDULEMEMOIZATIONTOSERVERPROGRAMSBYSPLITTINGCONTINUOUSREQUESTSTREAMSINTOWINDOWSOFREQUESTSOURTERNIMPLEMENTATIONRUNSONLINUXITOPERATESASUSERSPACESCHEDULERS,REQUIRINGNOCHANGESTOTHEOSANDONLYAFEWLINESOFCHANGESTOTHEAPPLICATIONPROGRAMSWEEVALUATEDTERNONADIVERSESETOF14PROGRAMSEG,APACHEANDMYSQLWITHREALANDSYNTHETICWORKLOADSOURRESULTSSHOWTHATTERNISEASYTOUSE,MAKESPROGRAMSMOREDETERMINISTICANDSTABLE,ANDHASREASONABLEOVERHEAD1INTRODUCTIONMULTITHREADEDPROGRAMSAREDIFFICULTTOWRITE,TEST,ANDDEBUGAKEYREASONISNONDETERMINISMDIFFERENTRUNSOFAMULTITHREADEDPROGRAMMAYSHOWDIFFERENTBEHAVIORS,DEPENDINGONHOWTHETHREADSINTERLEAVE35TWOMAINFACTORSMAKETHREADSINTERLEAVENONDETERMINISTICALLYTHEFIRSTISSCHEDULING,HOWTHEOSANDHARDWARESCHEDULETHREADSSCHEDULINGNONDETERMINISMISNOTESSENTIALANDCANBEELIMINATEDWITHOUTAFFECTINGCORRECTNESSFORMOSTPROGRAMSTHESECONDISINPUT,WHATDATAINPUTDATAARRIVESATWHATTIMEINPUTTIMINGINPUTNONDETERMINISMSOMETIMESISESSENTIALBECAUSEMAJORCHANGESININPUTSREQUIREDIFFERENTSCHEDULESHOWEVER,FREQUENTLYINPUTNONDETERMINISMISNOTESSENTIALANDTHESAMESCHEDULECANBEUSEDTOPROCESSMANYDIFFERENTINPUTS§22WEBELIEVENONESSENTIALNONDETERMINISMSHOULDBEELIMINATEDINFAVOROFDETERMINISMDETERMINISTICMULTITHREADINGDMTSYSTEMS13,22,41MAKETHREADSMOREDETERMINISTICBYELIMINATINGSCHEDULINGNONDETERMINISMSPECIFICALLY,THEYCONSTRAINAMULTITHREADEDPROGRAMSUCHTHATITALWAYSUSESTHESAMETHREADSCHEDULEFORTHESAMEINPUTBYDOINGSO,THESESYSTEMSMAKEPROGRAMBEHAVIORSREPEATABLE,INCREASETESTINGCONFIDENCE,ANDEASEBUGREPRODUCTIONUNFORTUNATELY,THOUGHEXISTINGDMTSYSTEMSELIMINATESCHEDULINGNONDETERMINISM,THEYDONOTREDUCEINPUTNONDETERMINISMINFACT,THEYMAYAGGRAVATETHEEFFECTSOFINPUTNONDETERMINISMBECAUSEOFTHEIRDESIGNLIMITATIONWHENSCHEDULINGTHETHREADSTOPROCESSANINPUT,THEYCONSIDERONLYTHISINPUTANDIGNOREPREVIOUSSIMILARINPUTSTHISSTATELESSDESIGNMAKESSCHEDULESOVERDEPENDENTONINPUTS,SOTHATASLIGHTCHANGETOINPUTSMAYFORCEAPROGRAMTOADVENTUREINTOAVASTLYDIFFERENT,POTENTIALLYBUGGYSCHEDULE,DEFEATINGMANYBENEFITSOFDETERMINISMWECALLTHISTHEINSTABILITYPROBLEMTHISPROBLEMISCONFIRMEDBYOURRESULTS§821FROMANEXISTINGDMTSYSTEM13INFACT,EVENWITHTHESAMEINPUT,EXISTINGDMTSYSTEMSMAYSTILLFORCEAPROGRAMINTODIFFERENTSCHEDULESFORMINORCHANGESINTHEEXECUTIONENVIRONMENTSUCHASPROCESSORTYPEANDSHAREDLIBRARYTHUS,DEVELOPERSMAYNOLONGERBEABLETOREPRODUCEBUGSBYRUNNINGTHEIRPROGRAMONTHEBUGINDUCINGINPUT,BECAUSETHEIRMACHINEMAYDIFFERFROMTHEMACHINEWHERETHEBUGOCCURREDTHISPAPERPRESENTSTERN,ASCHEDULECENTRIC,STATEFULDMTSYSTEMITADDRESSESTHEINSTABILITYPROBLEMUSINGANIDEACALLEDSCHEDULEMEMOIZATIONTHATMEMOIZESPASTWORKINGSCHEDULESANDREUSESTHEMFORFUTUREINPUTSSPECIFICALLY,TERNMAINTAINSACACHEOFPASTSCHEDULESANDTHEINPUTCONSTRAINTSREQUIREDTOREUSETHESESCHEDULESWHENANINPUTARRIVES,TERNCHECKSTHEINPUTAGAINSTTHEMEMOIZEDCONSTRAINTSFORACOMPATIBLESCHEDOURCHOICEOFSCHEDULEREPRESENTATIONINTERN§22,ANDWHYWECANREUSESCHEDULESACROSSINPUTS§2321THEINSTABILITYPROBLEMADMTSYSTEMIS,CONCEPTUALLY,AFUNCTIONTHATMAPSANINPUTITOASCHEDULESTHEPROPERTIESOFTHISFUNCTIONARETHATTHESAMEISHOULDMAPTOTHESAMESANDTHATSISAFEASIBLESCHEDULEFORPROCESSINGIASTABLEDMTSYSTEMSUCHASTERNHASANADDITIONALPROPERTYITMAPSSIMILARINPUTSTOTHESAMESCHEDULEEXISTINGDMTSYSTEMS,HOWEVER,TENDTOMAPSIMILARINPUTSTODIFFERENTSCHEDULES,THUSSUFFERINGFROMTHEINSTABILITYPROBLEMWEARGUETHATTHISPROBLEMISINHERENTINEXISTINGDMTSYSTEMSBECAUSETHEYARESTATELESSTHEYMUSTPROVIDETHESAMESCHEDULEFORANINPUTACROSSDIFFERENTRUNS,USINGINFORMATIONONLYFROMTHECURRENTRUNTOFORCETHREADSTOCOMMUNICATEEG,ACQUIRELOCKSORACCESSSHAREDMEMORYDETERMINISTICALLY,EXISTINGDMTSYSTEMSCANNOTRELYONPHYSICALCLOCKSINSTEAD,THEYMAINTAINALOGICALCLOCKPERTHREADTHATTICKSDETERMINISTICALLYBASEDONTHECODETHISTHREADHASRUNMOREOVER,THREADSMAYCOMMUNICATEONLYWHENTHEIRLOGICALCLOCKSHAVEDETERMINISTICVALUESEG,SMALLESTACROSSTHELOGICALCLOCKSOFALLTHREADS41BYINDUCTION,LOGICALCLOCKSMAKETHREADSDETERMINISTICHOWEVER,THEPROBLEMWITHLOGICALCLOCKSISTHATFOREFFICIENCY,THEYMUSTTICKATROUGHLYTHESAMERATETOPREVENTATHREADWITHASLOWERCLOCKFROMSTARVINGOTHERSTHUS,EXISTINGDMTSYSTEMSHAVETOTIETHEIRLOGICALCLOCKSTOLOWLEVELINSTRUCTIONSEXECUTEDEG,COMPLETEDLOADS41CONSEQUENTLY,ASMALLCHANGETOTHEINPUTOREXECUTIONENVIRONMENTMAYALTERAFEWINSTRUCTIONSEXECUTED,INTURNALTERINGTHELOGICALCLOCKSANDSUBSEQUENTTHREADCOMMUNICATIONSTHATIS,ASMALLCHANGETOTHEINPUTOREXECUTIONENVIRONMENTMAYCASCADEINTOAMUCHDIFFERENTEG,CORRECTVSBUGGYSCHEDULE22SCHEDULEREPRESENTATIONANDDETERMINISMPREVIOUSDMTSYSTEMSHAVECONSIDEREDTWOTYPESOFSCHEDULES1ADETERMINISTICORDEROFSHAREDMEMORYACCESSES13,22AND2ASYNCHRONIZATIONORDERIE,ATOTALORDEROFSYNCHRONIZATIONOPERATIONS41THEFIRSTTYPEOFSCHEDULESARETRULYDETERMINISTICEVENIFTHEREARERACES,BUTTHEYARECOSTLYTOENFORCEONCOMMODITYHARDWAREEG,UPTO10TIMESOVERHEAD13THESECONDTYPECANBEEFFICIENTLYENFORCEDEG,16OVERHEAD41BECAUSEMOSTCODEISNOTSYNCHRONIZATIONCODEANDCANRUNINPARALLELHOWEVER,THEYAREDETERMINISTICONLYFORINPUTSTHATLEADTORACEFREERUNS41,46TERNREPRESENTSSCHEDULESASSYNCHRONIZATIONORDERSFOREFFICIENCYANADDITIONALBENEFITISTHATSYNCHRONIZATIONORDERSCANBEREUSEDMOREFREQUENTLYTHANMEMORYACCESSORDERSCFNEXTSUBSECTIONMOREOVER,RESEARCHERSHAVEFOUNDTHATMANYCONCURRENCYERRORSARENOTDATAPROGRAMINPUTCONSTRAINTSFORSCHEDULEREUSEPBZIP2SAMENUMBEROFFILEBLOCKSNUMBLOCKSORBANDTHREADSPAPACHEFORGROUPSOFTYPICALHTTPGETREQUESTS,SAMECACHESTATUSANDRESPONSESIZESFFTSAMENUMBEROFTHREADSPLUSAMENUMBEROFTHREADSP,SIZEOFTHEMATRIXN,ANDBLOCKSIZEBBARNESSAMENUMBEROFTHREADSNPROCANDVALUESOFVARIABLESDTIMEANDTSTOPTABLE1INPUTCONSTRAINTSOFFIVEPROGRAMSTOREUSESCHEDULESIDENTIFIERSWITHOUTADASHARECONFIGURATIONVARIABLES,ANDTHOSEWITHADASHARECOMMANDLINEOPTIONSRACES,BUTATOMICITYANDORDERVIOLATIONS39THESEERRORSCANBEDETERMINISTICALLYREPRODUCEDORAVOIDEDUSINGONLYSYNCHRONIZATIONORDERSALTHOUGHDATARACESMAYSTILLMAKERUNSWHICHREUSESCHEDULESNONDETERMINISTIC,TERNISLESSPRONETOTHISPROBLEMTHANEXISTINGDMTSYSTEMS41BECAUSEITHASTHEFLEXIBILITYTOSELECTSCHEDULESIFITDETECTSARACEINAMEMOIZEDSCHEDULE,ITCANSIMPLYDISCARDTHISSCHEDULEANDMEMOIZEANOTHERTHISSELECTIONTASKISOFTENEASYBECAUSEMOSTSCHEDULESARERACEFREEINRARECASES,TERNMAYBEUNABLETOFINDARACEFREESCHEDULE,RESULTINGINNONDETERMINISTICRUNSHOWEVER,WEARGUETHATINPUTNONDETERMINISMCANNOTBEFULLYELIMINATEDANYWAY,SOWEMAYASWELLTOLERATESOMESCHEDULINGNONDETERMINISM,FOLLOWINGTHEENDTOENDARGUMENT23WHYCANWEREUSESCHEDULESTHISSUBSECTIONPRESENTSANINTUITIVEANDANEMPIRICALARGUMENTTOSUPPORTOURINSIGHTTHATWECANFREQUENTLYREUSESCHEDULESFORMANYPROGRAMS/WORKLOADSINTUITIVELY,SYNCHRONIZATIONOPERATIONSMAPTODEVELOPERINTENTSOFINTERTHREADCONTROLFLOWBYENFORCINGTHESAMESYNCHRONIZATIONORDER,WEFIXTHESAMEINTERTHREAD“PATH,”BUTSTILLALLOWMANYDIFFERENTINPUTSTOFLOWDOWNTHISPATHTHISOBSERVATIONISSIMILARLYMADEFORSEQUENTIALPATHS11,12,26TOEMPIRICALLYVALIDATEOURINSIGHT,WESTUDIEDTHEINPUTCONSTRAINTSTOREUSESCHEDULESFORFIVEPROGRAMS,INCLUDINGAPARALLELCOMPRESSIONUTILITYPBZIP2THEAPACHEWEBSERVERANDTHREESCIENTIFICPROGRAMSFFT,LU,ANDBARNESINSPLASH2TABLE1SHOWSTHERESULTSFORALLPROGRAMSSTUDIEDWEFOUNDTHATTHEINPUTCONSTRAINTSWEREOFTENGENERAL,ALLOWINGFREQUENTREUSESOFSCHEDULESFORINSTANCE,PBZIP2CANUSETHESAMESCHEDULETOCOMPRESSMANYDIFFERENTFILES,ASLONGASTHENUMBEROFTHREADSANDTHENUMBEROFFILEBLOCKSREMAINTHESAME3OVERVIEWOURDESIGNOFTERNADHERESTOTHEFOLLOWINGGOALS

簡(jiǎn)介：TRUSTANDPROTECTIONINTHEILLINOISBROWSEROPERATINGSYSTEMSHUOTANG,HAOHUIMAI,SAMUELTKINGUNIVERSITYOFILLINOISATURBANACHAMPAIGNABSTRACTCURRENTWEBBROWSERSARECOMPLEX,HAVEENORMOUSTRUSTEDCOMPUTINGBASES,ANDPROVIDEATTACKERSWITHEASYACCESSTOMODERNCOMPUTERSYSTEMSINTHISPAPERWEINTRODUCETHEILLINOISBROWSEROPERATINGSYSTEMIBOS,ANEWOPERATINGSYSTEMANDANEWBROWSERTHATREDUCESTHETRUSTEDCOMPUTINGBASEFORWEBBROWSERSINOURARCHITECTUREWEEXPOSEBROWSERLEVELABSTRACTIONSATTHELOWESTSOFTWARELAYER,ENABLINGUSTOREMOVEALMOSTALLTRADITIONALOSCOMPONENTSANDSERVICESFROMOURTRUSTEDCOMPUTINGBASEBYMAPPINGBROWSERABSTRACTIONSTOHARDWAREABSTRACTIONSDIRECTLYWESHOWTHATTHISARCHITECTUREISFLEXIBLEENOUGHTOENABLENEWBROWSERSECURITYPOLICIES,CANSTILLSUPPORTTRADITIONALAPPLICATIONS,ANDADDSLITTLEOVERHEADTOTHEOVERALLBROWSINGEXPERIENCE1INTRODUCTIONWEBBASEDAPPLICATIONSWEBAPPS,BROWSERS,ANDOPERATINGSYSTEMSHAVEBECOMEPOPULARTARGETSFORATTACKERSOFCOMPUTERSYSTEMSVULNERABILITIESINWEBAPPSAREWIDESPREADANDINCREASINGFOREXAMPLE,CROSSSITESCRIPTINGXSS,WHICHISEFFECTIVELYAFORMOFSCRIPTINJECTIONINTOAWEBAPP,RECENTLYOVERTOOKTHEUBIQUITOUSBUFFEROVERFLOWASTHEMOSTCOMMONSECURITYVULNERABILITY50VULNERABILITIESINWEBBROWSERSARELESSCOMMONTHANWEBAPPVULNERABILITIES,BUTSTILLOCCUROFTENFOREXAMPLE,IN2009INTERNETEXPLORER,CHROME,SAFARI,ANDFIREFOXHAD349NEWSECURITYVULNERABILITIES4,ANDATTACKERSEXPLOITBROWSERSCOMMONLY53,37,42,41,4VULNERABILITIESINLIBRARIES,SYSTEMSERVICES,ANDOPERATINGSYSTEMSARELESSCOMMONTHANVULNERABILITIESINBROWSERS,BUTARESTILLPROBLEMATICFORMODERNSYSTEMSFOREXAMPLE,GLIBC,GTK,X,ANDLINUXHAD114NEWSECURITYVULNERABILITIESIN20091,ANDIN2009THEMOSTCOMMONLYATTACKEDVULNERABILITYWASAREMOTECODEEXECUTIONBUGINTHEWINDOWSKERNEL4HOWEVER,NOTALLATTACKSONWEBAPPS,BROWSERS,ANDOPERATINGSYSTEMSAREEQUALLYVIRULENTATTHETOPOFTHECOMPUTERSTACK,ATTACKSONWEBAPPS,SUCHASXSS,OPERATEWITHINCURRENTBROWSERSECURITYPOLICIESTHATCONTAINTHEDAMAGETOTHEVULNERABLEWEBAPPMOVINGDOWNTHECOMPUTERSTACK,ATTACKSONBROWSERSCANCAUSEMOREDAMAGEBECAUSEASUCCESSFULATTACKGIVESTHEATTACKERACCESSTOBROWSERDATAFORALLWEBAPPSANDACCESSTOOTHERRESOURCESONTHESYSTEMATTHELOWESTLAYERSOFTHECOMPUTERSTACK,ATTACKSONLIBRARIES,SHAREDSYSTEMSERVICES,ANDOPERATINGSYSTEMSARETHEMOSTSERIOUSATTACKSBECAUSEATTACKERSCANACCESSARBITRARYSTATESANDEVENTS,GIVINGTHEMCOMPLETECONTROLOFTHESYSTEMOVERALL,THESETRENDSINDICATETHATVULNERABILITIESHIGHERINTHECOMPUTERSTACKAREMORECOMMON,BUTVULNERABILITIESLOWERINTHECOMPUTERSTACKPROVIDEATTACKERSWITHMORECONTROLANDAREMOREDAMAGINGINTHISPAPERWEFOCUSONPREVENTINGANDCONTAININGATTACKSONBROWSERS,LIBRARIES,SYSTEMSERVICES,ANDOPERATINGSYSTEMS–THELOWERLAYERSOFTHECOMPUTERSTACKCURRENTRESEARCHEFFORTSINTOMORESECUREWEBBROWSERSHELPIMPROVETHESECURITYOFBROWSERS,BUTREMAINSUSCEPTIBLETOATTACKSONLOWERLAYERSOFTHECOMPUTERSTACKTHEOPWEBBROWSER26,GAZELLE52,CHROME11,ANDCHROMEOS25PROPOSENEWBROWSERARCHITECTURESFORSEPARATINGTHEFUNCTIONALITYOFTHEBROWSERFROMSECURITYMECHANISMSANDPOLICIESHOWEVER,THESEMORESECUREWEBBROWSERSAREALLBUILTONTOPOFCOMMODITYOPERATINGSYSTEMSANDINCLUDECOMPLEXUSERMODELIBRARIESANDSHAREDSYSTEMSERVICESWITHINTHEIRTRUSTEDCOMPUTINGBASETCBEVENKERNELDESIGNSWITHSTRONGISOLATIONBETWEENOSCOMPONENTSEG,MICROKERNELS24,27,28ANDINFORMATIONFLOWKERNELS18,57,33STILLHAVEOSSERVICESTHATARESHAREDBYALLAPPLICATIONS,WHICHATTACKERSCANCOMPROMISEANDSTILLCAUSEDAMAGEHEREAREAFEWWAYSTHATANATTACKERCANSTILLCAUSEDAMAGETOMORESECUREWEBBROWSERSBUILTONTOPOFTRADITIONALOSES1HARDWAREBROWSERABSTRACTIONSIBOSKERNELL4REFERENCEMONITORUISTORAGEWEBPAGEINSTANCEWEBPAGEINSTANCEWEBPAGEINSTANCEMAPSCOMBINGCOMUIUCEDUUNIXLAYERTRADITIONALAPPLICATIONSLOCALHOSTMOUSE/KBDDRIVERNICDRIVERNETMANAGERNICMOUSE/KBDVIDEONETPROCESSESFIGURE1OVERALLIBOSARCHITECTUREOURSYSTEMCONTAINSUSERMODEDRIVERS,BROWSERSAPIMANAGERS,WEBPAGEINSTANCES,ANDTRADITIONALPROCESSESTOMANAGETHEINTERACTIONSBETWEENTHESECOMPONENTS,WEUSEAREFERENCEMONITORTHATRUNSWITHINOURIBOSKERNELSHADEDREGIONSMAKEUPTHETCB3MAINTAINCOMPATIBILITYWITHCURRENTBROWSERSECURITYPOLICIESOURPRIMARYGOALISTOIMPROVETHEENFORCEMENTOFCURRENTBROWSERPOLICIESWITHOUTCHANGINGCURRENTWEBBASEDAPPLICATIONS4EXPOSEENOUGHBROWSERSTATESANDEVENTSTOENABLENEWBROWSERSECURITYPOLICIESINADDITIONTOENFORCINGCURRENTBROWSERPOLICIES,WEWOULDLIKEOURARCHITECTURETOADAPTEASILYTOFUTUREBROWSERPOLICIES5AVOIDRULEBASEDOSSANDBOXINGFORBROWSERCOMPONENTSFUNDAMENTALLY,RULEBASEDOSSANDBOXINGISABOUTRESTRICTINGUNUSEDOROVERLYPERMISSIVEINTERFACESEXPOSEDBYTODAY’SOPERATINGSYSTEMSHOWEVER,SANDBOXINGSYSTEMSCANBECOMPLEXTHEUBUNTU1004SELINUXREFERENCEPOLICYUSESOVER104KLINESOFPOLICYCODEANDDIFFICULTTOIMPLEMENTCORRECTLY23,51IFOURARCHITECTUREREQUIRESOSSANDBOXINGFORBROWSERCOMPONENTSTHENWESHOULDRETHINKTHEARCHITECTURE22OVERALLARCHITECTUREFIGURE1SHOWSTHEOVERALLIBOSARCHITECTURETHEIBOSARCHITECTUREUSESABASICMICROKERNELAPPROACHWITHATHINKERNELFORMANAGINGHARDWAREANDFACILITATINGMESSAGEPASSINGBETWEENPROCESSESTHESYSTEMINCLUDESUSERMODEDEVICEDRIVERSFORINTERACTINGDIRECTLYWITHHARDWAREDEVICES,SUCHASNETWORKINTERFACECARDSNIC,ANDBROWSERAPIMANAGERSFORACCESSINGTHEDRIVERSANDIMPLEMENTINGBROWSERABSTRACTIONSTHEKEYBROWSERABSTRACTIONSTHATTHEBROWSERAPIMANAGERSIMPLEMENTAREHTTPREQUESTS,COOKIESANDLOCALSTORAGEFORSTORINGPERSISTENTDATA,ANDTABSFORDISPLAYINGUSERINTERFACEUICONTENTWEBAPPSUSETHESEABSTRACTIONSDIRECTLYTOIMPLEMENTBROWSERFUNCTIONALITY,ANDTRADITIONALAPPLICATIONSTRADITIONALAPPSUSEAUNIXLAYERTOACCESSUNIXLIKEABSTRACTIONSONTOPOFTHESEBROWSERABSTRACTIONS221THEIBOSKERNELOURIBOSKERNELISTHESOFTWARETCBFORTHEBROWSERANDINCLUDESRESOURCEMANAGEMENTFUNCTIONALITYANDAREFERENCEMONITORFORSECURITYENFORCEMENTTHEIBOSKERNELALSOHANDLESMANYTRADITIONALOSTASKSSUCHASMANAGINGGLOBALRESOURCES,CREATINGNEWPROCESSES,ANDMANAGINGMEMORYFORAPPLICATIONSTOFACILITATEMESSAGEPASSING,THEIBOSKERNELINCLUDESTHEL4KAPISTACHIO8MESSAGEPASSINGIMPLEMENTATIONANDMMUMANAGEMENTFUNCTIONSALLMESSAGESPASSTHROUGHOURREFERENCEMONITORANDARESUBJECTEDTOOUROVERALLSYSTEMSECURITYPOLICYSECTION4DESCRIBESTHEPOLICIESTHATTHEIBOSKERNELENFORCESANDTHEMECHANISMSITUSESTOIMPLEMENTTHESEPOLICIES222NETWORK,STORAGE,ANDUIMANAGERSTHEIBOSNETWORKSUBSYSTEMHANDLESHTTPREQUESTSANDSOCKETCALLSFORAPPLICATIONSTOHANDLEHTTPREQUESTS,NETWORKPROCESSESCHECKALOCALCACHETOSEEIFTHEREQUESTCANBESERVICEDVIATHECACHE,FETCHANYCOOKIESNEEDEDFORTHEREQUEST,FORMATTHEHTTPDATAINTOATCPSTREAM,ANDTRANSFORMTHATTCPSTREAMINTOASERIESOFETHERNETFRAMESTHATARESENTTOTHENICDRIVERSOCKETNETWORKPROCESSESEXPORTABASICSOCKETAPIANDSIMPLYTRANSFORMTCPSTREAMSTOETHERNETFRAMESFORTRANSMISSIONACROSSTHENETWORKONLYTRADITIONALAPPSCANACCESSOURSOCKETNETWORKPROCESSESTHEIBOSKERNELMANAGESGLOBALSTATES,LIKEPORTALLOCATIONTHEIBOSSTORAGEMANAGERMAINTAINSPERSISTENTSTORAGEFORKEYVALUEDATAPAIRSTHEBROWSERUSESTHESTORAGEMANAGERTOSTOREHTTPCOOKIESANDHTML5LOCALSTORAGEOBJECTS,ANDTHEBASICOBJECTSTOREINCLUDESOPTIONALPARAMETERS,SUCHASPATHANDMAXAGE,TOEXPOSECOOKIEPROPERTIESTOTHEREFERENCEMONITORTHESTORAGEMANAGERUSESSEVERALDIFFERENTNAMESPACESTOISOLATEOBJECTSFROMEACHOTHERWEBAPPSANDNETWORKPROCESSESSHAREANAMESPACEBASEDONTHEORIGINTHETUPLEOFAUNIFORMRESOURCELOCATORTHATTHEYORIGINATEFROM,ANDWEBAPPSANDTRADITIONALAPPSSHAREA“LOCALHOST”NAMESPACE,WHICHISSEPARATEFROMTHEHTTPNAMESPACEALLOTHERDRIVERSANDMANAGERSHAVETHEIROWNPRI3

簡(jiǎn)介：FLEXSCFLEXIBLESYSTEMCALLSCHEDULINGWITHEXCEPTIONLESSSYSTEMCALLSLIVIOSOARESUNIVERSITYOFTORONTOMICHAELSTUMMUNIVERSITYOFTORONTOABSTRACTFORTHEPAST30YEARS,SYSTEMCALLSHAVEBEENTHEDEFACTOINTERFACEUSEDBYAPPLICATIONSTOREQUESTSERVICESFROMTHEOPERATINGSYSTEMKERNELSYSTEMCALLSHAVEALMOSTUNIVERSALLYBEENIMPLEMENTEDASASYNCHRONOUSMECHANISM,WHEREASPECIALPROCESSORINSTRUCTIONISUSEDTOYIELDUSERSPACEEXECUTIONTOTHEKERNELINTHEFIRSTPARTOFTHISPAPER,WEEVALUATETHEPERFORMANCEIMPACTOFTRADITIONALSYNCHRONOUSSYSTEMCALLSONSYSTEMINTENSIVEWORKLOADSWESHOWTHATSYNCHRONOUSSYSTEMCALLSNEGATIVELYAFFECTPERFORMANCEINASIGNIFICANTWAY,PRIMARILYBECAUSEOFPIPELINEFLUSHINGANDPOLLUTIONOFKEYPROCESSORSTRUCTURESEG,TLB,DATAANDINSTRUCTIONCACHES,ETCWEPROPOSEANEWMECHANISMFORAPPLICATIONSTOREQUESTSERVICESFROMTHEOPERATINGSYSTEMKERNELEXCEPTIONLESSSYSTEMCALLSTHEYIMPROVEPROCESSOREFFICIENCYBYENABLINGFLEXIBILITYINTHESCHEDULINGOFOPERATINGSYSTEMWORK,WHICHINTURNCANLEADTOSIGNIFICANTLYINCREASEDTEMPORALANDSPACIALLOCALITYOFEXECUTIONINBOTHUSERANDKERNELSPACE,THUSREDUCINGPOLLUTIONEFFECTSONPROCESSORSTRUCTURESEXCEPTIONLESSSYSTEMCALLSAREPARTICULARLYEFFECTIVEONMULTICOREPROCESSORSTHEYPRIMARILYTARGETHIGHLYTHREADEDSERVERAPPLICATIONS,SUCHASWEBSERVERSANDDATABASESERVERSWEPRESENTFLEXSC,ANIMPLEMENTATIONOFEXCEPTIONLESSSYSTEMCALLSINTHELINUXKERNEL,ANDANACCOMPANYINGUSERMODETHREADPACKAGEFLEXSCTHREADS,BINARYCOMPATIBLEWITHPOSIXTHREADS,THATTRANSLATESLEGACYSYNCHRONOUSSYSTEMCALLSINTOEXCEPTIONLESSONESTRANSPARENTLYTOAPPLICATIONSWESHOWHOWFLEXSCIMPROVESPERFORMANCEOFAPACHEBYUPTO116,MYSQLBYUPTO40,ANDBINDBYUPTO105WHILEREQUIRINGNOMODIFICATIONSTOTHEAPPLICATIONS1INTRODUCTIONSYSTEMCALLSARETHEDEFACTOINTERFACETOTHEOPERATINGSYSTEMKERNELTHEYAREUSEDTOREQUESTSERVICESOFFEREDBY,ANDIMPLEMENTEDINTHEOPERATINGSYSTEMKERNELWHILE020004000600080001000012000140001600003050709111315SYSCALLIMPACTONUSERMODEIPCTIMEINCYCLESUSERMODEIPCHIGHERISFASTERSYSCALLEXCEPTIONLOSTPERFORMANCECYCLESFIGURE1USERMODEINSTRUCTIONSPERCYCLESIPCOFXALANFROMSPECCPU2006INRESPONSETOASYSTEMCALLEXCEPTIONEVENT,ASMEASUREDONANINTELCOREI7PROCESSORDIFFERENTOPERATINGSYSTEMSOFFERAVARIETYOFDIFFERENTSERVICES,THEBASICUNDERLYINGSYSTEMCALLMECHANISMHASBEENCOMMONONALLCOMMERCIALMULTIPROCESSEDOPERATINGSYSTEMSFORDECADESSYSTEMCALLINVOCATIONTYPICALLYINVOLVESWRITINGARGUMENTSTOAPPROPRIATEREGISTERSANDTHENISSUINGASPECIALMACHINEINSTRUCTIONTHATRAISESASYNCHRONOUSEXCEPTION,IMMEDIATELYYIELDINGUSERMODEEXECUTIONTOAKERNELMODEEXCEPTIONHANDLERTWOIMPORTANTPROPERTIESOFTHETRADITIONALSYSTEMCALLDESIGNARETHAT1APROCESSOREXCEPTIONISUSEDTOCOMMUNICATEWITHTHEKERNEL,AND2ASYNCHRONOUSEXECUTIONMODELISENFORCED,ASTHEAPPLICATIONEXPECTSTHECOMPLETIONOFTHESYSTEMCALLBEFORERESUMINGUSERMODEEXECUTIONBOTHOFTHESEEFFECTSRESULTINPERFORMANCEINEFFICIENCIESONMODERNPROCESSORSTHEINCREASINGNUMBEROFAVAILABLETRANSISTORSONACHIPMOORE’SLAWHAS,OVERTHEYEARS,LEDTOINCREASINGLYSOPHISTICATEDPROCESSORSTRUCTURES,SUCHASSUPERSCALARANDOUTOFORDEREXECUTIONUNITS,MULTILEVELCACHES,ANDBRANCHPREDICTORSTHESEPROCESSORSTRUCTURESHAVE,INTURN,LEDTOALARGEINCREASEINTHEPERFORMANCEPOTENTIALOFSOFTWARE,BUTATTHESAMETIMETHEREISAWIDENINGGAPBETWEENTHEPERFORMANCEOFEFFICIENTSOFTWAREANDTHEPERFORMANCEOFINEFFICIENTSOFTWARE,PRIMARILYDUETOTHEINCREASINGDISPARITYOFACCESSINGDIFFERENTPROCESSORRESOURCESEG,REGISTERSVSCACHESVSMEMORYSERVERANDSYSTEMINTENSIVEWORKLOADS,WHICHAREOFPARTICULARSYSCALLINSTRUCTIONSCYCLESIPCICACHEDCACHEL2L3DTLBSTAT49721358503732186660255921PREAD37391230003032294679216020PWRITE56893128501850373985316044OPENCLOSE66311916203447240900353428MMAPMUNMAP8977190790474123386939137OPENWRITECLOSE992132815030784811462510549TABLE1SYSTEMCALLFOOTPRINTOFDIFFERENTPROCESSORSTRUCTURESFORTHEPROCESSORSSTRUCTURESCACHESANDTLB,THENUMBERSREPRESENTNUMBEROFENTRIESEVICTEDTHECACHELINEFORTHEPROCESSORISOF64BYTESICACHEANDDCACHEREFERTOTHEINSTRUCTIONANDDATASECTIONSOFTHEL1CACHE,RESPECTIVELYTHEDTLBREPRESENTSTHEDATAPORTIONOFTHETLBKERNELSTACK,CHANGINGTHEPROTECTIONDOMAIN,ANDREDIRECTINGEXECUTIONTOTHEREGISTEREDEXCEPTIONHANDLERSUBSEQUENTLY,RETURNFROMEXCEPTIONISNECESSARYTORESUMEEXECUTIONINUSERMODEWEMEASUREDTHEMODESWITCHTIMEBYIMPLEMENTINGANEWSYSTEMCALL,GETTSCTHATOBTAINSTHETIMESTAMPCOUNTEROFTHEPROCESSORANDIMMEDIATELYRETURNSTOUSERMODEWECREATEDASIMPLEBENCHMARKTHATINVOKEDGETTSC1BILLIONTIMES,RECORDINGTHETIMESTAMPBEFOREANDAFTEREACHCALLTHEDIFFERENCEBETWEENEACHOFTHETHREETIMESTAMPSIDENTIFIESTHENUMBEROFCYCLESNECESSARYTOENTERANDLEAVETHEOPERATINGSYSTEMKERNEL,NAMELY79CYCLESAND71CYCLES,RESPECTIVELYTHETOTALROUNDTRIPTIMEFORTHEGETTSCSYSTEMCALLISMODESTAT150CYCLES,BEINGLESSTHANTHELATENCYOFAMEMORYACCESSTHATMISSESTHEPROCESSORCACHES250CYCLESONOURMACHINE322SYSTEMCALLFOOTPRINTTHEMODESWITCHTIME,HOWEVER,ISONLYPARTOFTHECOSTOFASYSTEMCALLDURINGKERNELMODEEXECUTION,PROCESSORSTRUCTURESINCLUDINGTHEL1DATAANDINSTRUCTIONCACHES,TRANSLATIONLOOKASIDEBUFFERSTLB,BRANCHPREDICTIONTABLES,PREFETCHBUFFERS,ASWELLASLARGERUNIFIEDCACHESL2ANDL3,AREPOPULATEDWITHKERNELSPECIFICSTATETHEREPLACEMENTOFUSERMODEPROCESSORSTATEBYKERNELMODEPROCESSORSTATEISREFERREDTOASTHEPROCESSORSTATEPOLLUTIONCAUSEDBYASYSTEMCALLTOQUANTIFYTHEPOLLUTIONCAUSEDBYSYSTEMCALLS,WEUSEDTHECOREI7HARDWAREPERFORMANCECOUNTERSHPCWERANAHIGHINSTRUCTIONPERCYCLEIPCWORKLOAD,XALAN,FROMTHESPECCPU2006BENCHMARKSUITETHATISKNOWNTOINVOKEFEWSYSTEMCALLSWECONFIGUREDANHPCTOTRIGGERINFREQUENTLYONCEEVERY10MILLIONUSERMODEINSTRUCTIONSSOTHATTHEPROCESSORSTRUCTURESWOULDBEDOMINATEDWITHAPPLICATIONSTATEWETHENSETUPTHEHPCEXCEPTIONHANDLERTOEXECUTESPECIFICSYSTEMCALLS,WHILEMEASURINGTHEREPLACEMENTOFAPPLICATIONSTATEINTHEPROCESSORSTRUCTURESCAUSEDBYKERNELEXECUTIONBUTNOTBYTHEPERFORMANCECOUNTEREXCEPTIONHANDLERITSELF3FORALLEXPERIMENTSPRESENTEDINTHISPAPER,USERMODEAPPLICATIONSEXECUTEIN64BITMODEANDWHENUSINGSYNCHRONOUSSYSTEMCALLS,USETHE“SYSCALL”X8664INSTRUCTION,WHICHISCURRENTLYTHEDEFAULTINLINUXTABLE1SHOWSTHEFOOTPRINTONSEVERALPROCESSORSTRUCTURESFORTHREEDIFFERENTSYSTEMCALLSANDTHREESYSTEMCALLCOMBINATIONSTHEDATASHOWSTHAT,EVENTHOUGHTHENUMBEROFICACHELINESREPLACEDISMODESTBETWEEN2AND5KB,THENUMBEROFDCACHELINESREPLACEDISSIGNIFICANTGIVENTHATTHESIZEOFTHEDCACHEONTHISPROCESSORIS32KB,WESEETHATTHESYSTEMCALLSLISTEDPOLLUTEATLEASTHALFOFTHEDCACHE,ANDALMOSTALLOFTHEDCACHEINTHE“OPENWRITECLOSE”CASETHE64ENTRYFIRSTLEVELDTLBISALSOSIGNIFICANTLYPOLLUTEDBYMOSTSYSTEMCALLSFINALLY,ITISINTERESTINGTONOTETHATTHESYSTEMCALLIMPACTONTHEL2ANDL3CACHESISLARGERTHANONTHEL1CACHES,PRIMARILYBECAUSETHEL2ANDL3CACHESUSEMOREAGGRESSIVEPREFETCHING23SYSTEMCALLIMPACTONUSERIPCULTIMATELY,THEMOSTIMPORTANTMEASUREOFTHEREALCOSTOFSYSTEMCALLSISTHEPERFORMANCEIMPACTONTHEAPPLICATIONTOQUANTIFYTHIS,WEEXECUTEDANEXPERIMENTSIMILARTOTHEONEDESCRIBEDINTHEPREVIOUSSUBSECTIONHOWEVER,INSTEADOFMEASURINGKERNELMODEEVENTS,WEONLYMEASUREDUSERMODEINSTRUCTIONSPERCYCLEIPC,IGNORINGALLKERNELEXECUTIONIDEALLY,USERMODEIPCSHOULDNOTDECREASEASARESULTOFINVOKINGSYSTEMCALLS,SINCETHECYCLESANDINSTRUCTIONSEXECUTEDASPARTOFTHESYSTEMCALLAREIGNOREDINOURMEASUREMENTSINPRACTICE,HOWEVER,USERMODEIPCISAFFECTEDBYTWOSOURCESOFOVERHEADDIRECTTHEPROCESSOREXCEPTIONASSOCIATEDWITHTHESYSTEMCALLINSTRUCTIONTHATFLUSHESTHEPROCESSORPIPELINEINDIRECTSYSTEMCALLPOLLUTIONONTHEPROCESSORSTRUCTURES,ASQUANTIFIEDINTABLE1FIGURES2AND3SHOWTHEDEGRADATIONINUSERMODEIPCWHENRUNNINGXALANFROMSPECCPU2006ANDSPECJBB,RESPECTIVELY,GIVENDIFFERENTFREQUENCIESOFPWRITECALLSTHESEBENCHMARKSWERECHOSENSINCETHEYHAVEBEENCREATEDTOAVOIDSIGNIFICANTUSEOFSYSTEMSERVICES,ANDSHOULDSPENDONLY12OFTIMEEXECUTINGINKERNELMODETHEGRAPHSSHOWTHATDIFFERENTWORKLOADSCANHAVEDIFFERENTSENSITIVITIESTOSYSTEMCALLPOLLUTIONXALANHASABASELINEUSERMODEIPCOF146,BUTTHEIPCDEGRADESBYUPTO65WHENEXECUTINGAPWRITEEVERY1,0002,000INSTRUCTIONS,YIELDINGANIPCBETWEEN058AND050

簡(jiǎn)介：EFFICIENTSYSTEMENFORCEDDETERMINISTICPARALLELISMAMITTAIAVIRAM,SHUCHUNWENG,SENHU,BRYANFORDYALEUNIVERSITYABSTRACTDETERMINISTICEXECUTIONOFFERSMANYBENEFITSFORDEBUGGING,FAULTTOLERANCE,ANDSECURITYCURRENTMETHODSOFEXECUTINGPARALLELPROGRAMSDETERMINISTICALLY,HOWEVER,OFTENINCURHIGHCOSTS,ALLOWMISBEHAVEDSOFTWARETODEFEATREPEATABILITY,ANDTRANSFORMTIMEDEPENDENTRACESINTOINPUTORPATHDEPENDENTRACESWITHOUTELIMINATINGTHEMWEINTRODUCEANEWPARALLELPROGRAMMINGMODELADDRESSINGTHESEISSUES,ANDUSEDETERMINATOR,APROOFOFCONCEPTOS,TODEMONSTRATETHEMODEL’SPRACTICALITYDETERMINATOR’SMICROKERNELAPIPROVIDESONLY“SHAREDNOTHING”ADDRESSSPACESANDDETERMINISTICINTERPROCESSCOMMUNICATIONPRIMITIVESTOMAKEEXECUTIONOFALLUNPRIVILEGEDCODEWELLBEHAVEDORNOTPRECISELYREPEATABLEATOPTHISMICROKERNEL,DETERMINATOR’SUSERLEVELRUNTIMEADAPTSOPTIMISTICREPLICATIONTECHNIQUESTOOFFERAPRIVATEWORKSPACEMODELFORBOTHTHREADLEVELANDPROCESSLEVELPARALLELPROGRAMINGTHISMODELAVOIDSTHEINTRODUCTIONOFREAD/WRITEDATARACES,ANDCONVERTSWRITE/WRITERACESINTORELIABLYDETECTEDCONFLICTSCOARSEGRAINEDPARALLELBENCHMARKSPERFORMANDSCALECOMPARABLYTONONDETERMINISTICSYSTEMS,ONBOTHMULTICOREPCSANDACROSSNODESINADISTRIBUTEDCLUSTER1INTRODUCTIONWEOFTENWISHTORUNSOFTWAREDETERMINISTICALLY,SOTHATFROMAGIVENINPUTITALWAYSPRODUCESTHESAMEOUTPUTDETERMINISMISTHEFOUNDATIONOFREPLAYDEBUGGING37,39,46,56,FAULTTOLERANCE15,18,50,ANDACCOUNTABILITYMECHANISMS30,31METHODSOFINTRUSIONANALYSIS22,34ANDTIMINGCHANNELCONTROL4FURTHERASSUMETHESYSTEMCANENFORCEDETERMINISMEVENONMALICIOUSCODEDESIGNEDTOEVADEANALYSISEXECUTINGPARALLELSOFTWAREDETERMINISTICALLYISCHALLENGING,HOWEVER,BECAUSETHREADSSHARINGANADDRESSSPACEORPROCESSESSHARINGRESOURCESSUCHASFILESYSTEMSAREPRONETONONDETERMINISTIC,TIMINGDEPENDENTRACES3,40,42,43USERSPACETECHNIQUESFORPARALLELDETERMINISTICEXECUTION8,10,20,21,44SHOWPROMISEBUTHAVELIMITATIONSFIRST,BYRELYINGONADETERMINISTICSCHEDULERRESIDINGINTHEAPPLICATIONPROCESS,THEYPERMITBUGGYORMALICIOUSAPPLICATIONSTOCOMPROMISEDETERMINISMBYINTERFERINGWITHTHESCHEDULERSECOND,DETERMINISTICSCHEDULERSEMULATECONVENTIONALAPISBYSYNTHESIZINGAREPEATABLEBUTARBITRARYSCHEDULEOFINTERTHREADINTERACTIONS,OFTENUSINGANINSTRUCTIONCOUNTERASANARTIFICIALTIMEMETRICDATARACESREMAIN,THEREFORE,BUTTHEIRMANIFESTATIONDEPENDSSUBTLYONINPUTSANDCODEPATHLENGTHSINSTEADOFON“REAL”TIMETHIRD,THEUSERLEVELINSTRUMENTATIONREQUIREDTOISOLATEANDSCHEDULETHREADS’MEMORYACCESSESCANINCURCONSIDERABLEOVERHEAD,EVENONCOARSEGRAINEDCODETHATSYNCHRONIZESRARELYTOMEETTHESOFTWAREDEVELOPMENT,DEBUGGING,ANDSECURITYCHALLENGESTHATUBIQUITOUSPARALLELISMPRESENTS,ITMAYBEINSUFFICIENTTOSHOEHORNTHESTANDARDNONDETERMINISTICPROGRAMMINGMODELINTOASYNTHETICEXECUTIONSCHEDULEINSTEADWEPROPOSETORETHINKTHEBASICMODELITSELFWEWOULDLIKEAPARALLELENVIRONMENTTHATAIS“DETERMINISTICBYDEFAULT”12,40,EXCEPTWHENWEINJECTNONDETERMINISMEXPLICITLYVIAEXTERNALINPUTSBINTRODUCESNODATARACES,EITHERATTHEMEMORYACCESSLEVEL25,43ORATHIGHERSEMANTICLEVELS3CCANENFORCEDETERMINISMONARBITRARY,COMPROMISEDORMALICIOUSCODEFORSECURITYREASONSANDDISEFFICIENTENOUGHTOUSEFOR“NORMALCASE”EXECUTIONOFDEPLOYEDCODE,NOTJUSTFORINSTRUMENTATIONDURINGDEVELOPMENTASASTEPTOWARDSUCHAMODEL,WEPRESENTDETERMINATOR,APROOFOFCONCEPTOSDESIGNEDAROUNDTHEABOVEGOALSDUETOITSOSLEVELAPPROACH,DETERMINATORSUPPORTSEXISTINGLANGUAGES,CANENFORCEDETERMINISTICEXECUTIONNOTONLYONASINGLEPROCESSBUTONGROUPSOFINTERACTINGPROCESSES,ANDCANPREVENTMALICIOUSUSERLEVELCODEFROMSUBVERTINGTHEKERNEL’SGUARANTEEOFDETERMINISMINORDERTOEXPLORETHEDESIGNSPACEFREELY,DETERMINATORTAKESA“CLEANSLATE”APPROACH,MAKINGFEWCOMPROMISESFORBACKWARDCOMPATIBILITYWITHEXISTINGKERNELSORAPISDETERMINATOR’SPROGRAMMINGMODELCOULDBEIMPLEMENTEDINALEGACYKERNELFORBACKWARDCOMPATIBILITY,HOWEVER,ASPARTOFA“DETERMINISTICSANDBOX”FOREXAMPLE9DETERMINATOR’SUSERLEVELRUNTIMEALSOPROVIDESLIMITEDEMULATIONOFTHEUNIXPROCESS,THREAD,ANDFILEAPIS,TOSIMPLIFYAPPLICATIONPORTINGDETERMINATOR’SKERNELENFORCESDETERMINISMBYDENYINGUSERCODEDIRECTACCESSTOHARDWARERESOURCESWHOSEUSECANYIELDNONDETERMINISTICBEHAVIOR,INCLUDINGREALTIMECLOCKS,CYCLECOUNTERS,ANDWRITABLESHAREDMEMORYDETERMINATORCONSTRAINSUSERCODETORUNWITHINAHIERARCHYOFSINGLETHREADED,PROCESSLIKESPACES,EACHHAVINGAPRIVATEVIRTUALADDRESSSPACETHEKERNEL’SLOWLEVELAPIPROVIDESONLYTHREESYSTEMCALLS,WITHWHICHASPACECANSYNCHRONIZEANDCOMMUNICATEWITHITSIMMEDIATEPARENTANDCHILDRENPOTENTIALLYUSEFULSOURCESOFNONDETERMINISM,SUCHASTIMERS,DETERMINATORENCAPSULATESINTOI/ODEVICES,WHICHUNPRIVILEGEDSPACESCANACCESSCATE,VERIFY,ORANALYZEAPROGRAM’SEXECUTIONHISTORYREPLAYCANBEEFFICIENTWHENONLYI/ONEEDBELOGGED,ASFORAUNIPROCESSORVIRTUALMACHINE22,BUTBECOMESMUCHMORECOSTLYIFINTERNALSOURCESOFNONDETERMINISMDUETOPARALLELISMMUSTALSOBEREPLAYED19,23DETERMINATORTHEREFORETRANSFORMSUSEFULSOURCESOFNONDETERMINISMINTOEXPLICITI/O,WHICHAPPLICATIONSMAYOBTAINVIACONTROLLABLECHANNELS,ANDELIMINATESONLYINTERNALNONDETERMINISMRESULTINGFROMPARALLELISMIFANAPPLICATIONCALLSGETTIMEOFDAY,FOREXAMPLE,THENASUPERVISINGPROCESSCANINTERCEPTTHISI/OTOLOG,REPLAY,ORSYNTHESIZETHESEEXPLICITTIMEINPUTS22ARACEFREEMODELFORSHAREDSTATECONVENTIONALSYSTEMSGIVETHREADSDIRECT,CONCURRENTACCESSTOMANYFORMSOFSHAREDSTATE,SUCHASSHAREDMEMORYANDFILESYSTEMS,YIELDINGDATARACESANDHEISENBUGSIFTHETHREADSFAILTOSYNCHRONIZEPROPERLY25,40,43WHILEREPLAYDEBUGGERS37,39,46,56ANDDETERMINISTICSCHEDULERS8,10,20,21,44MAKEDATARACESREPRODUCIBLEONCETHEYMANIFEST,THEYDONOTCHANGETHEINHERENTLYRACEPRONEMODELINWHICHDEVELOPERSWRITEAPPLICATIONSDETERMINATORREPLACESTHESTANDARDCONCURRENTACCESSMODELWITHAPRIVATEWORKSPACEMODEL,INWHICHDATARACESDONOTARISEINTHEFIRSTPLACETHISMODELGIVESEACHTHREADACOMPLETE,PRIVATEVIRTUALREPLICAOFALLLOGICALLYSHAREDSTATEATHREADMAYACCESS,INCLUDINGSHAREDMEMORYANDFILESYSTEMSTATEATHREAD’SNORMALREADSANDWRITESAFFECTONLYITSPRIVATEWORKINGSTATE,ANDDONOTINTERACTDIRECTLYWITHOTHERTHREADSINSTEAD,DETERMINATORACCUMULATESEACHTHREADS’SCHANGESTOSHAREDSTATE,THENRECONCILESTHESECHANGESAMONGTHREADSONLYATPROGRAMDEFINEDSYNCHRONIZATIONPOINTSTHISMODELISRELATEDTOANDINSPIREDBYEARLYPARALLELFORTRANSYSTEMS7,51,REPLICATEDFILESYSTEMS47,TRANSACTIONALMEMORY33,52ANDOPERATINGSYSTEMS48,ANDDISTRIBUTEDVERSIONCONTROLSYSTEMS29,BUTTOOURKNOWLEDGEDETERMINATORISTHEFIRSTOSTOINTRODUCEAMODELFORPERVASIVETHREADANDPROCESSLEVELDETERMINISMIFONETHREADEXECUTESTHEASSIGNMENT‘XY’WHILEANOTHERCONCURRENTLYEXECUTES‘YX’,FOREXAMPLE,THESEASSIGNMENTSRACEINTHECONVENTIONALMODEL,BUTARERACEFREEUNDERDETERMINATORANDALWAYSSWAPXWITHYEACHTHREAD’SREADOFXORYALWAYSSEESTHE“OLD”VERSIONOFTHATVARIABLE,SAVEDINTHETHREAD’SPRIVATEWORKSPACEATTHELASTEXPLICITSYNCHRONIZATIONPOINTFIGURE1ILLUSTRATESAMOREREALISTICEXAMPLEOFAGAMEORSIMULATOR,WHICHUSESANARRAYOF“ACTORS”PLAYERS,PARTICLES,ETCTOREPRESENTSOMELOGICAL“UNIVERSE,”ANDUPDATESALLOFTHEACTORSINPARALLELATEACHTIMESTEPTOUPDATETHEACTORS,THEMAINTHREADFORKSACHILDTHREADTOPROCESSEACHACTOR,THENSYNCHRONIZESBYJOININGALLTHESECHILDTHREADSTHECHILDTHREADCODETOUPDATEEACHACTORISSHOWN“INLINE”WITHINTHEMAINFUNCTION,WHICHSTRUCTACTORSTATEACTORNACTORSMAININITIALIZEALLELEMENTSOFACTORARRAYFORTIME0TIMEFORI0INACTORSIIFTHREADFORKIINCHILD//CHILDTHREADTOPROCESSACTORIEXAMINESTATEOFNEARBYACTORSUPDATESTATEOFACTORIACCORDINGLYTHREADEXITFORI0INACTORSITHREADJOINIFIGURE1CPSEUDOCODEFORLOCKSTEPTIMESIMULATION,WHICHCONTAINSADATARACEINSTANDARDCONCURRENCYMODELSBUTISBUGFREEUNDERDETERMINATORUNDERUNIXWORKSONLYWITHPROCESSLEVELFORKDETERMINATOROFFERSTHISCONVENIENCEFORSHAREDMEMORYTHREADSASWELL,ASDISCUSSEDLATERINSECTION44INTHISEXAMPLE,EACHCHILDTHREADREADSTHE“PRIOR”STATEOFANYORALLACTORSINTHEARRAY,THENUPDATESTHESTATEOFITSASSIGNEDACTOR“INPLACE,”WITHOUTANYEXPLICITCOPYINGORADDITIONALSYNCHRONIZATIONWITHSTANDARDTHREADSTHISCODEHASAREAD/WRITERACEEACHCHILDTHREADMAYSEEANARBITRARYMIXOF“OLD”AND“NEW”STATESASITEXAMINESOTHERACTORSINTHEARRAYUNDERDETERMINATOR,HOWEVER,THISCODEISCORRECTANDRACEFREEEACHCHILDTHREADREADSONLYITSPRIVATEWORKINGCOPYOFTHEACTORSARRAY,WHICHISUNTOUCHEDEXCEPTBYTHECHILDTHREADITSELFSINCETHEMAINTHREADFORKEDTHATCHILDASTHEMAINTHREADREJOINSALLITSCHILDTHREADS,DETERMINATORMERGESEACHCHILD’SACTORARRAYUPDATESBACKINTOTHEMAINTHREAD’SWORKINGCOPY,FORUSEINTHENEXTTIMESTEPWHILEREAD/WRITERACESDISAPPEARINDETERMINATOR’SMODEL,TRADITIONALWRITE/WRITERACESBECOMECONFLICTSIFTWOCHILDTHREADSCONCURRENTLYWRITETOTHESAMEACTORARRAYELEMENT,FOREXAMPLE,DETERMINATORDETECTSTHISCONFLICTANDSIGNALSARUNTIMEEXCEPTIONWHENTHEMAINTHREADATTEMPTSTOJOINTHESECONDCONFLICTINGCHILDINTHECONVENTIONALMODEL,BYCONTRAST,THETHREADS’EXECUTIONSCHEDULESMIGHTCAUSEEITHEROFTHETWOWRITESTO“WIN”ANDSILENTLYPROPAGATEITSLIKELYERRONEOUSVALUETHROUGHOUTTHECOMPUTATIONRUNNINGTHISCODEUNDERACONVENTIONALDETERMINISTICSCHEDULERCAUSESTHE“WINNER”TOBEDECIDEDBASEDONASYNTHETIC,REPRODUCIBLETIMEMETRICEG,INSTRUCTIONCOUNTRATHERTHANREALTIME,BUTTHERACEREMAINSANDMAYSTILLMANIFESTORVANISHDUETOSLIGHTCHANGESININPUTSORINSTRUCTIONPATHLENGTHS23ARACEFREESYNCHRONIZATIONAPICONVENTIONALTHREADSCANSTILLBEHAVENONDETERMINISTICALLYEVENINACORRECTLYLOCKEDPROGRAMWITHNOLOW

簡(jiǎn)介：；．’’．．分類號(hào)Ｒ２７１．１９密級(jí)公開，＾－單位代碼；１０７６０北＾學(xué)號(hào)１０７加２１２６４９４＾＾滬．－．Ａ亡＇．＇＇Ｖ＇，Ｖ，，．一．，－．，＇－Ｊ．，古．‘Ｖ．Ｒ－．一．．；＇．．．Ｖ＇．．．、．．‘－‘＇譜斯壤處種乂營(yíng)ＸＩＮＪＩＡＮＧＭＥＤＩＣＡＬＵＮＩＶＥＲＳＩＴＹ碩±學(xué)位論文ＴＨＥＳＩＳＯＦＭ乂ＳＴＥＲＤＥＧＲＥＥ臨床醫(yī)學(xué)專業(yè)學(xué)位（學(xué)歷教育）論文題目圍絕鐘期綜合征與西北燥證的相關(guān)性調(diào)查研究研究生李紫營(yíng)，置指導(dǎo)教師劉愛玲副教授主任醫(yī)師．，＾專業(yè)學(xué)位領(lǐng)域中醫(yī)婦科學(xué)總皆扣＞Ｆ．＇ＶＲ．‘Ｖ令研究方向中醫(yī)婦科病防治鉛Ｖ。．－，＇研究起止時(shí)間２〇１４．〇１２〇１５．〇１．茄違戸＾’．所在學(xué)院中醫(yī)學(xué)院’繫；項(xiàng)２０１５年０３月．＿，、，社，＜Ｔ＇－Ｓ－．＇．‘；；．．．－■．ＩＦ；；’產(chǎn)＇斯．．＇＇柏祭了．／；．；＇Ｋ．Ｎ＇廣游貨．午’驟ＫＲ－圍絕經(jīng)期綜合征與西北燥證的相關(guān)性調(diào)查研圍絕經(jīng)期綜合征與西北燥證的相關(guān)性調(diào)查研究2015年03月研究生研究生李紫萱李紫萱指導(dǎo)教師指導(dǎo)教師劉愛玲劉愛玲副教授教授主任醫(yī)師主任醫(yī)師專業(yè)學(xué)位領(lǐng)域?qū)I(yè)學(xué)位領(lǐng)域中醫(yī)婦科學(xué)婦科學(xué)研究方向研究方向中醫(yī)婦科病防治中醫(yī)婦科病防治